(header images)(menubar images)

Market Neutral Investing
By Mark T. Finn

This article was authored by Lindner Vice Chairman and Chief Operating Officer Mark T. Finn and is abridged from the full version of the text which appears in the August 1998 issue of Journal of Financial planning. Full copies are available from the (link)Financial Planning Association (formerly the Institute of Certified Financial Planning).

For those investors who believe in active management, a market neutral strategy may offer the most attractive asset allocation option available. While market neutral strategies have not been widely used to date, there are signs that the trend may be changing. These signs include recent articles in the popular press, such as Business Week, U.S. News & World Report, Forbes and The Wall Street Journal. In addition, several new market neutral mutual funds have become available and more are expected. (A change in tax regulations in 1997 lifted certain restrictions on short selling, thereby making a market neutral mutual fund possible for the first time.) In addition, there seems to be a significant increase in interest in market neutral strategies from institutional plan sponsors such as General Motors, Ameritech and Monsanto.

This interest is encouraged by such investment notables as William F. Sharpe. Sharpe, the recipient of the 1990 Nobel prize in economics and widely regarded as one of the foremost experts on investing, titled a presentation he gave on market neutral investing as, "A Modest Proposal to Revolutionize the Investment Management Industry."

This paper will describe market neutral investing and summarize the issues that financial planners should consider in determining the proper role for a market neutral strategy.

What is Market Neutral'?
Market neutral investing is an investment technique that combines the purchase of undervalued securities with the short sale of overvalued securities in such a way as to neutralize the impact of the overall market for that type of security. This strategy is also referred to as a "zero beta" strategy. This is because, since the portfolio is neutral to the overall market, the portfolio has no volatility or co-movement with the overall market.

Market neutral strategies have been employed using various asset classes such as fixed income, currency, risk arbitrage, linked-security arbitrage and hedged strategies. Some examples of linked securities (securities that are contractually connected to one another) are index futures and the underlying securities in the index, convertible securities and warrants and the underlying security; options on stocks and index futures, and depository receipts and the bundles of underlying common stock traded in other countries.

Market neutral strategies also are called absolute return strategies since the return is independent of the performance of the underlying market and hence a return may be generated whether the market goes up or down.

This paper will only address market neutral investing as implemented in a U.S. stock portfolio with equal dollar amounts in a long component and a short component.

The strategy is implemented as follows. The manager identifies undervalued stocks and overvalued stocks. He or she purchases a portfolio of undervalued stocks so that the portfolio of longs has the same risk characteristics as the portfolio of stocks the manager has identified as overvalued. The manager borrows (through an intermediary known as the "prime broker") the portfolio of overvalued stocks and sells them. The cash received from the sale is used as collateral and earns roughly a T-bill return until the manager "covers" the short sales. The manager does this by buying the shorted stocks after they have declined in price (or at least not appreciated as much as the long portfolio has) and returning them to the original lender. The portfolio's return is determined by the interest earned on the cash collateral plus the differential performance between the long portfolio component and the short portfolio component.

The return is not tied to the performance of the market. The long portfolio is expected to gain the market return plus some additional return from the selection of undervalued stocks. In addition, the short portfolio (of equal dollars) will lose the market return, but gain some additional return from having shorted overvalued stocks (which will either have gone down more than the market or at least not appreciated as much as the long portfolio component). Thus, the combined portfolio of the longs and shorts will cancel out the return from the market. All that is left will be the difference in performance between the long and the short portfolio component (plus the return earned on the cash collateral).

Negative Aspects
One potential negative aspect of market neutral investing concerns possible tax consequences. However, the tax implications of market neutral strategies are somewhat complicated. One important distinction has to do with the type of investment vehicle. In a partnership (the typical structure for a hedge fund), the expenses of the fund (including management fees and dividend "expense" on the shorts) are typically not deductible for tax purposes. In a mutual fund, these expenses are typically deductible, providing a major advantage for the after-tax returns of a market neutral mutual fund.

Another complication is that market neutral strategies will have higher turnover. This is due to the fact that there are two portfolio components (a long and a short) that will each have turnover. Also, the turnover on the long portfolio component is modestly increased to the extent that matching the risk characteristics of the short portfolio involves more turnover than the turnover required from a traditional long portfolio that simply needs to control the deviations from a benchmark. A good estimate is that for a market neutral manager with tight risk control, this might increase turnover by about 25 percent.

Another important issue is the distinction between ordinary income and long-term capital gains. The interest earned on the cash proceeds of the short sales, which is used as collateral, will be ordinary income and will be partially offset by the fund's expenses. (Typically, the dividends on the longs will be similar in magnitude to the dividends on the shorts, and will thus offset each other.)

