Ultrapar Participacoes SA

DISCLOSURE APPENDIX AT THE BACK OF THIS REPORT CONTAINS IMPORTANT DISCLOSURES, ANALYST CERTIFICATIONS, AND THE STATUS OF NON-US ANALYSTS. US Disclosure: Credit Suisse does and seeks to do business with companies covered in its research reports. As a result, investors should be aware that the Firm may have a conflict of interest that could affect the objectivity of this report. Investors should consider this report as only a single factor in making their investment decision.

22 June 2016 Americas/Brazil

Equity Research Oil & Gas Refining & Marketing

Ultrapar Participacoes SA (UGPA3) Rating OUTPERFORM Price (22-Jun-16,R$) 71.01 Target price (R$) 89.00 52-week price range 73.71 - 54.35 Market cap (R$ m) 39,510.33 Enterprise value (R$ m) 45,940.02 *Stock ratings are relative to the coverage universe in each

analyst's or each team's respective sector.

¹Target price is for 12 months.

Research Analysts

Andre Natal

55 11 3701 6299

[email protected]

Regis Cardoso

55 11 3701 6297

[email protected]

INITIATION

There Is Still Music in Them

■ Initiation: We are initiating coverage of Ultrapar with an Outperform rating

and a BRL89.00/share target price. Ultrapar is widely believed to be a good

but expensive company. We don't think it is expensive at all. There are still

many opportunities to add value. Rather than focusing solely on what is in

our numbers, we highlight what is not. Our target doesn't include Ipiranga's

expansion above cost of capital, the acquisition of Alesat, the increase in

convenience stores coverage, a possible acquisition of Liquigas, the ramp up

of Extrafarma, and so on, and yet we still see 25% upside potential. Investors

buying Ultrapar may get more, if a fraction of these opportunities materialize.

■ One Big Question. The company's track record is impressive. The question

then becomes, Can this music go on? To answer it, we went through the

fundamentals behind it, and we think they are not only still there but getting

stronger each day, in spite of the recent challenging macro environment.

■ Seven Answers To Be a Buyer: We unfold our main thesis into seven

essential questions that have not yet been fully addressed: (1) Why is the

distribution business so good in Brazil? (2) What are the prospects for

volume growth? (3) How do margins expand and what should we expect? (4)

How promising is the CONEN (Mid-West, North, Northeast) area? (5) How

much should we expect from fuel imports? (6) What are the prospects for the

other businesses? (7) But isn't it too expensive already? After going through

them, we believe the company owns businesses enjoying strong entry

barriers, in a market that will grow for many years, especially in CONEN,

allowing scale and margin expansion. Imports may also contribute in the

short term. We think Ultrapar can add shareholder value for years to come.

■ Valuation: We used five different valuation methods, and most point to

significant upsides. Our target is DCF based, assuming future returns will

decline. Our Bull case (BRL125/sh) includes the additional opportunities we

mapped. Our Bear case (BRL69/sh) assumes returns will decline faster.

Share price performance

On 22-Jun-2016 the SAO PAULO SE BOVESPA INDEX

closed at 50735.02804

Daily Jun23, 2015 - Jun22, 2016, 06/23/15 = R$67.55

Quarterly EPS Q1 Q2 Q3 Q4 2015A 0.69 0.59 0.53 0.89 2016E 0.69 0.62 0.63 0.80 2017E 0.59 0.73 0.70 0.91

Financial and valuation metrics

Year 12/14A 12/15A 12/16E 12/17E Revenue (R$ m) 67,736.3 75,655.3 79,816.7 89,223.4 EBITDA (R$ m) 3,174.4 3,964.2 4,133.9 4,606.9 EBIT (R$ m) 2,286.6 2,961.5 3,050.7 3,454.1 Net Income (R$ m) 1,241.6 1,503.5 1,446.9 1,612.2 EPS (CS adj.) (R$) 2.23 2.70 2.60 2.90 Prev. EPS (R$) - - - - Dividend yield (%) 2.0 2.2 2.1 2.4 P/E (x) 31.8 26.3 27.3 24.5 EV/EBITDA 13.7 11.3 11.1 10.0 P/B (x) 5.13 4.97 4.64 4.24 ROE stated-return on equity 17.5 19.2 17.6 18.1 ROIC (%) 13.25 14.91 13.84 14.26 Net debt (R$ m) 4,106 5,395 6,430 6,622 Net debt/EBITDA (12/15E, %) 53.1 67.7 75.1 70.7 Capex (R$ m) -1,216 -1,334 -1,792 -1,518

Source: Company data, Thomson Reuters, Credit Suisse estimates

22 June 2016

Ultrapar Participacoes SA (UGPA3) 2

Table of Contents

Introduction 4

Investment Summary ............................................................................................... 6

Ultrapar 10

Is It Too Good to Be True? ..................................................................................... 10

Ipiranga 13

Question #1: Why Is This Business So Good In Brazil? ........................................ 13

Question #2: What Are the Prospects for Volume Growth in Brazil? ..................... 21

Question #3: How Do Margins Expand and What Should We Expect? ................. 41

Question #4: How Promising Is the CONEN Area? ............................................... 48

Question #5: How Much Should We Expect From Fuel Imports? .......................... 56

Other Businesses 69

Question #6: What Are the Prospects for the Other Businesses? ......................... 69

Valuation 83

Question #7: But Isn't It Too Expensive Already? .................................................. 83

Risks 102

How Can We Be Wrong? ..................................................................................... 102

Appendix 105

Appendix 1: Ultrapar Through HOLT® Lens ......................................................... 105

Appendix 2: The HOLT® Framework .................................................................... 107

Appendix 3: Management Profile and Holdings ................................................... 112

Appendix 4: Ultrapar Financials ........................................................................... 116

Appendix 5: Financials Summary ......................................................................... 117

We would like to thank Gabriel Cordaro for his valuable contribution to the creation

of this report.

22 June 2016


Introduction "Opportunity meeting the prepared mind—that's the game" (Charlie Munger)

In 1915, the Russian painter, Kazimir Malevich painted a Black Square on a white canvas.

As simple as that. Some say he depicted the end of painting, a denial of art. Supposedly,

everything had already been painted, and therefore painting was over. But it so happens, it

was not. A lot has been done since then, and art never came to an end. It never will.

We think Ultrapar is like painting. We still see it fresh, with lots of possibilities. Many think

its value is already fully unlocked, completely known, entirely priced in, and that there is

nothing new to come. Some also think it's been covered by so many and for such a long

time, that there is nothing left to be said about it, as a Black Square on a white canvas.

We disagree. In this report we ask the question: Can the music go on? To answer that

question we derived seven smaller questions from it that we think lie in the inner core of

the investment case. By answering them, we intended to make clear what are the

fundamental principles that have guided the company's track record of value creation and

why we believe they are still in place, making this trend likely to continue for many years to

come. We tried to map not only what the company is doing, but also what the company is

not doing yet. And our view is that numerous opportunities to create value lie ahead.

We once heard that, "Musicians don't retire. They stop when there's no more music in

them." We think there is still music in Ultrapar. And our numbers suggest that the music is

playing loud, but that the market is not listening to it.

How to read this report?

This report will look too long for some investors. Hopefully, a few will read it thoroughly

and (who knows?) still wish for more. But some might think it deserves the words of

Winston Churchill, who once said: "This paper by its very length defends itself from the risk

of being read." The more we try to find the right balance between a concise view and an

adequate level of detailed analysis, the more we see there is no perfect formula with which

everybody would agree. The way to overcome that is to present our analysis in blocks that

can be entirely skipped. So, investors can jump directly to the topics they are most

interested in. And, to those who seek only a brief summary of our views, the three-page

investment summary may suffice. We think this way of organizing the ideas prevents us

from taking too much of the time of those who don't have any, but also from being too

shallow on relevant matters.

We think these seven fundamental questions were not yet answered but are essential for

an adequate understanding of the company's main value drivers, and the opportunities

that lie ahead. Questions are always a good way to address a complex issue. We tend to

agree with the professor and neurobiologist Stuart Firestein: "Questions are more relevant

than answers. Questions are bigger than answers. One good question can give rise to

several layers of questions, can inspire decades-long searches for solutions, can generate

whole new fields of inquiry, and can prompt changes in entrenched thinking. Answers, on

the other hand, often end the process". We hope our seven questions adequately address

investors' curiosity and interest.

22 June 2016


Figure 1: Our Fundamental Questions (An Illustrated Summary)

Source: Credit Suisse analysis

#1 - Economies of Scale # 2 - Logistics Infrastructure #3 - Network of Branded Stations #4 - Pricing Power

#5 - Access to Fuel Supplies #6 - Single Supplier #7 - Value of a Brand #8 - Playing Defensive and Offensive

Ultragaz Oxiteno Ultracargo Extrafarma

Fundamental Question: Can The Music Go On?

Q2) What Are The Prospects For Volumes Growth In Brazil? Q3) How do Margins Expand and What Should We Expect?

Q1) Why Is This Business So Good In Brazil?

Q4) How Promising Is The CONEN Area? Q5) How Much Should We Expect From Fuels Imports?

Q6) What Are The Prospects For The Other Businesses?

Q7) But Isn't It Too Expensive Already? Risks: How Can We Be Wrong?

0

500

1000

1500

2000

2500

3000

3500

4000

0 10 20 30 40 50 60 70 80 90 100

Income SSE 2014

Income CONEN 2014

Income CONEN 2004

1 mw

1 mw1/2 mw

2 mw

2 mw

3 mw

3

1/4 mw1/2 mw

5

1 minimum wage

2 mw

3 mw

5

1/2 mw

5

% of Population below each salary level

Ranges

of

sala

ry (

BR

L/

month

)

1.0

1.2

1.4

1.6

1.8

2.0

2.2

2.4

Q1 10 Q4 10 Q2 11 2011 Q3 12 Q1 13 Q4 13 Q2 14 2014 Q3 15

Economies of scale Inflation

Compound Inventory gains

Actual EBITDA margin

Model EBITDA margin

5,415 5,499 5,662 6,086 6,460 6,725 7,056

15% 15% 15%16%

16%17%

18%

2008 2009 2010 2011 2012 2013 2014

Ipiranga's # of stationsIpiranga's share of stations

0.8

1.0

1.2

1.4

1.6

Q1 10 2010 Q4 11Q3 12Q2 13Q1 14 2014 Q4 15

Average

Price exposure:

EBITDA margin

Branded

15

25

35

45

55

2008 2009 2010 2011 2012 2013 2014

Unbranded

1

1.05

1.1

1.15

1.2

1.25

1.3

1.35

Q1 10 2010 Q4 11Q3 12Q2 13Q1 14 2014 Q4 15

Volume sold

Economies of scale:

33% 35% 35% 35% 34% 33% 33% 32%

24% 21% 21% 21% 22% 22% 21% 22%

4% 4% 4% 4% 4% 4% 4% 4%

22% 24% 24% 23% 23% 23% 23% 23%

16% 16% 17% 17% 18% 18% 19% 19%

2008 2009 2010 2011 2012 2013 2014 2015BR Ipiranga Ale White flag Raízen

3.7%3.0%

0.0%

-5.0%-4.1%

5.7%

2.0% 1.7%

-1.8%

2.3%

-0.9%

-2.6%

-3.1%

-8.4% -8.3%

3.0%

1.3%

0.1%

-2.7%

-1.5%

1Q 15 2Q 15 3Q 15 4Q 15 1Q 16

Ipiranga

Raizen

BR

Market

0

500

1,000

1,500

2,000Jan-0

2

Dec-

02

Nov-

03

Oct

-04

Sep-0

5

Aug-0

6

Jul-

07

Jun-0

8

May-

09

Apr-

10

Mar-

11

Feb-1

2

Jan-1

3

Dec-

13

Nov-

14

Oct

-15

International Gasoline International Diesel

Domestic Diesel Domestic Gasoline

Oxiteno19%

Ultragaz8%

Ultracargo

3%1%

Extrafarma

74.5

9.1 4.1 1.6 3.3 12.51.4

11.1

92.6

Ipiranga Oxiteno Ultragaz Utracargo Extrafarma Corporate Net debt 12mo carryfwd

Ultrapar

-

2

4

6

8

10

0 20 40 60 80 100 120

Ipiranga's share of imports Other players' share of importsVolume (kbd)

Marg

in (

$/b

bl)

$ 303 mn$34mn $ 125 mn$14mn

Diesel Gasoline

-

1,000

2,000

3,000

4,000

5,000

19

96

19

98

20

00

20

02

20

04

20

06

20

08

20

10

20

12

20

14

20

16E

20

18E

20

20E

20

22E

20

24E

20

26E

20

28E

20

30E

20

32E

20

34E

20

36E

20

38E

20

40E

20

42E

20

44E

20

46E

20

48E

20

50E

Ave

rage

Sal

es p

er S

tati

on

(m

3/y

)

CS Forecast

USA Actuals

Ipiranga Actuals

22 June 2016


Ultrapar: There Is Still Music in Them

Investment Summary

■ Conservative Approach. We are initiating coverage of Ultrapar with an Outperform

rating and a 12-month forward target price of BRL 89.00/share. The company is widely

believed to be good quality, but it is also believed to be expensive and that current

prices would only be justifiable under aggressive assumptions for future growth and

margin behavior. However, we couldn't disagree more. We believe just the opposite is

true. To justify current prices, one would have to assume that capital will be allocated

at returns much lower than current levels for all the businesses Ultrapar owns. That's

exactly what we did in our base case. We conservatively assumed returns will be lower

and we still see a c.25% upside potential for the shares. But we also mapped a number

of additional opportunities that can lead to much higher upsides if at least part of them

is implemented. We think value creation is all about opportunity meeting a prepared

mind. That's why we say: There's still music in them.

■ Can the Music Go On? We chose to approach this question dividing it into seven

smaller questions that lie in the inner core of its businesses' fundamentals. (1) Why is

the distribution business so good in Brazil? (2) What are the prospects for volume

growth in Brazil? (3) How do margins expand and what should we expect? (4) How

promising is the CONEN area? (5) How much should we expect from fuel imports? (6)

What are the prospects for the other businesses? (7) But isn't it too expensive already?

We wouldn't be buyers if we just thought it's a good-quality company, because no good

business justifies whatever entry price. We wouldn't be sellers just because a good

track record can't go on forever. So, we thought it was essential to go through these

seven questions in order to understand the main value drivers behind the company's

outstanding track record of above average returns. After doing it, our conclusion was

that the fundamentals leading to such performance will not only remain in place, but

might get stronger for many years to come.

■ Lollapalooza Effects. When we see above-average returns on capital, we think

"moats" (or sustainable business protections) must be in place. Ultrapar is no different.

And particularly its main business—distribution—has strong moats, that, in our view,

become each day harder to cross. To put it simply, Ipiranga has a market position in

which: (1) there are enormous economies of scale; (2) logistics infrastructure acts as

an entry barrier; (3) more than 7,000 resellers are contracted; (4) pricing power is such

that the company can gradually and monotonically adjust margins; (5) access to fuel

supplies depends on previous volumes making it almost impossible for newcomers; (6)

the presence of a single refiner creates stability on the supply side; (7) brand power is

continually increasing and turning into better margins; and (8) the company can play

defensively and offensively by buying smaller distributors, which was the case with the

AleSat acquisition. In our view, these factors reinforce and combine with each other,

leading to hard-to-measure "lollapalooza effects" (non linear synergies).

■ "Opportunity Meeting the Prepared Mind—That's the Game." Rather than saying

only what is included in our numbers, we would also like to highlight what is not. We

think there are plenty of opportunities yet to be explored by the company to keep

creating shareholder value. These opportunities range from: deploying capital at least

in line with current return levels for longer (+BRL23.5/share); concluding the deal to

purchase Alesat (+BRL6.2/share); increasing coverage of convenience stores in its

service stations (+BRL6.4/share); possibly acquiring Liquigas (+BRL1.6/share);

ramping up Extrafarma to the initial plan of 100 stores/year, rather than 60 stores/year

(+BRL1.5/sh); exploring synergies among its different businesses; expanding the

22 June 2016


lubricants business; and so on. None of them were included in our base case and

target price. So, we believe that investors buying Ultrapar are buying 25% upside if the

future is just mediocre relative to current performance, and potentially a much higher

upside if at least a fraction of these additional options gets implemented.

Valuation and Sensitivities

■ Different Methodologies. We used five different valuation methodologies to assess

Ultrapar, and most of them point to a considerable upside to current share prices. Our

target price is based on a DCF model, in which we assume the returns on new

invested capital will decrease relative to today's figures. For the main business,

Ipiranga, that means return on invested capital will converge to 13% from current 25%.

The following table in Figure 2 summarizes the results of the different approaches we

used to define our base case. Figure 3 shows the DCF based valuation's sensitivity to

cost of capital. Figure 4 shows the main assumptions behind our base case. The main

point we would like to highlight is that we sought to apply conservative assumptions in

all of the methods, and none of the yet-to-explore opportunities were included in our

numbers. In the section about valuation (Question #7) we discuss the opportunities we

think lie ahead and might be explored by Ultrapar, creating additional upsides to our

numbers. Figure 5 shows a summary of the potential upsides leading to our bull case.

We also present a bear case, that assumes returns will converge to cost of capital

even faster.

Figure 2: Valuation Summary Table in BRL/share

Source: Credit Suisse estimates

Type Valuation MethodValue/share

in Mar-2017Description Preference

Level of

DetailTime Scale

Group two-parts DCF BRL 89/sh

Discounted Cash-flow approach with explicit model to 2020 and two periods of

continuing value '2020-2030' and '2030 onwards' with declining RONIC and

reinvestment assumptions

1st

SOTP BRL 93/sh

Business-by-business discounted cash-flow approach with explicit model to 2020

and two periods of continuing value '2020-2030' and '2030 onwards' with

declining RONIC and reinvestment assumptions

2nd

SOTP EV/EBITDA BRL 79/sh Implied EV/EBITDA multiples applied to each business's EBITDA '16 3rd

Group EV/EBITDA BRL 81/sh Historical EV/EBITDA multiple applied to group's consolidated EBITDA '16 4th

Group P/E BRL 66/sh Historical P/E multiple applied to group's consolidated earnings '16 4th

Multiples

DCF

Higher

Lower

Perpetuity

2016

22 June 2016


Figure 3: Target Price Sensitivity Analysis in BRL per share

Figure 4: Base Case Main Assumptions

Source: Credit Suisse estimates Source: Credit Suisse estimates

■ Assessing risk-rewards. In order to assess the balance between downside risks and

potential upside rewards, we have analyzed the Bear and Bull scenarios shown in

Figure 5. Bull scenario incorporates a non-exhaustive list of potential upsides which we

have valued (and discussed in greater detail in Figure 156). The Bear scenario

assumes that from 2020 onwards Ultrapar will never find new opportunities to deploy

capital in a value-accretive way. This could come as a result of fiercer competition,

exhaustion of good investment opportunities in the fuel distribution market, changes in

consumption habits, or other types of macro headwinds in the long-term. The Bull

scenario points to significant upside, c.81%, whereas Bear scenario points to only a

small downside, c.-3%. We think this is a compelling risk-reward profile.

Figure 5: Bear, Base, Bull Cases


Risks To Our Call

■ What If #1: Consumer Preferences and Solutions for Mobility. We might be wrong

in a number of ways. First, because we are dealing with businesses whose value

materializes over the long run, this assumes we are able to accurately understand how

consumer preferences will evolve in the decades to come. If new solutions for mobility,

to name one, are developed, or if people's habits require less and less mobility due to

increasingly virtual social interactions, demand for fuels might not continue to grow and

we might be proven wrong. We don't expect these things to happen during a time

frame that could materially affect Ultrapar's value, but we can of course be wrong.

■ What If #2: Macro Conditions. If macro conditions get worse and the current

economic recession proves to be long lasting, if inflation accelerates, if country risks

increase materially, or any other macro event takes place, our beliefs might have to be

reviewed.

R$ 89/sh 10% 12% 13.9% 14% 16% 18%

11% 853 361 232 228 166 130

12% 315 208 157 155 123 102

13% 192 146 118 117 98 84

14.3% 127 105 89 89 77 68

15% 106 90 78 78 68 61

16% 87 75 67 66 59 54

17% 73 65 58 58 52 48

Co

st

of

Eq

uit

y

Cost of Debt Assumption 2016E 2017E 2018E 2019E 2020E

GDP (YoY%) -3.5% 0.5% 2.0% 2.0% 2.0%

FX (avg) 3.60 3.76 3.92 4.10 4.28

Brazilian inflation (IPCA) (%) 7.5% 6.5% 6.5% 6.5% 6.5%

Selic (%) 14.25% 14.00% 13.75% 13.50% 13.25%

Ke (R$, nominal) (%) 15.6% 14.5% 14.4% 14.3% 14.3%

Cost of Debt (%) 15.0% 14.7% 14.4% 14.2% 13.9%

Wacc (R$, nominal) (%) 13.4% 12.6% 12.5% 12.5% 12.5%

Diesel demand growth (YoY%) -4.1% 0.5% 2.0% 2.0% 2.0%

Otto demand growth (YoY%) -3.0% 0.5% 2.0% 2.0% 2.0%

Valuation MethodValue/share

in Mar-2017

Upside

relative to BRL73.4/shDescription

Blue-sky scenario BRL 128/sh 81%

Same as base case plus upsides from i) higher RONIC at Ipiranga, ii) Alesat deal, iii)

Convenience stores expansion, iv) Possible M&A in LPG business and v) More

aggressive expansion in Extrafarma

Base case BRL 89/sh 25%Two parts DCF assuming return on new investments in 'the next decade', 2020-2030,

will average 18% in nominal terms.

Bear scenario BRL 69/sh -3%Assumes growth from 2020 onwards will not add value, but rather be made through

investments that return cost of capital.

22 June 2016


■ What If #3: Refineries Ownership. If Petrobras decides to sell a significant portion of

its refining business to private companies, it could lead to an increase in the need for

capital deployment on the part of distributors, as a more competitive refining segment

will likely be less prone to invest in logistics capacity in the country. Fluctuations in

refinery-gate prices would also very likely increase, meaning more volatility in

distribution margins. We are not assuming this as our base case scenario, but we

might be wrong.

■ What If #4: Consolidation May Change Competitive Landscape. We think

consolidation remains a trend in the fuel distribution and in the LPG distribution

businesses. The AleSat deal not being approved for any reason would be a risk to our

call. We also see a risk stemming from the potential sale of Petrobras's fuel distribution

arm, BR Distribuidora. Depending on the format the potential deal may take and on

who the buyers would be, it could materially change the competitive landscape and we

would have to reassess our call. Another similar risk relates to the sale of Liquigas,

Petrobras's LPG distribution subsidiary. Any risk this sort could result in higher

competitive pressure on Ultrapar from the company's peers, reducing the possibilities

for market share gains and economies of scale.

■ What If #5: Competition Gets More Aggressive. Finally, we consider price wars are

not in the best interest of any of the existing big distribution companies, but we might

be proven wrong if we see a self-destructive behavior like that, with companies willing

to sacrifice their margins for an additional market share percentage point.

■ What If #6: Ultrapar's Management, Governance and Capital Deployment. We

view Ultrapar as a good reference in terms of capital allocation capabilities, as was

once again evidenced by the AleSat deal. But we cannot rule out the risk of future

generations of management not being able to deploy capital as well. As we will see

ahead, fair value is extremely dependent on returns on new invested capital. We would

have to change our minds if this or any governance issues arise.

■ What If Other What Ifs Arise? The future reserves many unknowns. We are never

ready for them.

22 June 2016


Ultrapar

Is It Too Good to Be True?

"Mastery is a mindset." (Daniel Pink)

■ Some Say: The Music Can't Go On. Most of the arguments we see in the market

against Ultrapar's ability to keep going well seem to rely on an assumption that this is

just too good to be true. Ultrapar's amazing track record of margin expansion, along

with the continuous share price appreciation, makes investors cautious about paying

too much to own the business. We think this alertness is good, as no business justifies

whatever entry price. As Warren Buffett said: "What is smart at one price is dumb at

another." But, on the other hand, we don't typically like the arguments that come simply

on the back of an alleged "too good to be true" thesis. We think good businesses run

by good management can continue delivering good returns for longer than market

participants are willing to accept. We prefer the arguments that explore the true value

drivers of the businesses in order to understand if a sustainable competitive advantage

is in place. And we think this is the case here.

■ Some Say: It's Expensive but Has Good Quality. Most of the arguments we see in

the market in favor of investing in Ultrapar go like this: "We think it's expensive, but it is

a good-quality company with good management, and therefore it must be a good

investment." First of all, we don't agree that good-quality companies are necessarily

good investments, for obvious reasons. But we also don't agree that it is expensive in

the first place. We like the investment for a different reason: We believe the potential

value is bigger than prices, and this is what investing is all about. We never like

anything because it looks to have good quality, others like it, or news flow is favorable.

We like it when we think prices are lower than value, the latter measured

conservatively.

■ No Sacred Businesses, No Single Shots. Throughout the years, Ultrapar has kept

expanding its portfolio of businesses both organically and through acquisitions. The

company pretty much knows its circle of competence and keeps adding value to it. As

Thomas Watson once said: "I'm no genius. I'm smart in spots and I stay around those

spots." The gradual addition of businesses in a disciplined way is something we tend to

see as positive. It makes the company more able to deploy the cash from its operations

wherever most appropriate at each particular moment. We are not of the opinion that

diversification is always good or necessary. But we think value creation opportunities

sometimes become scarce within a specific business, and we like management teams

that are ready to play other games. Figure 6 shows the history and breakdown of the

group's EBITDA generation. It shows how the consolidated earnings power increased

in spite of the business cycles and specific challenges of each individual business

segment. Figure 7 shows the group's performance from a cash flow perspective,

comparing all the cash generated to the cash reinvested in the portfolio of businesses

since 2007.

22 June 2016


Figure 6: Ultrapar Historical EBITDA Composition

per Business in BRL millions

Figure 7: Free Cash Flow Growth vs. Invested

Capital Throughout the Years in BRL millions

Source: Company data, Credit Suisse analysis Source: Company data, Credit Suisse analysis

■ Metamorphosis. At the beginning, Ultrapar was pretty much an LPG distributor,

period. But the flexibility in capital allocation allowed the company to keep adding

businesses and also expand organically on different fronts, benefiting from the different

business cycles. Figure 8 depicts the EBITDA composition in relative terms among the

businesses. Figure 9 shows the capex profile throughout the years. Both charts clearly

show that Ultrapar is in constant metamorphosis. This culture is something we believe

builds up on itself and gives us confidence to believe in above-average capital

allocation. This is an important issue in valuing Ultrapar, as we will later discuss. It

always depends on the sustainability of its competitive advantages so that future

capital can keep being deployed at above its cost in the long run. The company is now

on the verge of consummating one more of these above-average capital allocation

initiatives, through the acquisition of AleSat, the fourth biggest fuel distributor in Brazil.

If that acquisition is approved, we think it will be yet another example of a trend that

has been in place for many years and that will likely be there for longer.

Figure 8: Historical % Contributions of Each

Business to the Group's EBITDA in %

Figure 9: Historical Capex Allocation Among the

Businesses within the Group in %

Source: Company data, Credit Suisse analysis Source: Company data, Credit Suisse estimates Note: Does not include capex in acquisitions

0 323 593 830

1,0

73

1,3

30

1,6

53

2,0

30

2,2

88

2,7

69

281

252

211

281

307

282

246

281

306

357

192

156

210

171

241

261

352

441

404

740

38

4351

104

111

118

143

158

167

26

30

29

0

500

1,000

1,500

2,000

2,500

3,000

3,500

4,000

2006 2007 2008 2009 2010 2011 2012 2013 2014 2015

Extrafarma

Ultracargo

Oxiteno

Ultragaz

Ipiranga

0

1,000

2,000

3,000

4,000

5,000

6,000

7,000

8,000

9,000

10,000

2007 2008 2009 2010 2011 2012 2013 2014 2015

Cumulative New Invested Capital Cumulative Cash from Operations

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

2010

2011

2012

2013

2014

2015

Oxiteno

Ultragaz

Ipiranga

Ultracargo

Extrafarma

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

2010

2011

2012

2013

2014

2015

Oxiteno

UltragazIpiranga

Ultracargo

Extrafarma

22 June 2016


■ Good Managements Add Value. We completely agree with Seth Klarman that "Good

managements add value. Good managements have lots of levers they can pull, they

can buy-back stocks when they are undervalued, they can use stocks as currency

when they are overvalued. Bad managements will only think about themselves first."

Ultrapar is a strong evidence of this, as it was able to keep adding value throughout the

years. In the long run, we believe skill overcomes luck, and discipline beats market

conditions. We have already seen good assets destroy value under managements not

focused on shareholder's returns. We have seen good managements add value out of

not-so-good assets and under tough business environments. But we think Ultrapar has

the best of both: good businesses and a management team with the appropriate

mindset. And as the psychologist and professor Daniel Pink puts it, "mastery is a

mindset."

■ Entry Points. Many investors wait for a significant price drop in Ultrapar's shares that

could offer them a good entry point. This happened in really very few instances in the

past five years. As we will hopefully make clear in the following pages, the company

owns well-positioned businesses that enjoy economies of scale and have a very

significant margin protection. In terms of volumes, despite the recent economic

downturn, long-term prospects remain good. So, waiting for prices to fall might be

frustrating, as the company has been able to consistently increase EBITDA every

single quarter, now for 39 consecutive quarters. Just to highlight this achievement, it

means 10 years with no instance of a quarterly EBITDA lower than in the year before.

We think every time the company's share prices reflect a convergence of future

margins to a flat level, as if the company's music were fading, the market is getting it

wrong. And the search for entry points becomes a pointless and long wait. The real

question becomes whether or not current prices already reflect Ultrapar's competitive

advantages, and our answer is no. In Question #7, we discuss the company's valuation

under different methodologies. They point toward higher equity value than current

prices indicate.

■ Monotonic Growth Sounds Boring. We think dancing to the music is much more fun

than dancing to the noise, but it seems this music is too long and monotonic for the

market to appreciate. People sometimes wait for earnings momentum, value triggers,

and positive news flow to indicate the right time to buy and enjoy a rally. But we believe

Ultrapar is a business to own. Its beauty is not in rallies, it's in the power of

compounding returns over the long run. The recent economic downturn has caused

worries about the company's ability to keep adding value. So, some have considered

selling it to avoid a decline. But, as Peter Lynch says: "It would be wonderful if we

could avoid the setbacks with timely exits, but nobody has figured out how to predict

them. Moreover, if you exit stocks and avoid a decline, how can you be certain you'll

get back for the next rally?" The company keeps gradually beating market consensus

estimates every quarter and also, time after time, surprising the market with timely

acquisitions of businesses with a good fit and also at good prices.

22 June 2016


Ipiranga

Question #1: Why Is This Business So Good In

Brazil?

"Frequently, you’ll look at a business having fabulous results. And the question is, ‘How long can this continue?’ Well, there’s only one way I know to answer that. And that’s to think about why the results are occurring now—and then to figure out what could cause those results to stop occurring." (Charlie Munger)

■ Are There Moats in This Business? In trying to assess the reasons a particular

business has been delivering good results for a long time, we like Charlie Munger's

approach of trying to find moats, or business barriers, that allow for sustained

performance. In capitalism, every time a company or sector is enjoying good margins

and returns, other players will try to join the party and, when it gets too crowded, all the

fun will be almost completely gone. So it is essential to understand how those margins

could be threatened. After analyzing the distribution business in Brazil, we concluded

that there are numerous and enormous barriers in place and that these barriers

reinforce and combine with each other. We not only think there are barriers in place,

but also that they are getting bigger.

■ The Party Is Not Open for Newcomers. Any company willing to assume a significant

portion of the distribution business within a certain geography will typically do that

through acquisitions. It makes no sense for a newcomer to build logistics infrastructure

from scratch next to an existing distribution network. This means a very high entry cost,

besides the marketing costs involved in building a brand. Figure 10 shows how this

segment has been gradually consolidating itself, concentrating in a few big companies.

These big four include AleSat, which is about to be purchased for BRL 2.168bn by

Ipiranga, if the anti-trust authority approves the deal announced on June, 12, 2016.

This is a business in which scale matters, as we will see ahead. And once a logistics

network is in place, it always makes much more sense for an established distribution

company to expand itself and supply the marginal demand than for another company

to build a new distribution base around the corner. This will always be the case, in our

view, and is a sustainable competitive advantage for those who are already well

positioned in the sector.

22 June 2016


Figure 10: History of Sector Consolidation Since 1990

Source: Sindicom, Credit Suisse Research Note: The potential purchase of AleSat by Ipiranga will leave the sector with now only three big distribution companies, which compares to eight companies in the 1990's. This is a material consolidation that allowed significant scale gains.

■ Moat #1: Economies of Scale. As we mentioned, the marginal investment to allow for

growth is always lower for the already established distribution company than it would

be for a newcomer. Building an additional tank in an existing facility is cheaper than

building an entire distribution facility. Also, using an existing tank with a higher turnover

is cheaper than building a new tank. On the operational side, fixed costs also allow for

economies of scale when volumes increase. So, this is a business in which scale

matters. There are many sectors in which size is a problem. As Warren Buffett says:

"Size is the enemy of performance to a significant degree." But we think the distribution

business enjoys the benefit of scalability, and it is not by chance that the survivors in

this sector in Brazil have made every effort to become bigger while keeping their

expansion strategies. In Question #3, we present a more detailed discussion of

economies of scale and how they have helped to boost margins.

■ Moat #2: Logistics Infrastructure. In order to play a major role in the distribution

business, a wide spectrum of well positioned assets is necessary. This includes a

combination of access to fuels, storage capacity, and the right distribution capability.

Throughout the years, after the already-mentioned consolidation process in the sector,

the few really big distribution companies accumulated assets with enough scale and

coverage that provide them with a sustainable competitive advantage. Having an

established country-wide network of distribution bases constitutes a big enough barrier

to discourage newcomers. Ipiranga is well positioned relative to the main sources of

fuels (mainly Petrobras's refineries) and has an established portfolio of assets,

including large-scale blending and storage capacity, truck-loading facilities, and a

transportation subsidiary with its own trucks fleet. Figure 11 shows Ipiranga's portfolio

of distribution bases that spread through all the high consuming regions over the whole

1993

2002

2006

2007

1990 Today

2003

2008

2001 2004 2009

1999

2009

2011

2008

Only the Service Stations was transferred or partially transferred

Distribution Companies members of Sindicom

22 June 2016


national territory and those of other players. The potential purchase of AleSat will add

to Ipiranga's current network ten additional distribution bases spread all over the

country. Ipiranga might be able to use its existing infrastructure to distribute part of the

volume after the deal goes through, if it does, meaning there might be future

economies of scale in the operation of distribution bases after AleSat joins Ipiranga's

network.

Figure 11: Network of Distribution Facilities Per Company in Brazil

Source: ANP, Credit Suisse Research Note: The size of the bases displayed in orange represents the sum of the smaller capacities of other players.

■ Moat #3: Network of Branded Stations. In Brazil, law does not allow distribution

companies to operate service stations. They may or not be owned by a distributor but,

in any case, another company must be in charge of their operations, typically small

independent private companies. The big distribution companies compete among

themselves to brand stations, through tailor-made contracts that involve the

deployment of capital from the distributor and the exclusive right to supply fuels. As

branded stations can only buy products from the same distribution company, this

network of branded stations constitutes an additional entry barrier for a newcomer

willing to play the distribution game. Terminating the existing contracts or establishing

new ones to build a significant client base would require a huge amount of capital.

Others

22 June 2016


Ipiranga currently has contracts with a network of 7,241 stations, or c.18% of all the

stations in Brazil. If the purchase of AleSat is approved, Ipiranga will expand its moat,

becoming the distributor with the greatest number of stations in the country, exceeding

BR, jumping to c.23% of the total. Figure 12 shows the historical evolution of Ipiranga's

network both by number of stations and share of the total. Figure 13 shows the

historical levels of total capital deployed by Ipiranga and its relation to the size of the

stations network throughout the years.

Figure 12: History of Ipiranga Service Stations and

Ipiranga's Share of Total Stations in Brazil Ipiranga stations in units, share of total stations in % of stations

Figure 13: Evolution of Ipiranga's Capex per Service

Station Added in BRL millions, Unitary capex in BRL thousand per station

Source: Sindicom, Credit Suisse analysis Source: Company data, Sindicom, Credit Suisse analysis

■ Moat #4: Pricing Power. When the owners of service stations decide to enter into

branding contracts with a big distributor, the retailers benefit from the franchise they

are adopting. So, not only does the station display the image code of the distribution

company but also inherit the customer service standards that typically allow them to

increase volumes relative to previous levels. The other side of this story is that these

contracts establish volume targets that retailers have to achieve but do not pre-

establish price formation. Therefore, through these exclusive contracts, Ipiranga gains

a pricing power that constitutes an invaluable asset. In Question #3, we present a

discussion of the effects of this pricing power that allows Ipiranga to keep gradually

expanding margins. Because of high inflationary pressures, any retail-related business

in Brazil should be particularly concerned about pricing mechanisms that allow them to

preserve margins. There are few businesses in which this sort of pricing power is

present, but we believe Ipiranga increases its power each day to become a

monotonically growing cash generator.

