Economic Forecasting: Learning from the Future

Dr. Olga Loiseau-Aslanidi

Asst. Director-Economist

Economic & Consumer Credit Analytics – EMEA

Moody’s Analytics

UPCES lecture , Prague, October 1, 2013

2 2

Lecture Outline

• Overview of Moody’s Corporation

• Basics of economic forecasting

• Macroeconomic forecasting: baseline and alternative scenarios

• Market risk forecasting (if time permits)

3 3

• Moody's Corporation provides credit ratings, research, tools and analysis that contribute to

transparent and integrated financial markets.

• Moody’s was founded more than 100 years ago by John Moody & Company.

• In 1900, Moody created the manual based on his assessment of the market’s needs at the

time.

• Moody's Corporation maintains a presence in 29 countries.

• Americas: Boston, Buenos Aires, Chicago, Dallas, Mexico City, New York, São Paulo, San

Francisco, Toronto, Alpharetta, Montreal, West Chester

• EMEA: Dubai, Frankfurt, Johannesburg, Limassol, London, Madrid, Milan, Moscow, Paris,

Grenoble, Saint Cloud, Port Louis, Prague, Warsaw

• Asia: Beijing, Shenzhen, Mumbai, Hong Kong, Seoul, Singapore, Sydney, Melbourne,

Tokyo

4 4

Moody's Analytics operates independently of the credit ratings activities of Moody's Investors

Service. We do not comment on credit ratings or potential rating changes, and no opinion or

analysis you hear during this presentation can be assumed to reflect those of the ratings agency.

5 5

Economics Group at Moody’s Analytics

Who we are

» 70+ economists, more than 40% of which

have PhD’s

» 20+ data specialists

» Located around the globe in London,

Prague, Sydney and West Chester

We focus on being THE LEADER at understanding local economies using the most

recent data from the best sources, analyzed by dedicated economists

What we do

» Maintain extensive database of economic,

financial and demographic data down to the

regional and city level with over 260 million time

series covering 200+ countries and 600+ cities

» Provide the highest frequency and most up-to-

date outlook with monthly updated forecasts of

national and regional economies worldwide

» Forecast alternative macroeconomic

scenarios globally for stress testing and risk

management

» Forecast and stress test clients’ consumer credit

portfolios with customized models

6 6

Why Markets Need Economic Forecasts?

• Banks, asset managers, corporates, insures, consumers,

governments need to measure, monitor, and manage risk.

• They need to adapt to changing financial markets, meet regulatory

requirements, and make informed decisions for allocating capital

and maximizing opportunities.

• Economic forecasts provide with greater ability to make the right

choices.

• Forecasts can incorporate baseline (most likely) and alternative (can

be regulatory-driven) scenarios.

• Market participants need forecasting at the global, macro, and

regional levels to evaluate impact on their businesses.

7 7

What Forecasting Involves?

• The forecast object: event outcome, event timing, time series

forecasts.

• The statement of time series forecasting: a point forecast, an interval

forecast or the probability density forecast.

• The forecast horizon: short, medium or long-term.

• Types of forecasts: qualitative or quantitative.

• Quantitative forecasting: model-building and forecasting stages.

• Important: the assumed data-generating process is never entirely

correct!

• Determine how well model describes the underlying relationships we

are trying to model, that are hidden in the data.

8 8

Forecasting Process

Drivers Identification and In-Sample Validation

Forecasting and Out-of-Sample

Validation

Phase 1: Model Building

Use of Model for Forecasting and Stress Testing

Phase 2: Implementation

• Identify both internal

and external drivers of

main macro variables

• Involves data

preparation and

modeling

• Ensure model forecasts

observed performance

accurately by running

several out-of sample

validation exercises

• Involves modeling and

fine-tuning to improve

forecast capabilities

• Models used to generate projections

• “What if “ analysis conducted under

various policy scenarios

• Economic data and alternative forecast

scenarios leveraged for further analysis

• Results refreshed as forecasts are

updated

9 9

Macroeconomic Forecasting Model

• Forecasting external forces that affect a particular agent: forecasting

economic environments.

• Examples of macro time series: GDP and its components, labor

force and unemployment rate, prices, interest rates, etc.

