Manager SelectionJanuary, 2018Scott Stewart, CFACornell University

Show of Hands

Do you think that institutional investors add value from their manager selection decisions?

Yes, No or I Don’t Know

2

Today’s Discussion

3

The Investment Business

1. Business people rewardedonce assets arrive, and stick

2. Portfolio managersrewarded once theyoutperform theirbenchmarks

4

What is Reward for Clients?

Wouldn’t it be interesting…

to collect all the hirings and firings of investment managers,

determine why these decisions were made,and whether or not they added value?

And use the results to improve the industry…5

Today’s Topics for Discussion

1. Scott’s research (2007 on)2. Key book observations (2013/14)3. Key takeaways

6

7

Scott’s Three Research Studies

1. Institutional manager selection process (1985-2000)

2. Performance of hiring/firing decisions (1985-2006)

3. Survey of institutional investor decision process (2004…)

8

Empirical Study—Research Design

Examine flows between managers

1. Explain flow activity2. Test subsequent performance

Informa database:Returns & Characteristics

on over 7000 Inst’l Products

First Two Studies—Research Design

9

Informa Database—Asset Levels

Assets in Billions

78.6211.9

420.7 354.6570.2

2,445.7

4,433.9

104.6202.4

1,488.6

1,857.6

472.5

0

500

1,000

1,500

2,000

2,500

3,000

3,500

4,000

4,500

5,000

1985 1990 1995 2000

BalancedEquityFixed

FYI: Asset Flows represent 10% per year on average

Total grew to exceed $10 trillion by 2007

10

Results of First Study: Drivers of flows

Asset and Account Flows: “Why do… Hire and Fire…”, JBES (2007)

Institutional investors1. rely on benchmark-relative performance, not simply

total return

2. are not overly focused on short term results

3. pay attention to style, but do not necessarily adjust for style extremeness

4. rely on return pattern more than simply cumulative returns

5. require more evidence before terminating an account

11

Part One of Second Study: Subsequent Performance

Financial Analysts Journal, 2009

Basic Question: Do institutional investors add value from changing manager allocations?

Statistical results suggest…

Managers who receive contributions tend to under-perform managers who experience withdrawals

Based on over 80,000 annual observations between 1985-2006

12

Subsequent Performance—WEIGHTED Results

% Difference in Performance: FLOW Weighted and ACCOUNT Weighted Portfolios of Managers

(1985-2006)

FLOW-WEIGHTED ACCOUNT-WEIGHTED

-4.0%

-3.0%

-2.0%

-1.0%

0.0%

1.0%

2.0%

3.0%

4.0%

5.0%

Pre-Flow Post Flow

Diff

eren

ce in

A

nnua

lized

Ret

urns

1-Year5-Year

-9-8-7-6-5-4-3-2-10

1-year 5-years 1-year 5-years

NOTE: Collectively, plan sponsors are losing billions of dollars a year through their manager allocation decisions!

13

Part Two of Second Study: Sources

Key Question: Which decisions lose value?

• Are they good at setting asset allocation but not at manager selection?

• Do they add value at the category or style level but destroy value once it is implemented?

Statistical results suggest…

Investors lose value at the style and mostly manager selection decision levels, and a little in the short term from asset allocation decisions

14

Sources of Loss of Value—One-Year Periods, 1986-2006

Brinson Analysis: Category versus Product Selection

-0.8-0.7-0.6-0.5-0.4-0.3-0.2-0.1

0Category Product Interaction

15

2004 Survey Study

1. Confirms many research results2. Identifies non-performance criteria

– Communication skills– Reputation– Consultant input

3. Mixed perceptions of investment performance

Current Study: Survey of Plan Sponsors

16

Summary of Other Survey Results

• Plans with consultants and higher education levels turn plans over to a greater extent

• More “functional” plans evaluate decisions to a greater extent, have fewer asset classes and higher turnover

• Tainted managers are terminated to a greater extent by public plans, yet only if performance is poor

Book’s Key Observations

17

18

1. Using institutional data– Confirm Scott’s results– Identify evidence of statistically significant

manager skill and persistence in consistency

2. Using mutual fund data– Limited evidence of statistically significant manager

skill (table 2.7)

– Limited evidence of persistence (table 2.9)– Some evidence of short-term manager

selection skill

Other Research Studies’ Results

.p f p M frp R R R ep

19

1. Active manager skill does exist2. It is very difficult to identify skillful managers

(after fees) in advance, especially in public markets

3. In some markets it may be worth the effort4. There are things we can do to improve

results

What did Scott learn from Manager Selection?