■ Moat #5: Access to Fuel Supplies. The barriers for anybody willing to enter the

distribution business are not limited to the exclusivity contracts with the client base but

also include the access to the very products they are supposed to distribute. Petrobras

requires that the distributors' orders be limited within a narrow range set according to

the historical volumes sold by that distributor. Therefore, even if we could conceive of a

player that would decide to invest big capital to create comparable infrastructure and

would also be willing to pay big capital to terminate current contracts with retailers, this

company would still not be able to purchase significant volumes from Petrobras out of

5,415 5,499 5,662 6,086

6,460 6,725

7,056

15%15%

15%

16%

16%

17%

18%

2008 2009 2010 2011 2012 2013 2014

Ipiranga's # of stations Ipiranga's share of stations

229 222

383

591

942

746

815

42 40

68

97

146

111 115

2008 2009 2010 2011 2012 2013 2014

Capex (BRLm) Capex/station (BRL'000/station)

22 June 2016


a sudden. The more scale current players gain, the more they are able to purchase

fuels from refineries and the less other players are able to hurt them in any meaningful

way. Figure 14 shows the level of concentration of Otto and Diesel volumes in the four

biggest companies and how the combined (Ipiranga + AleSat) would look like. Figure

15 shows the companies' market share per product.

Figure 14: Historical Market Share by Company In Diesel + Otto Cycle Fuels in %

Figure 15: 2015 Market Share by Company for Each

Product in %

Source: Sindicom, Credit Suisse analysis Source: Sindicom, Credit Suisse analysis

■ Moat #6: Single Supplier. The fact that almost 100% of refining capacity is

concentrated in a single player significantly differentiates the Brazilian distribution

sector from that of other countries. One such difference is that the existence of a single

state-owned refiner assures no disruption on the supply side, as Petrobras has

historically made all efforts to assure the market is completely served. If we think of a

scenario in which the supply of fuels is locally disrupted, this would likely create much

higher market share fluctuations among distributors, in our view. Another consequence

of a single supplier is that Petrobras has historically delivered a significant portion of its

products at advanced points rather than being limited to the refineries' gates. Although

this movement makes it easier for a smaller distributor to survive, it also reduces

overall cost pressure on the supply side, avoiding margin squeezes and allowing for

lower investments and working capital from distributors. Finally, a single supplier

assures that distribution players have similar supply prices and conditions, leaving

competitive pressures to the stages down the supply chain. Therefore, distributors

have been able to focus their attention on establishing their brands and services over

the years, rather than being pressured from both sides. This factor is very likely to stay

in place as long as Petrobras concentrates most of Brazilian refining operations.

■ Moat #7: The Value of a Brand. We are not in a position to put a number to the

Ipiranga brand, but we don't have to. We again recall the words of Charlie Munger,

when he said: "There are other factors that are terribly important, [yet] there is no

precise numbering you can put to these factors (…) Well practically everybody

overweighs the stuff that can be numbered, because it yields to the statistical factors

they are taught in academia." So, we won't do that. But, we understand that, in a

35% 35% 35% 34% 33% 33% 32%

21% 21% 21% 22% 22% 21% 22%

4% 4% 4% 4% 4% 4% 4%

24% 24% 23% 23% 23% 23% 23%

16% 17% 17% 18% 18% 19% 19%

2009 2010 2011 2012 2013 2014 2015

BR Ipiranga Ale White flag Raízen

20%28%

17%

37%

93%

57%

35%

19%

21%

14%

23%

3%

20%3%

5%

1%

3%3%

20%

20%

4%

19%

4%

32%19%

11%

1%38%27%

63%

18%

1% 0%

22%

HydrousEthanol

Gasoline GNV Diesel Fuel Oil Jet Fuel All Fuels

BR Ipiranga ALE Raízen AirBP Others

22 June 2016


distribution/retail market, brand matters and even more so in a market such as Brazil,

with a history of informality and fuel adulteration. The potential purchase of AleSat, if

approved, will boost Ipiranga's brand exposure to at least 2,000 additional service

stations in the country. In the United States, service stations are also mostly owned by

smaller private companies, rather than by the Oil Majors whose brands they display.

The widely known brands bring the stations a wealth of credibility and help attract

customers. Figure 16 shows the evolution of branded stations in Brazil in comparison

with the so-called white flag (unbranded) stations. Figure 17 compares the level of fuel

sales of branded stations in Brazil to those of unbranded stations. It clearly shows that

brand matters.

Figure 16: Number of Branded and White Flags

Stations in Brazil Throughout the Years in units

Figure 17: Average Fuel Sales Per Service Station of

Branded and White Flags Stations in Brazil in bbpd / station

Source: Sindicom, Credit Suisse analysis Source: Sindicom, Credit Suisse analysis

■ Moat #8: Playing Defensively and Offensively. While running a business in which

there is a moat, the main driver of managers should be to preserve the moat or even

widen it, if possible. Superior returns will attract other players, and moat preservation

should take most of managers' attention and efforts. Accordingly, if competition gets

fiercer despite the entry barriers so as to threaten or weaken the moat, managers can

act defensively by buying the new entrant. By buying smaller distributors that are

consolidating their positions as relevant players, Ipiranga is actually playing both

defensively and offensively, as it is at the same time eliminating the threat of a material

loss of share, and actually increasing its share and moat in the market relative to the

remaining players. We believe, Ultrapar's team, even after the AleSat deal is approved,

if approved at all, will keep looking carefully for smaller distributors to spot other good

opportunities.

■ When Moats Combine: Non-Linear Reinforcements. These barriers combine in a

way that reinforces each other by producing non-linear effects. The bigger a distributor

gets, the more its brand is in evidence and therefore the bigger the benefits for a

retailer to brand his station. This leads to more contracts and, thus, higher market

protection, bigger economies of scale, and higher pricing power. This combination

allows for better margins, which support further expansion. The higher the number of

Ipiranga stations, the more the negotiation power shifts toward the distributor. And the

21,882 22,703 23,524 24,250 25,073 26,109

15,834 15,635 15,709 15,533 14,255 13,884

58%

59%

60%

61%

64%

65%

2009 2010 2011 2012 2013 2014

Branded White Flags % Branded

Branded

Unbranded

15

25

35

45

55

2008 2009 2010 2011 2012 2013 2014

22 June 2016


stronger the franchise, the more final consumers will be attracted to Ipiranga stations.

There are many routes through which we can describe this reinforcement process,

because this is not a linear process. It means that the right combination of these

barriers may lead to hard-to-measure positive results.

■ Will These Moats Increase or Vanish? We think these barriers are increasing. By

observing, in Figure 10, the level of concentration in this sector in the last two decades,

it becomes clear that the power is now in the hands of three or four companies. They

keep branding white-flag stations, gaining scale, expanding margins, and increasing

their presence. And the more all this happens, the more powerful those moats become.

The very nature of the barriers involved is such that they get harder and harder to

overcome.

■ A Big Growing Market. All these entry barriers would be worth little if they existed in a

depressed market. We think, however, that in spite of the current economic downturn,

fuel consumption in Brazil is bound to grow. Fuel demand has been growing c.3.6% for

the last 45 years. We present in Figure 18 Brazil's GDP growth in this period, when

demand for oil products expanded consistently, almost in line with the overall economic

performance. We should highlight that these growth rates were achieved across

economic cycles and under different political regimes, despite global crises and

economic downturns. Figure 19 shows how emerging countries have seen oil demand

expand while developed countries have faced very slow growth or even consumption

declines over the same period. In Question #2, we provide more details on fuel

demand in Brazil and its yet-to-be-developed potential.

Figure 18: Oil Products Consumption Growth vs.

GDP in the last 45 years in Brazil index (1970 = 100)

Figure 19: Oil Consumption Growth in Different

Countries from 1970 to 2014 in kbpd

Source: EPE, BACEN, Credit Suisse analysis Source: Brazil from EPE, Other countries from BP Statistical Review, Credit Suisse analysis.

Note: Other countries' data biofuels, are included

■ How Long Will Consumers Depend on this Business? Oil products are essential

and will very likely be needed for a while, in spite of the threats from new energy

solutions, especially for mobility. Most of the alternatives are still expensive and will be

introduced gradually, with little impact in the near term. We think this might be

particularly true for emerging economies, where there will be a continuous challenge of

-

100

200

300

400

500

600

19

70

19

72

19

74

19

76

19

78

19

80

19

82

19

84

19

86

19

88

19

90

19

92

19

94

19

96

19

98

20

00

20

02

20

04

20

06

20

08

20

10

20

12

20

14

Oil Products GDP Index (1970=100)

GDPCAGR 3.9%

Oil Products DemandCAGR 3.6%

0.6%1.1%

-0.3% -0.4%-0.7% -0.7%

3.6%

5.3%

7.0%

US

Can

ada

Fra

nce

Germ

any

Ital

y

UK

Bra

zil

India

Chin

a

22 June 2016


making energy available at affordable prices to allow for an increase in standard of

living. Most of the future reduction in carbon emission will likely come from power

generation, which is the single biggest global emitter according to the IEA and the

cheapest one to reduce. Oil is a very competitive source of energy for transportation

purposes, and it will likely be around for many years. Even in IEA'ss most aggressive

scenario in terms of growth in alternative fuels, internal combustion engines will still

represent the majority of global engine sales over the next two decades. So, it would

take many decades for electric vehicles to get a big portion of the total fleet.

■ Why Is It So Good, Then? After getting into the details on how these moats are built

and how they become hard-to-overcome business protections, we get the sense that

they explain a lot of the past performance of the company in terms of profitability and

returns. It seems to us that the longer Ipiranga operates with such performance, the

harder it gets for the market to believe it can continue even further. This, of course,

reflects in the prices the market is willing to pay for the business. But we think that the

nature of these protections is such that time is more likely to reinforce them than to

weaken them. And their combination leads to what Charlie Munger calls "lollapalooza

effects," meaning that these moats produce synergic effects, as their combined impact

is stronger than the sum of their individual effect if each of them were present alone.

We believe this provides the basis for the discussion on margins presented in Question

#3.

22 June 2016


Question #2: What Are the Prospects for Volume

Growth in Brazil?

"Brazil is a teenager. If you have a teenager living in your house, you know it can be chaotic. But the best decades for a teenager are still ahead." (Howard Marks)

Fuel Consumption in Brazil

■ Extreme Short-Term Focus. The market has focused a lot of attention on the recent

demand decrease taking place since 2015. We've seen the market, time after time,

overreact to a momentary negative business environment, in spite of the long history of

favorable fundamentals and potential for long-term growth. This happens when

performance is not evaluated over a long enough horizon. It reminds us of the words of

the psychologist and Noble Prize winner, Daniel Kahneman: "Too much concern about

how well one is doing in a task sometimes disrupts performance by loading short-term

memory with pointless anxious thoughts." We think monthly market figures are a huge

distraction to businesses with strong long-term prospects, as is the case of the

distribution business.

Does It Matter? The impacts of the current economic slowdown are very limited, in our

view. As we will discuss in Question #3, there are significant possible economies of

scale in this business, and volume growth matters. But short-term shortfalls shouldn't

matter too much, as far as long-term growth prospects are preserved. In Figure 20, we

show the demand deceleration movement in recent quarters that made many investors

concerned. In the words of Lee Kuan Yew: "If you do not know history, you think short

term. If you know history, you think medium and long term." Figure 21 shows the

impact on Ipiranga's valuation of different short-term demand scenarios. It makes the

case that, although next year's demand matters, it is, of course, the long term that

carries most of the weight in justifying the value of the business. By analyzing this

chart, investors can see that if something jeopardizes the prospects for long-term

demand growth, by reducing it, let's say, from 3% to 1%, then the business value

would decrease by almost 50%. In order to have the same effect on value, the short-

term volumes would have to decrease by c.35% in the first year, which is completely

out of proportion. Our team of economists already sees a short-term rebound in GDP

beginning next year, with GDP still contracting 3.8% this year but starting to recover

(+0.5%) in 2017.

Figure 20: History of Demand Fluctuations for

Diesel and Otto Cycle Fuels in Brazil in % YoY

Figure 21: Value Per Share Sensitivity to Short Term

and Long-Term Demand Growth in x

Source: Brazilian Oil, Gas, and Biofuels Agency (ANP), Credit Suisse analysis Source: Credit Suisse analysis

Diesel YoY

Otto YoY

-10%

-5%

0%

5%

10%

15%

20%

Q1 07 Q2 08 Q3 09 Q4 10 Q1 12 Q2 13 Q3 14 Q4 15

1.0x 1.1x 1.1x

1.3x 1.4x 1.5x

1.7x 1.8x 1.9x

1%

1%

2%

2%

3%

3%

4%

-6% -4% -2% 0%

Lo

ng

Te

rm V

olu

mes G

row

th

2016 Volumes Decline (YoY)

22 June 2016


■ This Business Is Not for Short Termers. We believe that when investors are buying

Ultrapar, or particularly Ipiranga, they are buying an infrastructure and a strong position

in a market that will likely keep expanding over a long period. The business is capital-

intensive and, as such, materializes its value over the long run, especially in cases in

which scale grows over the years, consolidating an ever-higher business protection.

Even more so, if the company is successful in acquiring AleSat's network of stations

and logistics assets. The other side of this coin, and the good one, is that investors

don't have to bother trying to predict margin cycles or short-term demand dynamics, as

value is driven by long-term growth potential, as we've already discussed. As Howard

Marks once put it: "There is no such thing as THE time to go into a market. The

question is if this is A time to buy." So, the question becomes whether oil consumption

can still grow in Brazil for a prolonged period.

■ Developed versus Emerging. Over the past 25 years, consumption of oil products in

emerging economies has generally grown at significant rates, while we've seen this

trend actually in reverse in OECD countries, where demand has fallen as a result of

both lower economic growth and efficiency gains in fuel usage. In Figure 22, we depict

the historical variation (indexed in 1991) in consumption of oil products per capita in

some high consumer countries. In all of them, consumption either declined heavily or,

at best, stabilized at the same levels of 1991. Figure 23 depicts the same information

during the same period in some emerging economies, showing that per capita

consumption almost doubled in Brazil, and quadrupled in China in the period.

Increasing living standards promoted a sharp increase in consumption over a period of

more than 25 years.

Figure 22: History of Per Capita Consumption of

both Gasoline and Diesel (Selected Developed

Countries) Index (1991 = 1)

Figure 23: History of Per Capita Consumption of

both Gasoline and Diesel (Selected Emerging

Countries) Index (1991 = 1)

Source: EIA, World Bank, Credit Suisse analysis Source: EIA, World Bank, Credit Suisse analysis

■ But Per Capita Consumption Is Still Low. Despite more than two decades of high

demand growth rates, per capita consumption in emerging economies cannot yet even

approach the consumption levels of developed economies. As we can see in Figure

USA

France

Germany

Japan

0.7

0.8

0.9

1

1.1

1.2

1991 1994 1997 2000 2003 2006 2009 2012

Brazil

China

Mexico

0.7

1.2

1.7

2.2

2.7

3.2

3.7

4.2

1991 1994 1997 2000 2003 2006 2009 2012

22 June 2016


24, consumption in the US is around 40 barrels per day per thousand people, while

these same figures for other developed economies are in the high teens. The chart

also shows that all these countries decreased diesel and gasoline consumption per

capita since 2000. But, in spite of this decline, consumption in all those countries is still

much above what we see in Brazil, China, and Mexico, for example. In Figure 25, we

show that none of these three countries achieved 10bpd per thousand people yet.

More specifically, consumption in Brazil is still c.8bpd/thousand people, and in China,

c.4bpd/thousand people. We see this as a sign of undeveloped potential, as the

increase in consumption on a per capita basis takes many years to happen and comes

as a consequence of economic growth, social mobility, and increasing living standards.

Figure 24: Current Per Capita Consumption of both

Gasoline and Diesel (Selected Developed Countries) in bpd / thousand people

Figure 25: Current Per Capita Consumption of both

Gasoline and Diesel (Selected Emerging Countries) in bpd / thousand people

Source: EIA, World Bank, Credit Suisse analysis *Note: Last data available (US – 2014, remaining – 2013).

Source: EIA, World Bank, Credit Suisse analysis *Note: Last data available (Brazil – 2014, Mexico – 2013, China - 2012).

Diesel Consumption

■ Brazil Is Fueled by Diesel. One important driver of fuel consumption that we believe

will remain in place for a long time is the extremely high dependency of Brazil's

transportation matrix on road transportation. Considering other countries with large

territories, road transportation is more prevalent than other types of transportation

(such as railroads, pipelines, and waterways), since it allows for bigger scale and is

much cheaper for long distances. In Figure 26, we present the transportation matrix in

countries with large territories. Among the countries we analyzed, the share of road

transportation in the total cargo moved is generally much lower than 50%. In Brazil, it is

c.68%, with some variations depending on the information source. This level is closer

to what is observed in countries with much smaller areas, as shown in Figure 27.

42.5

24.1

21.1

17.4

40.6

18.5 17.5

14.0

USA Germany France Japan

2000

2014*

5.4

1.5

8.67.8

4.0

9.6

Brazil China Mexico

2000

2014*

22 June 2016


Figure 26: Transportation Matrix of Countries with

Territories Larger Than 7MM Km² as of 2014* in % of tkm transported

Figure 27: Transportation Matrix Countries with

Territories Smaller Than 0.6MM Km² as of 2014 in % of tkm transported

Source: Federal State Statistics Service of Russia, National Bureau of Statistics of China, BTS (USA), Department of Infrastructure and Regional Development (Australia), CNT and ILOS (Brazil), Credit Suisse analysis *Note: Last data available (US – 2011, Australia – 2012, remainder – 2014).

Source: Eurostat, Credit Suisse analysis Note: Information on share of pipelines was not available for these countries. Shares are ex-pipelines.

■ Small Railroads Coverage. It is remarkable how the usage of railroads for cargo

transportation is low in Brazil in comparison to other countries where there is equal

need to move goods over long distances. In Figure 28, we show this usage in volume

terms (tonne-kilometers, abbreviated as tkm) for other countries with large territories.

One related issue is of course the considerably small railroad infrastructure in Brazil

relative to other countries. In Figure 29, investors can find a metric of the railroad

coverage in different countries, with the figures in the circles representing the total area

of each country and the bars representing the ratio between the total length of railroads

available and the total area. Brazil has an area close to that of China and the USA, but

a "railroad coverage" of only c.10% of what is observed in those countries. The total

investment required to expand railroad infrastructure is, of course, very high, and that

is a barrier for this gap in the transportation matrix to be closed.

5%

32%

45%35%

68%

43%

18%

29% 46%

18%

3%

48%8%

19% 11%

49%

2%17%

3%

Russia China United States Australia Brazil

Road Rail Water Pipeline

80%

64%

85% 87%

16%

23%

15% 13%4%

12%

France Germany Italy United Kingdom

Road Rail Water

22 June 2016


Figure 28: Transportation of Goods in Different

Countries and Volumes Moved by Railroad in 2014* in Btkm

Figure 29: Railroad Coverage in Different Countries

as of 2014 Railroad coverage in km/th sq km, areas in sq km

Source: Russia from Federal State Statistics Service, China from National Bureau of Statistics of China, United States from BTS, Australia from Department of Insfrastructure and Regional Development, Brazil from CNT and ILOS, , Credit Suisse analysis *Note: Last data available (US – 2011, Australia – 2012, remaining – 2014).

Source: ANTT, CIA The World Factbook, Worldbank, Credit Suisse analysis

■ New Rail Infrastructure May Come. In spite of the higher investments involved in

railroads, an expansion of the current coverage or an increase in the capacity of

existing infrastructure may represent a risk to future diesel demand growth, as a single

composition may be able to move the equivalent of what c.300 trucks could move. For

that reason, diesel consumption by railroads (c.4 cu m/Mtkm) is much more efficient

than consumption by trucks (c.42 cu m) for the same amount of cargo transported. In

Figure 30, we depict the average diesel consumption required to move one tonne of

cargo over one kilometer versus the total volume (in Btkm) transported by both

railroads and trucks in 2014. The product of these two variables (on the horizontal and

vertical axes) leads to the areas of the two rectangles, representing the total diesel

consumption per transportation mode in that year.

■ Would It Risk Future Growth Significantly? In the face of this high efficiency in

diesel consumption by railroads in relation to trucks, we thought it would be useful to

understand the impact on total consumption if railroads got a bigger share of the

transportation matrix. We provide in Figure 31 a sensitivity analysis, in which we show

how we think the total diesel consumption would behave as a function of the share of

railroads in cargo transportation. We drew a vertical line that represents where Brazil

stands today in terms of diesel consumption for transportation purposes. The chart

shows that, for diesel consumption to be hurt by, let's say, 10%, railroads' share of

cargo transportation would have to increase by c.44% relative to today. We think an

increase of this magnitude would take time to materialize and would also require lots of

capital to be deployed in new and existing railroad infrastructure.

2,945 2,777 2,128 261 307

13,053

6,715

2,787

309

1,384

China United States Russia Australia Brazil

Rail Others

32.1

20.5

5.3 4.8 3.6

9.1 9.4

16.4

7.7 8.4

USA China Russia Australia Brazil

Railroad Coverage (Km/th Km²)

Area (mn Km²)

22 June 2016


Figure 30: Diesel Consumption by Road and Rail

transportation in 2014 in MCM

Figure 31: Diesel Consumption By Trucks And By

Railroads And Sensitivity To Higher Share of

Railroads in Transportation Matrix Consumption in MCM, share of railroads in % of tkm

Source: EPE, ANP, CNT, ILOS, Credit Suisse analysis Source: EPE, ANP, CNT, ILOS, Credit Suisse analysis

■ Would Local Rail Expansion Affect the Route of a Particular Crop? We were

particularly concerned about the potential reduction in diesel consumption that could

result from private investments in existing railways, especially those directly connected

to soy and corn production, which are concentrated in the Midwest region and

represent a high share of Brazilian agricultural exports. These two crops currently have

the two highest shares of rail transportation (after iron ore, of course, which alone

represents c.76% of total goods moved by rail). After new investments in capacity

expansion, rail might gain some share from trucks in the transportation of these two

crops. In Figure 33, we show how the total diesel consumed in transportation would be

affected by different shares of railroads in the transportation of soy and corn. It shows

that, only in an extreme and unrealistic scenario in which trucks would be left with none

of either product to transport, total diesel consumption would be affected by c.5%, at

most.

Figure 32: Share Of Different Cargoes In Railroad

Transportation (2011-2014) in % of tonnes

Figure 33: Diesel Consumption Sensitivity To

Higher Share of Railroads in Corn and Soy size of circles = diesel consumed in transportation in kbpd

Source: Brazilian Land Transportation Agency (ANTT), Credit Suisse analysis Source: Brazilian Energy Planning Corporation (EPE), Brazilian Oil, Gas, and Biofuels

Agency (ANP), Brazilian Land Transportation Agency (ANTT), Ministry of Development, Industry, and Foreign Trade (MDIC), Credit Suisse analysis

-

5

10

15

20

25

30

35

40

45

0 200 400 600 800 1,000 1,200

Consu

mptio

n p

er

freig

ht

tran

sport

ed

(m³/

mn t

on.k

m)

Freight transported (bn ton.km)

Rail

Road

47.7 mn m³

1.2 mn m³0

10

20

30

40

50

60

Die

sel c

onsu

mptio

n (

mn m

³)

Rail's share in freight transportation matrix (% of ton.km)

Consumption by RoadConsumption by Rail

0% 40% 60% 86%18% 26%

Current Total

Consumption

Consu

mptio

n b

y tr

uck

s

Consumption

by railroads

Scenario for 10%

Lower Diesel

Consumption

+44%

2.5%3.4%

76.0%

9.4%

3.7%

10.9%

Iron Ore Agricultural Metals & Mining Others

CornSoy

878 863 848 834 819

871 856 842 827 813

864 850 835 821 806

858 843 829 814 799

851 837 822 807 793

843

-25%

0%

25%

50%

75%

100%

-25% 0% 25% 50% 75% 100%

Rail

share

in c

orn

tra

nsp

ort

aio

n

(% o

f to

n.k

m)

Rail share in soy transportation (% of ton.km)

35%

56%

22 June 2016


■ Many Trucks, and Old Ones. Future economic growth will still depend a lot on trucks,

and therefore on diesel. Even more so if we consider that future growth will depend on

a truck fleet that is generally older than what we see in many countries. Figure 34

shows the age profile of different European countries and where the Brazilian fleet

stands. Figure 35 shows that the partial renewal of the Brazilian truck fleet resulted in a

slight reduction of the average age of the fleet. Especially in recent years, significant

public incentives were offered for the acquisition of new trucks in the country, but

goods are on average still transported by trucks older than 10 years. For fuel

consumption, efficiency obviously matters.

Figure 34: Age Profile of Trucks Fleet in Different

European Countries and % Less Than 10 Years in %

Figure 35: Historical Average Age of Brazilian Truck

Fleet in millions, unless otherwise stated

Source: Brazilian Association of Motor Vehicle Manufacturers (Anfavea), Eurostat, Credit Suisse analysis

Source: Brazilian Association of Motor Vehicle Manufacturers (Anfavea), Credit Suisse analysis Note: This chart was provided by courtesy of our LatAm Capital Goods research team.

■ GDP and Diesel, Hand In Hand. This strong dependence of the Brazilian economy on

trucks to sustain growth has translated in a correlation between GDP and diesel

consumption throughout the years. It has varied a bit across different periods; at times

diesel consumption has grown faster than the overall economy (such as in the past five

years), and at times diesel has lagged GDP. But, we think it is appropriate to observe a

longer period in this case; in the past c.45 years and in the past c.15 years, growth has

on average closely tracked GDP. For all the reasons we discussed above, we think the

fundamentals that lead to this close connection would take a lot of time and

investments to change. So we consider this a fundamental protection of the investment

case.

29%

35%

35%

36%

48%

56%

59%

67%

27%

42%

39%

51%

36%

35%

31%

28%

18%

18%

18%

13%

12%

8%

9%

4%

26%

6%

9%

5%

1%

1%

1%

Brazil

Spain

Portugal

Italy

EU

UK

France

Germany

Less than 5 years 5-9 years 10-14 years 15 years or overLess than 10 years

12.4 11.9 11.7 11.6 11.3 11.2

10.7 10.1 9.9 9.7 9.7

10.1

200420052006200720082009201020112012201320142015

22 June 2016


Figure 36: Diesel Consumption Growth and GDP

Growth Since 1970 in %

Figure 37: Diesel Consumption Growth and GDP

Growth Since 2000 in %

Source: Central Bank of Brazil, Brazilian Energy Planning Corporation (EPE), Credit Suisse analysis


Consumption of Otto Cycle Fuels

■ The Same Happens to Motorization Rates. We also see a lot of room for Otto cycle

fuels consumption to increase over many years to come. The motorization rates

observed in Brazil are still significantly low when compared to more developed

countries. Motorization has actually been growing very fast during the past decade as a

consequence of income growth and social mobility in Brazil, and also due to incentives

to purchase cars. Despite the high growth rates, the number of cars per thousand

people is far behind that of other developed economies, and even a few other

emerging countries. Figure 38 shows this metric in Brazil in comparison to different

countries. Figure 39 shows that in emerging economies the motorization rate has

increased sharply since 2005, while it has remained fairly stable or even declined in

developed economies. We believe this trend is likely to continue.

Figure 38: Motorization Rates for Different

Countries in 2014 (Passenger Cars Fleet Only) in Vehicles / 1000 People

Figure 39: Growth in Motorization Rates by Country,

from 2005 to 2014 (Passenger Car Fleet Only) in %

Source: International Organization of Motor Vehicle Manufacturers (OICA), World Bank, Credit Suisse analysis

Source: International Organization of Motor Vehicle Manufacturers (OICA), World Bank, Credit Suisse analysis

y = 0.99x + 0.01R² = 0.7

-5%

0%

5%

10%

15%

20%

-5% 0% 5% 10% 15%

Ye

arly

Die

sel C

on

sum

pti

on

Gro

wth

Yearly GDP Growth

y = 1.01x + 0.00R² = 0.50

-5%

0%

5%

10%

-5% 0% 5%

Ye

arly

Die

sel C

on

sum

pti

on

Gro

wth

Yearly GDP Growth

611 610

548

480 477

379

236 204

159

85

Can

ada

Ital

y

Germ

any

Fra

nce

Japan US

Arg

entin

a

Mexi

co

Bra

zil

Chin

a

9%

2%

-2%

1%

7%

-16%

77%

56%

63%

419%

Canada

Italy

Germany

France

Japan

US

Argentina

Mexico

Brazil

China

22 June 2016


■ Otto Consumption Hand in Hand with Income. The consumption of Otto cycle fuels

(gasoline and ethanol) has been highly correlated with the earnings of the country's

working population. Of course, this means that there is also a negative correlation

between Otto fuels consumption and unemployment rates. In the following charts,

displayed in Figure 40 and Figure 41, we show both correlations over recent years.

Investors might be concerned about the recent rise in unemployment rates in Brazil

and by the lack of improvement, which could take time. But, we also think that, 10 or 20

years from now, income will have increased, and that is what should matter most.

Figure 40: Consumption of Otto Cycle Fuels vs.

Real Earnings of Working Population in Brazil since

2003 consumption in TCM, earnings in BRL/year


Unemployment Rate since 2001 consumption in MCM, unemployment rate in %



■ More Income, More Cars, More Fuels. The advancement in income in Brazil, and

also in credit, has led to a clear increase in the number of cars, and thus higher

consumption. We've already shown how the per capita motorization rates evolved. The

charts in Figure 42 and Figure 43 show how these two variables have walked hand in

hand in the past. We think that long-term income growth will make it possible for more

and more people to purchase cars, and so the fleet is likely to keep expanding,

boosting consumption at Ipiranga's service stations. We still don't see this trend

reversing in the foreseeable future. There might be a time when living standards will

have increased to a point at which fuel consumption per capita will stabilize and even

start to decline, as we've seen in developed economies. But we think this would be

expected to happen only many years from now. Until then, we think the average

Brazilian will buy and use a car as soon as it becomes affordable.

R² = 0.98

25,000

30,000

35,000

40,000

45,000

50,000

55,000

60,000

65,000

20,000 22,000 24,000 26,000 28,000 30,000 32,000

Ott

o C

ycle

Fu

els

Co

nsu

mp

tio

n (

tho

usa

nd

m³)

Occupied People Real Earnings (BRL)

R² = 0.90

20

25

30

35

40

45

50

55

60

65

4% 6% 8% 10% 12% 14%

Ott

o C

ycle

Fu

els

Co

nsu

mp

tio

n (

mn

m³)

Unemployment Rate

22 June 2016



Light Vehicle Fleet since 1990 consumption in MCM, vehicles in million units


Light Vehicle Fleet since 1990 consumption in MCM, vehicles in million units

Source: Brazilian Association of Motor Vehicle Manufacturers (Anfavea), Brazilian Energy Planning Corporation (EPE), Credit Suisse Analysis

Source: Brazilian Association of Motor Vehicle Manufacturers (Anfavea), Brazilian Energy Planning Corporation (EPE), Credit Suisse Analysis

■ High versus Low Predictability. The strong correlation of Otto fuels consumption and

income gives rise to the idea of very high predictability of demand. But, in fact, this

seemingly high correlation disguises the high uncertainty of gasoline and ethanol

consumption. As most of the new light vehicles in Brazil are flex-fuel, consumers can

fill their tanks with any blend, ranging from 100% ethanol up to 100% gasoline. This

means much higher volatility for gasoline and ethanol consumption than for Otto fuels

as a whole. In Figure 44, we show the history of vehicle licensing in Brazil by type of

fuel since 1957. It shows the predominance of ethanol motor vehicle sales during the

1980s, the predominance of gasoline vehicle sales during the 1990s, and finally the

flex-fuel motors since the early 2000s. Figure 45 shows how the profile of the entire

fleet of light vehicles has changed substantially over the past decade as a result of the

introduction of flex-fuel vehicles.

Figure 44: Historical Vehicle Licensing in Brazil by

Type of Fuel in %

Figure 45: Change in Brazilian Fleet Profile, by Fuel

Type in %

Source: Brazilian Association of Motor Vehicle Manufacturers (Anfavea), Credit Suisse analysis

Source: Brazilian Association of Auto Parts Manufacturers (Sindipeças)

-

5

10

15

20

25

30

35

40

45

-

10

20

30

40

50

60

70

19

90

19

92

19

94

19

96

19

98

20

00

20

02

20

04

20

06

20

08

20

10

20

12

20

14

Gas

olin

e V

eh

icle

Fle

et

(mill

ion

s)

Ott

o C

ycle

Fu

els

Co

nsu

mp

tio

n (

mn

m³)

Otto consumption (thousand m³) Gasoline Vehicle Fleet (mn Units)

R² = 0.97

25

30

35

40

45

50

55

60

65

10 15 20 25 30 35 40

Ott

o F

ue

ls C

on

sum

pti

on

(m

illio

n m

³)

Light Vehicles Fleet (mn)

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

1957 1965 1973 1981 1989 1997 2005 2013

Gasoline

Ethanol

Flex

Fuel

Diesel

67%62%

58%53%

48% 45% 41% 37% 34%32%

11%18%

25%31%

37%42%

47% 51% 54%57%

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

2006 2007 2008 2009 2010 2011 2012 2013 2014 2015

Gasoline

Ethanol

Flex Fuel

Diesel

22 June 2016


■ A Very Particular Fleet Profile. This high flexibility in fuels consumption is a

characteristic unique to the Brazilian fleet profile. In the US, consumption of light

vehicles is focused on gasoline, while in European countries there is a migration

towards diesel powered engines (see Figure 46). This migration verified in the

European market represents a challenge for refiners to focus their production profile

towards diesel, and pressures gasoline cracks down. But it doesn't bring the same

level of uncertainty in demand that was created in Brazil as a consequence of flex-fuel

vehicles. In Figure 47, we show the many peaks and valleys that gasoline and ethanol

consumption has had over many decades. The negative effects of this unpredictability,

however, are much more disruptive for refiners than for distributors, as both gasoline

and ethanol are distributed through the same network. Distributors are responsible for

blending anhydrous ethanol with gasoline A to produce gasoline C and also for

distributing hydrous ethanol to service stations. So, demand for distributors is

somewhat protected from the oscillations of consumer preferences for each fuel.

Therefore, as long as liquid fuels are needed and require a distribution infrastructure,

there will be music for Ipiranga.

■ Volumes Fluctuate. Despite the overall protection distributors have in terms of the

Otto cycle fuels mix, there are indeed consequences for these companies when there

are shifts in consumer preferences. One relates to logistics. As ethanol has a lower

calorific power than gasoline, if consumers shift the mix towards more ethanol, there is

an increase in the overall volume, because more barrels of ethanol are needed to fulfill

the same demand for fuels. And, this is not negligible, as each barrel of gasoline

shifted to ethanol would require c.1.42 barrels of the latter. So, if there is a sudden

change in mix, the logistics network has to adapt very quickly in order to enable an

increase in the flow of products. So, the nominal capacity available approaches full

utilization rates faster if demand becomes predominantly for ethanol. In the long run,

this could mean higher capex needs.

Figure 46: Passenger Car Fleet by Fuel Type in

Different Countries in 2014 in %

Figure 47: Historical Consumption of Otto Cycle

Motors by Fuel Type in %

Source: Brazilian Association of Motor Vehicle Manufacturers (Anfavea) and Brazilian Association of Auto Parts Manufacturers (Sindipeças), BTS (USA), ACEA (other countries), Credit Suisse analysis

Source: Brazilian Energy Planning Corporation (EPE), Credit Suisse analysis

32% 37%42%

47% 47%

60% 65%

94%

64%

4%

56% 44%50%

35%33%

1%3% 2%

10%2% 5% 3% 5%

2%

58%

France Brazil Spain Italy Portugal UnitedKingdtom

Germany UnitedStates

Gasoline Diesel Other fuel Ethanol Flex fuel0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

1970 1976 1982 1988 1994 2000 2006 2012

Gasoline

Anhydrous

ethanol

Hydrous ethanol

22 June 2016


Profile and Capex of Stations

■ More and Bigger Stations. As a result of all this potential demand growth, the supply

side will have to keep adapting in two ways. One is through the addition of new fueling

stations to the existing network. The other is through the expansion of same-store

sales (SSS)—in other words, by expanding the average capacity of stations. Ipiranga

has been adding stations to its network by both branding existing stations and building

new ones. The current expansion rate is around 350 stations per year, of which c.50%

are new "greenfield" stations. We believe that not only Ipiranga but the entire market

will both increase the average volume per station and build new ones. Of course, if the

AleSat deal goes through, we will see an immediate jump of 28% in Ipiranga's number

of stations. It is interesting to observe, in Figure 48, that this has already been the case

in the Brazilian market, as demand has been expanding faster than the overall number

of stations, meaning that stations, on average, are selling higher volumes each day.

This trend also happened in the American market, where volumes increased while the

number of stations was actually reduced, meaning an accelerated concentration of

sales in fewer and bigger stations, as shown in Figure 49.

Figure 48: Brazil's Total Fuel¹ Consumption vs.

Number Of Service Stations (indexed) Index (2007=1)

Figure 49: United States Total Fuel Consumption vs.

Number Of Service Stations (indexed) Index (1996=1)

Source: Brazilian Association of Fuel and Lubricant Distributors (Sindicom), Brazilian Oil, Gas, and Biofuels Agency (ANP), Brazilian Energy Planning Corporation (EPE), Credit Suisse analysis 1 – Diesel/Gasoline/Ethanol

Source: EIA, NACS, Credit Suisse analysis

■ Winner Takes It All. Looking at businesses, we always seek potential "winner takes it

all" effects. We believe, though, that for service stations these effects are, of course,

limited to micro regions; the best stations end up enjoying some concentration benefits.