• Construct a macroeconomic model that that most accurately “fits”

the given historical data and which provides reasonable forecasts of

the future based on patterns “hidden” in the data.

• Systems of equations that define the relationships between different

sectors of the economy.

• Models need to be founded on the analysis of the statistical

properties of the data under study.

10 10

Forecasting Model Design: Theory at Work…

Exchange

rates

Investment

Wages and

salaries

Popula

tion

Prices

GDP

Monetary

policy rate

Imports

Government

Exports

Global GDP

Unemployment rate

Consumption

Labor force

Employment

Potential

GDP

Other

deflators

Import

prices

10-yr

yield

Global

prices

11 11

Global Macro Forecast Coverage

Americas Asia/Pacific EMEA

• Argentina

• Brazil

• Canada

• Chile

• Colombia

• Mexico

• Peru

• United States

• Uruguay

• Venezuela

• Australia

• China

• Hong Kong

• India

• Indonesia

• Japan

• Malaysia

• New Zealand

• Philippines

• Singapore

• South Korea

• Taiwan

• Thailand

• Austria

• Belgium

• Czech Republic

• Denmark

• Finland

• France

• Germany

• Greece

• Hungary

• Ireland

• Israel

• Italy

• Luxembourg

• Netherlands

• Norway

• Poland

• Portugal

• Russia

• Slovak Republic

• Slovenia

• South Africa

• Spain

• Sweden

• Switzerland

• Turkey

• United Kingdom

Forecasting national economies that account for more than 93% of

world output

12

Source: Moody’s Analytics

Expansion

In recession

At risk

Recovery

September 2013

Global Business Cycle Status

13

Source: Moody’s Analytics

Expansion

In recession

At risk

Recovery

September 2013

Europe Business Cycle Status

14 14

Stress Testing

• Outputs of forecasting: a detailed description of the current

economic situation; a fairly detailed forecast for the near future; an

outlook beyond that with a discussion of alternative scenarios.

• Helps market participants to answer "What if?" questions enabling

improved portfolio management.

• How will I meet regulatory requirements for stress testing?

• How should I adjust lending standards if the economy goes into a

double-dip recession?

• Could my portfolio withstand a sovereign default event in Europe?

• What impact would a prolonged deflation cycle have on current and

future loan performance?

15

Alternative Macroeconomic Scenarios

Stronger Near-Term Rebound S1

S2 Mild Second Recession

S3 Deeper Second Recession

Protracted Slump S4

Baseline / Most Likely BL

Standard

Below Trend Long Term Growth S5

Oil Price Shock S6

Fed Baseline FB

Fed Adverse Scenario FA

EBA Baseline EB

EBA Adverse Scenario ES

Regulatory Driven

FSA Anchor FSA

Fed Severely Adverse FS

16 16

-5 -4 -3 -2 -1 0 1 2 3 4 5

Global

North America

South America

Asia

Mid. East & Africa

Eastern Europe

Other Europe

Euro zone

S4: Protracted slump S3: Deeper Rec. S2: Moderate Rec. Baseline

Europe Suffers Across Downside Scenarios

Real GDP, 2013, % change yr ago

Sources: National statistical offices, Moody’s Analytics

17 17

Developed Markets: Unemployment Rate

Sources: National statistical offices, Moody’s Analytics

0

5

10

15

20

25

30

35

Euro Zone Japan Germany Spain UK US

2013 baseline 2013 S4 2013 S6

% of labour force

18 18

Italy Scenario Analysis: Standard Forecasts

Sources: National statistical offices, Moody’s Analytics

19

• Swaps and Sovereign Curves (term structure)

• Stock Market Returns, Historical and Implied Volatilities

• CDS Spreads by Sector and Rating category

• Credit Migration

•

From Macro Scenarios to Market Risk

Internal Actions

Consumer Loan Data

…

Past & Future

Business Decisions

Forecasts +

Alternative Scenarios

Economic Data

…

+ External Environment

Modeling Market Risk Variables

(Interest rates)

20

Interest Rates Modeling

Past & Future

Business Decisions

Forecasts +

Alternative Scenarios

+

21

EURO Swap Rate Curves

02

46

Rat

es,%

2000m1

2001m1

2002m1

2003m1

2004m1

2005m1

2006m1

2007m1

2008m1

2009m1

2010m1

2011m1

2012m1

2013m1

Maturities: 1M-360M

Euro Swap Rates

IRS- derivative contracts, which typically exchange (swap) fixed-rate interest

payments for floating-rate interest payments. Interest Rate Swap contracts may

have a duration from 2 years to 30 years and more.

