Is Diversification Dead

Solutions Timely investment ideas from Barclays Global Investors

MARCH 2009

Volume 1, Issue 2|

Is diversification dead? by VINCENT de MARTEL and KEVIN KNEAFSEY

EXECUTIVE SUMMARY Since the summer of 2007, asset-class returns have been headed in the same direction: down. With the exception of government bonds, no asset class—including alternatives— has protected investors from the credit crisis. On the surface, it looks as if diversification has failed investors. Our analysis reveals two factors behind these extreme returns: • A flight to quality, and • Extreme liquidity demand. These two fundamental risk factors— which permeate all asset classes—can explain current returns, not the end of diversification. As the credit crisis unfolds, significant opportunities exist for investors who take

advantage of oversold risk premia other than liquidity. Some of the current market dislocations cannot be sustained in the long run. Looking at market conditions through the lens of liquidity risk premia, signs are emerging that this is no longer the only determinant factor explaining returns. Investors seeking higher riskadjusted long-term returns should diversify not only across asset classes but also across risk premia.

INTRODUCTION Events of the last 18 months have caused many concepts of conventional wisdom in finance to be called into question. One of these concepts, which we rely on heavily, depends on the answe r to the question: Is diversification dead?

should portfolios adapt to this new environment? To answer these questions, we will briefly explore recent performance and then look to the future to apply these findings.

Is it possible that the one free lunch of investing has been taken from us? What fundamental changes have occurred that impact diversification? And how

To get a sense of how diversification has fared, we can look at recent correlation data. Figures 1 and 2 plot rolling 12-month correlations between developed equities and various asset classes over the last three years.

FIGURE 1: ROLLING 12-MONTH CORRELATIONS BETWEEN DEVELOPED EQUITIES AND VARIOUS BOND INDICES High-yield debt

Investment-grade debt

Developed sovereign debt

1.0 0.8 0.6 0.4 0.2 0 -0.2 -0.4 -0.6 -0.8 -1.0 Jan-06 Mar-06 May-06 Jul-06 Sep-06 Nov-06 Jan-07 Mar-07 May-07 Jul-07 Sep-07 Nov-07 Jan-08 Mar-08 May-08 Jul-08 Sep-08 Nov-08

Source: Bloomberg, 01/05 to 12/08. Indices include: MSCI World Standard Core Gross Index Local Currency, Citigroup WGBI 7–10 Year Local Currency Total Return, Barclays Capital Global Aggregate Corporates Total Return Hedged, Barclays Capital US Corporate High Yield Total Return.

Solutions Timely investment ideas from Barclays Global Investors 2

FIGURE 2: ROLLING 12- MONTH CORR ELATION S BETW EEN DEVELOPED EQUITIES AND VARIOUS REAL ASSETS Property

Ex-Energy

Energy

1.0 0.8 0.6 0.4 0.2 0 -0.2 -0.4 -0.6 -0.8 -1.0

Jan-06 Mar-06 May-06 Jul-06 Sep-06 Nov-06 Jan-07 Mar-07 May-07 Jul-07 Sep-07 Nov -7 Jan-08 Mar-08 May-08 Jul-08 Sep-08 Nov-08

Source: Bloomberg, 01/05 to 12/08. Indices include: MSCI World Standard Core Gross Index Local Currency, FTSE EPRA/NAREIT Global Total Return, Dow Jones–AIG ExEnergy Total Return, and Dow Jones–AIG Energy Total Return.

The graphs highlight several things over these three years:

• • •

Correlations are very unstable beasts, especially over short periods. Reliance on them as if they portray highly stable relationships is unwarranted. Correlations between the highest-quality investments— developed market sovereign debt—and riskier investments have been strongly negative. Correlations between risky assets have increased significantly, and in many cases are approaching near-perfect positive levels.

All risky asset classes appear to have moved in the same direction at the same time, which is expressed by their higher correlations through 2008. This is precisely what diversification was meant to avoid. Before we conclude that the theory of diversification is flawed, we must look beyond correlations.