The key issue then becomes the classification of the capital gains. Even though the turnover in market neutral accounts can exceed 200 percent a year (with the combined long and short turnover), the turnover is not the same for every security in the portfolio; therefore, some portion of the gains will be long-term. All transactions related to securities sold short are classified as short-term, so the breakdown between short-term and long-term is critically dependent on whether the market is up or down in a given year.

In an up market, the gains will tend to be long-term. That is because it is likely that both the longs and the shorts will increase in value. However, if the manager has stock selection skill, then the longs should appreciate more than the shorts (a liability.) In contrast, in a down market, the longs will likely decline so gains will not be generated. However, the longs would be expected to decline less than the shorts. The decline in value of the shorts is, of course, realized as gains-not losses, as in the case of the longs-through the mechanism of short selling. All these gains from short selling will be classified by (regulatory) definition as short-term.

The net result is that in a strong up market, a low percentage of gains could be classified as "ordinary." However, the flip side is that in a down market, a high percentage of gains could be classified as "ordinary."

One needs to balance the potential negative impact of taxes with the possibility that market neutral portfolios provide enough additional return to, at least partially, compensate for any added tax burden. While there is limited data on the performance of (equity long/short) market neutral managers, some of the studies that have been done are encouraging.

For example, a recent study done by the consulting firm BARRA looked at the performance of 14 quantitative market neutral managers over the period 1990-1997. All 14 managers had positive information ratios (a measure of skill). BARRA also compared the information ratios (after fees) produced by these market neutral managers with the information ratios produced by two large samples of typical long-only managers. One sample consisted of 367 institutional portfolios and the other sample consisted of 300 mutual funds. In both cases, the market neutral managers produced superior information ratios compared with the long-only managers.

Another study covering the five-year period ending November 30, 1997, found that its sample of market neutral managers had an average annual compound return (net of fees) of 14.08 percent and a standard deviation of 2.09 percent. That compared favorably with the Lehman Aggregate Bond Index return of 7.61 percent and standard deviation of 4.46 percent.

A study by the Frank Russell Company concluded that for the five-year period ending December 31, 1997, the average market neutral manager in their universe delivered over six percent a year in added value.

While the limited sample of market neutral managers studied and the potential of survivorship bias should make one cautious in relying on these and other performance studies, the number of different studies all showing positive results is at least intriguing.

Risks of Market Neutral
There are four main sources of risk that a financial planner needs to evaluate before investing with a particular market neutral manager.

The most obvious risk is that the manager may not possess any stock selection skill. However, even if the manager possesses no skill, the portfolio returns would be expected to equal the T-bill return. On the other hand, if the manager's stock selection is perverse and the stocks he or she selects as undervalued underperform the stocks the manager had identified as overvalued, then the manager could contribute a negative return that could reduce or swamp the expected T-bill return. (Obviously, if you did find a manager whose selection ability was perverse, you could just reverse their selections and make money.)

However, even if a manager has been able to identify undervalued stocks, it does not necessarily mean that they are equally adept at identifying overvalued stocks. Indeed, the biggest risk in this regard is that the manager's stock selection process may not normally generate both undervalued and overvalued candidates. Contrast two different managers.

The first is a quantitative manager who normally ranks the stocks in his selection universe from most attractive to least attractive. This manager would not have to change his selection process to implement a market neutral strategy. On the other hand, a manager who spends all her time searching out unique undervalued stock situations may have to double her efforts to also search out unique overvalued stock situations. And there is no guarantee the manager is equally adept at identifying overvalued stocks. One cannot assume the manager's skills will apply equally to undervalued and overvalued stocks.

If, for this manager, identifying overvalued stocks does require the same amount of effort as identifying undervalued stocks, then one needs to consider how all the manager's assets are invested.. If 90 percent of the manager's assets under management are invested in long portfolios and she is adding a market neutral strategy as one more product, then how many resources are going to be allocated to identifying overvalued stocks, given that most of the manager's fee income comes from identifying undervalued stocks?

Another example would be a manager whose skill is in identifying undervalued and overvalued industries. His performance may come from his overweighting the undervalued industries and underweighting or deleting the overvalued industries rather than from his stock selection skills within an industry.

Since the correlations among many industries are not high, it is unlikely that the industries the manager selected as undervalued would have the same risk characteristics as the industries the manager identifies as overvalued. Buying the semi-conductor industry and selling short the chemical industry may produce a market neutral portfolio in the sense that the beta may possibly be zero. However, the portfolio will still be subject to significant industry risk. (See discussion of common factor risk below.)