22

Stress Testing of Swap Rate Curves

-10

12

3

Term

Pre

miu

m

02

46

Rate

s (

%)

2000

m1

2002

m1

2004

m1

2006

m1

2008

m1

2012

m1

2014

m1

2016

m1

2018

m1

2010

m1

EURO Swap Curve: Baseline

-10

12

3

Te

rm P

rem

ium

02

46

Rate

s (

%)

2000

m1

2002

m1

2004

m1

2006

m1

2008

m1

2010

m1

2012

m1

2014

m1

2016

m1

2018

m1

Euro Zone Crisis

EUR Swap Curve vs Term Premium

23

Financial Models: Equity

• Stylized fact: stock index returns tend to co-move across country

• Multivariate exploratory analysis: how much of the variability of international co-

movements in global stock markets can we explain with a single component?

• Principal Component Analysis addresses this kind of question in a simple

descriptive statistics framework

• Let’s investigate stock index returns over the period 2000–2013:

Correlation Matrixukxindex sx5eindex spxindex omxindex smiindex nkyindex top40index mxwoindex obxindex

ukxindex 1

sx5eindex 0.7099 1

spxindex 0.9259 0.6663 1

omxindex 0.9381 0.5264 0.8458 1

smiindex 0.8781 0.8975 0.811 0.7789 1

nkyindex 0.666 0.9007 0.674 0.5228 0.8974 1

top40index 0.7326 0.14 0.6308 0.8337 0.4873 0.1637 1

mxwoindex 0.9415 0.7129 0.9525 0.8824 0.8812 0.7462 0.7066 1

obxindex 0.8723 0.3732 0.8011 0.9294 0.6823 0.4234 0.9474 0.8757 1

24

Financial Models: Equity

-80

-60

-40

-20

0

20

40

60

80

2002M8

2003M3

2003M10

2004M5

2004M12

2005M7

2006M2

2006M9

2007M4

2007M11

2008M6

2009M1

2009M8

2010M3

2010M10

2011M5

2011M12

2012M7

ukxindexsx5eindexspxindexomxindexsmiindexnkyindextop40indexobxindex

Because of the high cross-

correlation, we can reduce the

modelling space from the 8

original variables to 1

(unobserved) principal

component…

One single principal

component explains 91% of

the original variability!