20

Cool Stuff in CFA Book

In addition to information on the record of active management and manager selection…

• The arithmetic of active management• Index fund selection• Mixing fund managers• Lists of guidelines and key

recommendations• Excel tools• Bibliography

21

Cool Summary of Research on Qualitative Factors

22

Cool Summary of Research on Manager Success

Book’s Key Recommendations

23

24

1. Don’t follow fashion…seek valuation2. Know difference between deep value and

relative value3. If everyone wants you to fire manager, ask

yourself if you’re selling at the bottom4. Evaluate your process, not just your current

managers5. Look at managers’ portfolio construction

Recommendations for Plan Sponsors

25

Observations for Managers

• Your clients may select you simply because your track record (style may not be adjusted for fully) looks good

• They may give up on you when short term performance is poor

• There’s a good chance this decision is a mistake• Keys: know your client and develop good

communication

26

• Communicate frequently, begin by managing expectations

• Communicate more if performance weakens• Demonstrate, and then explain

– Why performance weak– Portfolio characteristics & performance consistent

with process– Performance tends to reverse

• Good followed by weak• Really good followed by weaker

Recommendations Regarding Communication

27

Business Recommendations

Pick an attractive asset classHire skillful managers

– Bright & knowledgeable– Focused

Structure appropriate incentives– Long term view– Independent decision making– Alignment of interests

Understand portfolio construction

28

Final Recommendation

NEW RESEARCH RESULTS:– Survey of Plan Sponsors

– Perceptions on confidence, importance to study process performance and control variables

– t-stat on “importance to study process performance” = 0.089

KEY RECOMMENDATION: Evaluate your process, not just your current managers

29

Manager Selection

Thank you

30

Scott’s Contact Information

EMAIL: sds58@cornell.edu

Appendix

31

32

More on Drivers of flows

1. Most assets flow to managers with good one, three and five-year numbers

2. Poor one-year number not a big problem

3. A really bad one-year number is

4. A poor 5-year number not a big problem if one and three-year numbers good

33

FAJ Study: DATA

• 1985-2006, over 80,000 annual observations

• Equity, fixed, international & global• Industry assets grew from $320 B to $13.5 T• Includes mutual fund data in later years• Tested for survivorship bias

34

Initial Analysis: Subsequent Performance—QUINTILE Results

These results are statistically significant.

% Difference in Performance: Highest Flow Quintile Managers minus Lowest Flow Quintile Managers

(1985-2006)

-4.0%

-3.0%

-2.0%

-1.0%

0.0%

1.0%

2.0%

3.0%

4.0%

5.0%

Pre-Flow Post Flow

Diff

eren

ce in

A

nnua

lized

Ret

urns

1-Year5-Year

-3.5

-3

-2.5

-2

-1.5

-1

-0.5

01-year 3-years 5-years

35

Subsequent Performance—DOLLARS

Collectively, plan sponsors are losing billions of dollars a year through their manager allocation decisions!

Flows (100’s of Billions) and Value Lost (Billions)(1985--2006)

020406080

100120140160180

Inflows (100's ofBillions)

$Billion Impactfrom 1-yearActive Flows

$Billion Impactfrom 5-yearActive Flows

36

Third, Survey Study test for perception on investment performance

Topic t-stat

Disagree Performance Deteriorates 4Believe Manager Performance Good 10Disagree Performance Improves 4

Results suggest apparent inconsistency between perception and reality.

37

So…What’s Going On?