The manager's ability to measure and control beta or volatility is an important risk factor. If the portfolio does not have a beta close to zero, it will not be market neutral. For example, if the long portfolio has a beta of 1.3 (it tends to exacerbate movements in the overall market by 30 percent), and the short portfolio has a beta of 0.90 (it tends to go up or down only 90 percent as much as the market), the combined portfolio of 100 percent in the longs and 100 percent in the shorts will not have a zero beta. The combined beta (1.3 minus .90) will be .40. Thus, the portfolio will be exposed to 40 percent of the market's movement rather than totally neutral.

While it is true that no one knows beforehand what the true beta of a portfolio is, there are better and worse ways of estimating it. The beta estimate is also dependent on the level of diversification in the portfolio. A beta estimate for an individual stock will have much more error than the beta estimate for a well-diversified portfolio.

Common factor risk is also critical. Remember, the key to controlling risk in a market neutral strategy is to make the long portfolio similar in risk profile to the short portfolio. This way, the returns will be completely dependent on the manager's stock selection skill and not any systematic or market-wide factors. While beta attempts to measure the systematic or market related risk, one cannot ignore exposures to common factors. These are factors that may not affect all securities but will affect groups of securities. The most obvious common factors are industries, value/growth and large/small capitalization. For a portfolio to be truly market neutral and only dependent on the manager's stock selection skills, the common factors of the long and short portfolio must be as closely matched as possible-while still allowing the manager enough flexibility to exploit his or her valuation insights.

Short selling also introduces a risk factor. Short selling, or shorting, is different from the typical long portfolio strategy in two ways. First, there must be someone who owns the stock who is willing to lend it. Second, since you must return some time in the future the equivalent number of borrowed shares of the stock, you could lose more than the maximum 100 percent loss of a normal (long) purchase. If the stock you have shorted goes up 400 percent before you purchase it to return to the lender, you have lost 400 percent of your investment in that stock.

Having a diversified portfolio with no individual large-stock short positions, and shorting only stocks with ample lending availability are two ways that the risks of shorting can be minimized. But it is incumbent on the financial planner to understand the specific characteristics of shorting and how the manager deals with it.

Benefits of Market Neutral
First, the returns are not dependent on the systematic returns of the underlying asset class. Thus, the manager is a likely to have a positive return in down markets as in up markets. This decoupling of the portfolio return from the underlying asset class return enhances the flexibility in using a market neutral strategy.

With the use of futures, the manager's alpha (risk-adjusted excess return) is portable to any asset class. For example, if you believe that it is less likely that a manager can add an alpha when investing in the larger stocks, you could buy S&P 500 futures and overlay it on the market neutral manager's portfolio. In this way, the benchmark can still be the S&P 500 return, but it may be more likely that the client will get an alpha. To the extent there is a liquid futures market, the financial planner can focus on finding the best market neutral manager and not be restricted to finding superior managers in every style or asset class.

Freeing up a manager so his or her investment decisions are no longer tied to the composition of a specific benchmark creates some potential trade-offs. On the one hand, the manager no longer needs to have an opinion about every industry represented in the benchmark. The manager can invest in just those industries or groups where he or she has opinions about the stocks within those specific industries or groups. The manager could go so far as to specialize in just one industry.

On the other hand, whatever industries or groups the manager focused on, they would have to offer a broad enough range of both undervalued and overvalued candidates so that the manager could create a long portfolio with the same risk characteristics as the short portfolio.

If implemented correctly, market neutral may be the most risk-efficient asset available. That claim is based on two unique features of market neutral. First is the correlation of market neutral with other assets-it's zero. Unlike fixed income and its interest rate risk, market neutral has no exposure to systematic risk. Indeed, while market neutral is expected to have zero correlation with the S&P 500, over the last five years, intermediate-term government bonds have had a correlation of 44 percent (R2 of about 20 percent). The import of this can be seen in Chart 1 (omitted). This chart shows the risk-reducing impact of adding different percentages of intermediate-term government bonds or a market neutral portfolio to a portfolio invested in the S&P 500. What this demonstrates is that even if the market neutral manager produced no alpha, it still would have been a superior risk-reducing asset to intermediate-term government bonds.

This exercise was repeated in Chart 2 (omitted), but using international equity as an asset class. Again, because there is significant correlation with the S&P 500 (MSCI EAFE had a correlation over the last five years with S&P 500 of 45 percent), using international stocks in a portfolio reduces risk, but not as much as the market neutral portfolio.

