What's So Absolute about Absolute Return Funds?

The term absolute return is a funny one. It literally only means the percentage gain or loss on an investment over time, but asset managers and media alike have mutated the term to describe a sublime method of investing, a sort of nirvana where, regardless of the macroenvironment, positive returns with low volatility are achievable.

For example, take Putnam's funds (Putnam Absolute Return 100  (PARTX), Putnam Absolute Return 300  (PTRNX), Putnam Absolute Return 500 , Putnam Absolute Return 700 ). The prospectus claims that the absolute return moniker "distinguishes the fund's goal and investment strategies from those of most other mutual funds available in the market place," which are "managed with the goal of outperforming an index." Instead, these funds seek "to earn a positive total return over a reasonable period of time regardless of market conditions or general market direction," and "investors should expect the funds to outperform the general securities markets during periods of flat or negative performance, to underperform during periods of strong market performance, and typically to produce less volatile returns than the general market." Other absolute return mutual funds promise more or less the same--positive returns, low correlations to the market, and low volatility.

If this goal seems a little lofty, it is. We examined 19 distinct absolute return mutual funds in our database, both dead and alive, and found that these funds have not, in fact, performed as planned. Furthermore, we examine the idea of an absolute return fund, the idea that there exists an investment strategy that allows positive returns with low correlations or sensitivities to general market movements, and find that, in fact, these funds do not follow such an investment strategy.

Absolute Return Mutual Fund Performance
The earliest fund touting an absolute return name tag, UBS Absolute Return Bond  , launched in April 2005, during the heyday of hedge funds, which first branded the absolute concept. Since then, four of these funds have liquidated ( , , , ), three with negative total returns since inception. The live funds haven't performed much better. Returns since inception of these funds range widely, from negative 20.81% to 17.58%, with five of the 15 falling on the negative end of the spectrum. So much for the promise of positive returns over time.

Absolute Return Funds Return and Standard Deviations as of May 31
	Inception Date	Obsolete Date	Total Return Incep-Cumulative (Mo-End) USD	Return 1/1/09 to 5/31/09 USD	Annual Return 2008 USD	Standard Deviation 6/1/08 to 5/30/09 USD	Standard Deviation 6/25/06 to 5/2/09 USD
CMG Absolute Ret Strat	5/1/2009		-0.40
Putnam Absolute Ret 100 (PARTX)	12/24/2008		1.12	1.10
Putnam Absolute Ret 300 (PTRNX)	12/24/2008		2.12	2.10
Putnam Absolute Ret 500	12/24/2008		2.60	2.60
Putnam Absolute Ret 700	12/24/2008		4.00	4.00
Absolute Opportunities	10/21/2008		17.58	15.00
Goldman Sachs Ab Ret Track (GARTX)	5/30/2008		-11.10	3.61		11.93
Aston/New Cent Ab Ret ETF	3/4/2008		-14.92	3.82		19.35
Dreyfus Global Ab Ret	12/18/2007		-9.24	2.48	-9.33	11.75
Credit Suisse Ab Ret Class	12/29/2006	12/29/2008	-5.12
Guerite Absolute Ret	12/29/2006	11/27/2007	-1.50
Nakoma Absolute Ret	12/18/2006		6.84	-3.35	-4.34	7.68
Western Asset Absolute Ret (WAARX)	7/6/2006		3.48	11.36	-14.62	11.87
JHancock2 Absolute Ret Port	6/23/2006	5/7/2009	-13.20		-27.36		16.04
RiverSource Ab Ret Ccy & Inc	6/15/2006		10.06	-1.52	1.07	5.27	4.78
QCM Absolute Return	12/20/2005	1/31/2008	8.76
Akros Absolute Return	9/30/2005		13.16	15.95	-2.89	21.55	13.18
Absolute Strategies (ASFIX)	7/11/2005		5.74	8.28	-13.54	12.56	7.55
UBS Absolute Return Bond	4/27/2005		-20.81	2.73	-23.05	4.27	3.65

Comparing these funds' returns with the market, we see that in 2008, when the S&P 500 lost 37% and the BarCap U.S. Aggregate Bond Index gained 5.4%, only one of eight "absolute return" funds produced a positive return, RiverSource Absolute Ret Ccy & Inc , but even this fund earned less than U.S. Treasuries. It's safe to say, then, that these funds have across the board failed to outperform in bad times. Let's look at good times. In 2009 (through the end of May), both stocks and bonds are up, yet seven of 14 funds reporting performance for that period are lagging stocks, and two of those laggards are actually losing investors' money. So in good times these funds have underperformed the market, as predicted, but that doesn't make investors feel any better.

