Clarendon Lectures in Finance: The Adaptive Markets Hypothesis · PDF file · 2013-06-17Number of fund launches 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007

Clarendon Lectures in Finance:

The Adaptive Markets Hypothesis

© 2013 by Andrew W. Lo All Rights Reserved

Andrew W. Lo, MIT

Lecture 3: Hedge Funds—The Galapagos Islands of Finance

June 14, 2013

TexPoint fonts used in EMF. Read the TexPoint manual before you delete this box.: AAAAAAAAAAAAAAA

Clarendon What Are Hedge Funds?

Unregulated Investment Companies For “qualified” (sophisticated) investors

Need not satisfy regulatory requirements (few investment constraints)

High fees, high performance (historically), and high attrition

Alfred Winslow Jones First “hedge fund” in 1949 (market exposure vs. stock selection):

Magnify stock selection (leverage). reduce market exposure (short positions)

Hence the term “hedge”

Charged 20% incentive fee

Eventually included several managers (fund of funds)

M ar k e t E x p o su r e =L o n g P o si t io n ¡ S h o r t P o si t io n

C a p i t a l

© 2013 by Andrew W. Lo

All Rights Reserved

Page 2 Lecture 3


Why Should We Care About Hedge Funds?

Hedge funds play a key role in the financial industry – During normal times, hedge funds are the “tip of the spear”

– During bad times, hedge funds are the “canary in the coalmine”

As unregulated entities, hedge funds innovate rapidly

Due to leverage, hedge funds have disproportionate impact on markets

Investors in hedge funds include: – Private investors – Fund of funds – Central banks and sovereign wealth funds – Insurance companies – Pension funds


All Rights Reserved

Page 3 Lecture 3


Hedge Funds Are The “Galapagos Islands” of Finance

Relatively low barriers to entry and exit

High levels of compensation (stakes are high)

Competition and adaptation are extreme

New “species” are coming and going constantly

Strategies wax and wane over time:

– Credit strategies are waxing

– Dedicated short bias is waning

Empirical evidence for Adaptive Markets Hypothesis


All Rights Reserved

Page 4 Lecture 3

Clarendon Dynamics of the Hedge Fund Industry


All Rights Reserved

Lecture 3 Page 5

Number of fund launches 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010

All Funds 499 570 610 646 662 793 850 1257 1472 1732 2139 2194 2372 2235 1880 1880 1370

CTA 111 111 72 64 64 66 34 54 54 53 43 51 39 58 54 48 60

ConvertibleArbitrage 16 10 23 14 20 14 24 33 34 24 32 21 19 16 11 14 18

DedicatedShortBias 3 1 4 4 1 8 3 1 1 4 4 7 3 9 4 1 5

EmergingMarkets 27 38 34 55 39 45 36 21 26 58 92 134 140 156 138 112 74

EquityMarketNeutral 12 12 17 21 36 49 25 68 78 79 85 81 72 87 54 48 28

EventDriven 25 38 42 45 43 48 67 76 97 76 96 69 101 63 41 51 44

FixedIncomeArbitrage 18 14 20 17 23 16 22 39 49 55 57 52 42 30 18 38 41

FundofFunds 106 121 120 145 168 207 275 492 602 725 920 902 995 821 720 873 572

GlobalMacro 11 30 19 29 29 30 22 26 49 57 58 85 84 95 73 163 74

LongShortEquityHedge 81 109 152 162 167 215 271 330 329 371 397 473 468 373 285 196 178

ManagedFutures 75 65 80 59 49 52 33 44 45 74 118 75 123 105 64 74 58

MultiStrategy 11 18 22 23 16 38 33 50 91 108 184 200 218 355 324 183 173

OptionsStrategy 0 0 0 4 0 1 0 5 1 3 4 4 6 2 5 7 7

Other 3 3 5 4 7 3 5 18 16 44 48 39 53 61 84 71 38

Undefined 0 0 0 0 0 1 0 0 0 1 1 1 9 4 5 0 0

catIgn1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0

Births 1994 to 2010 in Lipper TASS


Number of fund closings 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010

All Funds 240 294 337 302 379 405 456 521 634 636 843 1092 1288 1779 2846 2505 2210