Solutions Timely investment ideas from Barclays Global Investors 3

WHAT IS DIVERSIFICATION? The data in Figures 1 and 2 seem compelling evidence that diversification is dead, and yet we argue here that it is very much alive. To resolve this contradiction, first consider the definition of what it means to diversify. Merriam-Webster’s defines “diversify” as: 1. to make diverse: give variety to

It is probably easier to think of the exposure to each systematic risk factor as the product of (the asset’s loading on that risk factor) X (the price of that risk factor). Big loadings and big prices lead to large exposures. Other exposures are less obvious, because either the asset has a very low loading on a risk factor, or the factor commands little premia in the market, or both.

2. to balance (as an investment portfolio) defensively by dividing funds among securities of different industries or of different classes

FLIGH T TO QUALITY AND LIQUIDITY

3. to increase the variety of the products of ¹

• •

The key words here are “variety” and “different.” Diversification requires variety and exposure to different things, and this has been lacking recently. The conventional wisdom is that diversification is achieved by investing in different asset classes. But is it sufficient to look at the asset-class level? Asset-class investments expose investors to various systematic risks. It is clear, looking at the past 18 months, that asset classes are not the fundamental building blocks of investing. If they were, then surely private and public equity would have performed much differently than real estate and credit bonds or emerging market bonds and commodities. Yet they all seemed to be linked by something more fundamental. Each asset class provides exposure to some mix of systematic risk factors like interest rate risk, economic risk, inflation risk, liquidity risk, and the like. It is exposure to these risks that drives the returns of asset classes.

The primary drivers of price since the crisis began are: A flight to quality, and Extreme liquidity demand.

The first point, flight to quality, is a natural response to increased uncertainty. Market volatility, bank failures, economic weakness, and government responses to the situation have all significantly raised the level of uncertainty. In times of extreme uncertainty, it is natural to seek a safe haven. On the second point, liquidity is an interesting risk factor. It’s like air—you always need it, but you don’t notice it and won’t pay for it when it is abundant. When it is scarce, however, you will pay almost anything for it. What we have witnessed since the summer of 2007 is liquidity pricing driven to extremes, so much so that liquidity exposure (the product of each asset’s loading on liquidity risk and the price of liquidity risk) swamped all other systematic risk exposures and dominated the changes in asset prices. Many market participants were left gasping for breath.

1 From Merriam-Webster’s Online Dictionary. Accessed at www.merriam-webster.com/dictionary/diversify on 2/12/09.

Solutions Timely investment ideas from Barclays Global Investors 4

To see this, consider the following example of the swap spread—the yield difference between Treasury strips and like-duration interest rate swaps. Typically these interest rate swaps pay a higher yield than Treasuries; the yield premium reflects the counterparty risk of the banks on the other side of the swaps. What we have seen recently is that the swap spread has gone negative such that Treasury strips pay a higher rate of interest than the swaps (see Figure 3), even as creditworthiness of the banks backing these swaps

has deteriorated precipitously. This negative swap spread reflects a liquidity premium that the swaps command (reducing the yield they must pay) relative to Treasury strips; the swaps require little capital per unit of notional interest rate exposure, while the Treasury strips require full funding. In other words, the Treasury strips require more liquid capital for the same interest rate exposure. Liquid capital is scarce, so Treasuries must offer a yield premium to compete with capitalefficient interest rate swaps.²

FIGURE 3: 30-YEAR SWAP SPREADS (SWAPS-TREASURY YIELDS) 0.70 0.50

Swaps cheaper than Treasuries

Percent

0.30 0.10

-0.10

Swaps pricier than Treasuries

-0.30 -0.50 -0.70 Jan-08

Mar-08

May-08

Jul-08

Sep-08

Nov-08

Jan-09

Source: BGI as of 2/13/09.

