Post on 02-Oct-2020
Speakers:Jeff McGrew, CFA, Institutional Portfolio Manager, MFS
Christine Girvan, Managing Director – Canada, MFS
Defining and Demystifying Smart Beta:Implications for Active Management
34054.1
Agenda
• Smart beta around the world
• From theory to reality
• Managing factors in an integrated investment program
2
Which of the following best describes your organization:1. Have a smart beta strategy
2. Have evaluated smart beta but did not allocate
3. Currently evaluating or plan to evaluate
4. Do not plan to evaluate
3
23%38%
25%11%
2% 4%
16%
13%27%
7%19% 18% 21%
18%
19% 14%
23% 15%21%
5%
22%
19%14%
32%
23%
36%
5%
21%13%
21%27%
40%
21%
68%
Total <$1B $1-$10B $10B+ Total <$1-$10B $10B+
Have smart betaallocation
Evaluated and decidednot to implement
Currently evaluating smartbeta
Anticipate evaluatingsmart beta in the next 18months
Do not anticipateevaluating smart beta inthe next 18 months
North America Europe
Sample size for $10B+ for Europe is 19 and below preferred threshold of 30
Source: FTSE Russell – 2015 Global Survey Findings from asset owners
Europe leads in adoption of smart beta strategies
Who?
4
What?Low Volatility and Value are the most used strategies as part of a smart beta combination.What type of smart beta strategies are you using? (Combination of strategies)
Low volatility Value Fundamental Multi-factor combination
Minimum variance
Momentum High quality Maximum diversification
Equal weight
Risk parity Dividend/ income/
yield
Other (please specify)
Defensive
54% 51% 31% 29%
26% 26% 20% 17% 17% 17%11% 11%3%
Segment = Have smart beta allocation, AND using 2 or more strategies.Source: FTSE Russell – 2015 Global Survey Findings from asset owners
5
Do you consider smart beta strategies to be:1. Active investment management
2. Passive investment management
6
Return enhancement
Risk reduction
Improve diversification
Provide specific factor exposure
Cost savings
Income generation
Other
Multi-pick; Segment = Have smart beta allocation, Evaluated and decided not to implement, Currently evaluating smart beta
Smart beta adoption is not a replacement for passive
Why?What investment objective initiated the evaluation of smart beta strategies?
3%
5%
16%
24%
40%
52%
52%
Source: FTSE Russell – 2015 Global Survey Findings from asset owners
7
An improvement in explanatory power and a loss of simplicity
Evolution of Factors
Market Betaβ
Marketβ
Capital Asset Pricing Model (1960s)
Sizeβ
Valueβ Trading Yield Size NonLin Leverage
Value Volatility EY Er Var
Value Momentum Size Growth
Fama-French Three Factor Model (1990s)The market, company size and valuation Today's Multi-Factor Model Proliferation
Numerous models seek to explain a range of style, fundamental and geographic factors that drive returns
8
Source: Spence Johnson
Factor beta is being further refined into risk focused beta and return focused beta
The Evolution and Separation of Performance
'Traditional' Active
Portfolio Return
Alpha
Beta
Alpha
Regional Beta
'Return focused'
Beta
'Risk focused'
Beta
Alpha
'Broad market'
Beta
Factor Beta
Country Beta
Sector Beta
1970s 1980s 1990s 2000s
Smart Beta
'Traditional' Passive
9
Source: MSCI Harvesting Risk Premia for Large Scale Portfolios March 2013
Research illustrates structural and behavioral drivers for risk premia
Structural
Behavioural
Higher systematic risk Less flexibility
Market placeIndex sensitivity
Theories varyHigher turnover
LiquidityInformation uncertainty
VALUE
MOMENTUM
VOLATILITY
SIZE
This factor captures the excess performance of stocks with low prices relative to their fundamentals.
This factor captures excess returns to stocks with lower than average beta or volatility.
This factor captures excess returns to stocks with stronger past performance
This factor captures excess returns of smaller firms relative to their larger counterparts.
Risk Factors to Risk Premiums
10
Wor
st
perf
orm
ing
Best
pe
rfor
min
g
Market leadership varies from year to year
Source: ISI based on S&P 500 Unconstrained Index using ISI's quantitative research factor definitions to the group the underlying index constituents. As of December 31, 2014.
A Multitude of Potential Factors
11
Source: MSCI Barra US Equity Risk Model
• Returns can be volatile
• Underperformance can be prolonged
• Timing can make a big difference
The magnitude and timing of risk factor impact varies
Rolling 1-year returns for momentum factor 1973-2013Smart Beta Challenges: Drawdowns
12
Are you prepared to short or leverage?
Q1-Universe and Q1-Q5 compounded total return indices. Monthly rebalancing with monthly forward returns. Stock universe is MSCI World, data from 12/31/1997 through 12/31/2014. All factors ranked universe relative. Forward returns are capped at the stock level, +/- 75%/month. Quality = Last year Cash Flow from Operations minus Last Year's Net Income) / Average Assets.