Variability Explained, %

0

10

20

30

40

50

60

70

80

90

100

1 2 3 4 5 6 7 8

% E

xpai

ned

by e

ach

com

pone

nt

Principal Component #

25

Financial Models: Equity

-8

-6

-4

-2

0

2

4

6

2002M8

2003M3

2003M10

2004M5

2004M12

2005M7

2006M2

2006M9

2007M4

2007M11

2008M6

2009M1

2009M8

2010M3

2010M10

2011M5

2011M12

2012M7

2013M2

2013M9

2014M4

2014M11

2015M6

2016M1

2016M8

2017M3

2017M10

2018M5

Global Equity Factor

Global_Equity_Factor_bl

Global_Equity_Factor_fsa

Global_Equity_Factor_s4

26

Financial Models: Equity

-60

-40

-20

0

20

40

60

2001M9

2002M4

2002M11

2003M6

2004M1

2004M8

2005M3

2005M10

2006M5

2006M12

2007M7

2008M2

2008M9

2009M4

2009M11

2010M6

2011M1

2011M8

2012M3

2012M10

2013M5

2013M12

2014M7

2015M2

2015M9

2016M4

Equity Forecasts, Annual Returns

tyy_ukxindex_bl

tyy_ukxindex_fsa

tyy_ukxindex_s4

-60

-40

-20

0

20

40

60

2001M9

2002M4

2002M11

2003M6

2004M1

2004M8

2005M3

2005M10

2006M5

2006M12

2007M7

2008M2

2008M9

2009M4

2009M11

2010M6

2011M1

2011M8

2012M3

2012M10

2013M5

2013M12

2014M7

2015M2

2015M9

2016M4

Equity Forecasts, Annual Returns

tyy_spxindex_bl

tyy_spxindex_fsa

tyy_spxindex_s4-60

-40

-20

0

20

40

60

2001M8

2002M3

2002M10

2003M5

2003M12

2004M7

2005M2

2005M9

2006M4

2006M11

2007M6

2008M1

2008M8

2009M3

2009M10

2010M5

2010M12

2011M7

2012M2

2012M9

2013M4

2013M11

2014M6

2015M1

2015M8

2016M3

Equity Forecasts, Annual Returns

tyy_mxwoindex_bl

tyy_mxwoindex_fsa

tyy_mxwoindex_s4

Equity Indexes, annual growth rate

History and forecasts

27 27

Summary and Conclusions

Historical data provides comprehensive understanding of macroeconomic trends, risks

and opportunities worldwide

Market participants use alternative scenarios to meet regulatory requirements, evaluate

the impact of shocks, expose vulnerabilities, and develop strategic business plans

Economic historical data, forecasts, and alternative scenarios drive key business

processes

o Stress testing asset portfolios for effective risk management and regulatory compliance

o Managing exposure to potential risks in regions with weak economic fundamentals

o Researching and analyzing investment opportunities worldwide

o Developing strategic plans, determining demand, and forecasting business lines

o Measuring the impact of shocks on businesses

Importantly one should use sensible models and the most current data from the best

national, local, and multinational sources for the foundation for forecasts and alternative

scenarios.

www.economy.com

Moody’s Analytics

Olga Loiseau-Aslanidi

Asst. Director - Economist

Email: olga.loiseau-aslanidi@moodys.com

THANK YOU!

© 2013 Moody’s Analytics, Inc. and/or its licensors and affiliates (collectively, “MOODY’S”). All rights reserved. ALL INFORMATION CONTAINED HEREIN IS

PROTECTED BY COPYRIGHT LAW AND NONE OF SUCH INFORMATION MAY BE COPIED OR OTHERWISE REPRODUCED, REPACKAGED, FURTHER

TRANSMITTED, TRANSFERRED, DISSEMINATED, REDISTRIBUTED OR RESOLD, OR STORED FOR SUBSEQUENT USE FOR ANY SUCH PURPOSE, IN

WHOLE OR IN PART, IN ANY FORM OR MANNER OR BY ANY MEANS WHATSOEVER, BY ANY PERSON WITHOUT MOODY’S PRIOR WRITTEN CONSENT.

All information contained herein is obtained by MOODY’S from sources believed by it to be accurate and reliable. Because of the possibility of human or

mechanical error as well as other factors, however, all information contained herein is provided “AS IS” without warranty of any kind. Under no circumstances shall

MOODY’S have any liability to any person or entity for (a) any loss or damage in whole or in part caused by, resulting from, or relating to, any error (negligent or

otherwise) or other circumstance or contingency within or outside the control of MOODY’S or any of its directors, officers, employees or agents in connection with

the procurement, collection, compilation, analysis, interpretation, communication, publication or delivery of any such information, or (b) any direct, indirect, special,

consequential, compensatory or incidental damages whatsoever (including without limitation, lost profits), even if MOODY’S is advised in advance of the possibility

of such damages, resulting from the use of or inability to use, any such information. The ratings, financial reporting analysis, projections, and other observations, if

any, constituting part of the information contained herein are, and must be construed solely as, statements of opinion and not statements of fact or

recommendations to purchase, sell or hold any securities. NO WARRANTY, EXPRESS OR IMPLIED, AS TO THE ACCURACY, TIMELINESS, COMPLETENESS,

MERCHANTABILITY OR FITNESS FOR ANY PARTICULAR PURPOSE OF ANY SUCH RATING OR OTHER OPINION OR INFORMATION IS GIVEN OR MADE

BY MOODY’S IN ANY FORM OR MANNER WHATSOEVER. Each rating or other opinion must be weighed solely as one factor in any investment decision made

by or on behalf of any user of the information contained herein, and each such user must accordingly make its own study and evaluation of each security and of

each issuer and guarantor of, and each provider of credit support for, each security that it may consider purchasing, holding, or selling.