All else equal, better located stations, offering good convenience with a differentiated

customer service will tend to attract more customers. This will increase their scale, thus

reducing their fuels' purchase prices. So, slightly lower prices attract even more

customers from other stations. Besides that, while metropolitan areas get bigger, the

cost of real estate also tends to increase, thereby increasing the hurdle, in terms of the

volumes necessary for a station to survive. Better and higher scale stations are more

likely to stay in business.

Number of

service

stations

(index)

1.00

1.10

1.20

1.30

1.40

1.50

1.60

2007 2008 2009 2010 2011 2012 2013

Diesel and otto

consumption (index)

Number of

service

stations

(index)

0.75

0.85

0.95

1.05

1.15

1.251996

199

7

199

8

1999

200

0

200

1

200

2

200

3

200

4

200

5

200

6

200

7

200

8

200

9

2010

201

1

201

2

2013

Gasoline and diesel

consumption in retail (index)

22 June 2016


■ Volumes Per Station Will Keep Growing. This discussion has important implications

for long-term forecasts. It is relevant, when one is trying to determine the long-term

value of Ipiranga, to understand what capex levels should look like in order to enable

the addition of stations and the expansion of existing ones. We think the number of

stations will keep increasing for many years, although at a lower rate than the overall

volumes sold, maintaining the trend that we've seen in the past few years in Brazil. In

our models, we are assuming that SSS will on average be responsible for 50% of the

overall volumes growth each year, meaning that the remaining 50% of demand growth

will be met through the addition of stations. Figure 50 shows our forecasts for both

variables. As a sanity check, we extended our models to the long term in order to

understand where this assumption would lead the average sold volumes per station. In

Figure 51, it is possible to see that overall volumes per station would keep increasing

and, after 35 years, would still be slightly lower than what we currently observe in the

American market. It seems to us enough time for the effects of gradual concentration to

materialize in the retail segment in Brazil. And, it also made us feel more comfortable

that sales per station would not converge to a level that could be physically unrealistic.

Figure 50: CS Forecast of Brazilian Volumes in

Retail and for Total Number of Fueling Stations Index (2015 = 1)

Figure 51: Historical Volumes Per Station (Brazil vs.

USA) and CS Forecast for Brazilian Market in cu m/year

Source: Brazilian Oil, Gas, and Biofuels Agency (ANP), Brazilian Energy Planning Corporation (EPE), Brazilian Association of Fuel and Lubricant Distributors (Sindicom), Credit Suisse estimates

Source: Company data, EIA, NACS, Credit Suisse estimates

■ Will You Be There? As volumes resume growth in Brazil, we think Ipiranga will clearly

be well positioned to benefit from that. The company has very aggressively expanded

its network of branded stations, and this has contributed to both higher economies of

scale and a higher number of contracts through which Ipiranga's pricing power leads to

gradual margin expansion. Figure 52 shows that, in terms of network size, Ipiranga is

now very close to BR Distribuidora, and the number of stations still unbranded remains

high, meaning a yet-to-be-explored branding potential. Ipiranga was, by the way, the

company that was able to gain the highest share of total stations in recent years, in

percentage points, after increasing by more than 30% its number of stations in six

years (versus c.27% for Raízen and c.11% for BR).

0.5

1.0

1.5

2.0

2.5

3.0

2015

2017E

2019E

2021E

2023E

2025E

2027E

2029E

2031E

2033E

2035E

2037E

2039E

2041E

2043E

2045E

2047E

2049E

Index

(2015 =

1)

Retail Sales

# of Stations

0

1000

2000

3000

4000

5000

1996

1999

2002

2005

2008

2011

2014

2017E

2020E

2023E

2026E

2029E

2032E

2035E

2038E

2041E

2044E

2047E

2050E

Ave

rage S

ale

s per S

tatio

n (m

3/y

)

CS Forecast

USA

Actuals

Brazil

Actuals

22 June 2016


■ A Leap Forward. Now, Ipiranga is preparing for a sudden leap forward, potentially

expanding by an additional 28% its network of stations, through the acquisition of

AleSat. After the deal takes place, the company may reach somewhere close to 23% of

the total stations in Brazil. In terms of volumes per station, Figure 53 shows that

Ipiranga saw a sharp decline immediately after the Texaco acquisition in 2009 but has

continued to increase since then. The chart makes two points clear. One is that

Ipiranga's average volume per station has not increased at the same rates as the

average Brazilian station, as a consequence of its branding strategy over these years.

The other is that by acquiring AleSat's stations, Ipiranga's volumes per station will

decline, as happened after Texaco's deal. But we believe the Ipiranga franchise and

service standards will keep attracting higher volumes to both the recently branded

stations and to AleSat's converted stations, if the deal goes through.

Figure 52: Companies' Share Of Total Stations by

Number and Share After Potential Deal (Ipiranga +

AleSat) in %

Figure 53: Ipiranga's and AleSat's Historical

Average Volumes Per Station in Relation to Brazil in bpd / station

Source: Brazilian Association of Fuel and Lubricant Distributors (Sindicom), Brazilian Oil, Gas, and Biofuels Agency (ANP), Company data, Credit Suisse analysis

Source: Brazilian Association of Fuel and Lubricant Distributors (Sindicom), Brazilian Oil, Gas, and Biofuels Agency (ANP), Company data, Credit Suisse analysis

■ How Much Capex Is Necessary to Grow? One important though obvious issue is to

make sure that the capex levels for growth reflect both types of expansion so that no

free lunch is ever served. We attribute a capex level related to the addition of new

stations to Ipiranga's network (branding and construction combined), but we also

assume that, as volumes per station increase over time, it will gradually cost more to

renew existing contracts with retailers. This makes sense, in our view, because the

capex involved in the renewal of contracts should be somewhat connected to volumes

sold, as the investments and working capital required in the stations also rise. Figure

54 shows what we think capex should look like in order to sustain the expansion of the

number of stations as well as the sales per station. Of course, our intention here is not

to make any call on how much capex will be needed in 2050. We wouldn't have such

an excessive expectation. Our main purpose in extending the model that far is basically

understanding how capital will have to be deployed to sustain growth. What we want to

avoid is setting parameters for perpetuity that would allow for free growth. Figure 55

shows that capex will have to migrate from maintenance towards expansion, but most

4% 4% 4% 5% 5% 5%

15% 15% 16% 16% 17% 18%

19% 19% 19% 19% 20% 20%

12% 12% 12% 12% 12% 14%

51% 50% 50% 48% 46% 44%

2009 2010 2011 2012 2013 2014

Ale Ipiranga BR Raízen White flag and others

20

25

30

35

40

45

50

2008 2009 2010 2011 2012 2013 2014

Brazil Average

Ipiranga's Average

AleSat's Average

22 June 2016


of it would have to come from expansion in existing stations, and only a portion from

new stations, as we assume volumes per station will grow.

Figure 54: Long-Term Capex To Sustain Renewal Of

Existing Stations and Expand Current Network in BRL mn

Figure 55: Share of Long-Term Capex, Maintenance

vs. Expansion in %

Source: Company data, Credit Suisse estimates. Note: In this case, we are not including the additional capex that would be necessary after the deal with AleSat.

Source: Company data, Credit Suisse estimates. Note: In this case, we are not including the additional capex that would be necessary after the deal with AleSat.

■ Convenience Stores. Convenience stores represent additional potential for future

sales growth and market share gains in the years to come, in our view. Ipiranga is the

distributor with the highest share of the convenience stores market, both in terms of

number of stores and total revenues. The company currently has a network of 1,919

stores and revenues of c.BRL1.8bn. In Figure 56, we show that am/pm, Ipiranga's

convenience store brand, is almost 9 percentage points ahead of the second-biggest

network of convenience stores. The chart also shows that, after a potential AleSat deal,

Ipiranga will immediately have 30% of all convenience stores in the country. The

second place has 17%.

■ Just a Baby Sector. While the concept of combining convenience stores and service

stations is widespread in many countries, as shown in Figure 57, it is still in its early

stages in Brazil, with only a small portion of stations having convenience stores. This

clearly shows unexplored potential to boost fuel sales, gain market share, increase

scale, and increase brand exposure. Ipiranga had roughly 24% of stations with

convenience stores in 2014, up from 23% a year earlier. As of today, the coverage has

already increased to 27%, as reported by the company. AleSat, on the other hand, has

only c.13%. This unexplored potential means that, after the deal, Ipiranga might be

able to expand the am/pm franchise through the added stations and also that volumes

sold per station in AleSat's network may be boosted through a higher convenience

store coverage.

0

2,000

4,000

6,000

8,000

10,000

12,000

14,000

16,000

201

4

201

6E

201

8E

202

0E

202

2E

202

4E

202

6E

202

8E

203

0E

203

2E

203

4E

203

6E

203

8E

204

0E

204

2E

204

4E

204

6E

204

8E

205

0E

Network Maintenance

Churn

New

Expansion

Existing

Others

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

201

4

201

6E

201

8E

202

0E

202

2E

202

4E

202

6E

202

8E

203

0E

203

2E

203

4E

203

6E

203

8E

204

0E

204

2E

204

4E

204

6E

204

8E

205

0E

Network Maintenance

Maintenance on Churn

Expansion New Stations

Expansion Existing Stations

Others

22 June 2016


Figure 56: Share of Convenience Store Market in

Number (Inside) and in Total Revenues (Outside) as

% of Stores (Inside Chart), % of Revenues (Outside

Chart) in % of stores (Inside Chart), in % of Revenues (Outside Chart)

Figure 57: Coverage Of Convenience Stores In

Brazil vs. Other Countries In Relation to Stations

And Population in % of stations, thousand people / store

Source: Nielsen, Brazilian Association of Fuel and Lubricant Distributors (Sindicom), Credit Suisse analysis

Source: Mix Consultoria, Brazilian Association of Fuel and Lubricant Distributors (Sindicom), Credit Suisse analysis

■ Looks Good on the Costs Side. Detailed information on convenience stores is not

widely available, and therefore it is hard to measure the exact performance, results,

and impacts of their expansion. What seems to be obvious, though, is that their

presence attracts more customers to the service stations where they are installed.

Ipiranga estimates that the number of cars entering a gas station increases roughly by

15% because of them. Not all of this increase will, of course, translate into an

immediate and direct increase in volumes sold, but we think it might improve customer

loyalty and, therefore, later translate into more volumes. In terms of revenues,

Ipiranga's average store seems in line with the competitors', with a similar number of

transactions per year and similar average sales per customer (actually Raízen has a

slight advantage on that front). AleSat's stores have a slightly lower average

transaction value but apparently a similar number of annual transactions as that of the

competitors. On the cost side, we see both Ipiranga's and AleSat's average store

operating with fewer employees and requiring smaller areas than competitors. Figure

58 and Figure 59 present a summary of these figures.

17%

4%

26%

13%

40%

17%

3%

27%

17%

36%

Ipiranga

Ale

BR

Raizen

Others

19%

30%

60%

81%

88% 90%95%

26

37

53

7 74

0

5

10

15

20

25

30

35

40

45

Brazil Chile Canada US UK Uruguay Australia

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%Stations with Stores

th. People / Store

22 June 2016


Figure 58: Revenues of Average Convenience Store in BRL '000 / y, No. of transactions in '000 units/y, average sale in BRL

Figure 59: Required Labor and Area Per

Convenience Store, by Company area in sq m, number of employees

Source: Brazilian Association of Fuel and Lubricant Distributors (Sindicom), Company data, Credit Suisse analysis

Source: Brazilian Association of Fuel and Lubricant Distributors (Sindicom), Company data, Credit Suisse analysis

Competitive Environment

■ Curious Short-Term Dynamics. In the recent past, the Brazilian economic downturn

has made total fuel consumption decline. And, along with that, the main distributors,

affiliated with Sindicom, have seen their share of the market, on average, decline

slightly. As we demonstrate in Figure 60, the Otto fuels market has been decelerating

since the beginning of 2015 and in the 4Q15 reached its first year-on-year decline. The

chart shows that BR has been losing share since the beginning of 2015, while Ipiranga

and Raízen managed to keep their sales more resilient for a while. After the 4Q15,

though, even Ipiranga and Raízen saw their Otto fuel volumes decrease in year-on-

year terms for the first time, following what happened to the overall market. The chart

in Figure 61 shows year-on-year growth on a monthly basis. It is possible to note that

Sindicom companies have outperformed the market in times of growth, due to their

expansion and branding strategies, but have underperformed the market during this (in

our view) temporary demand decline.

921

729

946

1,227

11.1

8.5

10.6

12.7

0

2

4

6

8

10

12

14

16

18

-100

100

300

500

700

900

1100

1300

BR Ale Ipiranga Raizen

Revenues# of TransactionsAverage Sale

83 86 89 96

64

48

57

73

6.0

4.0

5.8

9.1

3

4

5

6

7

8

9

10

11

30

35

40

45

50

55

60

65

70

75

80

BR Ale Ipiranga Raizen

Area

Direct Employees

22 June 2016


Figure 60: Historical Quarterly Sales of Otto Cycle

Fuels, by Company in % year-on-year change

Figure 61: Historical Monthly Sales of Otto Cycle

Fuels, Sindicom in % year-on-year change

Source: Brazilian Association of Fuel and Lubricant Distributors (Sindicom), Brazilian Oil, Gas, and Biofuels Agency (ANP), Credit Suisse analysis


Less Diesel for Power Plants. The chart in Figure 62 shows that overall diesel

volumes have been declining consistently in recent quarters. A big portion of this

reduction though has been due to lower consumption by diesel-fired thermal power

plants. This affected BR much more than it did the market, as BR is the main supplier

of the power sector. Therefore, the average market decline since 2Q15 has been

steeper than what we see at Ipiranga and Raízen. As the chart in Figure 63 shows, the

Sindicom average has remained pretty much in line with the market, in spite of BR's

sharp decline in volumes, meaning that Ipiranga and Raízen gained some share here.

Figure 62: Historical Quarterly Diesel Sales, by

Company in % year-on-year change

Figure 63: Historical Monthly Diesel Sales, Sindicom in % year-on-year change



3.7%3.0%

0.0%

-5.0%

-4.1%

5.7%

2.0% 1.7%

-1.8%

2.3%

-0.9%

-2.6%

-3.1%

-8.4% -8.3%

3.0%

1.3%

0.1%

-2.7%

-1.5%

1Q 15 2Q 15 3Q 15 4Q 15 1Q 16

Ipiranga

Raizen

BR

Market

-20%

-15%

-10%

-5%

0%

5%

10%

15%

20%

Jan/1

3

Apr/

13

Jul/

13

Oct

/13

Jan/1

4

Apr/

14

Jul/

14

Oct

/14

Jan/1

5

Apr/

15

Jul/

15

Oct

/15

Jan/1

6

Apr/

16

Sindicom minus Market Market Sindicom

-2.9%

-1.1%

-1.8%

-0.7%-1.3%

-0.7%

-2.4%

-1.1%

-0.1%

3.5%

0.0%

-4.7%

-10.7%

-15.2%

-17.5%

-1.7%

-3.3%

-6.2%

-7.2%

-6.1%

1Q 15 2Q 15 3Q 15 4Q 15 1Q 16

Ipiranga

Raizen

BR

Market

-20%

-15%

-10%

-5%

0%

5%

10%

15%

20%

Jan/1

3

Apr/

13

Jul/

13

Oct

/13

Jan/1

4

Apr/

14

Jul/

14

Oct

/14

Jan/1

5

Apr/

15

Jul/

15

Oct

/15

Jan/1

6

Apr/

16

Sindicom minus Market Market Sindicom

22 June 2016


■ Somewhat Better as Far as Retail Is Concerned. The same deceleration in demand

can be observed when looking only at retail sales. In both Figure 64 and Figure 65, it is

possible to see that Ipiranga has not been declining as fast as Sindicom as a whole,

which of course means that the company has been gaining share relative to Sindicom.

It is also possible to note, however, that Raízen has declined less, and in gasoline has

even increased sales while overall demand and Sindicom sales have declined sharply.

Figure 64: Gasoline Sales, Retail Only in % year-on-year change

Figure 65: Diesel Sales, Retail Only in % year-on-year change

Source: Brazilian Association of Fuel and Lubricant Distributors (Sindicom), Credit Suisse analysis


■ Outperformance Seems Related to Imports. We think Raizen's outperformance in

volumes relative to Ipiranga in some of the recent quarters has been a result of a more

aggressive importing strategy, with Raízen importing higher volumes and selling them

possibly through unbranded stations. In Figure 66 and Figure 67, we show on a

monthly basis how the outperformance periods coincide with the periods in which price

premiums in Brazil were significant and therefore allowed distributors to import

additional volumes. The strategy of gaining market share by selling fuels through

unbranded stations is somewhat innocuous, as margins in those operations are close

to zero. But, when the fuels are imported, rather than bought from Petrobras, the

distributor may still sell through unbranded stations and get the import margins, so the

fact that distribution margins per se are close to zero through that channel becomes

less relevant. We think this recent movement is not likely to become a structural one.

Inasmuch as imports are possible, both Ipiranga and Raízen are likely to benefit from

this opportunity. But later, when the combination of oil prices and FX rate no longer

allow for imports, we think this temporary market share dynamics will return to the

normal trend. We provide a more detailed approach on imports in Question #5.

3.8%

3.0%

0.0%

-5.0%

-4.2%

5.6%

1.9% 1.8%

-1.8%

2.3%

-0.7%

-2.5%

-3.0%

-8.4% -8.4%

2.1%

-0.5%

-2.2%

-5.6%

-3.9%

1Q 15 2Q 15 3Q 15 4Q 15 1Q 16

Ipiranga

Raizen

BR

Sindicom

-2.2%-1.6%

-3.9%

-5.5%

-5.0%

0.3%

-2.0%

-2.9%

-3.7%

-1.7%

-4.0%

-4.8%

-5.3%

-7.3% -7.4%

-2.6%

-3.7%

-5.2%

-6.4%

-5.6%

1Q 15 2Q 15 3Q 15 4Q 15 1Q 16

Ipiranga

Raizen

BR

Sindicom

22 June 2016


Figure 66: Year-on-Year Growth in Gasoline

Volumes of Raízen ex Ipiranga vs. Gasoline Price

Premiums/Discounts in percentage points; price premiums in %

Figure 67: Year-on-Year Growth in Diesel Volumes

of Raízen ex Ipiranga vs. Diesel Price

Premiums/Discounts in percentage points; price premiums in %

Source: Brazilian Association of Fuel and Lubricant Distributors (Sindicom), Bloomberg, Credit Suisse analysis

Source: Brazilian Association of Fuel and Lubricant Distributors (Sindicom), Bloomberg, Credit Suisse analysis

■ What to Expect, Then? To invert the answer, we would say that the things that could

reverse the long-term demand growth trend would take many years to materialize and

would require significant capital. Investors would have to believe that either: (1) the

Brazilian economy and income will stagnate in the coming decades; (2) new alternative

cargo transportation infrastructure will suddenly be built and will capture the majority of

what is currently moved by trucks; (3) income growth in Brazil will not increase the

motorization and per capita consumption rates, as has happened in every other

country; (4) new distribution infrastructure will be built by new entrants in spite of the

current players' scale, brand power, and contracts with stations; (5) the current main

competitors will deploy capital better than Ipiranga and will be faster and smarter in

expanding; or (6) competitors will initiate a long-lasting self-destructive behavior,

through price wars and aggressive branding strategies, forcing Ipiranga to lose its

position. We believe none of these events are very likely. This last factor could happen

for a while but would not last very long, in our view, as this business leaves no room for

such behavior. Besides all of this, the potential purchase of AleSat, still to be approved,

could boost market share at once and lead Ipiranga to jump ahead of the competition

again. If this operation is approved, it would mean even more growth potential will be

added and more earnings power to make future expansion possible. So, we think the

words of C.S. Lewis would apply to this situation: "There are far, far better things ahead

than any we leave behind."

-30%

-20%

-10%

0%

10%

20%

30%

40%

50%

60%

70%

-10%

-5%

0%

5%

10%

15%

Jan/1

5

Feb/1

5

Mar

/15

Apr/

15

May

/15

Jun/1

5

Jul/

15

Aug/1

5

Sep/1

5

Oct

/15

Nov/

15

Dec/

15

Jan/1

6

Feb/1

6

Mar

/16

Apr/

16

Raízen minus Ipiranga Gasoline premium in domestic mkt

-40%

-30%

-20%

-10%

0%

10%

20%

30%

40%

50%

-6%

-4%

-2%

0%

2%

4%

6%

8%

10%

Jan/1

5

Feb/1

5

Mar

/15

Apr/

15

May

/15

Jun/1

5

Jul/

15

Aug/1

5

Sep/1

5

Oct

/15

Nov/

15

Dec/

15

Jan/1

6

Feb/1

6

Mar

/16

Apr/

16

Raízen minus Ipiranga Diesel premium in domestic mkt

22 June 2016


Question #3: How Do Margins Expand and What

Should We Expect?

"Simplicity is complexity resolved." (Steve Jobs)

■ The Value of Growth. Ipiranga is the single most important business within Ultrapar

and represents a big part of its value, as we will demonstrate later in this report.

Moreover, a big part of Ipiranga's value can be attributed to its growth prospects,

despite the fact that the company has already experienced an extended period of high

growth. Considering the period from 2010 to 2015 alone (since the acquisition of

Texaco's distribution network), EBITDA has grown 2.5x, from c.BRL1.1bn to

c.BRL2.7bn. In this chapter we take a look at the mechanisms behind Ipiranga's

growing profitability and make a judgment as to whether it can persist in the future.

A Very Simple Equation: EBITDA = EBITDA Margin x Volume. We can separate

EBITDA growth into two fundamental parts: volume growth and margin expansion. This

breakdown for Ipiranga is depicted in Figure 68. We note that, although volume

expansion has been an important factor behind the company's overall growth, there is

more to Ipiranga growing profitability than simply a growing market. In fact, a big part of

Ipiranga's growth can be attributed to margin expansion. We already discussed the

outlook for volumes in Question #2. In this section, we focus on assessing why margins

expand.

Figure 68: Ipiranga's EBITDA Growth: Volume Growth and Margin Expansion in %, year-on-year change

Source: Company data, Credit Suisse analysis. Note: The acquisition of Texaco stations took place in 2009.

■ The Million-Dollar Question. Explaining the mechanisms behind Ipiranga's past

margin expansion is key to understanding why this business is so profitable. This is

why we ask ourselves, Why do margins expand?. Naturally, investments are a matter

of future developments, rather than past events. Therefore, our objective is to

ultimately answer whether the factors that sustained growth before will be present in

the years to come. This can be framed by asking, How sustainable are the factors that

have allowed for margin expansion?"

■ Hypothesis testing. After phrasing this question, we should now move on and

develop our hypotheses to answer it. Before we proceed, however, we need to

understand how we will test whether our answers are sufficiently accurate. Our

proposed approach is to make hypotheses around the possible mechanisms behind

12%

31%

-2%

10%15% 15% 16%

9%

20%

6%

8%

43%17% 8% 8%

6%

3%

19%

42%40%

29%

24% 24%23%

13%

21%

2007 2008 2009 2010 2011 2012 2013 2014 2015

EBITDA margin growth

Volume growth

EBITDA growth

22 June 2016


the margin expansion, build models implementing each of these mechanisms and then

test the results assuming the hypotheses materialize. Finally, we compare our

mathematical experiment to the observable universe. In other words, we compare the

backtesting predictions of our models to the actual reported EBITDA. This is a sound

method when it is not possible to do an actual experiment.

■ The Wonder of Compounding. Fuel distribution is a business with narrow margins

and high turnover. This is a business in which small differences in key parameters can

result in a big difference in profitability. For example, a change of a fraction of a

percentage point in the margin profile over a long period can increase cash generation

twofold. This is the effect we want to capture in our models, so that overall growth

should be explained by compounding small changes over time. We have enough

evidence that Ipiranga strongly benefited from slight margin expansions over many

years.

■ Wondrous Mechanism #1—Economies of Scale. Almost every business enjoys

some degree of operational leverage from economies of scale. It certainly is no

different with Ipiranga. The company incurs costs for personnel, distribution bases,

advertising, marketing, lease of equipment, real estate, IT, and other corporate

expenses that are inelastic to volumes within a certain range. So our first hypothesis is

that economies of scale are a key driver of margin expansion for Ipiranga. In fact,

Figure 69 shows how we estimate Ipiranga's fixed costs diluted over time. It's only a

slight change of around 0.4 percentage points relative to the aggregate of cash costs,

which we think has helped to significantly boost margins.

■ Isolating Factors. In Figure 70 we illustrate the impact economies of scale alone

would have had on EBITDA per volume. In this analysis we exclude the seasonal

effects by plotting the modeled EBITDA margin (BRL/m3) indexed as 1.00 at each

quarter of 2010. The model assumes, hypothetically, zero inflation and that contribution

margins, the margins between selling price and variable costs for each product,

remained unchanged. In this period, volumes increased by 24% and the dilution of

fixed costs alone would have been sufficient to boost EBITDA margin (BRL/m3) by

c.20%. The margin expansion and volume expansion together would then explain

c.49% of EBITDA growth. However, actual EBITDA (excluding asset sales,

merchandising and other revenues) in the period grew by 160% in nominal terms, or

c.86% in real terms. This means that economies of scale alone don't explain the full

picture.

Figure 69: Breakdown of Ipiranga Costs: Fixed

Costs Diluted Over Time as a % of total cash costs

Figure 70: Economies of Scale: EBITDA/m3 Margin

Growth vs. Volume Increase Quarterly data, seasonally adjusted and indexed in 2010


95.4% 95.6% 95.5% 95.7% 96.1% 96.0%

0.8% 0.8% 0.8%0.8%

0.8% 0.8%0.6% 0.6% 0.6%0.6%

0.5% 0.5%1.2% 1.1% 1.0%

0.9% 0.8% 0.9%

2.0% 2.0% 2.1% 1.9% 1.9% 1.9%

2010 2011 2012 2013 2014 2015

COGS exposed to price and volume COGS exposed only to volume

SG&A exposed to volume COGS fixed portion

SG&A fixed portion

1.00

1.05

1.10

1.15

1.20

1.25

1.30

1.35

Q1 10 Q1 11 Q1 12 Q1 13 Q1 14 Q1 15

Volume sold

Economies of scale:

EBITDA margin (BRL/m3)

22 June 2016


■ Wondrous Mechanism #2—Prices Exposure. Our second hypothesis is that Ipiranga

has some degree of exposure to the price of the commodities it distributes. We think

this exposure affects the company in two different ways: (i) margin increases and (ii)

inventory gains.

■ Margin Increases. We hypothesize that distribution companies have ample room to

increase margins when there is a hike in the refinery-gate price of fuels. In practice, we

think this mechanism takes effect immediately after an increase in refinery-gate prices

or a tax hike (CIDE), when the distribution company raises its prices. In fact, it is only

the price increase in percentage terms that needs to be passed through along the

supply chain. Figure 71 shows a conceptual exercise to demonstrate how much this

mechanism can boost EBITDA. Based on the company's 2015 actual data, a

hypothetical 10% price hike across all fuels could lead to some 18% EBITDA growth.

Figure 71: Price Exposure: Margin Expansion Due to a Hypothetical Increase in

Refinery-Gate Price index

Source: Company data, Credit Suisse estimates. Note: 'Current' refers to 2015 data. EBITDA excluding asset sales and other revenues.

■ Isolating Factors. Conceptually, we have acknowledged that fuel distribution

companies are more exposed to commodity prices (gasoline and diesel) on the

revenue side than on the cost side. Operating expenses (COGS and SG&A) are

composed of different cost items, including but not limited to the cost of product

acquisition. Accordingly, if a price increase at the refinery gate is entirely passed

through to the supply chain, at the same magnitude in terms of percentage points, it

would be logical to conclude that its impact on revenues would be bigger than the

impact on total costs. What we do not know precisely, however, is how permanent

EBITDA*

margin (%)

Current cost

structure

After hypothetical

10% hike

Revenues

110

Revenues

100 (index)

4.0% 4.4%

EBITDA*

R$3.1bn2nd: +10%distributionprice hike

1st: +10%refinery gateprice increase

3rd:

COGS exposed

to price and volumeCOGS exposed only to volume

COGS fixed portion

SG&A exposed to volume

SG&A fixed portion

1.80.50.80.892.1

1.80.50.80.8101.3

EBITDA*

R$2.6bn

c.18% growth

22 June 2016


such margin gains are under a competitive environment. It could be a temporary gain,

in which companies benefit from the fact that the market temporarily loses its price

reference. On the other hand, if we were to assume that such gains are indeed

permanent, EBITDA margin would have expanded as shown in Figure 72. In this case,

our model assumes actual price increases and inflation, but no volume increases from

2010. As in mechanism #1, the plotted EBITDA margin (BRL/m3) data is seasonally

adjusted (each quarter is indexed to 2010). Our goal is to the growth can be attributed

solely to margin expansion due to what we called "commodity price exposure." In

2010-15, the average purchase price of the fuels Ipiranga distributes increased by

c.44%, whereas price exposure alone would have increased EBITDA margin by c.49%.

Figure 72: Price Exposure: EBITDA/m3 Margin

Growth Due to Price Increase Quarterly data, seasonally adjusted and indexed to 2010

Figure 73: Inventory Gains in BRL million

Source: Company data, Credit Suisse estimates Source: Company data, Credit Suisse estimates

■ Inventory Gains. Besides margin expansion, distribution companies' exposure to

commodity prices is also reflected in one-off inventory gains when prices increase, or

losses when prices fall. The reason behind such gains/losses is that inventories are

built at the price in effect before the price increase/reduction, which has a direct impact

on the cost of goods sold when this inventory is turned over and sold. The impact of

this non-recurring mechanism on EBITDA in 2010-15 is plotted on Figure 73 along with

announced price hikes for gasoline/diesel, including CIDE tax.

■ A Small Difference Is a Huge Difference. It is interesting to note that these two

effects are based on a simple assumption that revenues are somewhat more exposed

to changes in prices and volumes than costs. We think it is important to highlight how

small this difference can be (some 3-4 percentage points) and still produce such an

amazing boost to earnings. To illustrate this point, the chart on Figure 74 shows the

combined margin gains from mechanism #1, economies of scale, mechanism #2, and

price exposure. When put together, these two effects become even more powerful due

to the compound effect between the them. Once again the data presented is

seasonally adjusted and indexed to 2010. According to this model, EBITDA margin

(BRL/m3) would have almost doubled in the period. With the combined effect of margin

expansion and volume growth, we would have seen EBITDA grow by some 140% in

the 2010-2015 period, or c.19% CAGR. The output of our conceptual model at the

EBITDA level compared to the actual reported EBITDA is shown on Figure 75 (this

time data is not indexed and seasonality is preserved).

0.8

0.9

1.0

1.1

1.2

1.3

1.4

1.5

1.6

Q1 10 Q1 11 Q1 12 Q1 13 Q1 14 Q1 15

Average price

Price exposure:


6%

11%

8%

6%

3%

4%

7%

4%

3%

7%

6%

0

10

20

30

40

50

60

70

80

Q1 11 Q4 11 Q3 12 Q2 13 Q1 14 Q4 14 Q3 15

Inventory gains Diesel hike (%) Gasoline hike (%)

22 June 2016


Figure 74: Build-up of Backtesting EBITDA Margin

Assuming Wondrous Mechanisms #1 & #2 vs.

Actual EBITDA Margin Quarterly data, seasonally adjusted and indexed to 2010

Figure 75: EBITDA Backtesting Assuming

Wondrous Mechanisms #1 & #2 vs. Actual EBITDA Quarterly data, in BRL mn

Source: Company data, Credit Suisse analysis Note: Model EBITDA assumes the combination of the two described mechanisms applied with actual prices and volumes.

Source: Company data, Credit Suisse analysis Note: Model EBITDA assumes the combination of the two described mechanisms applied with actual prices and volumes.

■ Would an Alternative Model Explain It Better? We were quite impressed by how

accurate economies of scale and price exposure mechanisms were in predicting

margin expansion. However, we agree with the words of Michael Mauboussin: "If you

find a good idea, look for a better one—that's my motto. Good thinking is a matter of

making comparisons." So, we hypothesize a different mechanism behind margin

expansion to replace that of commodity exposure. The reason why we seek to replace

the previous mechanism is that margin gains from price increases may not be

permanent. Conceptually, we have conceived that distribution companies have better

bargaining power, hence higher margins, when demand for fuels is high. However,

changes in gasoline and diesel prices might not be directly linked to changes in

demand for such products. As the astrophysicist Lawrence Krauss once said: "I have

wonderful ideas, but the Universe was not smart enough to use them."

■ Wondrous Mechanism #3—Inflation Pass-Through. We make our third hypothesis

in the attempt to find a model that will explain margin expansion in our backtesting

experiment. This time we assume that Ipiranga will pass through small price increases

to compensate for cost inflation and maintain contribution margin stable in real terms.

Essentially, under this mechanism, we have to believe that competition in the fuel

distribution market allows for the companies to maintain profitability levels in real terms

in spite of inflation. This assumption seems very defensible to us, in light of the very

small adjustments needed for it to be true. In fact, because contribution margins are

narrow, a 0.6% price increase (less than BRL0.02/l) would be sufficient to pass-through

inflation of 7.5% to contribution margin, as shown on Figure 76. This level of price

increase is small enough not to hurt the competitiveness of service stations, while the

implementation of the price hike is facilitated by the contractual minimum volume

commitment that resellers have to distribution companies. The ability to pass-through

such price increases allows Ipiranga to monetize all its investment in differentiation and

building customer loyalty. Figure 77 shows how EBITDA margin (BRL/m3) in the period

would have been boosted by c.40% in nominal terms as a consequence of inflation

pass-through. As usual, data is presented on a quarterly basis, seasonally adjusted,

and indexed to 2010.

1.0

1.2

1.4

1.6

1.8

2.0

2.2

2.4

Q1 10 Q4 10 Q3 11 Q2 12 Q1 13 Q4 13 Q3 14 Q2 15

Economies of scale

Price exposure

Compound

Inventory gains

Model EBITDA

margin

Actual EBITDA

margin

200

300

400

500

600

700

800

900

Q1 10Q4 10Q3 11Q2 12Q1 13Q4 13Q3 14Q2 15

Actual EBITDA

Model EBITDA

22 June 2016


Figure 76: Effect of Inflation Pass-Through on

Contribution Margin: Small Price Increase Is

Required in BRL/m3, unless otherwise stated

Figure 77: Inflation Pass-Through: EBITDA/cu m

Margin Growth vs. Cumulative Inflation Quarterly data, seasonally adjusted and indexed to 2010

Source: Company data, Credit Suisse estimates. Note: 'Current' assumes 2015 data. Source: Company data, Credit Suisse estimates

■ Bull's Eye. Mechanism 3 consists of small price increases to offset inflation at the

variable cost level. This is important because it allows such gains to compound due to

economies of scale as volumes increase. We then find out that the combination of

mechanism 1, 3 and the non-recurring gains of mechanism 2 fit very accurately into our

backtesting experiment. Once again, the combination of the two mechanisms is more

powerful than each individually, as they compound. These simple and incremental

mechanisms, when put together, could track quite closely the historical EBITDA margin

(BRL/cu m) in 2010-15, as shown in Figure 78. Data is seasonally adjusted and

indexed to 2010. The combination of such margin expansion and actual sales volumes

results in the EBITDA curve shown in Figure 79.

Figure 78: Build-Up of Backtesting of EBITDA

Margin Assuming Wondrous #1 & #3 vs. Actual

EBITDA Margin Quarterly data, seasonally adjusted and indexed to 2010

Figure 79: Backtesting of EBITDA Assuming

Wondrous Mechanism #1 & #3 vs. Actual EBITDA Quarterly data, in BRL mn

Source: Company data, Credit Suisse analysis Note: Model EBITDA assumes the combination of the two described mechanisms adjusted for actual inflation and volumes.

Source: Company data, Credit Suisse analysis Note: Model EBITDA assumes the combination of the two described mechanisms adjusted for actual inflation and volumes.

2,340 2,340

32 34

169 181

2,540 2,555

Current After 7.5% inflation

PricePrice

+7.5%Contribution

margin

Raw materials

(no direct exposure

to inflation)

Variable costs

exposed to inflation

+0.6%

1.0

1.1

1.1

1.2

1.2

1.3

1.3

1.4

1.4

1.5

Q1 10 Q4 10 Q3 11 Q2 12 Q1 13 Q4 13 Q3 14 Q2 15

Accumulated inflation

Economies of scale:


1.0

1.2

1.4

1.6

1.8

2.0

2.2

2.4

Q1 10 Q4 10 Q3 11 Q2 12 Q1 13 Q4 13 Q3 14 Q2 15

Economies of scaleInflationCompoundInventory gains

Actual EBITDA

margin

Model EBITDA

margin

200

300

400

500

600

700

800

900

Q1 10Q4 10Q3 11Q2 12Q1 13Q4 13Q3 14Q2 15

Model EBITDA

Actual EBITDA

22 June 2016


■ Mechanism #4—Imports. We can't help but notice that our previous model closely

tracked the actual EBITDA margin, with the exception of two points in time (1Q15 and

4Q15), as shown in Figure 80. Probably not occasionally, these two specific quarters

are the ones in which Ipiranga posted temporary gains from imports, as domestic fuels

were priced above import parity. This observation then leads us to conceive of one last

mechanism for our model to account for import gains. We will delve into the imports

issue in much greater detail in Question #5 of this report. It is interesting to note,

however, that the final result of the combination of (i) economies of scale, (ii) temporary

gains from price exposure, (iii) inflation pass-through to contribution margin, and

(iv) imports has accurately predicted EBITDA in our backtesting experiment, Figure 81.