Implementation Matters
-10%
0%
10%
20%
30%
40%
50%
Dec
'04
Mar
'05
Jun'
05S
ep'0
5D
ec'0
5M
ar'0
6Ju
n'06
Sep
'06
Dec
'06
Mar
'07
Jun'
07S
ep'0
7D
ec'0
7M
ar'0
8Ju
n'08
Sep
'08
Dec
'08
Mar
'09
Jun'
09S
ep'0
9D
ec'0
9M
ar'1
0Ju
n'10
Sep
'10
Dec
'10
Mar
'11
Jun'
11S
ep'1
1D
ec'1
1M
ar'1
2Ju
n'12
Sep
'12
Dec
'12
Mar
'13
Jun'
13S
ep'1
3D
ec'1
3M
ar'1
4Ju
n'14
Sep
'14
Dec
'14
Quality Factor Q1-Q5
Quality Factor Q1-Universe
"Implementation Shortfall"
13
(30)
(20)
(10)
0
10
20
30
40
50
Yield Size Moment-um
Value Multi-factor
Funda-mentalIndex
Quality Beta Growth Event/Other
Volatility
%
Strong correlation between past performance and future flows
Smart Beta ETFs & Past PerformanceSmart-Beta ETFs¹
Correlation Between the Relative Return in Prior Quarter and Net New Money in the Subsequent One²2007 Through Q1 2015
Source: Strategic Insight Simfund, Empirical Research Partners Analysis. ¹ Based on underlying securities held being U.S.-listed.² Relative return is equally-weighted return of all ETFs in each category less the S&P 500 return. Relative net new money flow is the net flow into all ETFs in each category as a percent of start-of-period assets less that for all ETFs.
14
Source: Strategic Insights, Empirical Research Partners Analysis. ¹ Based on underlying securities held being U.S.-listed.² Relative return is equally-weighted return of all ETFs in each category less the S&P 500 return.
(30)
(25)
(20)
(15)
(10)
(5)
0
5
10
15
20
Size Multi-factor
Event/Other
Value Growth Funda-mentalIndex
Beta Moment-um
Volatility Quality Yield
%
Past & Future PerformanceU.S. Equity and Alternatives ETFs¹
Correlation Between Relative Return in Prior Quarter and Relative Return in Subsequent Quarter²2007 Through Q1 2015
Beware of backward looking bias15
Smart Beta ETFs Investment Horizon
0%
10%
20%
30%
40%
50%
60%
2007 2008 2009 2010 2011 2012 2013 2014
Average = 40%
U.S. Smart Beta ETFsAverage Monthly Turnover by Quarter1
2007 Through Mid-Q2 2015
Source: Empirical Research Partners Analysis. ¹ Monthly turnover computed as dollar value traded during month divided by the market capitalization. Quarterly turnover is average of monthly turnover over the quarter.
16
Source: MSCI World Index from 12/31/1997 to 12/31/2014; Quality = Last year Cash Flow from Operations minus Last Year's Net Income) / Average Assets; Momentum = 12M – 1M Momentum; Allocations to Factor Returns combines the return streams of the five factors in an asset allocation approach with monthly rebalancing (12m minus 1m Momentum; Trailing Earning/Price; Book/Price; Return on Equity; Quality). Multi-Factor Model is a composite of all of the factor returns; the mean monthly return is compounded to an annual return.
Multifactor More Robust Than Single Factor
5.0%6.0%7.0%8.0%9.0%
10.0%11.0%12.0%13.0%14.0%15.0%
Momentum Quality Multi-Factor Model
Annualized Cumulative Returns(1997-2014)
The Power Of A Multi-Factor Model
17
The Benefit Of Combining Factor Based Approach With Fundamental Research
Hypothetical annualized stock returns by MFS ratings – Feb 1995 – June 2015 (in USD). Past performance is no guarantee of future results. All annualized returns of MFS 1,2 and 3 rated stocks: 10.2%Period represents entire data set for stored MFS stock level ratings. Analysis assumes buy and hold with a monthly rebalance. Each month MFS calculates the equal weighted average of the monthly returns of all stocks rated 1,2 or 3 at the prior month end. The monthly returns are then linked and annualized. No transaction costs are assumed in this analysis.
Clear alpha benefit from combining fundamental and quantitative research
Per
cent
(%)
Buys Sells
11.6
3.2
-1.4
18.2
Buy rated stocks
Sell rated stocks
-5.9
21.4
Fundamental-Quant Intersections Quantitative Fundamental
18
Fundamental median manager return (%) minus quantitative median manager return (%)
Fundamental and quantitative approaches work in different environments
-0.8
-2.9
-0.4
-4.8 -4.8
1.5
5.8
4.3
-0.3 -0.1-1.4 -1.0 -0.8
2.0 2.0
4.8
-0.7
-3.6
-0.5-1.6 -2.0
Rel
ativ
e R
etur
n %
Different Approaches Benefit From Different Environments
Past performance is no guarantee of future results.Source: eVestment. Based on monthly returns. Products in large-cap U.S. core equity universe. Products are then split into those that identified themselves as having a "fundamental" approach or a "quantitative" approach
19
Sources: FactSet and MFS research. MSCI World holdings as of 31 March 2010. Forward total cumulative returns around the mean in USD from 31 March 2010 to 31 March 2015; 10th to 90th percentile range.
-125-100-75-50-25
0255075
100125
1D 1W 1M 3M 6M 1Y 2Y 3Y 4Y 5Y
Exce
ss re
turn a
roun
d mea
n (%
)
Lengthening time horizon creates greater alpha opportunities
Time Arbitrage OpportunityMSCI World total returns dispersion around the mean return (2010-2015)
20
Considerations For Investors • Factor investing involves active decisions
• Move from theory to implementation is complex and behaviouralbiases can be challenging
• Combining multiple factors with traditional fundamental analysis can create a more robust alpha stream while reducing risk overall
• Factor framework can also assist in better understanding your existing managers' exposures and where alpha is coming from
21
Thank You!
The views expressed are those of the author(s) and are subject to change at any time. These views are for informational purposes only and should not be relied upon as a recommendation to purchase any security or as a solicitation or investment advice from the Advisor.
Unless otherwise indicated, logos and product and service names are trademarks of MFS® and its affiliates and may be registered in certain countries.
Issued in Canada by MFS Investment Management Canada Limited. No securities commission or similar regulatory authority in Canada has reviewed this communication.