Now for volatility. Standard deviations also vary, from the low single digits to 20%-plus over 52- and 156-week periods through May 31, 2009. Only two funds (UBS Absolute Return Bond and RiverSource Absolute Ret Ccy & Inc) achieved bondlike or low volatility over these periods, and they happen to be bond funds, one of which lost 21% last year. Not exactly a "safe" investment. This leads us to our next topic: What exactly is an absolute return strategy, and does it exist?

The Absolute Return Investment Strategy
If you tried to find a common theme among absolute return mutual fund investment strategies, you wouldn't. The easiest way to see this is to examine categorization. These funds fall in six different categories (long-short, short-term bond, multisector bond, conservative allocation, moderate allocation, and world allocation), reflecting a wide variety of investment strategies, ranging from long only to long-short within one asset class, such as equities or bonds, or tactically allocating between them. Some even delve into less-traditional asset classes, such as currencies and commodities. A true, theoretical absolute return investment strategy, however, is much stricter.

Modern Portfolio Theory says that all returns are achieved through exposure to market risk (beta) and manager skill or luck (alpha) in timing the market or selecting securities. An absolute return strategy neutralizes this market risk by taking opposite (long and short) positions in securities with equal exposure to market risk. Absolute return, then, is theoretically a market neutral, or zero beta, strategy. This also results in zero correlation with the market (because the portfolio's zero beta is multiplied by the ratio of the market and portfolio volatility).

Betas to Stocks and Bonds
	Beta to S&P 500 6/1/08 to 5/30/09 USD	T-Stat to S&P 500 Index	R2 6/1/08 to 5/30/09 USD	Beta to BarCap US Agg Bond 6/1/08 to 5/30/09 USD	T-Stat to BarCap US Agg Bond Index	R2 6/1/08 to 5/30/09 USD
Goldman Sachs Ab Ret Tr (GARTX)	0.24	8.69	60.17	-0.12	-0.38	0.29
Aston/New Cent Ab Ret ETF	0.44	12.22	74.93	0.05	0.09	0.02
Dreyfus Global Ab Ret	0.20	5.93	41.32	0.49	1.60	4.87
Nakoma Absolute Return	0.01	0.37	0.28	-0.04	-0.21	0.09
Western Asset Ab Ret (WAARX)	0.20	5.65	38.97	0.96	3.29	17.80
RiverSource AbRetCcy&Inc	-0.02	-0.84	1.39	-0.07	-0.50	0.51
Akros Absolute Return	0.52	15.81	83.33	0.25	0.42	0.36
Absolute Strategies (ASFIX)	0.31	21.69	90.40	0.27	0.79	1.24
UBS Absolute Return Bond	0.05	3.16	16.66	0.06	0.49	0.48
Bold denotes statistical significance.

Do absolute mutual funds follow such a market neutral, zero beta investment strategy? The answer is resoundingly no. Using 52-week returns ended May 30, 2009, seven of nine funds show statistically significant betas to the S&P 500 Index, including bond fund UBS Absolute Return Bond, and exposure to this explains between 17% and 90% of these funds' returns. No wonder they racked up losses in 2008. One fund that does not show statistically significant exposure to the S&P 500 Index, Western Asset Absolute Return, shows instead significant exposure to the BarCap U.S. Aggregate Bond Index, which explains about 18% of its returns.

Is a zero beta, then, even possible? It is, theoretically, assuming that beta is proxied by very basic sources of market risk, such as the S&P 500 Index return, and transaction costs are not material. Let's take a very simple example of an equity market neutral fund. If my portfolio has two stocks, A and B, and A has a historical beta, or sensitivity to the S&P 500, of 0.5, and B exhibits a beta of 1.5, I could short one B stock for every three long A stock positions. If the portfolio dropped 2% one month, my portfolio should not react, because the losses due to market sensitivity (3%) in the long positions would be offset by the short positions.

So, then, why would I participate in such an investment strategy? Because I think that I can produce alpha, both on the long and short side.