Also consider that the risk of a market neutral manager is all risk specific to the particular manager. Thus, it can be diversified away. A portfolio of several market neutral managers might be the most efficient asset choice of all. For example, the market neutral manager shown in Chart 1 had a standard deviation about equal to intermediate government bonds. However, the risk of intermediate government bonds is almost entirely undiversifiable systematic risk-that is, interest rate risk. But the risk of the market neutral manager is not systematic. It arises from the supposedly unique stock selection method that particular manager employs. Hence, it can be effectively diversified away when combined with other market neutral managers. Therefore, by definition (no systematic risk), a portfolio of market neutral managers-assuming even a minimal level of stock selection skill-will be the most efficient risk/return asset available.

Another way a market neutral portfolio is efficient is that it is there to produce pure alpha. There is no diversification deadweight. When a manager has as his or her benchmark on index of the underlying asset class(such as the S&P 500), the manager's active portfolio is only the portion that deviates from the weights in the index. Indeed, based on such factors as tradition, risk control and the manager's business risk, the typical approach is for the manager to overweight or underweight individual stocks and industries relative to the benchmark. However, this usually results in high correlations with the benchmark. The typical institutional equity portfolio has an active risk of about 4.5 percent, according to BARRA. Even that level of active risk may be suspect, depending on how accurate the chosen comparison benchmark is. Other studies have shown that the majority of manager active risk is related to style and not stock selection. For example, in one study of 1,286 diversified U.S. equity funds, the average percent of active risk due to style was over 83 percent. That is, most of the typical active manager's risk could be duplicated with a passive portfolio representing a manager's style. What you are paying active fees for and hoping to benefit from-the manager's unique stock selection skill-tends to be the least significant portion of the risk/return package you get form a manager. However, market neutral reconfigures this frustrating situation. Instead of an insignificant portion of the manager's risk/return package coming from stock selection, the entire risk/return profile of market neutral is determined by the manager's stock selection skill (and the manager's a ability to control risk, of course).

With a market neutral portfolio, which has no systematic risk, the most appropriate benchmark would be an asset like U.S. 90-day T-bills. Hence, by definition, every stock position in the long and the short portfolio is an active position. This fact has interesting fee implications.

One could argue that the efficiency of the strategy translates to fee efficiency. This may be quite surprising to investors when they compare the typical market neutral fee schedule with the management fees charged by more traditional managers. There is a range of fees charged by market neutral managers, but the mot common is "1 and 20"; that is, the manager charges a flat 1 percent of assets plus 20 percent of the excess return above the T-bill benchmark.

How could that fee be more efficient than a traditional fee that might be in the range of 60 basis points? First, consider that the 60-basis point fee is charged on the entire portfolio, even though only 10 percent of it may be "active." Thus, to make the fees comparable to a market neutral portfolio where 100 percent of the assets are active, you would have to multiple the 60 basis points ninefold!

Second, consider that the fees for a traditional active manager are based on the value of the assets. For example, if the active manager didn't add any value above an index fund, the manager's fees would still have increased by the increase in the market. (Some adjustments would have to be made for the fact that most active fee schedules decrease as assets increase.) Over the last five years, fees would have increased (for a manager with the S&P 500 as the benchmark) by over 150 percent due just to the increase in portfolio value from an appreciating market.

On the other hand, the fees for a market neutral manager would not have increased at all due to market appreciation. The increase in fees would only come from the added value of the portfolio due to the manager's active decisions.

In summary, a market neutral investment may be an asset that financial planners want to consider for several reasons: returns do not depend on the systematic return of the underlying asset class, there is increased flexibility in exploiting a manager's stock selections kill, and there is potential for superior risk/return trade-off both in terms of the market neutral portfolio itself, as well as its impact on the overall risk/return of a client's total portfolio.

Disclosure Notes
Market Neutral Investing by Mark T. Finn
Journal of Financial Planning August 1998

Investment in shares of a market neutral fund is more volatile and risky than some other forms of investments. Since a market neutral fund has both a long and a short portfolio, an investment in the fund will involve risks associated with twice the number of investment decisions make for a typical stock fund. Thee types of fund typically have a high portfolio turnover that could increase transaction costs and cause short-term capital gains to be realized. While it may be the intent of the manager to take long positions in stocks that outperform the market and short positions in stocks that underperform the market, in various market climates, there is no assurance that they will be successful.

Since risk in a market neutral fund relates specifically to the manager's stock selection techniques and not to an systematic or economy-wide factors, the proper benchmark is an asset that also has the least exposure to systematic influences. U. S. Government 90-day T-bills are such an asset.

An investment in 90 day T-bills is different from an investment in a market neutral fund because T-bills are backed by the full faith and credit of the U.S. Government. T-bills have a fixed rate of return, and investors do no bear the risk of losing their investment. The income received from T-bills and U.S. government bonds is free from state income tax.

(link)back to Family of Funds (footer images)