Figure 80: Backtested vs. Actual EBITDA Margin:

Model Falls Short in 1Q15 and 4Q15 due to Imports Quarterly data, seasonally adjusted and indexed to 2010

Figure 81: Backtested vs. Actual EBITDA: Model

Tracks Very Closely Actual Data after Adjusting for

Imports in BRL mn, unless otherwise stated


■ There Is Music in Ipiranga. Our purpose in discussing and testing so many ways and

models to explain Ipiranga's margin expansion derives from our ultimate desire to

understand whether the company will continue to add value as it has in the past. And,

margin expansion has been a key driver behind Ipiranga's profitability growth. We have

conceived of a set of mechanisms that we think can explain past expansion, and we

acknowledge that, although the final solution is simple, the process to get there was

not. It reminds us of the words of Frederick Maitland: "Simplicity is the end result of

long, hard work, not the starting point." The nature of these mechanisms is such that it

gives us confidence that Ipiranga will likely continue to grow as it has in the past. We

are constructive on the growing demand of this market, as we discussed in Question

#2, especially after the deal with AleSat closes, which will allow Ipiranga to benefit from

additional economies of scale. Furthermore, we believe the company enjoys a sound

competitive advantage in its strategy to develop the loyalty in its client base that allows

for inflation pass-through to contribution margins. In addition to these perennial effects,

the company will likely continue to benefit from the optionality of extraordinary gains

through price increases and imports, whenever these opportunities are available.

1.0

1.2

1.4

1.6

1.8

2.0

2.2

2.4

Q1 10 Q4 10 Q3 11 Q2 12 Q1 13 Q4 13 Q3 14 Q2 15

Economies of scaleInflationCompoundInventory gains

Actual EBITDA

margin

Model EBITDA

margin

before imports

Imports

200

300

400

500

600

700

800

900

Q1 10 Q4 10 Q3 11 Q2 12 Q1 13 Q4 13 Q3 14 Q2 15

Model EBITDA

+ impots

Actual EBITDA

22 June 2016


Question #4: How Promising Is the CONEN Area?

"Greatness and near-sightedness are incompatible." (Daniel Pink)

■ Expanding Towards the CONEN. Ipiranga's sales are still highly concentrated in the

South/Southeast (SSE) regions in comparison to the overall market, mostly for

historical reasons. Total sales in the North, Northeast, and Central West regions of

Brazil, the so-called CONEN area, are lower than in the SSE, but on average the

market is less concentrated than Ipiranga is. Figure 82 and Figure 83 show that the

CONEN represents only 28% of Ipiranga's diesel sales and 23% of its gasoline

volumes. This region accounts for 41% of diesel and 33% of gasoline sales in the

overall market. This profile also applies to the Sindicom average. This concentration,

though, hides an important factor: Ipiranga has strongly focused on expanding to the

CONEN area, in recent years. The recent move to acquire AleSat is one more step

towards faster expansion in the CONEN region. As the chart shows, AleSat is one of

the companies with the lowest concentration in the SSE (only behind BR). AleSat's

diesel sales are 46% located in the CONEN area. This position makes the acquisition

by Ipiranga not only smart in terms of volumes and scale gains, but also in driving the

strategy of expansion to the CONEN. According to Ipiranga, the company was

particularly interested in AleSat's good position in the Northeast region.

Figure 82: CONEN vs. SSE Share of Diesel Sales by

Company in %

Figure 83: CONEN vs. SSE Share of Gasoline Sales

by Company in %

Source: Brazilian Oil, Gas, and Biofuels Agency (ANP), Brazilian Energy Planning Corporation (EPE), Brazilian Association of Fuel and Lubricant Distributors (Sindicom), Credit Suisse analysis


■ Looking Ahead. We regard Ipiranga's strategy to expand its network of branded

stations mainly throughout the CONEN as very forward-looking. In Figure 84 and

Figure 85, it is possible to see that Ipiranga has been adding stations to its branded

network in both the CONEN area and the South/Southeast (SSE) regions. In the

CONEN area, however, its expansion strategy becomes much more pronounced in

these charts, as the company managed to increase its share in the number of stations

by roughly 1pp every year since 2009, more than doubling its original share of stations.

Unbranded stations are gradually joining the networks of the four main brands over the

years, but no other company seems to have been as aggressive as Ipiranga in

59% 61%72%

64%54% 54% 52%

41% 39%28%

36%46% 46% 48%

Market Sindicom Ipiranga Raízen BR Ale White

Flag

S/SE CONEN

67% 69%77% 75%

59%66% 62%

33% 31%23% 25%

41%34% 38%

Market Sindicom Ipiranga Raízen BR Ale White

FlagS/SE CONEN

22 June 2016


expanding in the CONEN area, as these charts show. The charts also show that, in

spite of the branding of stations in the CONEN area over the years, a high percentage

of them are still unbranded (55%), as opposed to the SSE, where 36% are unbranded.

This may mean, in our view, an opportunity to find a considerable number of yet

unbranded fueling stations, with good locations and the potential to increase volumes,

once branded, besides the potential to rebrand AleSat's network, once the deal is

approved.

Figure 84: Share of Unbranded Service Stations and

Sindicom Companies in the CONEN Area in %

Figure 85: Share of Unbranded Service Stations and

Sindicom Companies in the SSE Area in %



■ Advancing Toward the Long Term. Sales per station is not an objective per se.

When Ipiranga brands a new station, that doesn't mean it is targeting an increase in

volumes per station. It is actually aiming to deploy capital that can deliver good returns

and that can be obtained either through stations with already high volumes sold or

through stations with lower volumes where Ipiranga sees a potential for improvement.

We like to know that returns have been the main driver of these decisions, rather than

volume growth per se. Even so, for the reasons mentioned before, we think scale

matters. In the charts in Figure 86 and Figure 87, we can see that, overall, the sales

per station don't differ much between the two regions. But one thing we noticed was

that, as a consequence of the aggressive branding strategy adopted by Ipiranga in the

CONEN area, the company's volumes per station in that region have not followed the

same trend as that of other Sindicom member companies. We can also observe that

the rebranding of AleSat's stations will lead to additional drops in volumes per station in

the short term.

■ AleSat and Branding Are Plausible Ways to Play CONEN. We think both the

acquisition of AleSat and the branding of unbranded stations in the CONEN regions

are perfectly plausible ways to increase exposure to the region. The immediate drop in

volumes, as stations with lower volumes on average are added to the company's

network, seems to us to be an investment in the future prospects of the region. It

doesn't worry us at all. Much to the contrary, we think Ipiranga is pursuing these

additions with returns above cost of capital and, at the same time, has doubled its base

of contracted stations. The benefits of the company's franchise will materialize in the

newly branded stations and result in increased share in the region.

4% 4% 4% 4% 4% 5%5% 6% 8% 9% 10% 11%

22% 21% 20% 20% 20% 20%

8% 8% 8% 8% 8% 9%

61% 61% 61% 59% 57% 55%

2009 2010 2011 2012 2013 2014


4% 4% 5% 5% 5% 5%

20% 20% 20% 21% 21% 22%

18% 18% 19% 19% 19% 20%

14% 14% 14% 14% 15% 16%

45% 44% 43% 41% 39% 36%

2009 2010 2011 2012 2013 2014


22 June 2016


Figure 86: Sales Per Station (Diesel + Otto) In The

CONEN Area in bpd per station

Figure 87: Sales Per Station (Diesel + Otto) In The

SSE Area in bpd per station



■ A Play on Prosperity. If we believe that there remains a lot of potential for

consumption growth in Brazil, this is even truer for the CONEN area, which is the

fastest-growing region in the country. Actually, as we are about to demonstrate, this

more accelerated growth has already been the case for the last decade or so and, in

our view, is still likely to be the case for many years to come as the country resumes

growth and living standards improve. Figure 88 shows how the CONEN area has

outpaced the South/Southeast (SSE) regions in terms of economic growth. Figure 89

shows that the consumption of motor fuels (diesel and Otto cycle) increased in the

CONEN at a compounded average annual rate of 7.7% from 2005 to 2014. In the

same period, consumption in the SSE area grew by c.5.5%p.a. And this, despite the

global financial crises in 2008-09.

Figure 88: Regional GDP Growth Rates in CONEN

vs. SSE (2000-2013) in %

Figure 89: Fuel Consumption Growth Rates in

CONEN vs. SSE index (2005 = 1)

Source: Central Bank of Brazil, Credit Suisse analysis Source: ANP, Brazilian Statistics Bureau (IBGE), Credit Suisse analysis

0

10

20

30

40

50

60

2009 2010 2011 2012 2013 2014

Ipiranga

Other Sindicom

White Flag

AleSat

0

10

20

30

40

50

60

2009 2010 2011 2012 2013 2014

Ipiranga

Other Sindicom

White Flag

AleSat

49%

79%

3.1%

4.6%

South/Southeast CONEN

Total Growth (2000-2013)

CAGR

Population

SSE

Population

CONEN

0.80

1.00

1.20

1.40

1.60

1.80

2.00

2005 2006 2007 2008 2009 2010 2011 2012 2013 2014

Consumption

CONEN

Consumption

SSE

22 June 2016


■ Per Capita Consumption Potential. We've already shown in previous sections how

Brazil's per capita fuel consumption is still low relative to that observed in developed

economies, and this means a significant potential for demand expansion in the

decades to come. If we split this trend between the CONEN and the SSE areas, it is

possible to observe that the average consumption per capita is still considerably lower

in the former than in the latter. Figure 90 shows the levels of motor fuel consumption

per thousand people in each area over a decade. Figure 91 shows the total and

average yearly growth in per capita consumption in the CONEN (6.3% p.a.) and in the

SSE (4.5% p.a.). We expect this trend to continue in the coming years, with the

expansion of income and economic activity.

Figure 90: Per Capita Diesel and Otto Liquid Fuel

Consumption, CONEN vs. SSE in cu m/ thousand people / year

Figure 91: Growth in Per Capita Consumption

CONEN vs. SSE (2005-2014) in %

Source: Brazilian Oil, Gas, and Biofuels Agency (ANP), Brazilian Statistics Bureau (IBGE), Credit Suisse analysis

Source: Brazilian Oil, Gas, and Biofuels Agency (ANP), Brazilian Statistics Bureau (IBGE), Credit Suisse analysis

■ Income Is Coming. Putting aside the current economic downturn, Brazilians have

seen their income expand in recent years, and this movement was more pronounced in

the CONEN area. Figure 92 shows the evolution of income in the SSE area between

2004 and 2014. On the horizontal axis, we depict the percentage of the population that

earned each level of income. The vertical axis expresses the levels of income in BRL

terms. Each level is defined as a multiple of the minimum monthly wage per person. As

the chart shows, there was a shift in the overall level of income in the SSE, as a

consequence of a real increase in the minimum wage. On the other hand, in Figure 93,

we can see what happened to the CONEN area. There was a clear shift in the salary

curve both upwards and to the left. So, besides the real increases in the minimum

wage, also verified in the SSE, the CONEN region benefited from an improvement in

the multiple of the minimum monthly wage each person was earning.

-

100

200

300

400

500

600

700

2005 2006 2007 2008 2009 2010 2011 2012 2013 2014

CONEN S/SE Brazil

49%

73%

4.5%

6.3%



CAGR

22 June 2016


Figure 92: Historical Real Wage in BRL and

Multiples of Minimum Wage 2004–2014 in SSE Area in BRL / month / person and in % of population

Figure 93: Historical Real Wage in BRL and

Multiples of Minimum Wage 2004–2014 in the

CONEN Area in BRL / month / person and in % of population

Source: Brazilian Statistics Bureau (IBGE), Credit Suisse analysis Source: Brazilian Statistics Bureau (IBGE), Credit Suisse analysis

■ Moving in the Right Direction, but Still Far to Go. In Figure 94, we show the

CONEN income curves moving over time and how these curves relate to the more

recent income situation in the SSE area. Two things become clear: (1) the CONEN is

moving towards increasing the multiple of minimum wages per person and is benefiting

from a nationwide increase in the minimum wage. (2) There is still a far way to go. For

the CONEN income profile to get closer to that of the SSE, a major shift to the left of

the curve is still necessary. If this happens, it will mean a huge boost to total

disposable income, which will inevitably lead to a bigger fleet of cars, higher gasoline

consumption, more diesel required to move products, etc. We're not making a call that

the CONEN area will equal the SSE area in terms of income and economic activity.

But, we do see a lot of room for the CONEN curve to keep moving left as well as up. In

Figure 95, we show how the average income (measured as the area below the curves

in BRL) increased faster in the CONEN (68%) than in the SSE (49%) area. We also

show how the average multiple of the monthly minimum wage (measured as the area

below the curves in multiples of the minimum wage) increased by almost 10% in the

CONEN, while it actually slightly decreased in the SSE (meaning that the SSE curve

has, on average, moved slightly to the right).

-

500

1,000

1,500

2,000

2,500

3,000

3,500

4,000

0 10 20 30 40 50 60 70 80 90 100

Income 2014 Income 2004

1 minimum wage

2 mw

1/2 mw

3 mw

1/2 mw1 mw

2 mw

3 mw

5 mw

5 mw

1/4


Ranges

of

sala

ry (

BR

L/

month

)

0

500

1000

1500

2000

2500

3000

3500

4000

0 10 20 30 40 50 60 70 80 90 100

Income 2014 Income 2004

1 minimum wage

1 mw

1/2 mw

2 mw

2 mw

3 mw

3

1/4 mw

1/2 mw

5 mw

5

Ran

ges

of

sala

ry (

BR

L/

month

)


22 June 2016


Figure 94: Historical Real Evolution In CONEN From

2004 To 2014 Compared with Current Wages In SSE in BRL / month / person and in % of population

Figure 95: Average Real Wage Increase between

2004 and 2014, CONEN vs. SSE in %

Source: Brazilian Statistics Bureau (IBGE), Credit Suisse analysis Source: Brazilian Statistics Bureau (IBGE), Credit Suisse analysis

■ Nine Cars Per 100 People. This lower, but moving income in the CONEN is reflected

directly in the number of cars in use in the region. We like to look at the "motorization

rate" as a basic metric of the expansion potential for demand in the region. We've

already showed how these rates are still low in Brazil relative to developed countries.

But, this is particularly true in the CONEN area, where there are still only c.9 passenger

cars for each 100 people. This is only 40% of the motorization rate observed in the

SSE area. Figure 96 shows the evolution of this indicator over the years for the

CONEN and the SSE areas, as well as the Brazilian average. We think the increasing

living standards are an additional lever, which will mean that consumption growth will

not be limited to population increase. Figure 97 shows the growth in the motorization

rates in both areas, evidencing this point.

Figure 96: Motorization Rate, CONEN vs. SSE in number of passenger cars / 1,000 people

Figure 97: Growth in Motorization Rates CONEN vs.

SSE in %

Source: Brazilian Association of Motor Vehicle Manufacturers (Anfavea), Brazilian Federation of Motor Vehicle Dealers (Fenabrave), Brazilian Statistics Bureau (IBGE), Credit Suisse

Source: Brazilian Association of Motor Vehicle Manufacturers (Anfavea), Brazilian Federation of Motor Vehicle Dealers (Fenabrave), Brazilian Statistics Bureau (IBGE), Credit Suisse

-

500

1,000

1,500

2,000

2,500

3,000

3,500

4,000

0 10 20 30 40 50 60 70 80 90 100

Income SSE 2014

Income CONEN 2014

Income CONEN 2004

1 mw

1 mw

1/2 mw

2 mw

2 mw

3 mw

3

1/4 mw1/2 mw

5

1 minimum wage

2 mw

3 mw

5

1/2 mw

5


Ran

ges

of

sala

ry (

BR

L/

month

)

49%

-0.5%

68%

9.8%

Growth in Average Income Variation in Average # of Salaries(minimum wages)

SSE

CONEN

-

50

100

150

200

250

2005 2006 2007 2008 2009 2010 2011 2012 2013 2014

CONEN S/SE Brazil

57%

89%

5.1%

7.3%



CAGR

22 June 2016


■ Every Day, 1,500 More Cars. Every Day, Each Car Consumes More. The observed

fleet expansion has, for obvious reasons, boosted consumption. However, apparently,

over the years, there has been a slight increase in the consumption per car. Figure 98

shows the fleet growth and the Otto cycle fuel consumption growth, comparing the two

regions. Figure 99 shows the average consumption per vehicle over the years. We

think this presents additional leverage for future demand growth in CONEN. Higher

income growth may promote higher fleet growth, boosted by a higher consumption per

vehicle.

Figure 98: Automobiles Fleet Growth and Otto Fuels

Consumption Growth - CONEN vs. SSE (2005-2015) CAGR %

Figure 99: Historical Otto Fuel Consumption per

Vehicle, CONEN vs. SSE (2005-2015) in m³ / vehicle / year

Source: Brazilian Association of Motor Vehicle Manufacturers (Anfavea), Brazilian Federation of Motor Vehicle Dealers (Fenabrave), ANP, Credit Suisse analysis


■ Ever More Trucks. Never Less Diesel. The truck fleet also expanded in the CONEN

in recent years much more rapidly than in the SSE. As we've already demonstrated,

Brazil's economic activity is to a great extent fueled by diesel. Faster GDP growth in

the CONEN area was followed by a significant increase in the truck fleet and also by

an increase in diesel consumption. Figure 100 shows how these rates compare to

those seen in the SSE area in the past ten years. Figure 101 shows the also important

fact that diesel consumption per truck over the years was quite stable. So, future

economic expansion will inevitably lead to more trucks moving products around the

region, and Ipiranga will very likely benefit from that.

Figure 100: Growth in Truck Fleet and in Diesel Fuel

Consumption, CONEN vs. SSE (2005-2015) CAGR %

Figure 101: Historical Diesel Consumption per

Truck, CONEN vs. SSE (2005-2015) in cu m / truck / year



9.2%

6.3%

8.6%

5.9%

CONEN SSE

Otto Consumption

Automobiles Fleet Growth

1.10

1.20

1.30

1.40

1.50

2005 2007 2009 2011 2013 2015

Ott

o F

uels

Consu

mptio

n

/ V

ehic

le (

m³/

year

)

Otto Consumption/Car

in Brazil

CAGR = +0.6%

5.0%

3.2%

5.3%

3.5%

CONEN SSE

Diesel Consumption Growth

Trucks Fleet Growth

20.0

22.0

24.0

26.0

28.0

2005 2007 2009 2011 2013 2015

Die

sel F

uels

Consu

mptio

n

/ V

ehic

le

(m³/

year)

Diesel Consumption/Truck

in Brazil

CAGR = -0.15%

22 June 2016


■ A Few More Trucks per Billion GDP. The Brazilian transportation matrix depends

heavily on trucks to move products throughout the country, be they from local industrial

activity or from agriculture, or even imports. If we measure the overall number of trucks

in use relative to the GDP in real terms, we see that this metric is closely aligned in the

two areas (CONEN and SSE) and also across time. Actually, over the last decade,

there was a slight increase in the number of trucks per billion reais of output in each

area. It is also possible to note, in Figure 102, that this metric increased slightly faster

in the CONEN area than in the SSE. We can assume this difference to be negligible,

as this is a rough ratio between two aggregate measures, and the main point in our

view is that Ipiranga is positioning itself to fuel the growing number of trucks that will

sustain future economic development.

Figure 102: Diesel Trucks Fleet Per BRL Billion GDP Over a Decade (CONEN vs.

SSE) in trucks per BRL billion GDP

Source: Brazilian Association of Motor Vehicle Manufacturers (Anfavea), Brazilian Federation of Motor Vehicle Dealers (Fenabrave), Central Bank of Brazil, Credit Suisse analysis

■ What to Expect, Then? After looking at the data, we got the impression that Ipiranga

is making a right and forward-looking move. There are still 55% of the service stations

in the CONEN area that are unbranded. And this way to play CONEN, through which

the company was able to basically double its presence in the region in recent years,

can now be complemented by an acquisition of a well-positioned player in the region

(AleSat). The fleet size and the per capita consumption standards explain the decades-

long volume growth in CONEN. We think that, through the lenses of the longer term,

income gains will continue to exist, increasing the regions above the national average

and propelling the expansion of car and truck fleets. And, Ipiranga will be there,

increasing its presence in this market, full of hard-to-quantify promising factors that will

likely be in place for years to come.

304

336

303

357

2005 2013

SSE CONEN

22 June 2016


Question #5: How Much Should We Expect From

Fuel Imports?

"You see things; and you say, ‘Why?’ But I dream things that never were; and I say, ‘Why not?’" (George Bernard Shaw)

■ Imports Are a Reality. We have enough evidence to show that imports helped boost

Ipiranga's margins during recent quarters, since the beginning of 2015, as a

consequence of the sharp drop in global oil prices with no price reduction from

Petrobras. Actually, prices were increased last September (diesel by 6% and gasoline

by 4%), leading domestic prices to wide premiums over import parity prices, and this, in

our view, constitutes a potential play for Ipiranga, at a time when it faces a temporarily

tough market environment of demand contraction combined with high inflation rates.

Distributors justifiably downplay the potential impacts of imports, as the conditions for

this to become a recurring business are limited, depending on a sustained premium

over import parity. In any case, we feel this is a question not yet answered in the

sector, which deserves some investigation. Rather than just agreeing that imports can't

make a big impact, we thought it was worth discussing how much they could add, and

under what conditions.

A Region Is a Region Is a Region. The logistics operations involved are local, and we

should first break down the supply/demand dynamics at least regionally to be able to

understand the costs involved. Therefore, we think dividing the country into major oil

corridors according to existing infrastructure fits well with the purposes of this analysis.

So, Figure 103 shows the 9 logistics corridors into which we chose to break down the

fuel balances and costs to reach those markets. We used as criteria to define these

regions the connection between the refining and logistics infrastructure, constituting a

kind of subsystem to supply the country.

Figure 103: Infrastructure Corridors for Supply/Demand Analysis

Source: Sindicom, Credit Suisse analysis

C. do SulRio

Branco

Vilhena

PortoVelho

Cuiabá

AltoTaquari

C. Grande

Itaituba

Altamira

Marabá

Manaus

CaracaraíOriximiná

Santana

Santarém

Barcarena

Açailândia

São LuisBelém

FortalezaSobral

Teresina

Crato

Juazeiro

Luiz EduardoMagalhães

Brasilia

Goiânia

Uberlândia

Pres. Prudente

Araçatuba Betim

Gov. Valadares

MontesClaros

Jequié

S.F.Conde

Madre de Deus

Aracajú

Maceió

Guamaré

Natal

Cabedelo

RecifeIpojuca (Suape)

Itabuna

Vitória

Campos

Duque de CaxiasMaringá

CascavelGuarapuava

Passo FundoIjuí

Sta. Maria

Uruguaiana

Bagé

Rio Grande

Canoas

Esteio

Cruz AltaFlorianópolisItajaí

Joinville

Araucária

Londrina

Ourinhos

Bauru

Uberaba

Rib.PretoPaulinia

Almeirim

Refinery

Primary Storage Facility

Secondary Storage Facility

Railway

Waterway

Road

Pipeline

Cabotage

•São Paulo

•Barueri

•Guarulhos

•São Caetano do Sul

•São José dos Campos

•Cubatão

S.J.R.Preto

Sul Paraná

Bahia

RJ/MGSP/CO

Suape

Ceará São Luís

Manaus

CorridorsInfrastructure

22 June 2016


■ How Much Demand Is out There for Imports? Local fuel supplies come basically

from Petrobras's refineries, with some additions from small capacity operations such as

RPR (in which Ultrapar has a 33.1% stake) and from other petrochemical processes

that have gasoline streams as by-products. For the purposes of this analysis, we had

to break down historical demand data and historical local production information into

the nine corridors. This analysis is relevant because there are regions in which local

production exceeds demand, and they are not very good candidates for any player to

try supplying from abroad. Figure 104 and Figure 105 show this analysis for the South

corridor for both diesel and gasoline, as an example. The region has been

oversupplied for a while and is very unlikely to become a preferred location for import

plays.

Figure 104: South Corridor Diesel Supply/Demand

Analysis in kbpd

Figure 105: South Corridor Gasoline

Supply/Demand Analysis in kbpd

Source: Brazilian Energy Planning Corporation (EPE), Brazilian Oil, Gas, and Biofuels Agency (ANP), Credit Suisse analysis


■ Where Are the Deficits, Then? In the following pictures we show the demand

balances for diesel (Figure 106) and gasoline (Figure 107), for all of the nine corridors.

They show that the biggest deficits in diesel are in the São Luís, RJ/MG, and Paraná

corridors. For gasoline, the two main deficits are the ones we called São Luís and

Suape. Also, the Paraná, RJ/MG, and Ceará corridors present deficits in their local

supplies.

-

10

20

30

40

50

60

70

80

90

100

2003 2005 2007 2009 2011 2013 2015

Balance Production Demand

(10)

-

10

20

30

40

50

60

70

2003 2005 2007 2009 2011 2013 2015

Balance Production Demand

22 June 2016


Figure 106: Diesel Supply/Demand Balances by Corridor from 2005 to 2015 in kbpd


Figure 107: Gasoline Supply/Demand Balances by Corridor from 2005 to 2015 in kbpd


Production Demand Superavit Deficit

São Luís

Ceará

Bahia

SP/CO

Paraná

MG/RJ

Sul

Suape

-5

-12 -11

2005 2010 2015

14

19

13 13

25 24

2005 2010 2015

5470

91

-54-70

-91

32 11

2005 2010 2015

283

315323

380

-11

334391 Manaus

2005 2010 2015

-8

-36 -32

88

110

7597

96

128

2005 2010 2015

8

30 3144

536052

8292

-15 -28 -41

2005 2010 2015

107118 118

122 146 159

2005 2010 2015

3747

5543

4538

8092 92

-31 -37

-16

2005 2010 2015

6 10

423747

58

2005 2010 2015

-10-14

-18

0 1 1

10

1518

Production Demand Superavit Deficit

São Luís

Ceará

Suape

Bahia

SP/CO

Paraná

Manaus

MG/RJ

Sul

34 4023

2005 2010 2015

130151

191164191

214

4

2005 2010 2015

-5-15

40

44

4452

48

67

2005 2010 2015

-1

-4-5

9

11

45

65

2005 2010 2015

-14-25

-40

1425

40

2005 2010 2015

-6-11

-17

711

17

2010 20152005

-19-28

-35

1030

45

110

0

34 30 33

2005 2010 2015

47 47

60

1317

27

75

5369

89

4

-14

11

2005 2010 2015

6473

4 5

14

2005 2010 2015

25

34

45

29

39

59

22 June 2016


■ Interregional Supplies May Close Some Gaps. After mapping all the regions running

at a deficit, we should remind investors that interregional (or inter-corridor) supplies will

also take place, changing the landscape for import opportunities. It is an economically

logical approach to consider that, if there is an oversupply in one region, this amount of

fuel will be reasonably allocated to close the gap in a neighboring region. Otherwise,

we would have to assume that Petrobras would have to export its regional

oversupplies, while bringing in, along with other players, products from abroad to fill the

gaps in other regions. In Figure 108 and Figure 109 we show how these neighboring

supplies would be able to fill the gaps in some corridors, but not in others, leaving the

North and Northeast with what we could call a 'structural deficit'.

Figure 108: Remaining Unmet Deficit in Diesel In

2016e after Interregional Supply Movements in kbpd

Figure 109: Remaining Unmet Deficit in Gasoline In

2016e after Interregional Supply Movements in kbpd



■ Structural Deficits versus Real Operations. Although it is possible to approach the

potential deficits through supply/demand analysis, the real world operations will always

involve many day-by-day intricacies that may cause interregional supplies to occur in

different ways. Besides, distributors and other players may import fuels to terminals

where they find more widely available infrastructure and where they can manage to

access faster the intended markets. So in spite of the structural deficits (after

accounting for inter-corridor supplies) being concentrated in the north and northeast of

the country, actual import operations made by players other than Petrobras are

typically aimed at south and southeast terminals, as shown in the charts in Figure 110

and Figure 111.

(84)

(9)

(100)

(80)

(60)

(40)

(20)

-

20

40

60

Sul

Para

ná

SP

/CO

RJ/M

G

Bahia

Suape

Ceará

São L

uís

Manaus

Surplus Sent to Other Regions

Deficit Supplied by Other Regions

Unaddressed Deficit

(35)

(4)

(50)

(40)

(30)

(20)

(10)

-

10

20

30

40

50

Sul

Par

aná

SP

/CO

RJ/

MG

Bah

ia

Suap

e

Cear

á

São

Luís

Man

aus

Deficit Supplied by Other Regions

Surplus Sent to Other Regions

Unaddressed Deficit

22 June 2016


Figure 110: Diesel Imports In 2015 By Players Other

Than Petrobras Per Destination in kbpd

Figure 111: Gasoline Imports In 2015 By Players

Other Than Petrobras Per Destination in kbpd

Source: Sindicom, Credit Suisse Source: Sindicom, Credit Suisse

■ Dislocating Local Supplies. When these volumes are imported to places which are

not the structural deficits, a portion of Petrobras's local oversupply will be shifted to

other more distant areas, or, alternatively, refineries will have to be run at lower

utilization rates. In recent quarters, utilization rates have indeed been lowered. Price

premiums have been high enough to justify imports taking a higher share of demand

(possibly even shifting some local supply). We already discussed in other reports that

this is good for Petrobras, from a cash flow perspective. The price lever is considerably

more powerful for boosting earnings than volumes are. And, for Ipiranga, this is also

good in the sense that it opens up a possibility of market-share gains and of margin

expansion during tough times for demand.

■ But What Is the Cost to Bring Products to Brazil? This is a question we hear a lot

from investors, and we think there are very few information sources in the market to

answer it. Investors, when trying to calculate import parity prices and also when trying

to estimate potential import gains and whether there is an import opportunity, always

face the problem of knowing how much the operation costs. We even suspect that

most market participants don't even consider logistics costs in their price parity

calculations. To address this issue, we had to gather cost information for many oil

terminals, pipelines, railways, ships, and road transportation in Brazil, so that we could

estimate the cost to supply different places with fuels from abroad. We could then

estimate the cost to move fuels from each corridor's entry terminals to reach each state

capital. We then weighted these costs by the consumption of each of those cities.

Figure 112 shows our estimates for the total freight cost (maritime + inland) for all 27

capitals of Brazil. Figure 113 consolidates both the maritime freight costs to reach each

corridor's terminals departing from the US Gulf coast (USGC) and the estimated

average inland freight costs to distribute the fuels throughout each corridor.

■ Maritime vs. Inland Trade-Off. In analyzing the chart in Figure 113, investors can see

that freight costs, be they maritime or inland, can vary a lot depending on which

corridor we consider. The maritime costs to reach each of the corridors, departing from

the USGC, can vary from c.USD3.2 to c.USD4.7/bbl. Clearly, the North and Northeast

15.1

5.6

1.6 1.2

0.7 0.1 0.1 24.4

0

5

10

15

20

25

30

SP/CO Paraná Suape RJ/MG Bahia Sul São Luis Total

2.7

1.5

1.7 0.0 6.0

0

1

2

3

4

5

6

7

SP/CO Paraná Suape Bahia Total

22 June 2016


corridors are the ones incurring the lowest maritime costs, due to their proximity to the

USGC. In our view, the inland freight costs can also vary a lot, ranging from c.USD0.5

to as much as c.USD7/bbl. Corridors in which demand spreads to the interior of the

country, farther from the terminals and with limited infrastructure, end up costing more

to supply than other corridors in which demand is highly concentrated close to the

coast. In that sense, we think the numbers expressed in the referred chart make sense;

the corridors in the South present high maritime costs, but low inland costs, while the

Southeast corridors face somewhat lower maritime costs, but enormous inland costs,

as they extend to the interior of the Central West region, and also to the interior of

Minas Gerais, two high consuming areas running supply deficits. As most of

consumption is concentrated in the Southeast corridors, the Brazilian weighted

average freight cost gets close to the figures in these corridors. We see the average

maritime cost currently at about USD3.5/bbl, while the inland average freight cost

would be about USD4.2/bbl. This combination creates a wide price premium of almost

USD8/bbl that the Brazilian market should be consistently running at.

Figure 112: Total Supply Costs (Freight + Inland) for

Each State Capital in Brazil in USD/bbl

Figure 113: Maritime and Inland Freight Costs for

Each of the Nine Corridors in Brazil in USD/bbl

Source: Antaq, Transpetro, Brasil Energia, Codeba, Decope-NTC, Credit Suisse analysis

Note: US Gulf Coast was adopted as a reference of surplus market with adequate liquidity

Source: Antaq, Transpetro, Brasil Energia, Codeba, Decope-NTC, Credit Suisse analysis

■ Relative Prices Are What Matters. Although freight costs vary depending on location,

fuel prices to the refiners also vary. Assuming the principle of opportunity costs, relative

prices should also respect the different combinations of maritime and inland costs

involved in supplying each region. Therefore, although some areas present lower costs

for an importer to supply (e.g., the Northeast), the prices in that region should also be

lower, reflecting this lower opportunity cost. So, it seems reasonable to assume that

import gains should not vary materially and perennially between regions, even in

periods when overall domestic prices are misaligned with international prices. We think

it is reasonable to assume that there should be no sustainable arbitrage between

regions, although there is currently a significant arbitrage in the overall level of

domestic prices relative to international prices.

■ Octane Specification Should Also Be Accounted For. The quality of gasoline sold

in Brazil to end consumers is very similar to what is called regular gasoline in the

-

2.0

4.0

6.0

8.0

10.0

12.0

14.0

16.0

18.0

0 5

10

15

20

25

30

Marit

ime +

Inla

nd F

reig

ht C

ost

($/b

bl)

Brazilian States Capitals Ordered According to Supply Cost

Share of Brazilian

Demand

-

1.0

2.0

3.0

4.0

5.0

6.0

7.0

8.0

3.0 3.5 4.0 4.5 5.0

Inla

nd F

reig

ht C

ost

s ($

/bbl)

Maritime Freight Costs ($/bbl)

North and

Northeast

Region

Corridors

Southeast/Midwest

Corridors

Brazil Weighted

Average

Share of Brazilian Demand

South Region

Corridors More Coastal

More to Interior

Farther from USGCloser to USG

22 June 2016


United States market. However, there is a fundamental difference. Gasoline in Brazil is

a blend with a mandatory 27% ethanol content. As the octane levels in ethanol are high

(actually much higher than the required in the final blend), the gasoline that goes into

the blend may have a lower octane specification and still make the final blend meet

regulations. We think most people give this issue less attention than the circumstances

deserve, because this lower octane requirement may mean a very significant discount

to the cost of imports, if any player buys gasoline with the correct specifications, but

without relinquishing higher quality.

■ Octane Is a Non-Linear Property of Fuels. In Figure 114 we show how the

necessary octane number (RON) combines with different ethanol contents to deliver a

certain octane result for the final blend. The chart makes it clear that the relation is not

linear. We highlighted (in yellow) the curve for the Brazilian requirement of 92 RON for

gasoline C (final blend). So, for the current 27% ethanol content, gasoline A could be

around 87 RON to meet the final specs. This means the gasoline that goes into the

blend can be as much as c.5 RON points lower than gasoline C. The potential discount

should not be neglected. Figure 115 shows the average implied market discounts in

the selling prices US refiners charge to retailers, considering a difference in RON of 5

points (equivalent to the gasoline A – gasoline C quality spread). This discount has

been accelerating in recent years. But it would be c.USD9/bbl, if a longer history is

considered. Of course, to value an operation in the coming quarters, the recent past

spreads should be more relevant. But, we think it is more conservative to use a lower

price discount, so that we don't inflate potential gains from these operations.

Furthermore, there is no certainty that the potential importer will be able to find in the

American refineries the exact desired specification and with the correct octane-

adjusted price. So, we should recommend some caution here. We are assuming

USD9.5/bbl for this quality spread.

Figure 114: Resulting Gasoline Octane as a

Function of Gasoline A Quality and Ethanol Content in RON Octane Number

Figure 115: Historical Price Discounts For Gasoline

A Assuming Current 27% Ethanol Content in

Gasoline C in USD/bbl

Source: EIA-DOE, Octane Mixture Calculator, Credit Suisse analysis Source: EIA-DOE, Credit Suisse analysis

■ Putting the Pieces Together. All in all, we see that, in spite of the recent recovery in

oil prices, as the BRL appreciated relative to the USD, import margins still exist at the

50

55

60

65

70

75

80

85

90

95

0% 10% 20% 30% 40% 50% 60% 70% 80%

Gas

olin

e A

req

uir

ed R

ON

to

mee

t sp

ecs

Ethanol content in Gasoline C

RON 84

RON 88

RON 92

RON 9627%

87

9.0 9.9

11.3

14.3

Jan/83-Jan/16

Jan/90-Jan/16

Jan/00-Jan/16

Jan/10-Jan/16

Longer History Recent Past

22 June 2016


current fuel prices, after accounting for the freight costs and quality discounts involved.

Figure 116 and Figure 117 show the import margins at current conditions (assuming

Brent @ USD50/bbl, Crack Spreads @ USD12/bbl, FX @ BRL 3.5/USD). Of course,

these margins being positive numbers, they theoretically mean there is a profit to be

made by bringing fuels to Brazil. On the other hand, these margins have to be wide

enough not only to cover the costs involved, but also to account for the risks to be

pursued by anyone willing to fill a tanker with diesel somewhere across the globe and

bring it to a Brazilian terminal, and still rely on existing infrastructure. Whether a certain

margin is enough to be worth the effort is a question each company will have a

different answer for. But we think the current levels of import margins should be

enticing for any distributor, and we are about to show why.