CTA 137 141 113 78 89 83 71 62 48 44 57 68 56 87 63 57 68

ConvertibleArbitrage 2 0 12 7 9 7 3 12 8 15 31 41 28 46 39 21 23

DedicatedShortBias 1 2 1 2 0 1 0 7 2 0 5 5 5 13 6 3 10

EmergingMarkets 2 3 11 18 34 21 35 36 18 17 26 33 50 76 149 129 125

EquityMarketNeutral 1 4 4 2 7 18 18 19 31 52 51 54 54 92 98 86 88

EventDriven 8 7 8 13 17 33 28 36 52 47 60 53 100 91 125 92 86

FixedIncomeArbitrage 4 6 7 4 14 8 12 9 16 12 19 36 37 90 78 51 32

FundofFunds 21 32 45 71 87 84 112 134 159 187 286 376 428 505 1138 1323 969

GlobalMacro 8 9 20 9 14 25 32 18 14 21 21 43 56 104 77 89 106

LongShortEquityHedge 25 33 42 48 53 66 91 146 210 177 201 252 313 444 581 334 415

ManagedFutures 28 45 72 41 50 49 46 33 45 34 31 53 62 75 155 67 69

MultiStrategy 3 12 2 8 4 8 8 7 24 24 39 60 66 118 265 201 157

OptionsStrategy 0 0 0 1 0 0 0 0 0 3 0 1 0 3 1 6 5

Other 0 0 0 0 1 2 0 2 7 3 15 17 33 24 61 46 57

Undefined 0 0 0 0 0 0 0 0 0 0 1 0 0 11 10 0 0

catIgn1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0


All Rights Reserved

Lecture 3 Page 6

Deaths 1994 to 2010 in Lipper TASS



All Rights Reserved

Lecture 3 Slide 7

Source: Credit Suisse 2012 Hedge Fund Market Review

Clarendon

Quantitative Equity Funds Hit Hard In August 2007

Specifically, August 7–9, and massive reversal on August 10

Some of the most consistently profitable funds lost too

Seemed to affect only quants

Lack of Transparency Is Problematic!

In Khandani and Lo (2007) we used a daily mean- reversion strategy to study these events:

Quant Meltdown of August 2007

Wall Street Journal September 7, 2007


All Rights Reserved

Page 8 Lecture 3

Clarendon Quant Meltdown of August 2007

Simulated Historical Performance of Contrarian Strategy


All Rights Reserved

Page 9 Lecture 3

Clarendon


All Rights Reserved

Lecture 3 Page 10


Simulated Historical Performance of Contrarian Strategy

Clarendon


All Rights Reserved

Lecture 3 Page 11


Clarendon

How Much Leverage Needed To Get 1998 Expected Return Level? In 2007, use 2006 multiplier of 4

8:1 leverage

Compute leveraged returns

How did the contrarian strategy

perform during August 2007?

Recall that for 8:1 leverage:

– E[Rpt] = 4 × 0.15% = 0.60%

– SD[Rpt] = 4 × 0.52% = 2.08%

2007 Daily Mean: 0.60%

2007 Daily SD: 2.08%

Year

Average

Daily

Return

Return

Multiplier

Required

Leverage

Ratio

1998 0.57% 1.00 2.00

1999 0.44% 1.28 2.57

2000 0.44% 1.28 2.56

2001 0.31% 1.81 3.63

2002 0.45% 1.26 2.52

2003 0.21% 2.77 5.53

2004 0.37% 1.52 3.04

2005 0.26% 2.20 4.40

2006 0.15% 3.88 7.76

2007 0.13% 4.48 8.96

Required Leverage Ratios For Contrarian Strategy To Yield 1998 Level of Average Daily Return