I think that Stock A, which I am long, will produce a positive alpha, or a positive risk-adjusted return, and Stock B, which I am short, will produce a negative alpha due to idiosyncratic, company-specific prospects or problems, such as top-selling product lines or poor management. If I am correct, due to luck or skill, I will make money on both my long and short positions, because Stock A will outperform the market after accounting for the stock's sensitivity to the market, and Stock B will underperform.

Only one of nine funds, RiverSource Ab Ret Ccy & Inc, showed insignificant betas (not statistically different than zero), to both the stock and bond indexes, but this fund also shows no significant alpha. Only one fund, Akros Absolute Return, shows significant alpha, but it is not market neutral. In conclusion, it appears that the funds calling themselves absolute return are, in fact, not.

Beyond Beta
Absolute return may be one of those grandiose ideas that is good in theory but unachievable in reality. One reason is transaction costs. Theory assumes they don't exist, but we know they do, namely in the form of trading commissions and shorting costs. Transaction costs reduce any alpha that a manager or a security may produce. Transaction costs are particularly high in a market neutral strategy because the portfolio must be constantly rebalanced to achieve beta neutrality. Beta is simply an estimate of a security's sensitivity to the market, based largely on historical returns. The estimate constantly changes. We can get an idea of the transaction costs in absolute return funds by looking at turnover ratio. QCM Absolute Return Fund, one of the defunct funds, sported an 862% turnover ratio, versus the average large-blend equity mutual fund's turnover ratio of 71%.

More important than transaction costs, though, are our estimates and proxies of beta, or market risk. Some sources of market risk are difficult to model or proxy and virtually impossible to hedge out, or neutralize. One such factor is liquidity. Liquidity is the ability to trade large quantities quickly and cheaply. Liquidity is always low in certain asset classes, such as small-cap stocks, convertible and high-yield bonds, emerging-markets, or structured securities. During a crisis, however, liquidity affects all asset classes. This is called a flight to liquidity, when investors suddenly sell less-liquid, riskier investments, such as stocks and bonds, and buy more-liquid, less-risky instruments, such as T-bills, causing large losses in all but the safest instruments. This happened in August 1998, when Russia defaulted on its bonds, and again very recently, because of the burst of the housing bubble.

While there is no widely available proxy to measure a fund's exposure to illiquidity, investors can gauge liquidity risk by looking at portfolio holdings. For example, as of March 31, 2009, 16% of now-defunct JHancock2 Absolute Return Portfolio's bond holdings were rated below B or unrated, while its 16% of its equity holdings were small cap, 2% were distressed, and 6% were emerging-markets stocks. Illiquidity also caused UBS Absolute Return Bond's 23% drop in 2008, per its prospectus. The fund invested in "various subsectors of the US securitized market," including collateralized debt obligations.

Besides liquidity, absolute return funds may be taking on other forms of hard-to-identify market risk. One such risk is insurance selling, a common risk in merger arbitrage strategies. These funds sell insurance against the completion of a merger or acquisition, collecting small premiums for a period of time, until the deal collapses and the fund incurs large losses. Finally, absolute return mutual funds may take on the risk of common holdings with highly leveraged hedge funds, which dump the same securities en masse when they are forced to delever. This happened to quantitative long-short equity strategies in August 2007 and convertible bonds in the fall of 2008.

The Alternative to Absolute Return
We know now that we can't have it all--a fund that has neutralized all types of market risk and produces consistently positive returns with low volatility. But we can have some parts. Take  Highbridge Statisical Market Neutral , for example. This equity market neutral fund doesn't advertise absolute returns, yet it has returned almost 15% since its December 2005 inception. Not many mutual funds can advertise this. Furthermore, it has achieved those returns with very low correlations to traditional stock and bond indexes, making it a good portfolio diversifier. Over the 52 weeks ended May 30, it shows insignificant betas to the S&P 500 and BarCap U.S. Aggregate Bond indexes, as well as a low standard deviation (5.7%). Yet this fund still has risks. During the week of Aug. 6, 2007, this fund lost 4.32% because of its common holdings with quantitative hedge funds.

The key here is that in absolute return funds, as in other investments, there is no free lunch. Investors thinking about an absolute return or market neutral mutual fund investment should take the time to investigate exactly what risks they face and if they are comfortable taking them on.