Figure 116: Import Margins for Diesel Assuming a

USD50/bbl Brent Price and USD12/bbl Crack

Spreads in USD/bbl

Figure 117: Import Margins for Gasoline Assuming a

USD50/bbl Brent Price and USD12/bbl Crack

Spreads in USD/bbl

Source: Credit Suisse analysis, Brasil Energia, Codeba, Brazilian Water Transportation Agency (Antaq), Decope-NTC, Transpetro rates

Source: Credit Suisse analysis, Brasil Energia, Codeba, Brazilian Water Transportation Agency (Antaq), Decope-NTC, Transpetro rates

■ These Margins Are Much Wider Than Typical Distribution Margins. The potential

volumes involved in these import operations are considerably lower than Ipiranga's

regular distribution operations. But, the potential margins involved are substantially

wider than typical distribution margins. And, the combination of unmet volumes and

import margins leads this opportunity to a non-negligible potential size. Figure 118

shows the potential volumes identified and the import margins for both diesel and

gasoline. The combination of these values leads the total size of the opportunities to as

much as c.USD470-480mn per year, for each product. Of course, it would not be

realistic to assume that Ipiranga would be responsible for all these operations. Actually,

Petrobras itself, as owner of most of the available infrastructure would likely get the

biggest share. Figure 119 shows the EBITDA that Ipiranga could potentially generate

with these operations, for different levels of "import share" depending on the company's

appetite to make this effort. We are using as a base case a 10% share, which is about

4% of Ipiranga's sales.

50.0

12.0

4.0

3.4

79.4

9.9

Bre

nt

Spre

ad

Fre

ight

Inte

rnatio

n

Marg

in

Dis

trib

utio

n

Price

50.0

12.0 4.14.0

70.2

9.5

9.6

Bre

nt

Spre

ad

Gas

olin

e

Qual

ity

Fre

ight

Inte

rnat

ion

Mar

gin

Dis

trib

utio

nP

rice

22 June 2016


Figure 118: Total Value of Imports and Ipiranga's

Potential Gains Assuming 10% Share In 2016 In USD million

Figure 119: Total Value of Imports for Different

Shares of Operations Held by Ipiranga In USD millions

Source: Credit Suisse analysis, Brazilian Oil, Gas, and Biofuels Agency (ANP), Brazilian Energy Planning Corporation (EPE), Ministry of Development, Industry, and Foreign Trade (MDIC), Brasil Energia, Codeba, Transpetro, Decope-NTC, Brazilian Water Transportation Agency (Antaq) Note: Scenario of USD50/bbl Brent price, gasoline crack spread of USD12, diesel crack spread of USD12, FX at 3.5 BRL/USD, gasoline demand decline of 3.5%, diesel demand decline of -3.5%, Ipiranga share of 10%, refineries utilization of 86%.

Source: Credit Suisse analysis, Brazilian Oil, Gas, and Biofuels Agency (ANP), Brazilian Energy Planning Corporation (EPE), Ministry of Development, Industry, and Foreign Trade (MDIC), Brazilian Association of Fuel and Lubricant Distributors (Sindicom), Brasil Energia, Codeba, Transpetro rates, Decope-NTC, Brazilian Water Transportation Agency (Antaq) Note: Scenario of USD50/bbl Brent price, gasoline crack spread of USD12, diesel crack spread of USD12, FX at 3.5 BRL/USD, gasoline demand growth of 0%, diesel demand decline of -3.5%, Ipiranga share of 10%, refineries utilization of 86%.

■ Low Volumes. High Margins. The following chart (Figure 120) puts into perspective

the relative size of the import opportunity for Ipiranga (assuming the company takes a

10% share) relative to the overall EBITDA it generates from its regular distribution

operations. What we think is worth highlighting is that, even with the opportunity

already diminished by the partial rebound in oil prices, the potential EBITDA could still

contribute to boost annual EBITDA by c.6%. This is not negligible at a time when fuel

demand is falling as a consequence of the economic slowdown.

Figure 120: EBITDA of Regular Distribution Operations vs. Potential EBITDA from Imports Margins in USD/bbl, volumes in kbpd, EBITDA in USD million

Source: Credit Suisse analysis, Company data, Brazilian Oil, Gas, and Biofuels Agency (ANP), Brazilian Energy Planning Corporation (EPE), Ministry of Development, Industry, and Foreign Trade (MDIC), Brasil Energia, Codeba, Transpetro rates, Decope-NTC, Brazilian Water Transportation Agency (Antaq) Note: Scenario of USD50/bbl Brent price, crack spreads of USD12, FX at 3.5 BRL/USD, gasoline and diesel demand growth of -3.5%, Ipiranga share of 10%, refineries utilization of 86%.

-

1

2

3

4

5

6

7

8

9

10

0 20 40 60 80 100 120

Ipiranga's share of imports Other players' share of imports

Volume (kbd)

Marg

in (

$/b

bl)

$ 303 mn$34mn$ 125 mn$14mn

DieselGasoline

33.6

67.3

84.1

100.9

134.6

13.9

27.734.7

41.6

55.5

10% 20% 25% 30% 40%

Diesel Gasoline

Pote

ntia

l pro

fit

from

import

s (U

SD

mn)

Ipiranga's share of opportunity volume (%)

-

1

2

3

4

5

6

7

8

9

10

0 50 100 150 200 250 300 350 400 450

Diesel import opportunity Gasoline import oportunity Diesel actual sales

Volume (kbd)

Marg

in (

$/b

bl) 594

$ 34 mn

$ 14 mn

$ 791 mn

EBITDA from Regular Distribution Business

Diesel Imports

Gasoline Imports

22 June 2016


■ Wider Impossible. We believe the size of this opportunity can widen or narrow every

day, depending on key variables such as oil prices and FX rate. It also depends on the

demand growth rate (as it expands or contracts unmet volumes) and on the

aggressiveness with which the company pursues this endeavor. In the following charts

we provide investors with a sensitivity analysis for a wide combination of these

variables, so that these charts can be consulted at any time. Figure 121 shows the

range of diesel import margins for different combinations of Brent and FX. Figure 122

shows diesel import volumes for different 2016 growth rates and different shares of the

opportunity. Investors may choose any combination of circles from the two charts to

calculate the size of the import EBITDA (just remembering that the chart on the right is

expressed in kbpd). Figure 123 and Figure 124 show the same for gasoline.

Figure 121: Diesel Import Margins Sensitivity To

Different Oil Prices and FX Rates in USD/bbl

Figure 122: Diesel Import Volumes Sensitivity To

2016 Demand Growth and Ipiranga's Shares of

Imports in kbpd

Source: Credit Suisse analysis, Brasil Energia, Codeba, Transpetro rates, Decope-NTC, Brazilian Water Transportation Agency (Antaq)

Source: Credit Suisse analysis, ANP, Brazilian Energy Planning Corporation (EPE), Ministry of Development, Industry, and Foreign Trade (MDIC)

Figure 123: Gasoline Import Margins Sensitivity To

Different Oil Prices and FX Rates in USD/bbl

Figure 124: Gasoline Import Volumes Sensitivity To

2016 Demand Growth and Ipiranga's Shares of

Imports in kbpd

Source: Credit Suisse analysis, Brasil Energia, Codeba, Transpetro rates, Decope-NTC, Brazilian Water Transportation Agency (Antaq)

Source: Credit Suisse analysis, ANP, Brazilian Energy Planning Corporation (EPE), Ministry of Development, Industry, and Foreign Trade (MDIC)

12 2

17 7

24 14 4

32 22 122

43 33 23 13 3

2.6

3

3.4

3.8

4.2

4.6

5

20 30 40 50 60 70

FX

9 22 44 66 88

11 27 54 81 107

13 32 63 95 127

15 37 73 110 146

17 41 83 124 166

-6%

-4%

-2%

0%

2%

4%

6%

0% 25% 50% 75% 100%

Share of the opportunity

Dem

and g

row

th

10%

14 4

19 9

24 14 4

32 22 12

2

41 31 21 11

1

2.6

3

3.4

3.8

4.2

4.6

5

20 30 40 50 60 70

FX

Brent price ($/bbl)

4 9 18 28 37

5 12 24 36 48

6 15 29 44 58

7 17 34 52 69

8 20 40 60 80

-6%

-4%

-2%

0%

2%

4%

6%

0% 25% 50% 75% 100%Share of the opportunity

Dem

and g

row

th

10%

22 June 2016


■ But, Can Ipiranga Use the Existing Logistics Infrastructure? ANP ensures open

access to any third party that wants to use a terminal or a 15km or longer pipeline,

within oil production areas. The pipeline's maximum capacity is composed of the part

reserved for the owner (owner's preference), the part already reserved for third parties

(contracted capacity), and the part available for new contracts (available capacity).

Therefore, by negotiating contracts (usually with a five-year term) and fees with the

operator of the facility, the interested party can have access to all the available

capacity, besides any idle contracted capacity. On the other hand, open access to

terminals must be requested from the operator before the 20th day of the month prior

to the month of transportation, in order to enter the facility's prior schedule. The other

alternative is to submit a request after the deadline to enter the "extemporaneous

schedule," depending on availability and other interested companies. The only limit for

this service is the case of an operator signing with a single shipper (who requests the

access) to use above 50% of the terminal capacity. All this open-access legislation in

Brazil is very important for players just like distributors, since it makes it possible for

them to move their own products through very large areas, such as the Central West

and Southeast regions.

■ What If Actual Capacity Is Not Enough? In terms of infrastructure expansion, a third

party is able to deploy capital to expand other players' existing facilities, including

pipelines, terminals, and even upstream infrastructure, except for the ones included in

the "Development Plan." After being granted construction authorization by the Brazilian

Oil, Gas, and Biofuels Agency (ANP), any player would be allowed to build new

facilities, improve existing facilities' capacities, and include new delivery points in

pipelines. Besides that, it is possible to operate a facility owned by other players as

well, by requesting an operation authorization from the ANP to operate new facilities,

expand facility capacity, transfer facility operations, and reactivate a facility whose

operation authorization had been revoked. Hence, the law makes it possible for a

distributor to invest in existing structure, by increasing its ability to bring additional

volumes to the demand points. Figure 125 shows the main terms and conditions for

each type of infrastructure and each type of use.

Figure 125: Details of Open-Access Regulations in Brazil for Each Type of Infrastructure and Use

Source: Brazilian Oil, Gas, and Biofuels Agency (ANP)

Pipelines open access Terminals open access Construction and Ampliation Operation

15km or longer pipelines with origin outside

oil production areas

Public or private water transportation terminals Pipelines and auxiliary facilities Pipelines and auxiliary facilities

Berth and landing systems Terminals Terminals

•Pipelines in terminal area, since essential to

transportation

LGN facilities LGN facilities

•Storage and complementary systems, since

essential to transportation

Upstream facilities, since not

included in the “Development Plan”

Upstream facilities, since not


Upstream pipelines, since not


Upstream pipelines, since not


Direct request to Carrier (who operates the

transport facilities)

Direct requests to Operator Construction Autorization requisition

to ANP

Operation Autorization requisition to

ANP

Request to Prior Schedule5, with 3 days for

reply period from Limit Date6

90 days reply period from

requisition date

60 days reply period from requisition

date

Request to “Extemporaneous Schedule”7, with

2 days for reply period from request date (when

there is disponibility)

Conditions

Third Parties Interested in Available Capacity1,

Available Operational Capacity2 and Idle

Contracted Capacity3

An only Shipper (who uses the transport

service) can not have utilization superior to 50%

of storage capacity

Registered company in ANP

regulation

Registered company in ANP

regulation

Fees

Negotiated between Carrier (who operates

the facility) and Shipper (who uses the

transport service)

Negotiated between Operator (who operates

the facility) and Shipper (who uses the transport

service)

Owner’s rightTransport ensured by the Owner’s Preference

4Transport ensured by the Owner’s Preference

Infrastructure

Authorization

1 Difference between the Maximum Capacity and the sum of Owner’s Preference with the Contracted Capacities2 Difference between the Operational Capacity and the sum of Owner’s Preference with the Contracted Capacities3 Difference between the sum of Contracted Capacities with the mensal volume of products effectively transported in a transportation facility4 Mensal volume of products, between reception and delivery points, that is granted to the transportation facility’s Owner Shipper8 to transport its own product5 Mensal schedule prepared by the private terminal’s operator to fulfill the access requests realized until the Limit Date6 Twentieth month day prior to the month in which the product will be transported in the private terminal7 Schedule prepared by the private terminal’s to fulfill the access requests realized after the Limit Date8 Company or consortium user of transportation service, owner of transported products and owner of the transportation facilities as well

22 June 2016


■ New Infrastructure Is Still a Very Distant Dream. Although the open-access

regulation lays the groundwork for any player to make marginal capacity expansions in

existing infrastructure, we don't believe this will happen in the foreseeable future.

Buying products abroad and trying to access the domestic market from Petrobras's or

third parties' existing infrastructure (via open access) is already risky, but the stakes

can be high enough to justify the high working capital commitment and the risks of

margins shrinking until the operation is over. But in order to justify why other players

would invest huge sums of money to create additional infrastructure, this opportunity

would have to be made very wide and sustainable in the long run.

■ A Chain of Mountains Stands in the Way. Brazilian geography poses an additional

difficulty for would-be infrastructure investors. Very close to the coast, there is a chain

of mountains, named Serra do Mar, extending from the Southeast all the way to the

South. This creates a very significant barrier, because it dramatically increases the

costs involved in any attempt to build inland logistics infrastructure. So, we don't see

room for committing hundreds of millions or even billions of dollars in additional

infrastructure. And, this is especially true because we don't think there should be a

sustainable wide premium in domestic prices above import parity that can be relied on

to be in place for decades to make this move at all smart. We're not saying that new

infrastructure will not be built. We're saying that nobody should do it just to become an

importer of fuels. There will be room for additional logistics investments in Brazil, but

for somebody willing to play the logistics game.

■ Past Operations Were Already Sizeable. We gathered data on past imports in the

last ten years for each of the main maritime terminals in Brazil and noted that the

volumes that were already imported in each region were, in most cases, higher than

the unmet deficits we identified. We also took care not to use the peak import levels as

a proxy for a sustainable import volume. So, we observed the monthly peaks as well as

the yearly averages. This is equivalent to assuming that the past monthly peak wouldn't

be replicated every month, but rather every other month. When we compare this level

of past operations (with this conservative approach) to the structural deficit, we see

there is room in the overall available infrastructure to get access the unattended

demand, as shown in Figure 126.

Figure 126: Historical Import Volumes vs. Total Structural Deficit In 2016 in kbpd

Source: Credit Suisse analysis, ANP, EPE, MDIC Note: Average between yearly peak and monthly peak imports, in the last five years.

163

90

77 20

19 7 7 4 1 387

132

São

Lu

ís

SP

/CO

Su

ape

Man

aus

Par

aná

Cea

rá

RJ/

MG

Bah

ia

Su

l

Bra

zil

Stru

ctu

ral

def

icit

22 June 2016


■ Opportunity Meets Capacity. Figure 127 shows these levels for different corridors

and also the levels of local deficits we see in those corridors for 2016. We see that the

Northeast terminals have been the ones with the highest levels of imports, although

most of these operations were made by Petrobras itself. It also shows that the SP/CO

corridor has received a high level of imports, as previously shown, although it

concentrates most of the domestic refining capacity. Some corridors like the Paraná

and the RJ/MG present deficits that are higher than historical imports. In our view, this

indicates these deficits have been attended not only through imports, but also through

the inter-corridor supplies, either inland or by ships. It also may mean there is room for

expansion in the inland logistics in those regions, given the size of their deficits. Figure

128 shows what the potential profits from these operations would look like, under our

assumptions for oil price, FX curves, and Petrobras's price adjustments.

Figure 127: Potential Volumes Per Corridor vs.

Historical Import Volumes in kbpd

Figure 128: Import EBITDA Opportunity Under Our

Base Case Assumptions Potential Profit in USD mn (LHS) and Margin in USD/bbl (RHS)

Source: Credit Suisse analysis, ANP, EPE, MDIC Source: Credit Suisse estimates, Brazilian Oil, Gas, and Biofuels Agency (ANP), Brazilian Energy Planning Corporation (EPE), Ministry of Development, Industry, and Foreign Trade (MDIC), Brazilian Association of Fuel and Lubricant Distributors (Sindicom), Brasil Energia, Codeba, Transpetro, Decope-NTC, Brazilian Water Transportation Agency (Antaq). Note: CS oil forecasts, gasoline spreads of USD12/b in '16; diesel spreads of USD12/b in '16; FX at 3.5 BRL/USD; Volumes growth: gasoline and diesel -3.5% in '16; +0.5% in '17, +2.5% later; Ipiranga's share of 10%; refineries utilization at 86%; refinery price increases of 0% in '16; 10% p.a. later.

■ What to Expect, Then? Will Ipiranga keep importing? Will Brent prices and FX rates

behave the way we see? We don't know, of course. But, under certain conditions, we

think there is room for imports to keep adding non-negligible earnings power to the

company. We think the market regards this as a minor issue and typically dedicates

little time to understanding the details of these operations. The amount of data

necessary to address the question is huge and makes this analysis unattractive to

many. But we don't appreciate preconceived ideas and wanted to check the evidence.

We hope this section was able to clarify what the conditions for this to happen are and

also how much can potentially come from that, in our view.

88

5

39 37

39

31

14 18

0

0

20

40

60

80

100

120

140

160

180

São Luís Suape RJ/MG Paraná SP/CO

Gasoline Opportunity beforeInterregional(kbd)

Diesel Opportunity beforeInterregional (kbd)

34 32

2

31

14

1310

-

2

4

6

8

10

12

14

16

18

20

-

5

10

15

20

25

30

35

40

45

50

2016 2017 2018 2019

Diesel Potential Profit (LHS) Gasoline Potential Profit (LHS)

Gasoline margin (RHS) Diesel margin (RHS)

22 June 2016


Other Businesses

Question #6: What Are the Prospects for the Other

Businesses?

"Diversification is not how many different things you own, but how different the things you own are in the risks they entail" (Seth Klarman)

■ Circle of Competence. We like Ultrapar's portfolio of businesses in general. Although

they belong to different sectors, they require similar skills to be run efficiently. Except

for Oxiteno, which is pretty much an industrial operation, all the other businesses

involve, in some sense, the need to move products and make them available to clients

where they are needed. The expansion of those businesses require a careful and

disciplined expansion of distribution infrastructure, as well as the addition of points of

sale to its network, be they fuel stations, LPG resale stores, or drugstores. So,

branding, client assistance, support to the resale network, and market share gains are

achieved through increased service quality and technology. On the other hand, these

businesses are exposed to completely different consumption segments, with different

growth rates and different market dynamics. In all these segments, Ultrapar's

subsidiaries are competitive and well-managed companies with leading positions and

decent returns on capital.

■ Focused Approach. From all the different ways to assess the prospects of a

company, the one we like the most is the value standpoint. And, in the case of Ultrapar,

we focused most of our attention on deeply exploring the reasons behind Ipiranga's

value-addition potential, because this business represents 87% of the value we see in

the group. The Ipiranga and, later, the Texaco acquisitions made the distribution

business deserve all this attention, and now the AleSat acquisition is likely to be a new

value driver for the company. But, of course, Ultrapar has also interesting stories

running in the other sectors that we will discuss more briefly in this section, focusing on

the main operational and financial performance metrics and how to project them.

Oxiteno

■ Petrochemicals. We expect c.19% of Ultrapar's 2016 EBITDA to come from its

petrochemicals business, Oxiteno, as shown in Figure 129. Oxiteno benefitted from the

BRL depreciation in 2015, which greatly increased its profitability level, as shown in

Figure 130, despite a sharp decline in domestic volumes due to the weaker economy.

We expect Oxiteno's profitability to decline slightly from levels observed in 2015 owing

to weakening margins in USD/t terms, partially offset by recovering volumes.

22 June 2016


Figure 129: Oxiteno's Share of 2016 Ebitda in % of 2016 CS forecasted EBITDA

Figure 130: Oxiteno's Historical EBITDA and CSe in BRL millions

Source: Credit Suisse estimates Source: Company data, Credit Suisse estimates

■ Monopoly in Ethylene Oxide. The petrochemical industry can be divided into three

"generations," depending on the stage of the transformation chain. First generation

companies, or "crackers," crack naphtha (Brazilian petrochemicals are predominantly

naphtha based) into ethylene, propylene, other olefins, and aromatics. Braskem

currently holds a monopolistic position in the production of first-generation

petrochemicals in Brazil, although it faces competition from imports in some products.

Second-generation companies, like Oxiteno, are situated within the same

petrochemical complexes as first generation plants and use products from the first-

generation plants to produce intermediate petrochemicals, such as polyethylene,

ethylene oxide, and others. The market for second-generation companies is less

concentrated (there were about 50 second-generation companies in Brazil in 2015),

but Oxiteno is the only producer of ethylene oxide in the country. Third-generation

companies buy intermediate chemicals from second-generation companies and

transform them into final products, such as polyester, plastics, nylon, and others, used

in several consumer and industrial goods. There are more than 11,000 third-generation

companies operating in Brazil.

■ Diversified Demand and Clients Base. Oxiteno is a second-generation chemical

company, which produces ethylene oxide and a variety of its by-products. Its products

are used in packaging, coatings, cosmetics, detergents, crop protection, and a number

of other products for different sectors of the economy, as shown in Figure 131.

Although the variety of sectors in Oxiteno's customer base provides a certain level of

protection against the cycles of each particular market, overall volumes are correlated

with economic growth. As shown in Figure 132, Oxiteno's volume growth was stronger

in years of economic expansion. Oxiteno's main customers in the domestic market in

2015 included Monsanto, Syngenta, and Unilever, whereas its main customers in the

international market included Unilever, P&G, Syngenta, and trading companies. In the

same period, the top 10 customers accounted for 32% of Oxiteno's net sales, and no

single customer accounted for more than 6%.

Ipiranga69%

Oxiteno19%

Ultragaz8%

Ultracargo3%

Extrafarma1%

41%

8%

35%

25%

-8%

83%

6%

-3% -2% -2% -2%

-

100

200

300

400

500

600

700

800

900

EBITDA EBITDA growth

22 June 2016


Figure 131: Oxiteno's Domestic Sales By Use % of volume sold in 2015

Figure 132: Total Volume Growth in %, year-on-year change

Source: Company data Source: Company data, Credit Suisse estimates

■ Specialty Chemicals and Commodities. Oxiteno's products can be classified into two

main groups: (i) specialty chemicals and (ii) commodity chemicals. Specialty chemicals

are designed for a specific end use, and customers often require products to be

certified to meet certain technical requirements. Specialties include a variety of

products, including surfactants, dispersants, and hydraulic fluid. These products are

normally produced in batches, and prices are negotiated individually. Commodities, on

the other hand, are standard specification and produced in continuous processes. The

main commodities sold by Oxiteno are ethylene oxide and ethylene glycol, used in the

production of polyester, PET, and antifreeze and brake fluids. The prices of these

products are referenced to international contract prices, and Oxiteno faces fiercer

competition from imports. Figure 133 and Figure 134 show how the sales mix between

specialty chemicals and commodities has evolved. As the chart shows, there was a

clear shift in volumes sold toward more specialized products, in the search for higher

margins. The company intends to optimize the sales mix towards the more value

added specialty chemicals; for this reason, the company maintains R&D centers to

develop products that meet clients' specific needs.

Figure 133: Oxiteno's sales volumes by product in kt

Figure 134: Oxiteno's volume mix in % of total sales volume

Source: Company data, Credit Suisse estimates Source: Company data

Polyester20%

Cosmetics and

detergents24%

Crop protection

17%

Distributors10%

Coatings9%

EO / DOT (brake fluids)

6%

Performance Products

5%

Glycols4%

Oil and Gas4%

Others1%

1%1%

4%

-3%

9%9%

1%

4%

19%

-12%

12%

8%

-4%

15%

2%0%

-7%-6%

2% 2% 2% 2%

-15%

-10%

-5%

0%

5%

10%

15%

20%

25%

1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

2010

2011

2012

2013

2014

2015

2016E

2017E

2018E

2019E

2020E

Volume growth Brazilian GDP, real

673618 591 600 610 620 631

107

107 160 163 166 169 172

2014 2015 2016E 2017E 2018E 2019E 2020E

Specialty Chemicals Commodities

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

2010

2011

2012

2013

2014

2015

Commodities

Specialty Chemicals

22 June 2016


Domestic Market and Exports. Oxiteno sells to both the domestic market, where it

enjoys higher margins, and the international market, through exports and plants

abroad. Although the domestic market has grown over time (Figure 135), Oxiteno has

continued to build capacity in order to maintain a certain level of oversupply in the

domestic market. The strategy to maintain production capacity in excess of domestic

demand has strategic value for the company, because it allows the company to

preserve margins in this market by discouraging new entrants. The excess in domestic

production is exported, albeit at lower margins. As a matter of fact, the mix between

domestic and foreign volumes has remained relatively stable over the past 10 years, as

shown in Figure 136.

Figure 135: Oxiteno's Sales, by Destination in kt

Figure 136: Oxiteno's Sales Mix, by Destination in % of total sales volume

Source: Company data Source: Company data

■ International Expansion. Since the early 2000s, Oxiteno has been expanding its

geographical footprint, with a focus on specialty chemicals in the Americas. In the

years that followed, Oxiteno completed a series of acquisitions, including the

acquisition of specialty chemicals companies in Mexico, Venezuela, and Uruguay, and

plants in Mexico and Pasadena (US). Figure 137 depicts the breakdown of the

company's sales volume outside Brazil, as well as the source of the products, in 2015.

Figure 137: Sources and Destinations of Oxiteno's Sales Volume in Foreign

Markets in % of foreign sales

Source: Company data. Note: Sales in foreign markets totaled 211.8 kt in 2015.

386458

397 430483 479

553 546 557522

157

189

170

205

201 181

208 230 222

203

2006 2007 2008 2009 2010 2011 2012 2013 2014 2015

Domestic market vols Foreign market vols

71% 71% 70% 68% 71% 73% 73% 70% 71% 72%

29% 29% 30% 32% 29% 27% 27% 30% 29% 28%

2006 2007 2008 2009 2010 2011 2012 2013 2014 2015

Domestic market share Foreign market share

46%

27%

4%

19%

4%

19%

5%

8%

6%

8%19%

5%

3%

3%

1%

7%

5%

0%

7%3%

Others

Others

Others

Others

Mercosur

ex-Brazil

NAFTA

Europe

Asia

Legend:

Destination

Source

22 June 2016


■ Capacity Expansion. In November 2015, Ultrapar announced the expansion of

Oxiteno’s specialty chemicals’ capacity in the United States by building an ethoxylation

unit with 170,000 tonnes per year of capacity at its current site in Pasadena, Texas.

The project is expected to be concluded by the end of 2017 and will require a total

investment estimated at USD113 million. The project will benefit from competitive

feedstock from shale gas.

■ Raw Materials. Oxiteno's main raw material is ethylene, which accounts for c. 24% of

its COGS. The company buys ethylene from Braskem at terms defined in a supply

agreement establishing a minimum ethylene purchase requirement at prices linked to

the international market. Supply of ethylene constitutes an entry barrier in Oxiteno's

market, as current ethylene production capacity is committed to existing second-

generation producers. Other raw materials used in the company's productive

processes include palm kernel oil, C4, butyl alcohol, primary fatty amine, ethanol, and

others. These materials account for c. 28% of COGS. Utilities used by Oxiteno plants

include electric power, steam, and natural gas. Utilities account for c. 3% of COGS.

■ Margins. Oxiteno sells commodity products in the domestic market at prices similar to

import parity. That is, international prices plus the costs of importing the products.

Domestic commodity volumes, then, have higher margins than exports, which are sold

at export parity. For specialty chemicals, the company enjoys higher pricing power and

is normally able to pass through price increases in the raw materials. Oxiteno's margins

therefore are, to a great extent, dollarized. Moreover, specialty chemicals margins are

much less sensitive to industry cycles than commodities, as shown in Figure 138. We

forecast margins that revert to the historical mean from current levels, which ultimately

affects EBITDA margins, as shown in Figure 139.

Figure 138: Oxiteno's Margins over Raw Materials in USD per tonne

Figure 139: Oxiteno's EBITDA Margin Per Tonne in USD per tonne


Ultragaz

■ LPG Distribution. We expect Ultragaz, the group's LPG distributor, to provide c. 8% of

Ultrapar's EBITDA in 2016, as shown in Figure 140. Ultragaz achieved significant

growth in EBITDA in the past three years owing to increases in LPG prices, as shown

(600)

(400)

(200)

-

200

400

600

800

1,000

1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

2010

2011

2012

2013

2014

2015

2016E

2017E

2018E

2019E

2020E

Polyester and Glycols margin Spec chems margin

228

169 169 184

167

278

235

162

124

202

135

200

236 236

263

220

306

285

266

246

227 210

1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

2010

2011

2012

2013

2014

2015

2016E

2017E

2018E

2019E

2020E

22 June 2016


in Figure 141. In 2015, prices in the bottled segment increased by 10%, and prices in

the bulk segment also increased by 21%; as a result, EBITDA grew by 17%, in spite of

the slight decline in sales volume. In the next two years we expect Ultragaz to keep

profitability close to the current levels and to grow in line with inflation.

Figure 140: Ultragaz's Share of 2016 EBITDA in % of 2016 CS forecasted EBITDA

Figure 141: Ultragaz's Historical EBITDA and CS

Forecast in BRL million


■ Pioneering Spirit and Acquisitions. Ultragaz was founded in 1937 and introduced

LPG as cooking gas in Brazil. Thirty years later, together with two other distributors, it

founded Utingás, a company with LPG storage facilities in the states of São Paulo and

Paraná. In 1980, it acquired 90% of Bahiana, which operated in the Northeast, and in

1999 acquired the remaining 10%. The next big moves were in August 2003, with the

acquisition of Shell Gás, Royal Dutch Shell's LPG operation in Brazil, and in 2011 with

the acquisition of Repsol's LPG business in Brazil. Ultragaz is currently comprised of

three operating subsidiaries: Companhia Ultragaz (99%), Bahiana (100%), and Utingás

(57%).

■ Ultragaz's Company Description. Since the acquisition of Shell Gás, Ultragaz stands

as the Brazilian market leader in LPG, with a 23% market share, and operates

nationwide in the distribution of both bottled and bulk LPG. The bottled LPG is used

primarily by residential consumers for cooking, whereas the bulk LPG is used by (i)

commercial consumers, such as hotels, restaurants, and hospitals, for heating water;

and (ii) by industrial consumers for furnace heating, asphalt production, and other

applications. The majority of the bottled LPG is sold through independent dealers,

responsible for 94% of the bottled sales in 2015, as well as through its own truck fleet

and retail stores. In total, the company posted 1.17 Mt in bottled sales last year (shown

in Figure 173). On the other hand, bulk LPG is serviced only through Ultragaz's own

trucks and contributed 31% to Ultragaz's sales, which share has remained flat over the

past 10 years, as Figure 143 shows.

Ipiranga69%

Oxiteno19%

Ultragaz8%

Ultracargo3%

Extrafarma1%

9%

-8%

-13%

14%

9%

17%

1% 0%

7% 7% 7%

-

50

100

150

200

250

300

350

400

450

500


22 June 2016


Figure 142: Ultragaz's Sales, by Segment in kt

Figure 143: Ultragaz's Sales Mix, by Segment in % of total sales volume

Source: Company data Source: Company data

■ Supply of LPG. Currently, all distributors in Brazil purchase nearly all LPG from

Petrobras and with similar procedures for ordering, which consist of sending an

estimate of the needs to the company four months in advance and a more precise

estimate one month in advance. Therefore, Petrobras freely prices both LPG for

residential use and LPG for commercial and industrial use in the domestic market. For

the bottled product, though, the refinery prices remained unchanged from 2003 to

2015, when average price reached USD 331/ton due to a 15% increase in September,

compared with USD 254/t outside Brazil. On the other hand, Petrobras has increased

bulk LPG refinery price seven times since January 2008, after five years without price

changes. Therefore, the average bulk price became USD 425/t in 2014, still 28% lower

than the international price of USD 544/t. Just as in fuel distribution, Petrobras's

practice has been not to immediately reflect in its LPG prices in Brazil the volatility of

international prices.

■ Storage of LPG. Of the Ultragaz's 17 filling plants, 9 receive LPG through pipelines

from Petrobras's refineries (primary units) and the other 8 receive LPG transported by

tanker trucks (secondary units). Besides that, Ultragaz also has 19 satellite plants,

storage bases for supplying its bulk trucks. With this infrastructure, the company's

storage capacity is c. 19.4 kt, or 3.5 days' sales, including c. 4.2 tonnes corresponding

to its stake in Utingás. As for its filling capacity, in 2015 Ultragaz used 68% of 125.6 kt

per month at its filling plants.

■ Distribution of LPG. Ultragaz sold 1.7 Mt of LPG in 2015, of which 69% went to the

residential segment and 31% to the commercial and industrial segment. In the first

segment, Ultragaz uses three distribution channels: (i) 5,100 independent dealers,

which carry out extensive home delivery and represent 95% of Ultragaz's bottled LPG

sales; (ii) 93 vehicles of its own fleet; and (iii) 52 retail stores, normally located at the

company's distribution centers or bottling plants. Recently, Ultragaz rapidly reduced the

share of its own fleet and retail stores in its volume sales, consolidating the

independent dealers even more as its main distribution channel, in addition to investing

in more phone orders rather than door-to-door delivery. In the commercial and

industrial segments, Ultragaz distinguishes the service by the size of the client. Large

bulk distribution, classified as consumption of more than five tonnes per month and

1,059 1,061 1,106 1,114 1,115 1,134 1,133 1,134 1,155 1,168

485 511 495 475 493 518 548 562 556 529

2006 2007 2008 2009 2010 2011 2012 2013 2014 2015

Bottled Bulk

69% 68% 69% 70% 69% 69% 67% 67% 68% 69%

31% 32% 31% 30% 31% 31% 33% 33% 32% 31%

2006 2007 2008 2009 2010 2011 2012 2013 2014 2015

Bottled share Bulk share

22 June 2016


composed mostly of industrial users, is made by tanker trucks that deliver the LPG

directly to the storage tanks located at the customers' premises. Small bulk distribution,

classified as consumption under five tonnes per month and comprised of residential

buildings, commercial users, and smaller industrial users, is made primarily by bobtail

trucks.

■ Many Similar Moats… Many of the characteristics of the fuels distribution segment

are also present in the case of Ultragaz. Ultragaz's main markets are geographically

concentrated. Therefore, distribution to this market can be carried out with great

economies of scale resulting in lower distribution costs to the company (Moat #1:

Economies of Scale). For these residential consumers, most of the company's bottled

LPG sales (95%) are made through a network of resellers, which have compensations

to display Ultragaz's brand on their storefronts, trucks, and uniforms (Moat #7: The

Value of a Brand). The brand is also stamped in approximately 23 million 13Kg bottles

in the market. The company establishes contracts with the resellers, through which

brand loyalty and volume exclusivity are assured, but does not predetermine prices

(Moats #3 and #4: Network and Pricing Power). Logistics infrastructure is also an entry

barrier as the business depends on the construction of filling plants to fill the bottles or

the bulk trucks (Moat #2: Logistics Infrastructure). On the supply side, the presence of

a single player assures stability for the market demand to be completely met (Moat #6:

Single Supplier). Therefore, the last significant interruption in supply was in 1995, due

to a 15-day strike by Petrobras's employees. In addition, the 17 filling plants present in

4 different regions provide a protection for Ultragaz, as it is able to purchase from

refineries to supply the growing demand (Moat #5: Access to Fuel Supplies).

■ …But with Much Lower Growth. Despite having many similar moats as fuel

distribution, the LPG market has grown much more slowly than fuel consumption and

Ultragaz's sales follow this trend (as shown in Figure 145). Thus, Ultragaz is also

expanding towards the North and Northeast, where faster economic growth is likely to

allow for better demand prospects. Besides that, the company is looking to expand in

niche markets, aiming at the industrial and agribusiness segments and new small and

medium businesses. The first target consists of expanding LPG usage to local heating,

such as pre-heating of industrial furnaces, grain drying and pest control. The second is

related to sales to new clients, such as laundries, restaurants, and bakeries.

■ Business Competitiveness. Currently, the LPG distribution sector in Brazil consists of

15 LPG distribution companies or groups of companies and is regulated by the

Brazilian Oil, Gas, and Biofuels Agency (ANP). Since per capita consumption in Brazil

is low and all distributors purchase nearly all of their LPG from Petrobras, with the

same procedures and same prices, they largely compete on the basis of low costs and

efficiencies in distribution and delivery. In addition, considering that the market for LPG

is a mature market with relatively low consumption growth, the competition is largely

based on attempts by the players to increase market share at the expense of their

competitors, focusing mainly on brand awareness, reliability of delivery and service

quality (Figure 144 shows how the market share scenario moves slowly in the LPG

sector). This perspective guides Ultragaz's strategy, with programs such as

UltraSystem, UltraPonto, and the Reseller Qualification Program, and targets

acquisitions as a source of volume growth. UltraSystem consists of a small bulk

distribution system to the residential, commercial, and industrial segments. It was

founded in 1997 and is receiving a considerable amount of investments to date,

delivering cost reductions in bulk distribution. UltraPonto is an innovative service

started in 2015 in an attempt to deliver a more rapid and complete service to clients.

Finally, the Reseller Qualification Program is a tool based on performance rankings

that aims to standardize Ultragaz's resellers' best practices, including brand use,

management quality and compliance with industry laws. In 2015, approximately 4.700

22 June 2016


resellers participated in the program (compared with 700 in 2008), which represents

more than 90% of total company's resellers.