All Rights Reserved

Page 12 Lecture 3


Clarendon

Daily Returns of the Contrarian Strategy In August 2007


All Rights Reserved

Page 13 Lecture 3


Clarendon Quant Meltdown of August 2007 Daily Returns of Various Indexes In August 2007


All Rights Reserved

Page 14 Lecture 3



All Rights Reserved

Lecture 3 Page 15

Source: John B. Taylor



All Rights Reserved

Lecture 3 Page 16

Source: Sengupta and Tam (2008, St. Louis Fed)

3-Month LIBOR/OIS Spread August 2006 to October 2008

Clarendon Comparison with August 1998


All Rights Reserved

Lecture 3 Page 17

Daily Returns of the Contrarian Strategy In August and September 1998

Clarendon Comparison with August 1998


All Rights Reserved

Page 18 Lecture 3

Daily Returns of the Contrarian Strategy In August and September 1998

Clarendon Market-Making Profits During August 2007 Cumulative m -Min Returns of Intra-Daily Contrarian Profits for Deciles 10/1 of

S&P 1500 Stocks July 2 to September 30, 2008

0.00

0.50

1.00

1.50

2.00

2.50

3.00

3.50

4.00

4.50

7/2/07

12:00:00

7/11/07

12:00:00

7/19/07

12:00:00

7/27/07

12:00:00

8/6/07

12:00:00

8/14/07

12:00:00

8/22/07

12:00:00

8/30/07

12:00:00

9/10/07

12:00:00

9/18/07

12:00:00

9/26/07

12:00:00

Cu

mu

lati

ve

Re

turn

60 Min

30 Min

15 Min

10 Min

5 Min


All Rights Reserved

Page 19 Lecture 3

Clarendon The Unwind Hypothesis

Khandani and Lo (2007, 2011) Conjecture That: Losses due to rapid and large unwind of quant fund (market-

neutral), and liquidation is likely forced because of firesale prices (sub-prime?)

Initial losses caused other funds to reduce risk and de-leverage, and de-leveraging caused further losses across broader set of equity funds

Friday rebound consistent with liquidity trade, not informed trade, and rebound due to quant funds, long/short, 130/30, long-only funds

How and Why Did This Happen?


All Rights Reserved

Page 20 Lecture 3

Clarendon


All Rights Reserved

Lecture 3 Slide 21

The Financial Crisis

0

100

200

300

400

500

600

700

800

900

1000

0

50

100

150

200

250

1880 1900 1920 1940 1960 1980 2000 2020

Po

pu

lati

on

in M

illio

ns

Re

al H

om

e P

rice

Ind

ex

U.S. Real Home Price Index, 1890 –2012

Source: Robert J. Shiller

Home Prices

Population

Clarendon


All Rights Reserved

Lecture 3 Slide 22

Corporate Federal Agency Municipal Treasury1 Mortgage-Related2 Debt3 Securities Asset-Backed Total

1996 185.2 612.4 479.7 343.7 277.9 168.4 2,067.2 1997 220.7 540.0 577.6 466.0 323.1 223.1 2,350.5 1998 286.8 438.4 1,118.1 610.7 596.4 286.6 3,336.9 1999 227.5 364.6 985.4 629.2 548.0 287.1 3,041.8 2000 200.8 312.4 660.0 587.5 446.6 281.5 2,488.8 2001 287.7 380.7 1,663.9 776.1 941.0 326.2 4,375.6 2002 357.5 571.6 2,283.0 636.7 1,041.5 373.9 5,264.2 2003 382.7 745.2 3,084.3 775.8 1,267.5 461.5 6,717.0 2004 359.8 853.3 1,879.0 780.7 881.8(4) 651.5 4,524.3 2005 408.2 746.2 2,182.4 752.8 669.0 753.5 5,512.1 2006 386.5 788.5 2,088.8 1,058.9 747.3 753.9 5,823.9 2007 429.3 752.3 2,186.2 1,127.5 941.8 509.7 5,946.8 2008 389.5 1,037.3 1,362.2 707.2 984.5 139.5 4,620.2 2009 409.8 2,185.5 2,041.4 901.8 1,117.0 150.9 6,806.4 2010 433.1 2,303.9 1,742.7 1,062.7 1,032.6 109.4 6,684.5