• Respondents agree they evaluate subsequent performance of decisions and believe their decisions are appropriate and effective

• Yet the More Experienced See ItSome respondents seem to appreciate performance reversals to a greater extent

38

Performance Chasing May be a Problem

1. 98% believe returns are important2. 85% require minimum of 3-yr record3. Anticipated changes in asset class

allocations correlated with trailing returns

0.0

0.2

0.4

0.6

0.8

1.0

1.2

-10.0% -5.0% 0.0% 5.0% 10.0% 15.0% 20.0%

3-Year Trailing Return

Incr

ease

/(In

crea

se+D

ecre

ase)

Regression Line

39

6.4%

4.8%

0.0%

1.0%

2.0%

3.0%

4.0%

5.0%

6.0%

7.0%

1 2

12.0%

10.0%

9.0%

10.0%

11.0%

12.0%

13.0%

1 2

Average Manager Turnover Percentage Disappointed with Supplier Performance

1.

High Level of Performance Chasing

Low Level of Performance Chasing

High Level of Performance Chasing

Low Level of Performance Chasing

t = 2.4 Not statistically significant

Influence of Performance Chasing on Turnover and Performance

Sample ReferencesPresentation based upon:

Stewart, S., Heisler, J., Knittel, C., Neumann, J. (2009). “Absence of Value: an Analysis of Investment Allocation Decisions by Institutional Plan Sponsors.” Financial Analysts Journal, 65(6), 34-51, Nov/Dec.

Heisler, Jeffrey, Christopher R. Knittel, John J. Neumann, and Scott D. Stewart. 2007. “Why Do Institutional Plan Sponsors Hire and Fire Their Investment Managers?” Journal of Business and Economic Studies, vol. 13, no. 13 (Spring):88–115.

References:Barberis, Nicholas, and Andrei Shleifer. 2003. “Style Investing.” Journal of Financial Economics, vol.

68, no. 2 (May):161–199. Busse, Jeffrey, Amit Goyal, and Sunil Wahal. 2006. “Performance Persistence in Institutional

Investment Management.” Working paper, Arizona State University (July).Dalbar, Inc. 2005. “QAIB 2005: Quantitative Analysis of Investor Behavior.” Del Guercio, Diane, and Paula A. Tkac. 2002. “The Determinants of the Flow of Funds of Managed

Portfolios: Mutual Funds versus Pension Funds.” Journal of Financial and Quantitative Analysis, vol. 37, no. 4 (December):523–557.

Fama, Eugene F., and Kenneth R. French. 1992. “The Cross-Section of Expected Stock Returns.” Journal of Finance, vol. 47, no. 2 (June):427–465.

40

References (cont)Goyal, Amit, and Sunil Wahal. 2008. “The Selection and Termination of Investment Management

Firms by Plan Sponsors.” Journal of Finance, vol. 63, no. 4 (August):1805–1847.Grinblatt, Mark, and Sheridan Titman. 1993. “Performance Measurement without Benchmarks: An

Examination of Mutual Fund Returns.” Journal of Business, vol. 66, no. 1 (January):47–68. Grinold, Richard C. 1989. “The Fundamental Law of Active Management.” Journal of Portfolio

Management, vol. 15, no. 3 (Spring):30–37. Gruber, Martin J. 1996. “Another Puzzle: The Growth in Actively Managed Mutual Funds.” Journal of

Finance, vol. 51, no. 3 (July):783–810. Lakonishok, Josef, Andrei Shleifer, and Robert W. Vishny. 1992. “The Structure and Performance of

the Money Management Industry.” Brookings Papers: Microeconomics:339391.Odean, Terrance. 1998. “Are Investors Reluctant to Realize Their Losses?” Journal of Finance, vol.

53, no. 5 (October):1775–1798.Skoulakis, Georgios. 2008. “Panel Data Inference in Finance: Least-Squares vs. Fama-MacBeth.”

Mimeo, University of Maryland.Stewart, Scott. 1998. “Is Consistency of Performance a Good Measure of Manager Skill?”

Journal of Portfolio Management, vol. 24, no. 3 (Spring):22–32.Teo, Melvyn, and Sung-Jun Woo. 2004. “Style Effects in the Cross-Section of Stock Returns.” Journal

of Financial Economics, vol. 74, no. 2 (November).Treynor, Jack. 1990. “The Ten Most Important Questions to Ask in Selecting a Money Manager.”

Financial Analysts Journal, vol. 46, no. 3 (May/June):4–5.Zheng, Lu. 1999. “Is Money Smart? A Study of Mutual Fund Investors' Fund Selection Ability.” Journal

of Finance, vol. 54, no. 3 (June) 41