Figure 144: Market Share Of Main Players In The

LPG Distribution Segment In Brazil in %

Figure 145: Ultragaz's Total Sales Volume Growth in %, year-on-year change


■ Potential M&A. Acquisitions in the LPG distribution segment may be one of the

avenues available to Ultrapar to further deploy capital effectively. Petrobras is going

through an important divestment process, and one of the assets that may be up for

sale is its LPG distribution arm, Liquigás. For Ultragaz, buying and absorbing Liquigas

would translate into significant economies of scale. A deal would also make sense from

a defensive perspective, in order to avoid having any of its existing competitors buying

that asset and growing much bigger than Ultragaz by size of operations. Furthermore,

a merger would represent an important consolidation in the sector. However, a

combined Ultragaz and Liquigas would have a market share in excess to 45%. For this

reason, we believe the competitive landscape could require the antitrust authority to

impose some degree of restrictions in terms of geographical overlaps, not to overly

concentrate market share in specific regions.

Ultracargo

■ Storage Services. We expect c. 3% of Ultrapar's 2016 EBITDA to come from its

storage services business, Ultracargo, as shown in Figure 146. In 2Q15, a fire

impacted operations for 22% of Ultracargo's storage, which caused the sharp decline in

the company's EBITDA, as shown in Figure 147. We expect the company to continue

to receive insurance payments related to the accident during 2016 and to normalize

operations at the Santos terminal throughout 2017, partially restoring EBITDA to prior

levels.

23.2% 23.2% 23.5% 23.1% 23.0% 23.2%

22.3% 22.8% 22.6% 22.6% 22.5% 22.6%

22.1% 21.2% 20.9% 21.1% 21.1% 20.4%

18.7% 18.9% 19.0% 18.8% 18.8% 19.2%

13.8% 13.9% 13.9% 14.3% 14.6% 14.6%

2010 2011 2012 2013 2014 2015

Ultragaz Liquigás Supergasbras Nacional Gás Butano Others

-1%

1%

2% 2%

-1%

1%

3%

2%

1% 1%

-1%

0%

1%

1% 1% 1%

22 June 2016


Figure 146: Ultracargo's Share of 2016 EBITDA in % of 2016 CS forecasted EBITDA

Figure 147: Ultracargo's Historical EBITDA and CS

Forecast in BRL million, unless otherwise stated


■ Ultracargo, Another Leader. Ultracargo provides storage services for liquid bulk,

specially chemicals, fuels, and vegetable oil, besides offering ship loading and

unloading services, the operation of pipelines and logistics programming. With 630

TCM of capacity at 6 storage facilities, the company is the largest provider of storage

for liquid bulk in Brazil. Figure 148 shows these facilities and their storage capacities,

utilization rates and product lines. Among these facilities, the Aratu terminal should be

highlighted, since Ultracargo accounts for 67% of all tank capacity for liquids at the

facility, which serves South America's largest petrochemicals complex. On the financial

side, Ultracargo's ten largest clients are responsible for 77% of its revenues, with its

three largest clients, Braskem, Petrobras, and Oxiteno accounting for 21%, 16% and

8%, respectively.

Figure 148: Ultracargo's facilities And Their Storage Capacity, Average

Utilization And Main Products

Source: Ultrapar

Ipiranga69%

Oxiteno19%

Ultragaz8%

Ultracargo3%

Extrafarma1%

7% 6%21% 10% 6%

-84%

354%

0% 7% 7% 7%

-

20

40

60

80

100

120

140

160

180


Facility Capacity (in m³) Average utilization (%) Product Lines

Aratu (Bahia) 218,190 96%

Chemicals, vegetable oils,

corrosives, and fuels

Suape (Pernambuco) 157,910 117%

Chemicals, ethanol,

corrosives and fuels

Itaqui (Maranhão) 55,280 134% Fuels

Santos (São Paulo) 152,800 78%

Chemicals, lubricants,

fuels, corrosives, ethanol

and vegetable oils

Rio de Janeiro (Rio de Janeiro) 17,247 108% Corrosives and lubricants

Paranaguá (Paraná) 28,262 58% Corrosives, vegetable oils

Total 629,689 97%

22 June 2016


■ Fire at Storage Facilities in Santos. In April 2015, part of Ultracargo's operations in

Santos were hurt by a fire in six ethanol and gasoline tanks, which represented 4% of

Ultrapar's overall capacity in Brazil at the time. The total capacity for the operations

halted is higher: 22% of total capacity, since it includes neighboring storage facilities in

the same area affected by the fire. We forecast average storage recovering from

current depressed levels and then growing in line with GDP, as shown in Figure 149.

Figure 149: Average Storage and Storage Growth in MCM and % year-on-year change, unless otherwise stated

Source: Company data, Credit Suisse estimates

■ Margins. In July 2010, Ultrapar sold Ultracargo's in-house logistics, solid bulk storage,

and road transportation businesses, in a transaction worth BRL82mn. This sale

allowed Ultracargo to focus exclusively on its liquid bulk storage business, a segment

in which it has a market leadership position, sharply increasing its margin in the

following years, from 41% in 2009 to 60% in 2012 (as shown in Figure 150). However,

in 2015 the fire at the Santos facilities brought the margin back to 52% and the EBITDA

margin per cu m down 83%, to BRL40/cu m, as the Figure 151 shows. We assume in

our forecasts that margins will normalize at levels similar to those before the event.

19.9%

5.4% 5.4%

13.4%

2.7%

-8.4%

-3.8%

8.0%

2.0% 2.0% 2.0%

-

100,000

200,000

300,000

400,000

500,000

600,000

700,000

800,000

2010 2011 2012 2013 2014 2015 2016E 2017E 2018E 2019E 2020E

Average Storage Storage growth

22 June 2016


Figure 150: Ultracargo's Gross Margin in %

Figure 151: Ultracargo's EBITDA Margin Per Cu M in BRL million per cu m


Extrafarma

■ Drugstores. We expect c. 1% of Ultrapar's 2016 EBITDA to come from its drugstore

business, Extrafarma, as shown in Figure 152:. We expect the company's EBITDA to

recover substantially, as shown in Figure 153, on the back of a growing number of

stores and better margins, once integration costs are concluded and stores added in

the accelerated expansion begin to mature.

Figure 152: Extrafarma's Share of 2016 EBITDA in % of 2016 CS forecasted EBITDA

Figure 153: Extrafarma's Historical EBITDA and CS

Forecast in BRL million


0%

10%

20%

30%

40%

50%

60%

70%

1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

2010

2011

2012

2013

2014

2015

2016E

2017E

2018E

2019E

2020E

202 203

233 226 234

40

190

171 178

185 193

Ipiranga69%

Oxiteno19%

Ultragaz8%

Ultracargo3%

Extrafarma1%

-4% -3%

165%

41%33%

28%

-

20

40

60

80

100

120

140

160

180

200

2014 2015 2016E 2017E 2018E 2019E 2020E


22 June 2016


■ The Transaction. Despite its 50 years of activity, Extrafarma joined the Ultrapar's

business only in February 2014. According to the terms of the agreement, Ultrapar and

Extrafarma entered into a share merger, pursuant to which Ultrapar acquired 100% of

the shares of Extrafarma in exchange for up to 2.9% of shares issued by Ultrapar to

Extrafarma's shareholders. Thus, the drugstore business became a wholly owned

subsidiary of Ultrapar on February 1, 2014, after closing the transaction that consisted

of the issuance of up to 16,028,131 shares in Ultrapar and the assumption of

Extrafarma's net debt of BRL106mn by Ultrapar.

■ Operation. Extrafarma operates 254 drugstores in six states in the North and

Northeast regions of Brazil, with the most significant activity in Pará, Ceará, and

Maranhão, as shown in Figure 154. Besides that, Extrafarma operates two distribution

centers, one in Pará and the other in Ceará, which supply all its stores. With this

structure, the subsidiary is active both in pharmaceutical retail and wholesale. The

former is Extrafarma's main focus, representing 87% of its 2015 revenue of

BRL1.14bn. In retail, its product mix consists of all the main types of pharmaceutical

products, such as brand-name drugs, generic drugs, brand-name similar drugs, and

OTC drugs. On the wholesale side, the company operates as a distributor of

pharmaceuticals and personal care products, purchasing them from manufacturers and

selling them to other drugstore chains and independent retailers, which are serviced

through its own and leased truck fleet.

■ Expansion Strategy. Just as in the fuel distribution, the drugstore business is focused

on consolidating its presence in the North and Northeast regions of Brazil, whose GDP

and household income are growing above the average for Brazil. Besides that,

Extrafarma's expansion is through (i) increased investment capacity, (ii) access to retail

space at Ipiranga's service stations and Ultragaz's resellers, with over 12,000 potential

retail outlets; and (iii) strengthening Extrafarma's management through the

implementation of Ultrapar’s recognized corporate governance. Therefore, we assume

the business will grow its number of stores in the coming years, achieving c.500 in

2020, as Figure 155 shows.

Figure 154: Extrafarma Drugstores by Region

Figure 155: Store Openings, Closings and Number

of Total Stores Per Year in units

Source: Ultrapar Source: Company data, Credit Suisse estimates

Location # of stores

North 114

Pará 103

Amapá 11

Northeast 139

Ceará 63

Maranhão 57

Rio Grande do Norte 11

Piauí 6

Pernambuco 1

Paraíba 1

Southeast 1

São Paulo 1

Total 254

60 60 60 60 60

-(6)

-

37

(2) (3) (3) (4) (4)

223 254

312

369

426

482

538

2014 2015 2016E 2017E 2018E 2019E 2020E

Store closures Store openings Total number of stores

22 June 2016


Valuation

Question #7: But Isn't It Too Expensive Already?

"It’s far better to buy a wonderful company at a fair price than a fair company at a wonderful price" (Warren Buffett)

■ Valuation Concerns. Many investors are convinced of Ultrapar's high-quality profile

and promising outlook. However, "What is smart at one price is dumb at another," as

Warren Buffet puts it in one of his letters. Investors, then, are naturally concerned

about whether they are paying a fair price for Ultrapar or whether the shares are

overvalued.

■ The Value of Growth. Investors' concerns regarding Ultrapar's valuation are well

founded. In our view, the next five years of cash flows represent only slightly above

11% of Ultrapar's current BRL47bn enterprise value. Moreover, the same five years

and a perpetuity of zero growth from 2020 onwards would still represent some 15%

downside to current share prices. In fact, a big part of Ultrapar's value can only be

justified by its prospects of further growth in free cash flows in the future and beyond

2020. It is important to highlight that growth is only valuable in economic terms if it

comes through investments with returns in excess of the cost of capital.

■ Isn't It Too Expensive? Based on the rating and target price on the cover of this

report, the attentive reader probably already knows our humble answer to such

question. In this section we intend to disentangle the issue using a two-part approach.

First, a qualitative and important discussion about Ultrapar's return on equity and its

competitive advantages. Last, but not least, we present our take on the company's fair

value, as well as consistency checks using different valuation methodologies.

■ Opportunities Lie Ahead. Rather than focusing solely on our assumptions, we think

it's also worth highlighting what is not in our numbers. Investors in Ultrapar are buying

a business with good results and many protections, but are also buying a portfolio of

opportunities that will be available for a shareholder-oriented management team.

These opportunities are not included in our numbers but if a fraction of them gets done,

shareholders will be much better off than we are initially considering. This, by the way,

has been the case for many years in Ultrapar. Every time investors expected returns to

fade, they later found out that management was able to gradually and continually open

new value creation fronts and deliver more. Figure 156 summarizes some of these

potential initiatives. The AleSat acquisition was already announced by the company

and, therefore, is discussed again later in this section.

22 June 2016


Figure 156: Summary of optionalities to base case valuation


Valuation Thoughts

■ Return on Investment and the Importance of the Entry Point. From an equity

investor's perspective, it doesn't matter how good a company's returns may be relative

to its cost of capital if it's already priced in the shares. A company may continue to

deliver returns in excess of cost of capital, yet its share prices may decline and it may

turn out not to be a good investment simply because it was initially overpriced. In

theory, risk-adjusted returns in excess of the appropriate benchmark can only be

achieved by buying securities below their fair value. Therefore, the entry point has a

major role in determining whether an investment will be successful or not.

■ The Problem with Bargains. Conversely, a company may perform poorly and yet turn

out to be a great investment simply because it was initially underpriced. In fact, buying

businesses that are at discount to their fair value, regardless of their quality, is a fair

strategy. However, in our view, the long-term payoff of this strategy depends on some

inherent risks that can't be ignored.

■ Risk #1: Value Trap. Return on investment will depend on the time market prices take

to converge to fair value. To exemplify, should an investor find a bargain deal in which

upside is 10% to fair value, return may be of 10% p.a. in excess of cost of equity if

market prices converge to fair value in a year's time. However, from a longer-term

perspective, the benefit of the bargain diminishes. This single deal will only deliver c.

1% p.a. return in excess of cost of equity in 10 years' time. Although that's a fair return

on top of cost of equity, there is very limited safety margin. Furthermore, the strategy

becomes riskier when buying businesses that are mediocre, or worse, because poor-

performing companies often deliver returns below the cost of capital, which can rapidly

erode the supposed excess return from a deal at bargain price.

■ Risk #2: False Bargain. Investment decisions, by their very nature and dependency

on future developments, are made using limited information. Therefore, investors

Valuation MethodValue/share

in Mar-2017Description

Ipiranga Returns BRL 23.5/sh

Assumes new investments will be made at 25% IRR in the 'next decade, 2020-2030' time horizon. Return levels compare

to CS's target price base case which conservatively assumes 18% in the same period. CS calculates 29% IRR for a single

Ipiranga station, when accounting for fixed costs dilution, which allows 4% overhead in corporante investments to maintain

Ipiranga's ROIC of 25% in 2015.

Alesat deal BRL 6.2/sh

Upside from the announced deal not included in target price. Economic value calculated as the SOTP of i) revaluation of

ALE's 2015 EBITDA at Ipiranga's 13x EV/EBITDA implied multiple, for BRL 2.5/sh; ii) present value of synergies, BRL

2.8/sh; and iii) 12-mo carry fwd.

Convenience stores BRL 6.4/sh

Ipiranga will grow average volume per station in its network of branded stations by increasing penetration of AMPM

convenience stores. Present value calculation assumes 10% increase in volume per station at convenience store opening

and penetration in the existing network increasing from 27% to 50% by 2020.

Liquigas BRL 1.6/shAssumes Ultragaz buys Liquigas at fair price. Economic value is created through sinergies of 80% of Liquigas G&A

implemented in three years time.

Extrafarma expansion BRL 1.5/shUpside to valuation assuming Extrafarma will increase store openings to 100 stores/year from 2018 onwards. CS's base

case assumes constant rate of 60 stores/year.

Extrafarma/Ipiranga sinergies N/A.Sinergies to be captured between Extrafarma and Ipiranga/Ultragaz by opening drugstores in service stations and LPG

resellers.

Acquisition of Lubricants plants N/A. Potential inorganic growth in the lubricants market

Total Upside BRL 39.2/sh Upsides not included in target price

22 June 2016


cannot rule out the possibility of entering into a deal that turns out not to be much of a

bargain after all, especially if it involves poor-performing companies. In Buffet's words,

"In a difficult business, no sooner is one problem solved than another surfaces—never

is there just one cockroach in the kitchen." If this is the case, the lower the quality of

the underlying business, the higher are the odds of a substantial loss. This is important

because not losing money is key to achieving a high payoff in the long run. As an

example and assuming a theoretical 17% cost of equity in BRL as reference, only after

ten years of 25% p.a. returns (8pps above cost of equity) would a portfolio that made a

bad investment decision that costed a 40% loss in the first year reach the same payoff

expected from a portfolio of same risk level.

■ Risk #3: Consistently Finding Bargains. The third risk is also possibly the most

important. Because gains from buying at a bargain price relative to fair value get

diluted over time, sustaining returns on excess of cost of capital depends on

consistently finding new bargains and, at the same time, avoiding bad deals. As the

capital base that needs to be allocated grows bigger over time, this can turn out to be a

challenging task even for the most successful investors.

■ Better Buy a Wonderful Business at a Fair Price. These risks are inherent to the

investing activity and simply cannot be eliminated. However, we think investing in

"wonderful businesses" is an efficient way to mitigate many of them. A wonderful

business is one that enjoys sound competitive advantages that allow it to constantly

produce returns in excess of cost of capital and yet protects it from economic

downturns. These businesses are often good investments, even at an ordinary time, in

the absence of particularly strong earnings momentum or when they are not at a

discount to their historical multiples.

■ Revolving Bargains. In fact, wonderful businesses often tend not to be fully priced at

fair value. This is so because the market hardly ever pays in advance for abnormal

returns extending too far into the future. However, generally speaking, there is also no

particular reason to believe the benefits of a strong brand, entry barriers, and other

competitive advantages will terminate in any specific year in the future. So, at any point

in time, the market may continue to grant such abnormal returns for an arbitrary period,

e.g., the next ten years. If that is the case, then this simple effect of extending for an

additional year of returns in excess of cost of capital renews the bargain price upside

each year. In this case, by remaining invested in the same business for an extended

period of time, it also reduces the risks and costs associated with reallocating capital.

In this sense, wonderful businesses are like revolving bargains.

■ Fair Value and Fair Price. So how can we be talking about abnormal returns on an

investment made at a fair price, if we said ourselves that by its very definition

investments made at fair value will only return cost of equity? At this point, we beg

forgiveness for inventing our own definitions. In fact, what we mean by fair price is not

the price that captures all present value of the entire magnitude of future reserves for

any given company, but rather only the value we would expect from a conservative set

of assumptions. So, we don't want to buy a wonderful company at fair value, because if

we did so we would be paying in advance for performance that is just too strong for a

period of time that is too far ahead, which would only leave us the downside. We want

to buy it at a fair price, meaning a price in which a fair future is embedded, not a stellar

one.

■ Is Ultrapar a Wonderful Business? Only time will tell. But, at this point, we have good

reasons to believe it is. Ultrapar has sustaining moats that only grow stronger and

synergically as each of its businesses expand. In light of these advantages, we also

believe that the company will continue to deliver returns in excess of the capital for an

extended period of time. Moreover, Ultrapar also has a characteristic that is common to

22 June 2016


wonderful businesses: a wonderful track record, evidenced by Figure 157. This,

incidentally, takes us back to our first comment.

■ Return on Investment and the (Limited) Importance of the Entry Point. We started

the discussion by saying that entry point was key to investing successfully because it

defined whether any specific investment was made below, or above fair value.

Subsequently, we mentioned that importance of entry point gets diluted over extended

holding periods. Finally, we made the case that wonderful businesses are, for the most

part, undervalued. When these three assertions are put together, we conclude that

importance of the entry point is rather limited when it comes down to investing in

wonderful businesses for long holding periods. Indeed, over the past decade, Ultrapar's

total shareholder return has been 27% per year. Had an investor paid an arbitrarily

higher price in 2006, for example 20% more, the returns would still have been

significantly above cost of equity.

Figure 157: Ultrapar's Total Shareholder Return in BRL/sh and index 2006

Source: Bloomberg, Central Bank, Credit Suisse analysis. Note: Share prices adjusted for share split that took place in September 2011. Total return assumes dividends are reinvested in UGPA shares.

Valuing Ultrapar

■ We Don't Want To Pay Too Much. So, we have analyzed Ultrapar's businesses and

concluded that it enjoys many moats that are likely to sustain strong operational

performance in the future. But we first need to know whether or not these moats are

too expensive. One important thing to keep in mind as we carry out our valuation,

though, is our ultimate goal: We don't want to pay a share price that is just too high.

■ AleSat Deal: Plug and Play Valuation. The acquisition of AleSat by Ipiranga is, at this

point in time, still contingent on approval by the Brazilian antitrust authority, Cade. The

authority may take 8-11 months to issue a final decision and, only then, will the

companies be allowed to merge operations. For this reason, we ought to treat the

value from the deal as an optionality, by valuing it separately from the remainder of

Ultrapar's operations, rather than incorporating it directly into our operational forecasts.

That way, we have a plug-and-play optionality, which we can easily add to our

0.0

1.0

2.0

3.0

4.0

5.0

6.0

7.0

8.0

9.0

10.0

11.0

12.0

0.0

8.5

17.1

25.6

34.1

42.6

51.2

59.7

68.2

76.7

85.3

93.8

102.3

May

-06

Oct

-06

Mar

-07

Au

g-0

7

Jan

-08

Jun

-08

No

v-0

8

Ap

r-0

9

Sep

-09

Feb

-10

Jul-

10

Dec

-10

May

-11

Oct

-11

Mar

-12

Au

g-1

2

Jan

-13

Jun

-13

No

v-1

3

Ap

r-1

4

Sep

-14

Feb

-15

Jul-

15

Dec

-15

Ind

ex

20

06

Shar

e p

rice

(B

RL/

sh)

Total return+27% p.a.

Share price +24% p.a.

Accrued CDI +11% p.a.

22 June 2016


valuation of Ultrapar's existing businesses. In our view, this optionality is worth some

BRL3.0bn, or BRL5.4/share, as shown in Figure 158. In our base case and target

price, we are not accounting for the economic value added by the deal. Although we

would attribute a very high probability that the deal will be concluded, we adopted this

stance because our valuation philosophy of conservativeness.

Figure 158: Valuation of the AleSat deal

Source: Credit Suisse Analysis

■ Risks to the Deal. Although we would attribute a low probability, the main risk to the

deal is that it may not be completed. Assuming it will, however, there are still risks to

the economic value that it will ultimately generate to Ultrapar's shareholders. Such risks

are for the most part related to the integration of operations. One possible downside we

highlight to our valuation concerns the relevance of AleSat's sales to white-flag

stations. Assuming AleSat sells c.30% of its volume to white flag stations, whereas

Ipiranga only some 4% to 5%, one can say that the actual volume added by the

acquisition is limited to some 3.2Mm³/year and, consequently, that synergy gains would

then be BRL115mn, resulting in reduction in upside from synergies by BRL0.7/sh, to

BRL 2.1/sh. We acknowledge this risk, but we do not think it materially changes our

valuation of the deal. Figure 159 shows a simple pro-forma financial statement of the

combination between Ipiranga and Alesat.

Revaluation of ALE’s EBITDA at

Ipiranga’s implied multiple

Total deal value BRL 3bn or BRL 5.4/sh

ALE’s EBITDA (BRL mn) 275

Ipiranga’s EV/EBITDA (x) 13

Value to Ultrapar (BRL mn) 3,575

Deal (BRL mn) 2,168

Revaluation value (BRL mn) 1,407

Shares 556

Value from revaluation BRL 2.5/sh

x

-

/

Synergies

Ipiranga’s EBITDA/m³ (BRL/m³) 100

ALE’s EBITDA/m³ (BRL/m³) 64

Gain in margin EBITDA/m³ (BRL/m³) 36

ALE volumes (mn m³/y) 4.3

EBITDA gains in sinergies (BRL mn) 155

Ipiranga’s EV/EBITDA 13x

EV/EBITDA in 2 years time 10x

Value of sinergies (BRLmn) 1,550

Shares 556

Sinergies per share BRL 2.8/sh

x

-

(a)

(b)

(a)x(b)

22 June 2016


Figure 159: Pro-forma financial statements of Ipiranga and Alesat combination


■ Various Methodologies. We want to minimize the chances of committing mistakes.

Seasoned practitioners know very well how sensitive DCF-based valuations can be to

certain assumptions, such as discount rates and continuing value. In order to check

and cross-check the implied beliefs behind our valuations, we have designed and used

several different methodologies. They are: (i) the group's two-part DCF, (ii) SOTP of

company-by-company DCF, (iii) multiples, and (iv) SOTP of multiples by business.

Each of these valuation methodologies are discussed in more detail in the paragraphs

below.

■ Setting a Target Price. The results of our valuation methodologies are presented in

Figure 160. Furthermore, we highlight that our target price is not intended to capture all

present value there may be in Ultrapar's equity, but rather to serve as a reference entry

price that, we believe, limits downside. Finally, we set our target price to BRL89/share

based on our two-part discounted cash flow of Ultrapar and not accounting for any

value from the AleSat deal. Still, the target price represents 25% upside potential from

our reference share price of BRL71.0/share.

Ipiranga Alesat Ipiranga+Alesat

Financial

Revenues BRL mn 75,655 11,351 87,006

COGS BRL mn 68,934 10,776 79,709

Sales expenses BRL mn 2,517 97 2,614

G&A expenses BRL mn 1,321 244 1,565

EBITDA BRL mn 3,953 275 4,228

Volumes

Gasoline volume sold mn m³ 8.55 2.16 10.71

Ethanol volume sold mn m³ 3.44 0.53 3.97

Diesel volume sold mn m³ 13.10 1.62 14.72

Share

Otto market share % 20.4% 4.7% 25.2%

Diesel market share % 22.9% 2.8% 25.7%

Stations

# of convenience stores units 1,910 288 2,198

# of stations units 7,230 2,037 9,267

CONEN units 1,900 772 2,672

S/SE units 5,330 1,265 6,595

22 June 2016


Figure 160: Summary of Different Valuation Methodologies Applied to Ultrapar in BRL per share

Source: Credit Suisse estimate. Note: Valuations presented in 12-month time, consistent with the definition of target price.

Two-Parts Discounted Cash Flow

■ Diminishing Marginal Returns, a/k/a the Grim Reaper. In Michael Mauboussin's

words "reversion to the mean is the microeconomic equivalent of the grim reaper: all

high-return companies succumb to it sooner or later (great managers make it later.)".

We think Ultrapar stands a good chance to make it later. Although we don't know

precisely when that will be, we believe the company will likely continue to enjoy returns

above cost of capital beyond our explicitly modeled horizon to 2020. To capture this

effect in our valuation and yet not exaggerate this performance going into perpetuity,

we separated the continuing value of our DCF into two periods: "the next decade" and

"2030 onwards." During "the next decade" we assume Ultrapar will continue to grow by

creating economic value, whereas from 2030 onwards returns conservatively converge

closer to cost of capital so that present value of perpetuity becomes much less

sensitive to growth rate.

■ There Is No Free Lunch. In a paper discussing 110 common errors in valuation,

Professor Pablo Fernández presents the use of perpetuity growth rates inconsistent

with free cash flow forecasts as one of the common mistakes. Errors often consist of a

missing link between growth rates and the capital reinvestment needs that undermine

free cash flows. This is an issue we think is important to be attentive to, especially in

light of the relevance of the continuing value in Ultrapar's DCF (c.90% of the present

value is due to cash flows from years beyond 2020). For this reason we opt to make

assumptions on value drivers, reinvestment rate, and the internal rate of return on new

investments (return on new invested capital, RONIC), to establish a reasonable link

between free cash flows and growth rates.

■ The Next Decade, More of the Same. So we have divided our continuing value

assumptions into two periods, the first one being the next decade, from 2020 to 2030.

We think it is fairly conservative to assume that, in this period, the company will

continue to operate and make investments as it has in recent years. To translate this

expectation into numbers, we have turned our attention to a pair of assumptions:

RONIC and reinvestment rate.

■ Marginal Returns, RONIC. In order to define our assumption of return on new

invested capital, we can come up with different estimates. A first possible alternative

BRL 89/shBRL 93/sh

BRL 80/sh BRL 81/sh

BRL 66/sh

Two-parts DCF SOTP: DCF SOTP: Multiples UGPA EV/EBITDAmultiple

UGPA PE Multiple

22 June 2016


would be to use the internal rate of return that is representative of the marginal

investment, for example the 20%-25% IRR on the investment to add a gas station to

the Ipiranga network. A second option, more conservative and the one we have used,

is to make a rough estimate of RONIC using historical company data. This latter

approach is illustrated in Figure 161, in which the growth in net operating profit less

adjusted tax (NOPLAT) is compared to the cumulative new invested capital in the

period. This approach results in an RONIC estimate of 18%. It is important to bear in

mind that the resulting return on investment may be underestimated if cash flows from

new investments are not constant, but rather growing, or if investments were made but

cash flows have not yet materialized, which we think was the case, by the way.

■ Reinvestment Rate. Concerning the reinvestment rate, we have also relied on

historical levels, as shown in Figure 162. In recent years, excluding big acquisitions

such as the Texaco stations in 2009, the reinvestment rate has averaged at 43%.

Furthermore, in our explicit model to 2020 we have forecasted operational assumptions

by business that imply an overall 39% reinvestment rate for Ultrapar. Therefore, we

assume the company will continue to reinvest 40% of NOPLAT on average during the

next decade. In the sum-of-the parts analysis that follows, we discuss in more details

what this rate of investment means in terms of operational metrics for Ipiranga. Our

objective then is to assess whether it is reasonable to assume it is possible to allocate

this much capital to existing businesses and the conclusion we have arrived at is that

not only is it possible, but that it is also fairly conservative. Under our assumption, then,

we do not have to account for new venues for capital allocation, such as M&A or

acceleration in capital allocation in the drugstore business, for example.

Figure 161: Ultrapar's rough RONIC calculation:

capital invested and NOPLAT added in recent years. in BRL million, unless otherwise stated

Figure 162: Ultrapar's Historical Reinvestment Rate

and Base-Case Assumption in BRL million, unless otherwise stated


■ There May Be Additional Upside. At the beginning of our valuation talk, we discussed

how "wonderful companies" can turn out to be great investments even when bought at

fair price. Then, we set the grounds for our valuation by saying that we wanted to

define our estimate of fair price using a conservative set of assumptions, in order to

reduce downside risk and preserve upside. At this point, it is probably a good

14% 15%

19%

14%

18%

-

1,000

2,000

3,000

4,000

5,000

6,000

2011 2012 2013 2014 2015

New Invested Capital (BRL mn)

NOPLAT growth (BRL mn)

RONIC

1.1 1.21.5 1.6

2.0

2.6

5.4

8.6

0.8 0.6 0.5 0.3

0.81.0

2.1

2.6

73%

50%

37%

16%

42%39% 40%

30%

2011 2012 2013 2014 2015 avg2016-20

avg2021-30

avg2031+

NOPLAT (BRL bn)

New Invested Capital (BRL bn)

Reinvestment rate

Forecast average

Avg = 43%

22 June 2016


opportunity to quantify part of this upside. That is because we think the two-parts DCF

at the holding company level allows for a good compromise between the simplicity that

facilitates sensitivity analyses and the complexity that helps to understand real-

economy implications. Figure 164 depicts the sensitivity of our target price to our return

on marginal investments and level of capital allocation assumptions. An interesting

reading of that chart we think is the additional upside/downside we would find if we had

made a different pair of assumptions. For example, had we assumed 22% RONIC and

the same 40% reinvestment rate, we would find 14% additional upside to our target

price. Additionally, Figure 166 illustrates how much growth is implied in each case,

assuming long-term inflation of 6.5%. For example, by choosing an 18% RONIC and a

40% reinvestment rate, we are assuming free cash flows will grow by 11% p.a.,

whereas had we assumed a 22% return rate, the same capital allocation would lead to

13% p.a. growth.

Figure 163: Sensitivity of DCF target to 2020-2030

RONIC and reinvest. rates (relative to base case) in multiple relative to base case target price

Figure 164: Growth at Different RONIC and

Reinvestment Rates for 2020-2030 in annual growth rate

Source: Credit Suisse estimates Source: Credit Suisse estimate. Note: assumes 5% inflation

■ 2030 Onwards, Little Value from Growth. In the long term, companies experience

diminishing returns, which undermines economic value created by growth.

Furthermore, growth converges to GDP growth plus inflation. With these restrictions in

mind, we assume return on invested capital will converge to 13%, roughly in line with

cost of capital. When this is the case, growth adds much less economic value than it

does at higher return rates. As for the reinvestment rate assumption, we chose 40% as

the necessary rate for the company to grow by 9.0% (of which 6.5% is for inflation and

2.5% is for GDP growth). Figure 165 shows the target price sensitivity to these

parameters, whereas Figure 166 shows the implied growth rate in each case. The

charts, as before, can be read as additional upside/downside to our target price, had

we used different premises. We highlight an interesting outcome from our sensitivity

analysis: had we assumed the company doesn't allocate any capital from 2030

onwards, the upside in our target price (22%) would be reduced by only 5 percentage

points. That still preserves a reasonable margin of safety.

0.81x 0.79x 0.77x 0.74x 0.72x

0.83x 0.85x 0.88x 0.90x 0.94x

0.84x 0.91x 1.00x 1.10x 1.22x

0.85x 0.98x 1.14x 1.33x 1.57x

0.87x 1.05x 1.30x 1.61x 2.02x

0.88x 1.13x 1.48x 1.95x 2.57x

6%

10%

14%

18%

22%

26%

30%

-20% 0% 20% 40% 60% 80%

Reinvestment rate

RO

NIC

7% 7% 8% 9% 9%

7% 8% 10% 11% 13%

7% 9% 11% 13% 16%

7% 10% 13% 16% 19%

7% 10% 14% 18% 22%

7% 11% 16% 21% 25%

6%

10%

14%

18%

22%

26%

30%

-20% 0% 20% 40% 60% 80%

Reinvestment rate

RO

NIC

22 June 2016


Figure 165: Sensitivity table of DCF target to 2030

RONIC and growth (relative to base case) in multiple relative to base case target price

Figure 166: Growth at different RONIC and

reinvestment rates for 2030 onwards in annual growth rate

Source: Credit Suisse estimates Source: Credit Suisse estimates Note: assumes 5% inflation

■ Upside Potential versus Downside Risks. Our two-parts DCF for Ultrapar results in

our target price of USD89/share in 12 months' time, 25% upside relative to current

levels. Furthermore, what is more interesting is that in this target price we have only

priced in performance that is comparable to recent years for the next 10 years. We

know that, should the average return on new investments be dramatically worse, for

example 4pps lower than our assumption, then the downside to our target price would

be 12pps, which would still leave us with a comfortable margin of safety on our

investment. What's more, the defensive characteristics of Ultrapar's businesses, as we

have discussed in this report, add to confidence that the potential upside/downside will

come on top of a return on the equity investment that is at least similar to the cost of

equity. In our view, upside potential is sizeable and downside risks is rather limited.

Sum of the Parts: Discounted Cash Flow

■ Holding versus Sum of the Parts. So far we have designed our valuation

assumptions and framed our investment case based on performance expectation for

the Ultrapar's consolidated businesses. This approach has the advantage of allowing

us not to worry about where the new investments will be allocated. However, the

negative side of making our assumptions at the holding level is that we can't know

precisely how that impacts its underlying businesses' performance. Therefore, in order

to understand whether the performance we expect from Ultrapar is compatible with its

existing businesses, we propose a second valuation approach using the same ideas of

marginal returns and reinvestments rates discussed before, applied on a company-by-

company basis.

■ Value Driver Assumptions for Ipiranga. Throughout this report we have discussed

moats that Ipiranga enjoys in its business and that suggest it will continue to enjoy

returns above cost of capital. Once again, to translate this understanding into numbers,

we should make assumptions on returns of marginal investments and reinvestment

rate. On the RONIC front, we rely on our model, built as discussed in Chapter 2, to

calculate the internal rate of return of a new Ipiranga gas station in the long term, when

0.95x 0.91x 0.85x 0.75x 0.59x

0.95x 0.94x 0.92x 0.88x 0.80x

0.95x 0.97x 1.00x 1.09x 1.53x

0.95x 1.00x 1.10x 1.49x

0.95x 1.04x 1.23x 2.50x

0.95x 1.08x 1.41x

10%

11%

12%

13%

14%

15%

16%

-20% 0% 20% 40% 60% 80%

Reinvestment rate

RO

NIC

6.5% 7.4% 8.2% 9.2% 10.1%

6.5% 7.6% 8.6% 9.8% 10.9%

6.5% 7.8% 9.0% 10.4% 11.7%

6.5% 8.0% 9.4% 11.0% 12.5%

6.5% 8.2% 9.7% 11.6% 13.3%

6.5% 8.4% 10.1% 12.2% 14.1%

10%

11%

12%

13%

14%

15%

16%

-20% 0% 20% 40% 60% 80%

Reinvestment rate

RO

NIC

22 June 2016


short-term fixed costs are considered variable costs. This analysis results in an internal

rate of return of 18% (assuming 5% inflation), which we then assume will be the return

on new invested capital Ipiranga will receive through 2030. Thereafter, as we did

before, we assume nominal RONIC will be 13%, roughly in line with cost of capital.

Figure 167 shows how ROIC fades over time under these assumptions, as new capital

is invested at lower return rates and how it compares to our explicit model for Ipiranga

extended to 2050. The quickly diminishing return rates illustrate the conservativeness

of our premises. This set of assumptions applied to our two-parts DCF results in an

enterprise value of BRL41.5bn, or c. BRL74.5/share as shown in Figure 168. The same

exhibit compares the enterprise calculated using the two-parts DCF methodology to

that obtained from the methodologies we will discuss below. At this point, we note that

the results are quite similar. One quick note is that in the near future it might be

reasonable to assume there is scalability in building a new station. So, if full scale

gains are considered, the IRRs of a new station may increase to as much as 29%. We

are conservatively not assuming that.

Figure 167: Ipiranga's ROIC Assumed to Diminish

Over Time Due to Conservative Assumptions in %

Figure 168: Ipiranga's EV per share assessed using

different valuation methodologies in BRL per share

Source: Company data, Credit Suisse estimate. Source: Credit Suisse estimates. Note: Values at 1Q16.

■ Value Drivers in Other Businesses. For the remaining businesses we followed our

initial valuation framework of using conservative premises and assumed a return on

new invested capital of 13% (6.1% real) in both 2020-2030 and from 2030 onwards.