U.S. Bond Market Debt Issuance ($Billions)

1 Interest bearing marketable coupon public debt. 2 Includes GNMA, FNMA, and FHLMC mortgage-backed securities and CMOs and private-label MBS/CMOs. 3 Includes all non-convertible debt, MTNs and Yankee bonds, but excludes CDs and federal agency debt. 4 Beginning with 2004, Sallie Mae has been excluded due to privatization. Source: SIFMA

The Financial Crisis

What Could Possibly Go Wrong?

Clarendon


All Rights Reserved

Lecture 3 Slide 23

Who Benefited From This Trend?:

Commercial banks Credit rating agencies (S&P, Moody’s, Fitch) Economists Government sponsored enterprises Homeowners Insurance companies (multiline, monoline) Investment banks and other issuers of MBSs, CDOs, and CDSs Investors (hedge funds, pension funds, mutual funds, others) Mortgage lenders, brokers, servicers, trustees Politicians Regulators (CFTC, Fed, FDIC, FHFA, OCC, OTS, SEC, etc.)

“A Rising Tide Lifts All Boats” (Everybody Benefits)

How Could This Have Happened?

Clarendon No Negative Feedback In The System


All Rights Reserved

Slide 24 Lecture 3

Pain and negative feedback produce a response

– Fear learning, preventive measures, a new narrative

– Many regulations arise from pain

– Fire codes and the Triangle Shirt Waist Factory fire of March 25, 1911 (146 died, 60 new laws enacted between 1911 and 1913)

What if we felt no pain?

– Anesthetic (recall methamphetamine)

– Disrupts critical negative feedback control loop

– No learning, or incorrect learning

Clarendon No Negative Feedback In The System


All Rights Reserved

Slide 25 Lecture 3

Negative Feedback Is Unpleasant But Useful

Clarendon Why?

Adaptive Markets Perspective

We develop many narratives that guide our behavior

These narratives are shaped by natural selection, but only if there is accurate feedback: pain for less beneficial narratives, pleasure for more beneficial narratives

The only reliable way to obtain accurate feedback on our narratives is the systematic and objective empirical analysis of hypotheses, i.e.,

The Scientific Method


All Rights Reserved

Lecture 3 Slide 26

Clarendon Many Narratives Have Emerged

Popular Narratives of the Crisis

Crisis started with a “run on repo”

Bankers didn’t have enough “skin in the game”

Wall street bonuses were too high

Predatory lending created the subprime crisis

No one saw the crisis coming

Devotion to market efficiency caused the crisis

Changes in regulation allowed huge increases in leverage

But Are They True?


All Rights Reserved

Lecture 3 Slide 27

Clarendon What Can Be Done?

The NTSB Model

No regulatory authority

Investigates accidents and issues reports

Investigative teams include industry reps

Conducts forensic examinations

Publicly available searchable database

Http://www.ntsb.gov/ntsb/query.asp

Example: USAir flight 405


All Rights Reserved

Lecture 3 Slide 28

Clarendon What Can Be Done?

NTSB Report AAR–93/02, p. vi:

The National Transportation Safety Board determines that the probable cause of this accident were the failure of the airline industry and the Federal Aviation Administration to provide flightcrews with procedures, requirements, and criteria compatible with departure delays in conditions conducive to airframe icing and the decision by the flightcrew to take off without positive assurance that the airplane's wings were free of ice accumulation after 35 minutes of exposure to precipitation following de-icing. The ice contamination on the wings resulted in an aerodynamic stall and loss of control after liftoff. Contributing to the cause of the accident were the inappropriate procedures used by, and inadequate coordination between, the flightcrew that led to a takeoff rotation at a lower than prescribed air speed.