2 Another example where the markets are clearly pricing liquidity at the extreme comes from the investment-grade bond market, in which the same economic exposures can be achieved with a derivatives contract (not requiring cash) and a physical instrument (requiring cash). For instance, the reward for taking the risk of default for IBM by buying a five-year bond rose to 3.0% per year, against only 1.5% for investing in a credit default swap on IBM senior debt. Taking advantage of this seeming arbitrage opportunity requires selling the credit default swaps and buying the corporate bond. In other words, cash is needed to make a profit on the difference. Liquid capital is scarce, which explains why the discrepancy has not been resolved.

Solutions Timely investment ideas from Barclays Global Investors 5

Normally it is each asset’s different exposures to systematic risks that lead to diverse exposure and the benefits of diversification. When one risk factor dominates all others—in this case, liquidity risk— there is no diversity and hence there are no benefits of diversification. The extreme liquidity demand is driven by falling asset values and the need to repay debts or post more collateral. Because consumers and the financial system as a whole are excessively leveraged, they are scrambling to sell assets and realize liquid capital to pay down debt. The need and competition for liquid capital has driven the price of liquidity to extremes.³ What we’ve witnessed, over the last 18 months, is a one-two punch to diversification. First, heightened uncertainty led to a flight to quality in which all risky assets were treated as unattractive, and the highestquality assets were treated as extremely attractive (i.e., the market focused on what was alike about these assets—they were risky—and not their differences). Second, liquidity risk pricing dominated all other systematic risk exposures and led asset prices to move very much in line with each other. When will diversification return to the markets? As uncertainty is resolved about the severity of the crisis, and as liquid capital returns to the markets and works to drive differences in the risk pricing (beyond liquidity), we expect to see the return of diversification. The challenge for investors today is identifying the right barometers to gauge this change.4

BUYING HURRIC ANE INSURANCE AFTER THE HURRICANE HITS Just as natural disasters tend to precede increases in the cost of property insurance, we are now seeing a similar increase in the cost of owning less-risky assets. The domination of liquidity as a factor in the pricing of assets has caused a strong increase in the value of assets offering the highest liquidity. This is particularly the case for Treasury bills and very short-dated cash strategies (such as repurchase agreements) rolled daily. In December, the U.S. Treasury auctioned fourweek T-bills at a yield of zero; on the secondary market, the yield on T-bills even became negative, meaning that some investors were effectively prepared to pay for the right to lend to the US Government over a short period rather than be paid. At the other end of the investment spectrum, the prices of less-liquid or higher-risk assets have fallen. Investors with cash on hand and a long-term investment horizon—arguably a much reduced number—could benefit from higher expected returns. The aversion to illiquidity has pushed the rewards of other risk premia to extremes. Figure 4 shows the yield on US high-yield bonds, which shot up as liquidity became more dear and as the reward for taking outright credit risk increased. The yield reached 22%. If one ignores the liquidity premia, this implies an annual default probability of over 30%, a scenario that could only be justified in a 1929-style depression. For reference, during the spectacular junk bond crisis of the early 1990s, the realized default rate reached 12%.5

3 This is consistent with Hyman Minsky’s “The Financial Instability Hypothesis,” published in 1992 as The Jerome Levy Economics Institute Working Paper No. 74. 4 Investors looking for signs of the renewed power of diversification may consider these barometers of increasing liquidity: swap spreads; yield spreads for off-the-run (previously issued) versus on-the-run (newly issued) Treasury bonds; the TED spread (the difference between 3-month LIBOR and Treasury bills); and the spread between credit default swaps and credit bonds of the same maturity on the same entity. 5 Source: Barclays Capital.

Solutions Timely investment ideas from Barclays Global Investors 6

FIGURE 4: YIELD SPREADS ON HIGH-YIELD BONDS US high-yield minus US government bond yields 25

20

Percent

15

10

5

0 1986

1988

1990

1992

1994

1996

1998

2000

2002

2004

2006

2008

Source: BGI and Datastream, as of 2/16/09.