The following exhibit shows how the valuation under these assumptions compares with

multiples (discussed in more detail below).

0%

5%

10%

15%

20%

25%

30%

2010

2012

2014

2016E

2018E

2020E

2022E

2024E

2026E

2028E

2030E

2032E

2034E

2036E

2038E

2040E

2042E

2044E

2046E

2048E

2050E

Model to 2050 Two-parts DCF

74.5

81.2

66.0

Two-parts DCF Model to 2050 EBITDA multiple

22 June 2016


Figure 169: EV per share by business assessed using different valuation

methodologies in BRL per share

Source: Credit Suisse estimate. Note: Values as of 1Q16.

■ Real Economy Implications, Model to 2050. We generally prefer DCF valuations

over multiples, as only then can we understand what real economy assumptions are

implied in share prices. In order to understand what our continuing value assumptions

for Ipiranga mean in terms of volume growth, market share, new gas station additions,

and other operational metrics, we have extended the outlook of our explicit model to

2050 (with adjustments to transform fixed costs into variable costs in the long term).

Below, we discuss the operational metrics that support the value drivers we discussed

before.

■ Moderate Volume Growth. The volume growth rates we forecast are nothing like the

ones enjoyed by Ipiranga recently, as shown in Figure 170. Over the past ten years,

volumes in the overall fuel distribution market grew by a CAGR of c. 8% in Otto cycle

fuels and c. 4% in diesel. Ipiranga grew even faster than the market. In 2010-2014,

which is after the acquisition of Texaco stations and before the economic recession,

Ipiranga grew volumes sold to service stations at a pace 2pps above the market due to

the expansion of its network of gas stations. Going forward we expect the market to

grow 2.5% in the long term, in line with GDP, and Ipiranga continues to gain share over

unbranded service stations as it adds stations to its own network, as shown in Figure

171. If we assume that other players in Sindicom will have similar performance, then

unbranded stations would be left with a market share of around 8-10% in 2050. Market

share figures in the charts do not account for the incorporation of AleSat stations.

9.18.5

4.1 3.9

1.6 1.3

3.32.3

Two-partsDCF

EBITDAmultiple

Two-partsDCF

EBITDAmultiple

Two-partsDCF

EBITDAmultiple

Model to2050

EBITDAmultiple

Oxiteno Ultragaz Ultracargo Extrafarma

22 June 2016


Figure 170: Ipiranga's Forecasted Volume Growth in %, year-on-year change

Figure 171: Ipiranga's Forecast Market Share in % of total market volume

Source: Company data, Credit Suisse estimates Source: Company data, Credit Suisse estimate. Notes: Assumes overall market grows in line with GDP of 2.5% in the long term

■ New Stations. To support the volume growth of the scenario described above, we

assume Ipiranga will open 450 stations per year from 2020 onwards. This level of

gross additions to the Ipiranga-branded gas stations network is not unprecedented. In

fact, it's about the same level of additions as in 2014 and below 2011 and 2012, when

the company added 537 and 496 stations, respectively. Furthermore, as the network

grows over time, and churn in the branded base remains 2% of total network, the

additions represent ever-smaller relative growth in number of stations, as shown in

Figure 172. Moreover, we assume that the volume per station in the following 35 years

will continue to converge to the average volume per station in the United States in the

early 2000s, as shown in Figure 173. These assumptions are, in our view,

conservative, and there may be additional upside here.

Figure 172: Forecasted Number of Ipiranga Gas

Stations thousand stations (LHS) and % year-on-year change (RHS)

Figure 173: Forecasted Volume per Station In cu m per station per year

Source: Company data, Credit Suisse estimates Source: Company data, Credit Suisse estimates, EIA, NACS

-5%

0%

5%

10%

15%

20%

20

10

20

12

20

14

20

16E

20

18E

20

20E

20

22E

20

24E

20

26E

20

28E

20

30E

20

32E

20

34E

20

36E

20

38E

20

40E

20

42E

20

44E

20

46E

20

48E

20

50E

Diesel Otto

0%

5%

10%

15%

20%

25%

30%

20

10

20

12

20

14

20

16E

20

18E

20

20E

20

22E

20

24E

20

26E

20

28E

20

30E

20

32E

20

34E

20

36E

20

38E

20

40E

20

42E

20

44E

20

46E

20

48E

20

50E

Diesel Gasoline Ethanol

0%

1%

2%

3%

4%

5%

6%

7%

8%

9%

10%

-

2

4

6

8

10

12

14

16

20

07

20

10

20

13

20

16E

20

19E

20

22E

20

25E

20

28E

20

31E

20

34E

20

37E

20

40E

20

43E

20

46E

20

49E

Gro

wth

in n

um

be

r o

f st

atio

ns

(Yo

Y %

)

Nu

mb

er

of

gas

stat

ion

s (t

ho

usa

nd

s)

# of stations # of stations growth

-

1,000

2,000

3,000

4,000

5,000

19

96

19

99

20

02

20

05

20

08

20

11

20

14

20

17E

20

20E

20

23E

20

26E

20

29E

20

32E

20

35E

20

38E

20

41E

20

44E

20

47E

20

50E

Ave

rage

Sal

es p

er S

tati

on

(m

3/y

)

CS Forecast

USA Actuals

IpirangaActuals

22 June 2016


■ Most Value In Ipiranga. An additional advantage of the sum-of-the-parts approach is

that it allows one to understand how value is distributed among the businesses.

Ipiranga's share of Ultrapar's overall value then becomes clear, as shown in the build-

up of the BRL93/share target price using the SOTP two-part DCF methodology

depicted in Figure 174.

Figure 174: Ultrapar's Valuation Using Sum-of-the-Parts DCF in BRL per share


■ What Is Priced In? We initially used conservative value driver assumptions to carry

out DCF valuations on a business-by-business basis. We then asked ourselves how

those premises translated into operational metrics in the long term for the most

important business within the group, Ipiranga. This exercise then led us to believe that

the operational outlook priced in it seems reasonable and may actually reserve some

upside in the future. Finally, we note that the valuation level according this

methodology is also similar to the DCF carried out at the group level, which allows for

the conclusion that Ultrapar's existing portfolio justifies the assumptions we had initially

adopted at the holding level, hence we are not pricing-in any future transformational

change.

Multiples

■ Sanity Check. We are skeptical about valuation by multiples because we cannot know

precisely what kind of operational outlook is embedded in them. However, this does not

mean multiples are of no use to us. In fact, we believe a valuable tool is to run sanity

checks on our DCFs. So we took a look at what UGPA's valuation would look like using

EV/EBITDA and P/E multiples, and then compared the results to our DCF.

■ EV/EBITDA Multiple. Ultrapar has traded from slightly below 10x to slightly above 11x

EV/EBITDA in recent years, as shown in Figure 175. Furthermore, volatility has been

higher recently due to the uncertainties brought by recent economic crisis. Our model

uses our forecasted EBITDA for 2016 and assumes a 11x multiple would result in a

BRL81/sh target price in 12 months' time.

74.5

9.1 4.1 1.6 3.3 12.5

1.4

11.1

92.6

Ipiranga Oxiteno Ultragaz Utracargo Extrafarma Corporate Net debt 12mo carryfwd

Ultrapar

22 June 2016


Figure 175: Ultrapar's EV/EBITDA Multiple Relative to Consensus in multiples

Source: Bloomberg, Credit Suisse Analysis

■ P/E Multiple. From an earnings perspective, valuation has oscillated between 20x and

24x P/E. Our model uses our forecasted earnings for 2016, which is 4% below

consensus, and assumes a 22x times PE multiple would result in a target price of

BRL66/sh in 12 months. It is worth mentioning, however, that 2016 is a tough year in

terms of the earnings growth in our forecast. This is why the uninspiring result of the

valuation based on 2016 earnings is not cause for concern.

Figure 176: Ultrapar's P/E Multiple Relative to Consensus in multiples

Source: Bloomberg, Credit Suisse Analysis

■ A Brief History of Ultrapar's Multiples. Looking solely at current multiples does not

capture the full picture, however. When we analyze a longer time horizon, we note that

Ultrapar's multiples have expanded quite dramatically over time. Figure 177 shows how

the market price has evolved relative to valuation levels that would have been

expected at different EV/EBITDA bands (depicted in six months' moving averages to

11x

7x

8x

9x

10x

11x

12x

13x

Mar-10 Mar-11 Mar-12 Mar-13 Mar-14 Mar-15 Mar-16

22x

12x

14x

16x

18x

20x

22x

24x

26x

28x

Mar-10 Mar-11 Mar-12 Mar-13 Mar-14 Mar-15 Mar-16

22 June 2016


reduce noise). The same analysis through an earnings perspective yields results that

are no different. Figure 178 depicts how Ultrapar's market value went from lower to

higher P/E bands. It is worthwhile to mention that this expansion in multiples has

happened since the acquisition of Ipiranga and has been more intense since the

acquisition of Texaco-branded stations in Brazil. This is no surprise. Ipiranga is a fast-

growing business that enjoys higher multiples than Ultragaz (LPG distribution), Oxiteno

(chemicals), and Ultracargo (liquid bulk storage). In fact, we expect this trend in

multiples expansion to continue as Ipiranga outgrows the other businesses within

Ultrapar and gains even more relevance within the holding's overall EBITDA and

earnings. The same rationale applies to Extrafarma, despite its limited importance for

the consolidated results at this time.

Figure 177: Ultrapar's Market Cap vs. EV/EBITDA

Bands in USD bn

Figure 178: Ultrapar's Market Cap vs. P/E Bands in USD bn

Source: Bloomberg, Credit Suisse analysis. Note: Six-month moving average. Source: Bloomberg, Credit Suisse analysis. Note: Six-month moving average.

■ Multiples Price-In Relatively Conservative Assumptions. The exercise of using

valuation by multiples has led us to, incidentally, find valuation levels that are similar to

the ones yielded by our DCF approach. This is interesting because it gives us a sense

that the operational assumptions implied in market multiples are similar to the ones we

had assumed before. We remind the reader that we have made our DCF assumptions

conservative in order to minimize downside risks. Therefore, if we were to incorporate

more optimistic assumptions in our valuation, we would find additional upside to market

prices. We think this corroborates the thesis that there is potential for renewed upside,

should recent past performance repeat itself.

Sum of the Parts: Multiples

■ Build-Up of Business Values. Finally, we propose one last approach to assessing

Ultrapar's value by combining the sum-of-the-parts approach and valuation by

multiples. The idea again is that the holding value must be the sum of the value of the

subsidiaries and that higher-growth businesses like Ipiranga and Extrafarma deserve

higher multiples. Once again, this is not meant to be a precise assessment of the

0

2

4

6

8

10

12

14

16

EV/EBITDA 11x

EV/EBITDA 9x

EV/EBITDA 7x

Mkt Cap

Ipiranga

acquisition

Texaco stations

acquisition

0

2

4

6

8

10

12

14

16

18

PE 25x

PE 20x

PE 15x

PE 10x

Mkt Cap

Ipiranga

acquisition

Texaco stations

acquisition

22 June 2016


company's value, but rather a sanity check. We have assumed 13x EV/EBITDA for

Ipiranga, 20x times for Extrafarma and 6x for the remaining businesses. The build-up

for Ultrapar value is shown in Figure 179. This valuation methodology leads to a fair

price of BRL 80/share in 12 months, in line with the previous methodologies, and

implies an EV/EBITDA multiple of c. 11x for the consolidated Ultrapar, also in line with

historical multiples.

Figure 179: Ultrapar's SOTP Multiples, Valuation Build-Up in BRL per share

Source: Credit Suisse estimates. Note: assumes normalized level of EBITDA for Ultracargo.

Scenarios Tool

■ Interactive Tool. We are making an effort to avoid paying too much for Ultrapar's

shares. In this regard, we tried a number of scenarios to observe the different results

within a range of assumptions and make sure all the possibilities would be covered. To

make this a bearable task, we created an interactive tool that Credit Suisse's clients

may find valuable. It allows one to input a variety of assumptions and easily observe

the sensitivities of Ultrapar's metrics.

■ User Friendly Inputs and Visual Outputs. The tool basically considers a set of the

most relevant inputs for each business and, after the user presses a button, provides

the adjusted forecasts for the next five years. This way, it is possible to create any

scenario with the user's own prediction of economics and industrial events, and

observe the impact they would have in each business of the company, based on

financial and operational metrics and charts. For example, Figure 180 shows the tool

displaying the financial metrics of Ultrapar, after setting a group of assumptions, and

Figure 181 shows the tool displaying the financial charts of Ipiranga.

■ Library of Scenarios. We also wanted an easy way to recover scenarios we had

tested previously, so we created a function in the tool to save and recover the

scenarios that the user found most valuable. After opening the tool with the CS

assumptions set, users can recreate a number of expected scenarios, and then save,

delete, or recover them whenever they want. Besides that, the tool generates a

different sheet with the values of the saved scenario, to make it easier to print out.

66.0

8.5

3.9 1.32.3

10.9

1.4

11.1

80.4

13.0x

6.0x 6.0x 6.0x

20.0x

10.7x

Ipiranga Oxiteno Ultragaz Utracargo Extrafarma Corporate Net debt 12mo

carry fwd

Ultrapar

EV/EBITDA multiple

22 June 2016


Figure 180: Screen Print of Interactive Tool to Create Different Scenarios for Ultrapar Forecasts

Source: Credit Suisse Research.

Figure 181: Screen Print of Interactive Tool to Create Different Scenarios for Ultrapar Forecasts

Source: Credit Suisse Research.

22 June 2016


Risks

How Can We Be Wrong?

"Risk means more things can happen than will happen." (Elroy Dimson)

■ A Different Time Scale. One of the main challenges in evaluating businesses, in our

view, is to try to figure out what the country will look like a few decades from now, and

also how consumer preferences and habits might evolve in the period. Although the

exercise is amusing, it is also difficult because the short-term fluctuations and setbacks

sometimes prevent us from appreciating the broad picture and more general trends

taking place on a different time scale. In 1960, for instance, Brazil had a nearly 50%

illiteracy rate (among people above 7 years of age). Back then, the population was

about 70 million people. In the four decades thereafter, this rate was reduced to 39%,

32%, 24%, and then to less than 10% in 2010. These numbers are still too high, of

course, but the general trend, measured on a scale of decades, meant an addition of c.

145 million people to the group that can read. This has enormous implications for

consumer preferences, the way things are sold, the way purchase decisions are made

and so on. So, identifying the bigger trends is mandatory as well as challenging, and

we try to find out where we may be wrong. As Shane Parrish recently wrote: "The way

to stand on the shoulders of giants is to start the day by telling yourself 'I can't wait to

correct my bad ideas, because then I'll be one step closer to reality.'"

■ What If There Is a Material Change In Consumer Preferences? Some habits

become too strong for us to even consider the possibility of them being broken. But,

sometimes a new generation comes and those habits are suddenly gone. When we

talk about mobility, we have a sense that, as income grows, Brazilians will keep

purchasing cars when they can afford them. As we have discussed at length, this is a

trend that happened in all developed economies we were able to analyze. However,

we could also conceive of a scenario in which new generations will gradually need to

drive less, as home offices and virtual social interactions may reduce the need for

mobility. So far this is not yet the case, as online technology has not hurt in any

meaningful way the consumption of oil products, but we might be wrong here.

■ What If Macro Conditions Get Worse? The current economic slowdown has already

affected Ultrapar in a number of ways, including above all other impacts a tougher

scenario for demand expansion. As we discussed at length in this report, volume

growth is a key variable to determine future value creation potential. We think that the

Brazilian consumption of Ultrapar's products and services will resume growth at some

point and that the company's moats will keep being defensive and preserving above-

average returns. But, we could also have conceived a scenario in which Brazil's long-

term economic activity gets impaired in a way that materially affects long-term demand.

In that case, sustaining high returns might become more challenging for the company.

We may have to reconsider our assumptions if we see any sign that this will indeed be

the case.

■ What If Electric or Other Alternative Vehicles Materialize Faster? Disruptive

technologies might anticipate the realities we expect only for the distant future. We

think the aggressive substitution of an entire fleet of combustion engine vehicles will

take decades and the establishment of a new infrastructure that makes this possible

will also require lots of capital. Besides, the whole industry built upon the consumption

of oil products also adapts to postpone its substitution for new energy types. But, we've

seen other technologies become viable, affordable, and more widespread than

22 June 2016


anybody would have anticipated. So, we might be also wrong if we see signs that other

types of vehicles, that no longer need service stations or distribution networks of any

sort, will suddenly be massively adopted.

■ What If Refineries Are Sold? The structure of the downstream sector in Brazil is very

unique. The concentration of most of the refining capacity in the country into a single

company makes the sector behave in very particular ways. If, for example, Petrobras

decides to sell a significant portion of its refining business, attracting different private

companies to it, we could see, for instance, an increase in the need for capital

deployment on the part of distributors, as a more competitive refining segment will

likely be less prone to invest in logistics capacity in the country. This would increase

the pressure on distributors, but would also mean new opportunities for these

companies to wisely create infrastructure and as a result expand market share. Price

fluctuations would also likely increase at the refinery gate, meaning more volatility in

distribution margins. We do not assume this as our base-case scenario, but we might

be wrong.

■ What If an Aggressive Company Comes In Through an Acquisition? Petrobras's

need to divest assets is widely known, and one of the initiatives involves the potential

sale of BR Distribuidora. If the company decides to sell a controlling stake, the new

controlling shareholder might adopt a more aggressive branding strategy, open a

greater number of convenience stores, or even engage in market share expansion and

accept lower distribution margins. We didn't take into account any irrational pricing or

practice that departs from the current competitive ones, as we think it is not in the

interest of any of the big distribution companies to pursue price wars or make moves

that could reduce the industry returns, eroding the benefits of the oligopoly. However,

we might be wrong in that assumption.

■ What If Railroads Take On a Big Portion of Cargo Transportation? We think

railroads might receive new invested capital in the coming years, and this should

definitely lead the existing railroads, and occasionally new ones, to capture part of the

cargo currently moved by trucks. As we discussed, railroads are much more efficient in

diesel consumption relative to trucks in proportion to the cargo moved. This means a

potential risk for future demand growth for diesel. In our view, a higher share of

railroads in the transportation matrix will be contingent on lots of investments actually

taking place and will also be limited to long-distance transportation of low value-added

goods. But, we don't see the transportation matrix changing dramatically and rapidly in

a timeframe that would hurt the potential for volume growth we see in the distribution

business. But we might be wrong.

■ What If New Taxes or Other Regulations Go Against Ultrapar? We cannot

anticipate whether a new administration will significantly increase taxes in such a way

as to impair Ultrapar's pricing power in the markets it serves. Regulations and

inspections can also affect the company, by either reducing its moats or failing to avoid

adulteration, tax evasion, and other anti-competitive practices from taking place in the

sectors in which Ultrapar operates.

■ What If Competitors Merge and Get Stronger? We think consolidation continues to

be a trend in the fuels distribution and in the LPG distribution businesses. The AleSat

deal not being approved for any reason would be bad news, but not exactly a risk to

our call, because we didn't incorporate that value to our numbers. The prices involved

in AleSat's deal make a lot of sense to us, by the way, and, if the deal is not approved,

it will be a missed chance of a leap ahead of competition. We didn`t consider that

incremental value in our valuations, but we like the prospects of the acquisition

anyway. Another similar risk concerns the sale of Liquigas, Petrobras's LPG

distribution subsidiary, in the case one competitor buys it, creating a bigger player. Any

22 June 2016


risk of this sort could result in higher competitive pressure over Ultrapar from the

company's peers, reducing the possibilities for market share gains and economies of

scale. We think Ultrapar has been able to deal very well with the strategic and

commercial moves from its main competitors in different fields. We see no point in

assuming it won`t be the case in the future, at least until we see a clear sign that this

could be so. But we might be wrong.

■ What If Ultrapar Deploys Capital Poorly? We've already seen that Ultrapar's track

record as capital allocator was extremely good. This, in our view, does not depend only

on the talent of individual managers but also on a business culture in which economic

value added is a key driver of decisions. So, we think this lends more resilience to the

company's ability to keep deploying capital very well. Still, we cannot anticipate

whether at some point in the future the company may pursue an investment

opportunity that later proves to have destroyed value for shareholders. If that happens,

we would have to reevaluate the investment case.

■ What If Other What Ifs Arise? We've seen countless times new facts arising where no

one could anticipate, and great business environments turn into lousy ones. So, we

know the future might reserve changes that we can't imagine today that could

transform Ultrapar into a not-so-great investment. For these, we can't be ready and

can't be sorry, we can only be alert.

22 June 2016


Appendix

Appendix 1: Ultrapar Through HOLT® Lens

■ What Is HOLT? HOLT is Credit Suisse's proprietary valuation framework based on

future discounted cash flows. It uses cash flow returns on investment (CFROI®) as a

key performance and valuation metric. A key attractiveness of HOLT, in our view, is the

ability of the methodology to bring together performance and valuation of companies in

a wide number of geographies, time periods, and different accounting standards under

single and comparable metrics. Another useful feature of HOLT is that it is easy to

compare what type of returns are embedded in current share prices for different stocks

and compare those returns with what the consensus is expecting for the next few

years. We provide further detail on the HOLT methodology in Appendix 2.

■ What Is CFROI? CFROI is an approximation of the economic return, or an estimate of

the average real internal rate of return, earned by a firm on the portfolio of projects that

constitute its operating assets. A firm's CFROI can be directly compared against its real

cost of capital (the investors' real discount rate) to see if the firm is creating economic

wealth.

■ Valuation: Ultrapar's CFROI has risen every year since 2008 (approximately eight

continuous years of CFROI expansion), something that has only been attained by a

tiny fraction of all global companies covered by HOLT (only 1.5% have reached this

growth for five consecutive years or more). Therefore, the company has managed to

consolidate its CFROI levels above its cost of capital (green line in Figure 182) for six

consecutive years now, a trend that is supposed to continue according to HOLT

forecasts (pink bars in chart). They show a CFROI increase from 10.18% in 2015 to

10.52% in 2017 and a decrease in cost of capital, suggesting that Ultrapar is expected

to continue creating value.

■ Ultrapar Screens Expensive: However, Ultrapar screens expensive in HOLT

Economic PE, Figure 183. HOLT Economic P/E is the ratio of market enterprise value

to net operating cash flow. Similar to the traditional PE, which reflects the price paid

per dollar of earnings, the HOLT Economic PE reflects the price paid by all capital

providers per dollar of net operating cash flow.

Figure 182: Ultrapar's CFROI In %

Figure 183: Economic PE multiple

Source: Credit Suisse HOLT Lens™. Source: Credit Suisse HOLT Lens™.

■ Sell-Side vs. Buy-Side Expectations: Holt gives an interesting possibility of

comparing the market-implied expectations (buy side) over time with the expectations

22 June 2016


embedded in consensus estimates (sell side). In the case of Ultrapar, the current

market-implied expectations (green dots on Figure 184) is at 12.2%, the highest value

since 2005. On the other side, the sell-side consensus (pink bars on the chart) stands

slightly below, at 9.8%, suggesting the shares may still be overpriced.

Figure 184: UGPA's Market-Implied CFROI (Green Dots) vs. One-Year CFROI Levels Implied in Consensus

(Pink Bars) in %

Source: Credit Suisse HOLT Lens™.

22 June 2016


Appendix 2: The HOLT® Framework

Overview

The HOLT Valuation Framework is based on two fundamental principles:

■ The market pays for economic (cash) performance and not accounting performance.

■ The value of a company is determined by its discounted future cash flows over its life

cycle.

The HOLT methodology uses a proprietary performance measure known as Cash Flow

Return on Investment (CFROI®).

The HOLT Valuation Framework uses CFROI to estimate future cash flows and applies a

unique notion of life cycle fade to reflect the position of any individual company on its

industrial life cycle.

Figure 185: HOLT Framework

Source: Credit Suisse HOLT®.

Cash Flow Return on Investment® (CFROI®)

Why use CFROI? Accounting statements often present a distorted view of underlying

economic performance. In order to better define a cash measure, HOLT's economic

CFROI corrects for the distortions found in traditional accounting-based measures of

performance by adjusting for inflation, off-balance sheet assets (e.g., leased property),

depreciation, LIFO & FIFO accounting, asset mix, asset holding gains or losses, asset life,

acquisition accounting, deferred taxes, pensions, investments, revaluations, special

reserves, research & development, and others.

As a result, CFROI provides comparability over time, among companies, and across

industries and national borders. This proprietary measure focuses on the cash economics

of businesses. Once the economics of the company are understood, we can more

accurately determine value by taking into account expected future cash flows, asset

growth rates, discount rates and life cycles.

HOLT's CFROI is calculated in two steps. First, compare the inflation-adjusted (current

dollar) cash flows available to all capital owners in the company to the inflation-adjusted

(current dollar) gross investment made by those capital owners.

Next, translate the ratio of gross cash flow to gross investment to an internal rate of return

(IRR) by recognizing the finite economic life of depreciating assets and the residual value

of non-depreciating assets such as land and working capital. The process is identical to

calculating the yield to maturity for a bond. As a percent per year IRR, CFROI is directly

comparable to the return investors expect to receive (i.e., the cost of capital or discount

rate).

Accounting Cash Value

Income Statement

Balance Sheet

EPS, ROE, ROCE

Cash Flow Return on

Investment (CFROI)

Economic Performance

CFROI

Asset Growth

Life Cycle Fade

Discount Rate

22 June 2016


Figure 186: From Accounting to CFROI


Value Creation

Companies can create wealth for shareholders by making the right decisions with respect

to CFROI and asset growth.

Wealth is created when companies:

■ Improve CFROI

■ Grow assets when CFROI is above the discount rate (positive spread)

■ Shrink assets when CFROI is below the discount rate (negative spread)

$100

Inflation Adjusted

Gross Investment

13 - Year Asset Life

$10

$25

Gross Cash Flow

CFROI = 6.0%

Non-Depreciating

Assets

Cash+ Land+ Accounts Receivable+ Inventory

Book Assets

+ Accumulated Depreciation+ Inflation Adjustment to Gross Plant+ Inflation Adjustment to LIFO Inventory+ Operating Leases+ Capitalized R&D- Goodwill

+/ - Unrealized Gains/Losses on Investments- Non -debt Monetary Liabilities & Deferred Tax Asset

Net Income (Before Extraordinary Items)+/ - Special Items (after tax)+ Depreciation/Amortization Expense+ R&D Expense+ Interest Expense+ Minority Interest Expense+ Rental Expense+/ - Holding Gains (Losses)- FIFO Profits

22 June 2016


Figure 187: Managing for Shareholder Value


HOLT Valuation

The HOLT valuation model, at its foundation, is a type of DCF (discounted cash flow)

model. Among the model’s distinguishing features, along with the CFROI metric, is the

way in which the forecasted stream of net cash receipts (NCRs) is generated and the

method in which the firm’s discount rate (DR) is estimated.

From a beginning asset base, key variables that drive the forecast NCR stream are

variables that actually generate cash flows, namely, economic returns (CFROIs),

reinvestment rates (growth), and their expected patterns of change over time due to

competition (fade). The competitive life cycle is covered in the following section.

The discount rate is the rate of return investors demand for making their funds available to

the firm. DRs used in our model are real rates, not nominal rates, so they are consistent

with CFROIs. The DRs are also consistent with other aspects of our model, since base

DRs are mathematically derived from known market values and from NCR streams

consistent within our model. Adjustments (positive or negative) to the base rate are made

for company-specific financial and liquidity risk characteristics.

The result of discounting the NCRs at the market-derived DR is what is referred to as a

warranted value or warranted price. Essentially, this is the valuation that results (or is

“warranted”), given the default (or users’ own) assumptions built into the forecast and the

resulting present value of the NCRs plus the value of any non-operating investments.

Cash Flow

Return (%)

(CFROI)

Strategic

Options

Discount Rate

(Cost of Capital)

Positive

Spread

Increase or hold CFROI

Grow Assets

Neutral

Increase CFROI

Then Grow

Negative

Spread

Increase CFROI

Contract Assets

22 June 2016


Figure 188: Major Components of the HOLT Valuation Model


Economic Life Cycles

In evaluating CFROI and asset growth rates, HOLT found that over long periods of time,

companies tend to follow an industrial life cycle. Competition tends to force firms’ real

economic returns toward the corporate sector’s average CFROI. (See Figure 189.)

HOLT empirical research has found that in the US and other industrialized economies,

these aggregate CFROIs have been averaging about 6%. (See Figure 190.) Thus, a

warranted price or market price can be viewed as implying a firm’s potential future life

cycle of CFROIs and growth rates that will eventually regress to the average aggregate

economic level.

The benefits of using the HOLT’s framework can be summarized as follows:

■ The framework eliminates accounting distortions, allowing investors to understand

levels and changes in stock prices worldwide.

■ The framework provides comparability over time, among companies, and across

international borders.

■ HOLT users benefit from a common language for measuring track records, making

forecasts, and calibrating market expectations.

Warranted

Price

creates

+

+

–

–

÷

NPV of Existing Assets

NPV of Future Invests

NPV of Net Cash Receipts

MV of Non-Op. Assets

Total Enterprise Value

MV of Debt

Total Equity Value

Minority Interest

Common Equity Value

Adjusted Shares

Com Equity / Share

n

t=1

= Σ

Revenue Growth

Operating Margins

Asset Turns

Net Cash Receipts

(1+Disc. Rate) t

Country Base Rate

Size Differential

Leverage Differential

Growth

CFROI

Fade

Asset Base

22 June 2016


Figure 189: A Company’s Typical Industrial Life Cycle


Figure 190: Today’s Price Recognizes Performance Fades Under Competitive

Pressures


Value/Cost =

This Research Allows Us to Accurately Analyze the Expected Pattern of Growth and Fade Built into a

Company’s Share Price

Discount Rate

(Investor’s Required

Rate of Return)

Below-Average

CFROIs

Average

CFROIs

High Premium Premium DiscountPar

SickMatureFading

Above-Average

but Fading CFROIs

Growth

Increasing CFROIs &

High Reinvestment

CFROI

CFROI Competitive Fade Toward Corporate

Average

Growth Competitive Fade Toward Corporate

Average

Conclusion:

Competition causes CFROIs and real asset growth rates to regress to the mean (Benchmarks:

CFROIs = 6.0% and Asset Growth = 2.5%)

Positive and negative surprises vs fade expectations cause significant changes in stock prices

High Growth Firm

Low Growth Firm

High CFROI Firm

Low CFROI Firm

2.5%6.0%

22 June 2016


Appendix 3: Management Profile and Holdings

Board of Directors

Ultrapar's board of directors is composed of a minimum of five and maximum of nine

members, all of whom with a unified term of office of two years, reelection permitted. The

members are elected at the annual general shareholders' meeting.

The company's bylaws require that at least 30% of the members of the board of directors

be independent, in accordance with the independence requirements of the rules of the

Novo Mercado segment. In addition, the bylaws set forth that the election of the board

members must be made through the nomination of a slate of candidates, unless

cumulative voting is requested. Only the following slates of candidates will be eligible: (i)

those nominated by the board of directors; or (ii) those nominated by any shareholder or

group of shareholders.

Figure 191: Ultrapar's Board of Directors

Source: Ultrapar

■ Paulo Guilherme Aguiar Cunha: Mr. Cunha was CEO of Ultrapar from 1981 to 2007

and now has been the chairman of the company's board of directors since 1998.

Furthermore, he has been a member of the board of directors of Monteiro Aranha since

1997, a member of the consulting board of the Brazilian Association of Chemical

Industries (Abiquim), of the board of BNDESPAR, of Insper – IBMEC Business School,

of the Technological Research Institute (IPT), among others. In addition, Mr. Cunha is

currently president of Brazilian Association of Technical Standards and president of

Brazilian Petroleum Institute (IBP).

■ Lucio de Castro Andrade Filho: Mr. Andrade Filho has been at Ultrapar for 39 years,

serving as Ultrapar's vice president from 1982 to 2006 and becoming vice chairman of

the board of directors of the company in 1998. Previously, he was the CEO of GLP –

Qualidade Compartilhada and a member of the board of directors of the Brazilian

Petroleum Institute.

■ Alexandre Gonçalves Silva: Mr. Silva has been a member of Ultrapar's board of

directors since April 2015, chairman of the board of directors of Embraer S.A. since

2012, and member of the board of directors of Fibria Celulose S.A. since 2011.

Previously, he was CEO of General Electric in Brazil.

■ Carlos Tadeu da Costa Fraga: Mr. Fraga has been a member of board of directors

since April 2015. Prior to that, he was a member of the Technological Park Council of

the Federal University of Rio de Janeiro, a member of the advisory board of the

Technological Research Institute of São Paulo, and executive manager of exploration

and production of the pre-salt of Petrobras.

Name Position Years with the Company Age Independent member

Paulo Guilherme Aguiar Cunha Chairman 49 76 No

Lucio de Castro Andrade Filho Vice Chairman 39 71 No

Alexandre Gonçalves Silva Director 1 71 No

Carlos Tadeu da Costa Fraga Director 1 58 No

Jorge Marques de Toledo Camargo Director 1 61 Yes

José Maurício Pereira Coelho Director 1 49 Yes

Nildemar Secches Director 14 67 Yes

Olavo Egydio Monteiro de Carvalho Director 13 74 Yes

Pedro Wongtschowski Director 38 70 Yes

22 June 2016


■ Jorge Marques de Toledo Camargo: Mr. Marques has been a member of the board

of directors of Prumo Logística S.A. since 2014, a senior consultant at McKinsey & Co.,

Inc. since 2012, and a member of the board of directors of Mills Estruturas e Serviços

de Engenharia S.A. since 2011. Previously, he was a member of the board of directors

of Deepflex do Brasil Ltda. from 2010 to 2013.

■ José Maurício Pereira Coelho: Mr. Coelho joined Ultrapar as a member of the board

of directors in April 2015. He is also vice president of financial management and

investor relations of Banco do Brasil S.A., a member of the board of directors of Grupo

Neoenergia, a member of the board of directors of BB Securities LLC, New York, and a

member of the board of directors of Cielo S.A.

■ Nildemar Secches: Mr. Secches is also a member of the board of directors of Itaú

Unibanco Holding S.A., of WEG S.A., and of Cia Suzano Papel and Celulose.

Previously, he was the CEO of Empresas Perdigão S.A., of Grupo Iochpe-Maxion, and

of Brazilian Association of Chicken Producers and Exporters (Abef). He also served as

an executive officer of the National Bank for Economic and Social Development

(BNDES) and as Chairman of the board of directors of Brasil Foods S.A.

■ Olavo Egydio Monteiro de Carvalho: Mr. Monteiro de Carvalho is also chairman of

the board of directors of Monteiro Aranha S.A. and a member of the board of directors

of Klabin S.A. In addition, he is a member of the Municipal Development Advisory

Board, member of the board of directors of Agência Rio-Negócios, among others.

■ Pedro Wongtschowski: Mr. Wongtschowski joined Ultrapar in 1985 and held the

positions of president and CEO of the company from January 2007 to December 2012.

Currently, he is also chairman of the boards of the National Center for Energy and

Materials Research, the National Association for Research, Development and

Engineering of Innovative Companies, and the Brazilian Association of Industrial

Research and Innovation (Emprapii). Furthermore, Mr. Wongtschowski is a member of

the board of directors of Embraer S.A., Companhia Nitroquímica Brasileira, among

other privately held companies and non-profit organizations.

Executive officers

The Ultrapar's executive committee is comprised of a minimum of four and a maximum of

eight members, including the chief executive officer. Members of the executive committee

are appointed for a two-year term of office and can be reelected.

Figure 192: Ultrapar's Officers

Source: Ultrapar

■ Thilo Mannhardt: Mr. Mannhardt has been CEO of Ultrapar since January 2013, after

almost two years as a member of the company's board of directors. Previously, he was

a senior partner and director of McKinsey & Co. from 1985 to 2012.

Name Position Years with the Company

Thilo Mannhardt Chief Executive Officer 5

André Pires de Oliveira Dias(2) Chief Financial and Investor Relations Officer, Ultrapar 0

João Benjamin Parolin Officer, Oxiteno 30

Leocadio de Almeida Antunes Filho Officer, Ipiranga 8

André Covre Officer, Extrafarma 12

Pedro Jorge Filho Officer, Ultragaz 38

Ricardo Isaac Catran Officer, Ultracargo 36

22 June 2016


■ André Pires de Oliveira Dias: Mr. Pires has been chief financial and investor relations

officer since July 2015. Previously, he worked in the financial and capital markets at

Banco Geral de Comércio, Montgomery Asset Management, Banco BBA Creditanstalt,

and BNP Paribas. Later, he joined Gerdau S.A. in 2004 and served as executive vice

president of the executive committee and CFO from 2013 to 2015.

■ João Benjamin Parolin: Mr. Parolin joined Ultrapar in 1986, serving as Oxiteno's

marketing manager, sales manager and sales director before becoming chief operating

officer in 2007. Prior, he worked in the sales division at Dow Química S.A.

■ Leocadio de Almeida Antunes Filho: Mr. Antunes Filho has been an officer of

Ultrapar since May 2008. He has held a number of positions at Ipiranga since 1987,

when he joined Fertisul S.A., and has also worked at Ipiranga Serrana Fertilizantes

S.A. Currently Mr. Antunes is also a member of the board of Brazilian Association of

Fuel and Lubricant Distributors (Sindicom), of the board of directors of the Brazilian

Petroleum Institute, and of the Association of Convenience and Fuel Retailers.