All Rights Reserved

Lecture 3 Slide 29

Clarendon


All Rights Reserved

Lecture 3 Slide 30

What Can Be Done? Break up banks and broker/dealers that are “too big to fail”

Create exchanges for CDSs and other large OTC contracts

Create financial NTSB for analyzing all blow-ups

Require confidential disclosure regarding “network” exposures

Implement counter-cyclical leverage constraints for bank-like entities

Enforce “suitability” requirements for mortgage-broker advice

Require certification for mgmt. and boards of complex financial institutions

Impose better mark-to-market accounting and risk controls

Impose capital adequacy requirements for all bank-like entities

Create new discipline of “risk accounting”

Consolidate insurance regulation at the federal level

Impose small derivatives tax to fund financial engineering programs

Revise laws to allow “pre-packaged” bankruptcies for finance companies

Change corporate governance structure (compensation, CRO role, etc.)

Teach economics, finance, and risk management in high school

Clarendon


All Rights Reserved

Lecture 3

In the beginning… Implications: Correlation matters; diversification Cost of capital for fundamental analysis Benchmarks, performance attribution Passive investing Indexation and hedging Portable alpha overlays Portfolio construction and risk budgeting Framework for discharging fiduciary duty

The Traditional Investment Framework

Slide 31

Clarendon


All Rights Reserved

Lecture 3

But This Framework Requires Several Key Assumptions:

Relationship is linear

Relationship is static across time and circumstances

Parameters can be accurately estimated

Investors behave rationally

Markets are stationary (static probability laws)

Markets are efficient

What If Some of These Assumptions Don’t Hold?


Slide 32

Clarendon


All Rights Reserved

Lecture 3


Slide 33

But Do They Still Hold Today??

Cumulative Return of S&P 500 (log scale) January 1926 to July 2011

Clarendon The Traditional Investment Framework


All Rights Reserved

Lecture 3 Slide 34

Clarendon

Long-Term Risk Premia of Various Asset Classes January 1929 to December 2009

Asset Class Geometric

Mean

Arithmetic

Mean

Standard

Deviation

Small Company Stocks 11.9% 16.6% 32.8%

Large Company Stocks 9.8% 11.8% 20.5%

Long-Term Corporate Bonds 5.9% 6.2% 8.3%

Long-Term Government Bonds 5.4% 5.8% 9.6%

Intermediate-Term Government Bonds 5.3% 5.5% 5.7%

U.S. Treasury Bills 3.7% 3.7% 3.1%

Inflation 3.0% 3.1% 4.2% Source: Summary statistics computed from Ibbotson’s Stocks, Bonds, Bills, and Inflation series from January 1926 to December 2009 (2010, Table 2–1).

Risk/Reward Relationship Seems To Apply Over Long Periods of Time

Implications


All Rights Reserved

Slide 35 Lecture 3

Clarendon

-30%

-20%

-10%

0%

10%

20%

30%

40%

50%

1250-Day Annualized Return 1250-Day Annualized Volatility

Risk/Reward Relationship Need Not Apply Over Shorter Periods of Time!

Implications


All Rights Reserved

Slide 36 Lecture 3

Clarendon Implications

Efficient Markets

Long-only constraint

Diversify across stocks and bonds

Market-cap-weighted indexes

Manage risk via asset allocation

Alpha vs. market beta

Markets are efficient

Equities in the long run

Adaptive Markets

Long/short strategies

Diversify across more asset classes and strategies

Passive transparent indexes

Manage risk via active volatility scaling algorithms

Alphas multiple betas

Markets are adaptive

“In the long run we’re all dead”, but make sure the short run doesn’t kill you first

Lecture 3 Slide 37 © 2013 by Andrew W. Lo

All Rights Reserved

Clarendon


All Rights Reserved

Lecture 3 Slide 38

Conclusion

Many Disciplines Focus On Human Behavior Anthropology Biology Computer Science Economics Neuroscience Organizational science Political Science Psychology Sociology

What Is The Common Denominator? Homo sapiens

Clarendon

Lecture 3

Conclusion

Consilience (E.O. Wilson, 1998):

The Consilience of Inductions takes place when an Induction, obtained from one class of facts, coincides with an Induction, obtained from another different class. This Consilience is a test of the truth of the Theory in which it occurs.