This significantly improved environment for offering rewards for risk exposure is illustrated by David Viniar, the CFO of Goldman Sachs. He provides an interesting perspective at the end of 2008: “Across many of our businesses, trading margins are robust and the premium on risk capital is higher than we’ve seen in years.”6 Investment banks and hedge funds have traditionally been well placed to benefit from market dislocations. But apart from a few exceptions, both their appetite for taking risk and their access to capital have been substantially eroded. Pension funds and sovereign wealth funds now have an opportunity to take advantage of these large risk premia. Since the extreme risk aversion and focus on liquidity caused diversified strategies to offer little protection against the market turmoil, what can current market conditions tell us about the liquidity risk premium?

ANALYZING CURRE NT CONDITIONS THROUGH THE ILLIQU IDITY LENS Any improvement in the flight-to-quality-driven pricing will likely be reflected in improved investor sentiment. In periods of high stress and risk aversion, investors tend to flock to the safest investments and divest from high-risk assets, causing prices of less-risky assets to increase and prices of higher-risk assets to fall. When low-risk asset classes also become the highest-performing ones, investor sentiment becomes highly negative—investors in risky assets are rewarded poorly or even punished by the market. Figure 6 illustrates the evolution of investment sentiment from 1994 to 2008. It is calculated as the correlation of risk and return over one year, for 13 global asset classes. We see that investor sentiment is currently strongly negative; risk taking has not been well rewarded.

6 Source: Goldman Sachs Q4 2008 Earnings Call, December 16, 2008.

Solutions Timely investment ideas from Barclays Global Investors 7

WHAT CAN PENSION PLANS DO? It may be tempting to conclude that the best place to be today is out of the market. Either by choice or by coincidence, many investors find themselves in this very position. Although it may be comforting in the short term, divesting leaves open the question of when to re-enter the market, which requires at least as great a skill as the exit decision. One way of quantifying the opportunity cost is to look at by how much investors would have gained (or lost) by remaining uninvested following certain runs of underperformance. The Dow Jones Industrial Average has daily price history over a sufficiently long history to permit this analysis. Here we look at the worst 90-day returns for each of the major stock market crashes, including the Great Depression. Figure 5 compares the returns of the DJIA invested for the five years immediately following the 90-day periods of underperformance to sitting out of the market for the first six months following the bad 90-day period and then being fully invested the remaining 4½ years. F I G U R E 5 : R A NG E O F O P T I O NS Start date

End date

Performance

Five-year performance with lag 0 days

6 months

Performance gap

Annualized (X 5 years)

Mar 4, 1932

July 8, 1932

–52%

318%

174%

144%

20%

Aug 13, 1931

Dec 17, 1931

–47%

147%

262%

–115%

–17%

July 10, 1929

Nov 13, 1929

–42%

–51%

–64%

14%

3%

July 22, 1937

Nov 25, 1937

–38%

1%

6%

–5%

–1%

July 17, 2008

Nov 20, 2008

–34%

?

?

?

?

July 31, 1987

Dec 4, 1987

–31%

86%

58%

27%

5%

Mar 19, 2002

July 23, 2002

–28%

81%

65%

16%

3%

June 4, 1974

Oct 8, 1974

–27%

47%

18%

29%

5%

May 18, 2001

Sep 21, 2001

–27%

40%

11%

30%

5%

Source: BGI, as of 12/31/08. Past performance is no guarantee of future results.

In 1931, staying uninvested for six months would have been very beneficial, as the market plunged again in early 1932. During other periods in which the market recovered after six months, the annual return gap was 3–5% due to being uninvested for six months. (We find similar results over different time periods.) The opportunity cost can be substantial. Investing in a well-constructed portfolio and sticking to it, even in difficult times, may be a more practical option than trying to buy low and sell high. The decision to invest or not in diversified strategies may be summarized by an observation from Peter L. Bernstein. Investors must ask “What are the consequences if I’m wrong?”7 In the case of diversification, there are two decisions to be made: •

For investors mostly in cash or in very low-risk assets (by choice or not), it is a balance of the risks of experiencing more negative returns and further destroying wealth, against the risk of missing out on positive returns and never meeting long-term return objectives.

•

For investors already in risky assets, it is the risk of being wrong to invest in a single asset class (such as betting on a recovery in equities) against the risk of investing in a diversified portfolio of asset classes (which may have a lower return than equities).