■ André Covre: Mr. Covre joined Ultrapar in 2003 and has been the chief operating

officer of Extrafarma since July 2015, after serving as chief financial and investor

relations officer of the company. Prior to that, he was the chairman and vice chairman

of the Latin American Corporate Governance Roundtable and Companies Circle, after

serving as treasurer of Pepsi Cola Engarrafadora in Brazil and as director of ABN

AMRO Capital in Amsterdam.

■ Pedro Jorge Filho: Mr. Jorge Filho has been in Ultrapar since 1977 and held a

number of positions at the company, including officer at Utingás and director of

engineering and marketing at Ultragaz before becoming Ultragaz's chief operating

officer in 2002. In addition, he is an officer at the Brazilian Association of Distributors of

Liquefied Gas (Sindigás) and at Associación Iberoamericana de Gás Licuado de

Petróleo (AIGLP), first vice president of the World LP Gas Association – Paris, and

CEO of GLP – Qualidade Compartilhada.

■ Ricardo Isaac Catran: Mr. Catran has been Ultracargo's chief operating officer since

2008. He joined the company in 1980 and since then has held a number of positions,

including sales officer at Transultra and Tequimar for the Northeast region.

Shareholders

In accordance with Ultrapar's bylaws, its authorized and outstanding capital stock is

comprised solely of common shares. They entitle their holders to voting rights on any

matter. As of December 31, 2015, Ultrapar’s capital stock was composed of 556,405,096

common voting shares, as shown below:

Figure 193: Ownership of Ultrapar's Total Shares as of December 31, 2015

Source: Ultrapar

Shareholders Shares %

Ultra S.A. Participações 119,760,601 22%

Aberdeen Asset Management PLC 65,205,340 12%

Parth do Brasil Participações 42,833,956 8%

Others 328,605,199 59%

Total 556,405,096 100%

22 June 2016


As of December 31, 2015, Ultra S.A. owned 22% of Ultrapar's voting shares. Therefore, its

equity structure was as follows:

Figure 194: Capital stock of Ultra S.A. as of December 31, 2015

Source: Ultrapar

Shares % Shares % Shares %

Shareholders

Ana Maria Levy Villela Igel 9,764,689 17% 12,395,100 52% 22,159,789 27%

Fabio Igel 7,518,770 13% 1,768,275 7% 9,287,045 11%

Marcia Igel Joppert 6,201,602 11% 2,062,988 9% 8,264,590 10%

Joyce Igel de Castro Andrade 5,916,246 10% 1,262,989 5% 7,179,235 9%

Rogério Igel 3,634,187 6% 130,519 1% 3,764,706 5%

Christy Participações Ltda 4,049,199 7% 4,990,444 21% 9,039,643 11%

Others 4,140,702 7% 1,332,571 6% 5,473,273 7%

Total Shareholders 41,225,395 72% 23,942,886 100% 65,168,281 80%

Directors and officers

Paulo Guilherme Aguiar de Cunha 10,654,109 19% - 0% 10,654,109 13%

Lucio de Castro Andrade Filho 3,775,470 7% - 0% 3,775,470 5%

Pedro Wongtschowski 1,606,301 3% - 0% 1,606,301 2%

Total Directors and Officers 16,035,880 28% - 0% 16,035,880 20%

Total 57,261,275 100% 23,942,886 100% 81,204,161 100%

Common Preferred Total

22 June 2016


Appendix 4: Ultrapar Financials

Figure 195: Ultrapar Summary P&L, CF and BS – Credit Suisse estimates


Group P&L 2013 2014 2015 2016E 2017E 2018E 2019E 2020E

Revenues BRLm 60,940 67,736 75,655 79,817 89,223 100,729 114,042 125,679

COGS BRLm (56,165) (62,305) (68,934) (72,838) (81,456) (92,154) (104,559) (115,273)

Gross profit BRLm 4,775 5,432 6,722 6,979 7,767 8,574 9,483 10,406

SG&A BRLm (2,769) (3,289) (3,838) (4,100) (4,451) (4,881) (5,356) (5,889)

Others BRLm 98 107 51 143 110 115 120 126

Asset sales BRLm 40 37 27 28 28 28 28 28

EBIT BRLm 2,144 2,287 2,962 3,051 3,454 3,837 4,276 4,671

Financials BRLm (338) (445) (703) (895) (983) (1,000) (1,023) (1,043)

Associates BRLm (5) (16) (11) (14) (14) (14) (14) (14)

PBT BRLm 1,801 1,825 2,247 2,142 2,457 2,823 3,238 3,614

Taxes and tax benefits BRLm (573) (573) (734) (685) (835) (960) (1,101) (1,229)

PAT BRLm 1,229 1,251 1,513 1,456 1,622 1,863 2,137 2,385

Minorities BRLm (4) (10) (10) (10) (10) (10) (10) (10)

Net income BRLm 1,225 1,242 1,503 1,447 1,612 1,853 2,128 2,376

Cash-flow statement 2013 2014 2015 2016E 2017E 2018E 2019E 2020E

EBITDA BRLm 2,918 3,158 3,953 4,119 4,592 5,033 5,538 6,019

Tax BRLm (573) (573) (734) (685) (835) (960) (1,101) (1,229)

Other non cash-adjustments BRLm (38) (53) 705 (386) 0 0 0 0

Cash earnings BRLm 2,308 2,532 3,924 3,048 3,757 4,073 4,437 4,790

Working capital BRLm (186) 99 (455) (371) (585) (661) (766) (675)

CFO BRLm 2,122 2,631 3,469 2,677 3,172 3,412 3,671 4,115

Capex less disposals BRLm (1,102) (1,216) (1,334) (1,792) (1,518) (1,629) (1,815) (1,954)

Acquisitions BRLm (30) (29) (41) (41) (41) (41) (41) (41)

Others BRLm 0 0 0 0 0 0 0 0

CFI BRLm (1,131) (1,244) (1,375) (1,833) (1,559) (1,670) (1,856) (1,996)

FCFF BRLm 991 1,387 2,094 844 1,612 1,741 1,815 2,119

Debt issuance BRLm 1,446 1,816 2,385 1,098 3,394 3,166 1,156 1,156

Debt repayment BRLm (761) (931) (2,825) (1,098) (3,394) (3,166) (1,156) (1,156)

Net financials BRLm (308) (273) (429) (629) (710) (724) (745) (762)

Dividends BRLm (711) (783) (832) (868) (822) (982) (1,144) (1,313)

Others BRLm (245) (368) (820) (115) 0 0 0 0

CFF BRLm (579) (539) (2,521) (1,612) (1,532) (1,707) (1,889) (2,074)

FX and other adjustments BRLm 0 (0) 0 0 0 0 0 0

Change in cash BRLm 412 847 (427) (769) 80 35 (74) 45

Balance sheet 2013 2014 2015 2016E 2017E 2018E 2019E 2020E

Cash and equivalents BRLm 3,425 4,269 3,506 2,738 2,818 2,853 2,778 2,823

Receivables BRLm 2,322 2,604 3,167 3,436 3,879 4,379 4,959 5,467

Inventories BRLm 1,593 1,925 2,495 2,707 3,056 3,450 3,907 4,307

Others BRLm 565 704 743 743 743 743 743 743

Current assets BRLm 7,904 9,502 9,911 9,624 10,495 11,425 12,388 13,340

PP&E BRLm 7,029 8,250 8,733 9,483 9,889 10,350 10,930 11,564

LT receivables BRLm 124 144 152 152 152 152 152 152

Associates BRLm 59 70 104 104 104 104 104 104

Others, inc financial investments BRLm 1,262 1,514 2,066 2,066 2,066 2,066 2,066 2,066

Non-current assets BRLm 8,475 9,978 11,055 11,805 12,211 12,671 13,252 13,885

Total assets BRLm 16,379 19,480 20,966 21,429 22,707 24,096 25,639 27,225

Current debt BRLm 1,830 3,442 1,098 1,098 1,098 1,098 1,098 1,098

Suppliers BRLm 969 1,280 1,461 1,571 1,777 2,011 2,282 2,514

Payroll BRLm 298 295 404 404 404 404 404 404

Others BRLm 668 676 871 871 871 871 871 871

Current liabilities BRLm 3,764 5,692 3,833 3,944 4,150 4,383 4,655 4,887

Non-current debt BRLm 5,140 4,933 7,804 8,069 8,342 8,617 8,896 9,177

Pension BRLm 99 108 113 113 113 113 113 113

Others BRLm 828 1,020 1,242 740 740 740 740 740

Non-current liabilities BRLm 6,067 6,062 9,159 8,923 9,195 9,471 9,749 10,030

Equity BRLm 6,520 7,698 7,945 8,524 9,314 10,185 11,169 12,232

Minorities BRLm 27 29 29 39 48 58 67 77

Total equity BRLm 6,547 7,727 7,974 8,562 9,362 10,242 11,236 12,308

FX and other djustments BRLm 0 0 0 0 0 0 0 0

Total equity + liabilities BRLm 16,379 19,480 20,966 21,429 22,707 24,096 25,639 27,225

22 June 2016


Appendix 5: Financials Summary

Figure 196: Ultrapar Financial Summary

Source: Company data, Credit Suisse estimates (Prices as of 14/06/2016)

BASICS COMPANY DESCRIPTION

Sector Fuel Distribution / Others Ticker UGPA3

Price (R$) 71.0

Target (R$) 89.0

Recommendation OUTPERFORM

Mkt. cap. (R$ mn) 39,510 POSITIVES

Average daily volume (R$ mn) 79

SHAREHOLDING STRUCTURE Shares %

Ultra S.A. Participações 119.7 21.5%

Path Investments Company 42.8 7.7%

Previ 32.0 5.8% NEGATIVES

Aberdeen Asset Management PLC 24.2 4.4%

Others 337.3 60.7%

TOTAL 556.0

HISTORICAL 1Y FWD PE EBITDA BY SEGMENT

P&L (R$ mn) 2013A 2014A 2015A 2016E 2017E 2018E SELECTED METRICS 2013A 2014A 2015A 2016E 2017E 2018E

Revenues 60,940 67,736 75,655 79,817 89,223 100,729 Ipiranga total volumes (th m3) 24,758 25,614 25,725 25,456 26,052 26,879

COGS (56,165) (62,305) (68,934) (72,838) (81,456) (92,154) Ipiranga EBITDA/m3 (R$/m3) 82 89 108 111 126 136

SG&A and other expenses (2,631) (3,145) (3,760) (3,929) (4,313) (4,738) Ipiranga EBITDA (R$m) 2,030 2,288 2,769 2,826 3,276 3,666

EBIT 2,144 2,287 2,962 3,051 3,454 3,837 Oxiteno EBITDA (R$m) 441 404 740 786 762 750

EBIT margin 3.5% 3.4% 3.9% 3.8% 3.9% 3.8% Ultragaz EBITDA (R$m) 281 306 357 359 359 383

EBITDA 2,918 3,158 3,953 4,119 4,592 5,033 Ultracargo EBITDA (R$m) 158 167 26 119 120 128

EBITDA margin 4.8% 4.7% 5.2% 5.2% 5.1% 5.0% Extrafarma EBITDA (R$m) - 30 29 28 74 104

Net financial results (338) (445) (703) (895) (983) (1,000) LEVERAGE 2013A 2014A 2015A 2016E 2017E 2018E

Associates (5) (16) (11) (14) (14) (14) Adj. Net debt / EBITDA 1.2 1.3 1.4 1.6 1.4 1.4

Taxes (573) (573) (734) (685) (835) (960) Net debt / (ND + Equity) (%) 35.2% 34.8% 40.4% 43.0% 41.6% 40.3%

Minorities (4) (10) (10) (10) (10) (10) EBITDA / Interest expense (8.6) (7.1) (5.6) (4.6) (4.7) (5.0)

Net income 1,225 1,242 1,503 1,447 1,612 1,853 Capex / Cash from operations (0.5) (0.5) (0.4) (0.7) (0.5) (0.5)

Number of shares (millions) 544 556 556 556 556 556 RETURN / YIELD 2013A 2014A 2015A 2016E 2017E 2018E

EPS (R$/shares) 2.25 2.23 2.70 2.60 2.90 3.33 ROA 9.6% 8.8% 10.6% 11.0% 11.5% 11.9%

CASHFLOW (R$ mn) 2013A 2014A 2015A 2016E 2017E 2018E ROIC 14.7% 13.6% 15.4% 14.8% 15.2% 15.7%

Cash from operations 2,122 2,631 3,469 2,677 3,172 3,412 ROE 18.8% 16.1% 18.9% 17.0% 17.3% 18.2%

Cash from investments (1,131) (1,244) (1,375) (1,833) (1,559) (1,670) Dividend Yield 1.8% 2.0% 2.1% 2.2% 2.1% 2.5%

Cash from financing (579) (539) (2,521) (1,612) (1,532) (1,707) FCF Yield 1.7% 2.8% 4.2% 0.5% 2.3% 2.6%

FCFF 991 1,387 2,094 844 1,612 1,741 Wacc (R$, nominal) 13.8% 13.8% 16.2% 13.3% 12.5% 12.4%

FCFE 683 1,114 1,665 215 902 1,017 Ke (R$, nominal) 15.7% 15.6% 20.4% 15.6% 14.5% 14.4%

BALANCE SHEET (R$ mn) 2013A 2014A 2015A 2016E 2017E 2018E VALUATION 2013A 2014A 2015A 2016E 2017E 2018E

Cash and equivalents 3,425 4,269 3,506 2,738 2,818 2,853 PE 32.2 31.8 26.3 27.3 24.5 21.3

Total assets 16,379 19,480 20,966 21,429 22,707 24,096 P/B 6.1 5.1 5.0 4.6 4.2 3.9

Debt 6,970 8,375 8,902 9,167 9,440 9,715 EV / EBITDA 14.8 13.8 11.4 11.2 10.0 9.2

Equity 6,520 7,698 7,945 8,524 9,314 10,185 EV / IC 4.3 3.7 3.4 3.1 2.9 2.7

Ultrapar is a company with 75 years of history, and leading positions in the markets where it operates: fuel

distribution via Ipiranga and Ultragaz, specialty chemicals via Oxiteno, liquid bulk storage services via

Ultracargo, and drugstores with the recent acquisition of Extrafarma.

Proven track record of value creation (27% p.a. total shareholder return past 10 years).

High quality defensive business.

Exposed to growing businesses.

High levels of corporate governance practices.

High quality and well aligned management.

Valuation levels dependent on growth.

Natural difficulty in allocate capital at higher returns as the business grows strongly.

Risks of tougher competition.

Ipiranga68%

Oxiteno19%

Ultragaz9%

Ultracargo3%

Extrafarma1%

22x

12x

14x

16x

18x

20x

22x

24x

26x

28x

22 June 2016


Companies Mentioned (Price as of 22-Jun-2016) Petrobras (PBR.N, $6.92) Ultrapar Participacoes SA (UGPA3.SA, R$71.01, OUTPERFORM, TP R$89.0)

Disclosure Appendix

Important Global Disclosures Andre Natal and Regis Cardoso each certify, with respect to the companies or securities that the individual analyzes, that (1) the views expressed in this report accurately reflect his or her personal views about all of the subject companies and securities and (2) no part of his or her compensation was, is or will be directly or indirectly related to the specific recommendations or views expressed in this report.

3-Year Price and Rating History for Ultrapar Participacoes SA (UGPA3.SA)

UGPA3.SA Closing Price Target Price

Date (R$) (R$) Rating

03-Oct-13 57.15 NR

26-Dec-13 55.68 50.00 N *

03-Dec-14 52.34 50.00 *

* Asterisk signifies initiation or assumption of coverage.

N O T RA T ED

N EU T RA L

The analyst(s) responsible for preparing this research report received Compensation that is based upon various factors including Credit Suisse's total revenues, a portion of which are generated by Credit Suisse's investment banking activities

As of December 10, 2012 Analysts’ stock rating are defined as follows: Outperform (O) : The stock’s total return is expected to outperform the relevant benchmark* over the next 12 months. Neutral (N) : The stock’s total return is expected to be in line with the relevant benchmark* over the next 12 months. Underperform (U) : The stock’s total return is expected to underperform the relevant benchmark* over the next 12 months. *Relevant benchmark by region: As of 10th December 2012, Japanese ratings are based on a stock’s total return relative to the analyst's coverage universe which consists of all companies covered by the analyst within the relevant sector, with Outperforms representing the most attractive, Neut rals the less attractive, and Underperforms the least attractive investment opportunities. As of 2nd October 2012, U.S. and Canadian as well as European ratings are based on a stock’s total return relative to the analyst's coverage universe which consists of all companies covered by the analyst within the relevant sector, with Outperforms representing the most attractive, Neutrals the less attractive, and Underperforms the least attractive investment opportunities. For Latin American and non -Japan Asia stocks, ratings are based on a stock’s total return relative to the average total return of the relevant country or regional benchmark; prior to 2nd October 2012 U.S. and Canadian ratings were based on (1) a stock’s absolute total return potential to its current share price and (2) the relative attractiv eness of a stock’s total return potential within an analyst’s coverage universe. For Australian and New Zealand stocks, the expected total return (ETR) calculation includes 12 -month rolling dividend yield. An Outperform rating is assigned where an ETR is greater than or equal to 7.5%; Underperform where an ETR less than or equal to 5%. A Neutral may be assigned where the ETR is between -5% and 15%. The overlapping rating range allows analysts to assign a rating that puts ETR in the context of associated risks. Prior to 18 May 2015, ETR ranges for Outperform and Underperform ratings did not overlap with Neutral thresholds between 15% and 7.5%, which was in operation from 7 July 2011. Restricted (R) : In certain circumstances, Credit Suisse policy and/or applicable law and regulations preclude certain types of communications, including an investment recommendation, during the course of Credit Suisse's engagement in an investment banking transaction and in certain other circumstances.

Volatility Indicator [V] : A stock is defined as volatile if the stock price has moved up or down by 20% or more in a month in at least 8 of the past 24 months or the analyst expects significant volatility going forward.

Analysts’ sector weightings are distinct from analysts’ stock ratings and are based on the analyst’s expectations for the fundamentals and/or valuation of the sector* relative to the group’s historic fundamentals and/or valuation: Overweight : The analyst’s expectation for the sector’s fundamentals and/or valuation is favorable over the next 12 months. Market Weight : The analyst’s expectation for the sector’s fundamentals and/or valuation is neutral over the next 12 months. Underweight : The analyst’s expectation for the sector’s fundamentals and/or valuation is cautious over the next 12 months. *An analyst’s coverage sector consists of all companies covered by the analyst within the relevant sector. An analyst may cover multiple sectors.

22 June 2016


Credit Suisse's distribution of stock ratings (and banking clients) is:

Global Ratings Distribution

Rating Versus universe (%) Of which banking clients (%) Outperform/Buy* 56% (39% banking clients) Neutral/Hold* 34% (18% banking clients) Underperform/Sell* 10% (40% banking clients) Restricted 0% *For purposes of the NYSE and NASD ratings distribution disclosure requirements, our stock ratings of Outperform, Neutral, and Underperform most closely correspond to Buy, Hold, and Sell, respectively; however, the meanings are not the same, as our stock ratings are determined on a relative basis. (Please refer to definitions above.) An investor's decision to buy or sell a security should be based on investment objectives, current holdings, and other individual factors.

Credit Suisse’s policy is to update research reports as it deems appropriate, based on developments with the subject company, the sector or the market that may have a material impact on the research views or opinions stated herein. Credit Suisse's policy is only to publish investment research that is impartial, independent, clear, fair and not misleading. For more detail please refer to Credit Suisse's Policies for Managing Conflicts of Interest in connection with Investment Research: http://www.csfb.com/research-and-analytics/disclaimer/managing_conflicts_disclaimer.html Credit Suisse does not provide any tax advice. Any statement herein regarding any US federal tax is not intended or written to be used, and cannot be used, by any taxpayer for the purposes of avoiding any penalties. This material is intended for your use only and not for general distribution. This material is not intended to promote or procure a particular outcome in the UK referendum on membership of the European Union (the “Referendum”). Credit Suisse does not promote or endorse either campaign in the Referendum. This material does not constitute, and should not be interpreted as, a recommendation by Credit Suisse as to the merits of a particular outcome of the Referendum.

Target Price and Rating Valuation Methodology and Risks: (12 months) for Ultrapar Participacoes SA (UGPA3.SA)

Method: Our target price of R$89/sh to Ultrapar shares is based on on two-parts DCF in two time horizons: (i) 2020-230: performance similar to recent years, 18% RONIC and 40% reinvestment; and (ii) 2030 onwards: diminishing returns, 13% RONIC (compares to 12.5% wacc) and c.40% reinvestment rate, resulting in growth rate of c.9.0% in line with the nominal growth in GDP.

Risk: Risks to our R$89/sh target price and OUTPERFORM rating include, but are not restricted to: (1) tougher competitive environment; (2) significant changes in consumer preference and consumption of fuels, for instance due to the fast pace adoption of electric vehicles; (3) worse macro conditions; (4) foreign exchange rate variations; (5) significant change in market fuels market dynamics, as for example caused by a possible sale of Petrobras' refineries; (6) unkown governance issues that may arise; (7) execution risk on the acquisition front, as Ultrapar seeks innorganic expansion; (8) future growth and ability of Ipiranga to protect its gross margins in the Brazilian fuel distribution market and (9) unfavorable changes in the regulatory framework.

Please refer to the firm's disclosure website at https://rave.credit-suisse.com/disclosures for the definitions of abbreviations typically used in the target price method and risk sections.

See the Companies Mentioned section for full company names

Important Regional Disclosures Singapore recipients should contact Credit Suisse AG, Singapore Branch for any matters arising from this research report. The analyst(s) involved in the preparation of this report may participate in events hosted by the subject company, including site visits. Credit Suisse does not accept or permit analysts to accept payment or reimbursement for travel expenses associated with these events. Restrictions on certain Canadian securities are indicated by the following abbreviations: NVS--Non-Voting shares; RVS--Restricted Voting Shares; SVS--Subordinate Voting Shares. Individuals receiving this report from a Canadian investment dealer that is not affiliated with Credit Suisse should be advised that this report may not contain regulatory disclosures the non-affiliated Canadian investment dealer would be required to make if this were its own report. For Credit Suisse Securities (Canada), Inc.'s policies and procedures regarding the dissemination of equity research, please visit https://www.credit-suisse.com/sites/disclaimers-ib/en/canada-research-policy.html. As of the date of this report, Credit Suisse acts as a market maker or liquidity provider in the equities securities that are the subject of this report. Principal is not guaranteed in the case of equities because equity prices are variable. Commission is the commission rate or the amount agreed with a customer when setting up an account or at any time after that. Andre Natal & Regis Cardoso each certify that (1) The views expressed in this report solely and exclusively reflect my personal opinions and have been prepared independently, including with respect to Banco de Investimentos Credit Suisse (Brasil) S.A. or its affiliates ("Credit Suisse"). (2) Part of my compensation is based on various factors, including the total revenues of Credit Suisse, but no part of my compensation has been, is, or will be related to the specific recommendations or views expressed in this report. In addition, Credit Suisse declares that: Credit Suisse has provided, and/or may in the future provide investment banking, brokerage, asset management, commercial banking and other financial services to the subject company/companies or its affiliates, for which they have received or may receive customary fees and commissions, and which constituted or may constitute relevant financial or commercial interests in relation to the subject company/companies or the subject securities. Andre Natal is the responsible analyst for this report according to Instruction CVM 483 To the extent this is a report authored in whole or in part by a non-U.S. analyst and is made available in the U.S., the following are important disclosures regarding any non-U.S. analyst contributors: The non-U.S. research analysts listed below (if any) are not registered/qualified as research analysts with FINRA. The non-U.S. research analysts listed below may not be associated persons of CSSU and therefore may not be subject to the

22 June 2016


NASD Rule 2711 and NYSE Rule 472 restrictions on communications with a subject company, public appearances and trading securities held by a research analyst account. Banco de Investments Credit Suisse (Brasil) SA or its affiliates. ............................................................................ Andre Natal ; Regis Cardoso

Important Credit Suisse HOLT Disclosures With respect to the analysis in this report based on the Credit Suisse HOLT methodology, Credit Suisse certifies that (1) the views expressed in this report accurately reflect the Credit Suisse HOLT methodology and (2) no part of the Firm’s compensation was, is, or will be directly related to the specific views disclosed in this report. The Credit Suisse HOLT methodology does not assign ratings to a security. It is an analytical tool that involves use of a set of proprietary quantitative algorithms and warranted value calculations, collectively called the Credit Suisse HOLT valuation model, that are consistently applied to all the companies included in its database. Third-party data (including consensus earnings estimates) are systematically translated into a number of default algorithms available in the Credit Suisse HOLT valuation model. The source financial statement, pricing, and earnings data provided by outside data vendors are subject to quality control and may also be adjusted to more closely measure the underlying economics of firm performance. The adjustments provide consistency when analyzing a single company across time, or analyzing multiple companies across industries or national borders. The default scenario that is produced by the Credit Suisse HOLT valuation model establishes the baseline valuation for a security, and a user then may adjust the default variables to produce alternative scenarios, any of which could occur. Additional information about the Credit Suisse HOLT methodology is available on request. The Credit Suisse HOLT methodology does not assign a price target to a security. The default scenario that is produced by the Credit Suisse HOLT valuation model establishes a warranted price for a security, and as the third-party data are updated, the warranted price may also change. The default variable may also be adjusted to produce alternative warranted prices, any of which could occur. CFROI®, HOLT, HOLTfolio, ValueSearch, AggreGator, Signal Flag and “Powered by HOLT” are trademarks or service marks or registered trademarks or registered service marks of Credit Suisse or its affiliates in the United States and other countries. HOLT is a corporate performance and valuation advisory service of Credit Suisse.

For Credit Suisse disclosure information on other companies mentioned in this report, please visit the website at https://rave.credit-suisse.com/disclosures or call +1 (877) 291-2683.

22 June 2016


References in this report to Credit Suisse include all of the subsidiaries and affiliates of Credit Suisse operating under its investment banking division. For more information on our structure, please use the following link: https://www.credit-suisse.com/who-we-are This report may contain material that is not directed to, or intended for distribution to or use by, any person or entity who is a citizen or resident of or located in any locality, state, country or other jurisdiction where such distribution, publication, availability or use would be contrary to law or regulation or which would subject Credit Suisse AG or its affiliates ("CS") to any registration or licensing requirement within such jurisdiction. All material presented in this report, unless specifically indicated otherwise, is under copyright to CS. None of the material, nor its content, nor any copy of it, may be altered in any way, transmitted to, copied or distributed to any other party, without the prior express written permission of CS. All trademarks, service marks and logos used in this report are trademarks or service marks or registered trademarks or service marks of CS or its affiliates. The information, tools and material presented in this report are provided to you for information purposes only and are not to be used or considered as an offer or the solicitation of an offer to sell or to buy or subscribe for securities or other financial instruments. CS may not have taken any steps to ensure that the securities referred to in this report are suitable for any particular investor. CS will not treat recipients of this report as its customers by virtue of their receiving this report. The investments and services contained or referred to in this report may not be suitable for you and it is recommended that you consult an independent investment advisor if you are in doubt about such investments or investment services. Nothing in this report constitutes investment, legal, accounting or tax advice, or a representation that any investment or strategy is suitable or appropriate to your individual circumstances, or otherwise constitutes a personal recommendation to you. CS does not advise on the tax consequences of investments and you are advised to contact an independent tax adviser. Please note in particular that the bases and levels of taxation may change. Information and opinions presented in this report have been obtained or derived from sources believed by CS to be reliable, but CS makes no representation as to their accuracy or completeness. CS accepts no liability for loss arising from the use of the material presented in this report, except that this exclusion of liability does not apply to the extent that such liability arises under specific statutes or regulations applicable to CS. This report is not to be relied upon in substitution for the exercise of independent judgment. CS may have issued, and may in the future issue, other communications that are inconsistent with, and reach different conclusions from, the information presented in this report. Those communications reflect the different assumptions, views and analytical methods of the analysts who prepared them and CS is under no obligation to ensure that such other communications are brought to the attention of any recipient of this report. Some investments referred to in this report will be offered solely by a single entity and in the case of some investments solely by CS, or an associate of CS or CS may be the only market maker in such investments. Past performance should not be taken as an indication or guarantee of future performance, and no representation or warranty, express or implied, is made regarding future performance. Information, opinions and estimates contained in this report reflect a judgment at its original date of publication by CS and are subject to change without notice. The price, value of and income from any of the securities or financial instruments mentioned in this report can fall as well as rise. The value of securities and financial instruments is subject to exchange rate fluctuation that may have a positive or adverse effect on the price or income of such securities or financial instruments. Investors in securities such as ADR's, the values of which are influenced by currency volatility, effectively assume this risk. Structured securities are complex instruments, typically involve a high degree of risk and are intended for sale only to sophisticated investors who are capable of understanding and assuming the risks involved. The market value of any structured security may be affected by changes in economic, financial and political factors (including, but not limited to, spot and forward interest and exchange rates), time to maturity, market conditions and volatility, and the credit quality of any issuer or reference issuer. Any investor interested in purchasing a structured product should conduct their own investigation and analysis of the product and consult with their own professional advisers as to the risks involved in making such a purchase. Some investments discussed in this report may have a high level of volatility. High volatility investments may experience sudden and large falls in their value causing losses when that investment is realised. Those losses may equal your original investment. Indeed, in the case of some investments the potential losses may exceed the amount of initial investment and, in such circumstances, you may be required to pay more money to support those losses. Income yields from investments may fluctuate and, in consequence, initial capital paid to make the investment may be used as part of that income yield. Some investments may not be readily realisable and it may be difficult to sell or realise those investments, similarly it may prove difficult for you to obtain reliable information about the value, or risks, to which such an investment is exposed. This report may provide the addresses of, or contain hyperlinks to, websites. Except to the extent to which the report refers to website material of CS, CS has not reviewed any such site and takes no responsibility for the content contained therein. Such address or hyperlink (including addresses or hyperlinks to CS's own website material) is provided solely for your convenience and information and the content of any such website does not in any way form part of this document. Accessing such website or following such link through this report or CS's website shall be at your own risk. This report is issued and distributed in Europe (except Switzerland) by Credit Suisse Securities (Europe) Limited, One Cabot Square, London E14 4QJ, England, which is authorised by the Prudential Regulation Authority and regulated by the Financial Conduct Authority and the Prudential Regulation Authority. This report is issued and distributed in Europe (except Switzerland) by Credit Suisse International, One Cabot Square, London E14 4QJ, England, which is authorised by the Prudential Regulation Authority and regulated by the Financial Conduct Authority and the Prudential Regulation Authority. This report is being distributed in Germany by Credit Suisse Securities (Europe) Limited Niederlassung Frankfurt am Main regulated by the Bundesanstalt fuer Finanzdienstleistungsaufsicht ("BaFin"). This report is being distributed in the United States and Canada by Credit Suisse Securities (USA) LLC; in Switzerland by Credit Suisse AG; in Brazil by Banco de Investimentos Credit Suisse (Brasil) S.A or its affiliates; in Mexico by Banco Credit Suisse (México), S.A. (transactions related to the securities mentioned in this report will only be effected in compliance with applicable regulation); in Japan by Credit Suisse Securities (Japan) Limited, Financial Instruments Firm, Director-General of Kanto Local Finance Bureau (Kinsho) No. 66, a member of Japan Securities Dealers Association, The Financial Futures Association of Japan, Japan Investment Advisers Association, Type II Financial Instruments Firms Association; elsewhere in Asia/ Pacific by whichever of the following is the appropriately authorised entity in the relevant jurisdiction: Credit Suisse (Hong Kong) Limited, Credit Suisse Equities (Australia) Limited, Credit Suisse Securities (Thailand) Limited, regulated by the Office of the Securities and Exchange Commission, Thailand, having registered address at 990 Abdulrahim Place, 27th Floor, Unit 2701, Rama IV Road, Silom, Bangrak, Bangkok 10500, Thailand, Tel. +66 2614 6000, Credit Suisse Securities (Malaysia) Sdn Bhd, Credit Suisse AG, Singapore Branch, Credit Suisse Securities (India) Private Limited (CIN no. U67120MH1996PTC104392) regulated by the Securities and Exchange Board of India as Research Analyst (registration no. INH 000001030) and as Stock Broker (registration no. INB230970637; INF230970637; INB010970631; INF010970631), having registered address at 9th Floor, Ceejay House, Dr.A.B. Road, Worli, Mumbai - 18, India, T- +91-22 6777 3777, Credit Suisse Securities (Europe) Limited, Seoul Branch, Credit Suisse AG, Taipei Securities Branch, PT Credit Suisse Securities Indonesia, Credit Suisse Securities (Philippines ) Inc., and elsewhere in the world by the relevant authorised affiliate of the above. Credit Suisse (Hong Kong) Limited ("CSHK") is licensed and regulated by the Securities and Futures Commission of Hong Kong under the laws of Hong Kong, which differ from Australian laws. CSHKL does not hold an Australian financial services licence (AFSL) and is exempt from the requirement to hold an AFSL under the Corporations Act 2001 (the Act) under Class Order 03/1103 published by the ASIC in respect of financial services provided to Australian wholesale clients (within the meaning of section 761G of the Act). Research on Taiwanese securities produced by Credit Suisse AG, Taipei Securities Branch has been prepared by a registered Senior Business Person. Research provided to residents of Malaysia is authorised by the Head of Research for Credit Suisse Securities (Malaysia) Sdn Bhd, to whom they should direct any queries on +603 2723 2020. This report has been prepared and issued for distribution in Singapore to institutional investors, accredited investors and expert investors (each as defined under the Financial Advisers Regulations) only, and is also distributed by Credit Suisse AG, Singapore branch to overseas investors (as defined under the Financial Advisers Regulations). By virtue of your status as an institutional investor, accredited investor, expert investor or overseas investor, Credit Suisse AG, Singapore branch is exempted from complying with certain compliance requirements under the Financial Advisers Act, Chapter 110 of Singapore (the "FAA"), the Financial Advisers Regulations and the relevant Notices and Guidelines issued thereunder, in respect of any financial advisory service which Credit Suisse AG, Singapore branch may provide to you. This information is being distributed by Credit Suisse AG (DIFC Branch), duly licensed and regulated by the Dubai Financial Services Authority (“DFSA”). Related financial services or products are only made available to Professional Clients or Market Counterparties, as defined by the DFSA, and are not intended for any other persons. Credit Suisse AG (DIFC Branch) is located on Level 9 East, The Gate Building, DIFC, Dubai, United Arab Emirates. This research may not conform to Canadian disclosure requirements. In jurisdictions where CS is not already registered or licensed to trade in securities, transactions will only be effected in accordance with applicable securities legislation, which will vary from jurisdiction to jurisdiction and may require that the trade be made in accordance with applicable exemptions from registration or licensing requirements. Non-U.S. customers wishing to effect a transaction should contact a CS entity in their local jurisdiction unless governing law permits otherwise. U.S. customers wishing to effect a transaction should do so only by contacting a representative at Credit Suisse Securities (USA) LLC in the U.S. Please note that this research was originally prepared and issued by CS for distribution to their market professional and institutional investor customers. Recipients who are not market professional or institutional investor customers of CS should seek the advice of their independent financial advisor prior to taking any investment decision based on this report or for any necessary explanation of its contents. This research may relate to investments or services of a person outside of the UK or to other matters which are not authorised by the Prudential Regulation Authority and regulated by the Financial Conduct Authority and the Prudential Regulation Authority or in respect of which the protections of the Prudential Regulation Authority and Financial Conduct Authority for private customers and/or the UK compensation scheme may not be available, and further details as to where this may be the case are available upon request in respect of this report. CS may provide various services to US municipal entities or obligated persons ("municipalities"), including suggesting individual transactions or trades and entering into such transactions. Any services CS provides to municipalities are not viewed as "advice" within the meaning of Section 975 of the Dodd-Frank Wall Street Reform and Consumer Protection Act. CS is providing any such services and related information solely on an arm's length basis and not as an advisor or fiduciary to the municipality. In connection with the provision of the any such services, there is no agreement, direct or indirect, between any municipality (including the officials, management, employees or agents thereof) and CS for CS to provide advice to the municipality. Municipalities should consult with their financial, accounting and legal advisors regarding any such services provided by CS. In addition, CS is not acting for direct or indirect compensation to solicit the municipality on behalf of an unaffiliated broker, dealer, municipal securities dealer, municipal advisor, or investment adviser for the purpose of obtaining or retaining an engagement by the municipality for or in connection with Municipal Financial Products, the issuance of municipal securities, or of an investment adviser to provide investment advisory services to or on behalf of the municipality. If this report is being distributed by a financial institution other than Credit Suisse AG, or its affiliates, that financial institution is solely responsible for distribution. Clients of that institution should contact that institution to effect a transaction in the securities mentioned in this report or require further information. This report does not constitute investment advice by Credit Suisse to the clients of the distributing financial institution, and neither Credit Suisse AG, its affiliates, and their respective officers, directors and employees accept any liability whatsoever for any direct or consequential loss arising from their use of this report or its content. Principal is not guaranteed. Commission is the commission rate or the amount agreed with a customer when setting up an account or at any time after that. Copyright © 2016 CREDIT SUISSE AG and/or its affiliates. All rights reserved.

Investment principal on bonds can be eroded depending on sale price or market price. In addition, there are bonds on which investment principal can be eroded due to changes in redemption amounts. Care is required when investing in such instruments. When you purchase non-listed Japanese fixed income securities (Japanese government bonds, Japanese municipal bonds, Japanese government guaranteed bonds, Japanese corporate bonds) from CS as a seller, you will be requested to pay the purchase price only.

https://www.credit-suisse.com/who-we-are

Ultrapar Participacoes SA

Documents

Transcript of Ultrapar Participacoes SA