— William Whewell, 1840, Philosophy of

the Inductive Sciences, 1840.


All Rights Reserved

Slide 39

Clarendon


All Rights Reserved

Lecture 3 Slide 40

Conclusion

Even Samuelson (1947) Raised Concerns: …[O]nly the smallest fraction of economic writings, theoretical and applied, has been concerned with the derivation of operationally meaningful theorems. In part at least this has been the result of the bad methodological preconceptions that economic laws deduced from a priori assumptions possessed rigor and validity independently of any empirical human behavior. But only a very few economists have gone so far as this. The majority would have been glad to enunciate meaningful theorems if any had occurred to them. In fact, the literature abounds with false generalization.

We do not have to dig deep to find examples. Literally hundreds of learned papers have been written on the subject of utility. Take a little bad psychology, add a dash of bad philosophy and ethics, and liberal quantities of bad logic, and any economist can prove that the demand curve for a commodity is negatively inclined.

Clarendon Conclusion

A New Narrative

Efficient Markets is not wrong; it is just incomplete

Human behavior has been stable for 60,000 years

Our environment has changed rapidly

The mismatch can create challenges

Evolution determines dynamics

Competition, selection, innovation

AMH: How Adaptive Are You?


All Rights Reserved

Lecture 3 Slide 41

Thank You!

Clarendon Further Reading Brennan, T. and A. Lo, 2011, “The Origin of Behavior”, Quarterly Journal of Finance 1, 55–108.

Brennan, T. and A. Lo, 2012, “An Evolutionary Model of Bounded Rationality and Intelligence”, PLOS ONE 7: e34569. doi:10.1371/journal.pone.0050310.

Farmer, D. and A. Lo, 1999, “Frontiers of Finance: Evolution and Efficient Markets”, Proc. Nat. Acad. Sci. 96, 9991–9992.

Hasanhodzic, J. and A. Lo, 2007, “Can Hedge-Fund Returns Be Replicated?: The Linear Case”, Journal of Investment Management 5, 5–45.

Lo, A., 1999, “The Three P’s of Total Risk Management”, Financial Analysts Journal 55, 13–26.

Lo, A., 2001, “Risk Management for Hedge Funds: Introduction and Overview”, Financial Analysts Journal 57, 16–33.

Lo, A., 2004, “The Adaptive Markets Hypothesis: Market Efficiency from an Evolutionary Perspective”, Journal of Portfolio Management 30, 15–29.

Lo, A., 2005, “Reconciling Efficient Markets with Behavioral Finance: The Adaptive Markets Hypothesis”, Journal of Investment Consulting 7, 21–44.

Lo, A., 2008a, Hedge Funds: An Analytic Perspective. Princeton, NJ: Princeton University Press.

Lo, A., 2008b, “Hedge Funds, Systemic Risk, and the Financial Crisis of 2007–2008: Written Testimony for the House Oversight Committee Hearing on Hedge Funds (November 13, 2008)”, Available at SSRN: http://ssrn.com/abstract=1301217.

Lo, A., 2012, “ Adaptive Markets and the New World Order, Financial Analysts Journal 68, 18–29.

Lo, A., 2012, “Fear, Greed, and Financial Crises: A Cognitive Neurosciences Perspective”, to appear in J.P. Fouque and J. Langsam, eds., Handbook of Systemic Risk, Cambridge University Press.

Lo, A. and C. MacKinlay, 1999, A Non-Random Walk Down Wall Street. Princeton, NJ: Princeton University Press.


All Rights Reserved

Lecture 3 Slide 43

http://ssrn.com/abstract=1301217

Documents

Clarendon Lectures in Finance: The Adaptive Markets Hypothesis · PDF file · 2013-06-17Number of fund launches 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007