7 From “Lessons from a Crisis,” Pensions & Investments, December 8, 2008.

Solutions Timely investment ideas from Barclays Global Investors 8

FIGURE 6: INVESTOR SENTIMENT 100

Highest-risk assets have highest returns = Feeling good about taking risks

80 60

Percent

40 20 0 -20 -40 -60 -80

Highest-risk assets have worst returns = Not wanting to take risks

-100 Jan-94

Jan-96

Jan-98

Jan-00

Jan-02

Jan-04

Jan-06

Jan-08

Source: BGI, as of 2/11/09.

In the final weeks of 2008 and the first weeks of 2009, there has been a sign of tentative return to lessvolatile market conditions, as some uncertainty is being resolved. We see this in the equity markets, with the VIX index remaining well below the maximum points reached in October and November.

should come as no surprise to anyone who has analyzed data and observed that asset-class diversification offers little protection when extreme risk premia pricing (most recently, liquidity) swamps risk exposure differences across all asset classes.

Liquidity also shows signs of an easing. In the bond markets, the difference in price between corporate bonds and credit default swaps has been narrowing. In the cash markets, we have seen the difference between LIBOR rates and overnight rates falling to levels last seen before the Lehman Brothers bankruptcy. Alas, all is not perfect, as investor sentiment remains low, and some signs of dislocation persist (e.g., 30-year government bonds are still cheaper than 30-year swaps).

The evidence remains, however, that portfolios invested predominantly in equities with the goal of providing future growth are overly exposed to risks in an economic cycle that can affect all equity markets simultaneously. Diversified portfolios provide a greater risk-adjusted return potential over time than equities because they are exposed to different risk premia. Investors should attempt to create more-efficient portfolios with a genuine long-term view. Here we can take a lesson from civil engineers: We wouldn’t build a bridge assuming that the past three years of weather reports are a good indication of future weather patterns.

AN ANSWER TO OUR QUESTION The events of 2008 have proved that diversification across asset classes will not offer consistent positive returns and low risks in all market conditions. The search for that financial rosetta stone is not over. That

So to answer the original question, is diversification dead? Our answer is no, diversification is not dead; it was just hibernating. And we’re hopeful that spring is near.

Solutions Timely investment ideas from Barclays Global Investors 9

Solutions from Barclays Global Investors Please direct questi ons and comments on this topic to Vincent de Martel, senior investment strategist: Telephone 415 597 2492 Facsimile 916 861 8045 [email protected] Please direct questions and comments on this publication to Marcia Roitberg, editor: Telephone 850 893 8586 Facsimile 415 618 1455 [email protected]

For ease of reference, “BGI” may be used to refer to Barclays Global Investors, N.A. and its affiliates. Barclays Global Investors, N.A., a national banking association operating as a limited purpose trust company, manages the investment strategies and other fiduciary services referred to in this publication and provides fiduciary and trust services to various institutional investors. Strategies maintained by Barclays Global Investors are not insured by the Federal Deposit Insurance Corporation and are not guaranteed by BGI or its affiliates. BGI does not provide investment advice regarding any security, manager or market. The information is not intended to provide investment advice. BGI does not guarantee the suitability or potential value of any particular investment. Past performance is no guarantee of future results.

Investing involves risk, including possible loss of principal. Asset allocation and diversification do not promise any level of performance or guarantee against loss of principal. The information included in this publication has been taken from trade and other sources we consider to be reliable. We do not represent that this information is accurate and complete and should not be relied upon as such. Any opinions expressed in this publication reflect our judgment at this date and are subject to change. No part of this publication may be reproduced in any manner without the prior written permission of Barclays Global Investors. This material is not an offer to sell, nor an invitation to apply for any particular product or service. FOR INSTITUTIONAL USE ONLY— NOT FOR PUBLIC DISTRIBUTION ©2009 Barclays Global Investors, N.A. All rights reserved. BGI-0260-0309

Solutions Timely investment ideas from Barclays Global Investors 10