Negative Alpha Is Built Into 130/30 Funds


Short-extension strategies, also known under 130/30, 120/20 or even 150/50 monikers, are the latest fad to hit money management. Amidst the excitement over this new product, it has been lost on most of its advocates that the implementation actually starts with a negative alpha.

The logic behind these strategies is that, in addition to placing 100% of an investment into an index, a further 30% (or 20%, or 50%) of the invested amount are sold short, and the proceeds of the short sales are used to acquire an offsetting long position. The net exposure is still only 100% and generates pure beta, while the long/short component of the portfolio is supposed to generate some alpha. So much for the theory.

The reality of implementing 130/30 strategies is a little more complex that the theory. The difficulty lies in the implementation of the short side of the alpha-generating part of the strategy. Myron Scholes coined the phrase that markets look more efficient from the banks of the Charles River than the banks of the Hudson River. Far from being as frictionless as its academic protagonists proclaim, the long/short part of the strategy has a net cost that must be covered before any positive alpha can be generated.

In theory, the proceeds of the short sale are used to purchase a long position. Unfortunately, that is not possible in practice. Instead, at least some of the proceeds of the short sale must be deposited with the broker that holds the short sale. The only way to acquire an offsetting long position is to borrow an equivalent amount of what remains at the broker. The difference between the the borrowing cost and the interest earned from the broker is a cost that simply is ignored by the 130/30 literature. It leads to a negative alpha in the long/short portion of a 1X0/X0.

For readers who are confused by this problem, there is another, more intuitive way of looking at it: an investor in a 130/30 fund is fully invested in an index fund, and acquires an additional alpha-generating overlay. Because the investor is already fully invested, she can not create something out of nothing without incurring a cost. The source of this cost is the bid/offer spread inherent in the cost of the leverage incurred by the investor. In the ivory tower’s frictionless world, there are no bid/offer spreads, and an investor can get alpha exposure for free. In the real world, there is no free lunch, and investors in this construct will incur a cost before alpha can be generated. That cost is a negative alpha. It is the leverage that creates this effect; therefore, an unleveraged long/short fund does not have this problem.

The exact amount of negative alpha depends on the extent of the long short portion, the interest rate spread, and the amount of short sale proceeds withheld by the broker. In our experience, the interest rate paid by brokers on the proceeds of short positions, known as short rebate in industry parlance, is highly variable and can be as low as 0%, while the portion of short proceeds held as collateral by the broker can be as high as 100%. It depends on the bargaining power of the client, and willingness of the clients’ management to bargain. An manager affiliated with a large financial services firm probably has little incentive to bargain on behalf of the fund against his employer’s affiliated brokerage.

Unfortunately, the plethora of sales presentations, articles and other literature touting short extension strategies does not address the cost of leverage required for the long/short component. Even fund prospectuses are not vague about it. Take, for example, the Russell Investment Company Quantitative Equity Fund, a 130/30 offering that was actually a 104/04 fund at the time of its annual report. If you make it to page 53 of the Statement of Additional Information (how many investors even bother to ask for the SAI?), you can find a very cryptic description of the mechanics of the short sale:

Until a Fund replaces a borrowed security in connection with a short sale, the Fund will: (a) maintain daily a segregated account, containing cash, cash equivalents, or liquid marketable securities, at such a level that the amount deposited in the segregated account plus the amount deposited with the broker as collateral will equal the current value of the security sold short; or (b) otherwise cover its short position in accordance with positions taken by the staff of the SEC (e.g., taking an offsetting long position in the security sold short.) [Emphasis added by author]

(There is nothing special about Russell’s 130/30 fund, it just happens to be one we came across.)

In the end, what matters to most investors is the impact of all this on return. The following listing shows the negative alpha for different 1X0/X0 levels, interest rate spreads and percentage of collateral retained by the broker. It is reasonable to assume that a large short extension fund can negotiate a collateral deposit with the broker equivalent to 30% of the short sale proceeds, and can pay a 200 bp spread between the short rebate and the interest rate paid by the fund to borrow for its long position. Our calculation shows that this will result in a 18 bp per year performance drag for a 130/30 fund, and a whopping 30 bp in the case of a 150/50 fund:

 

 

Short Proceeds Retained By Broker

120/20

 

30%

50%

100%

Spread

1.00%

0.06%

0.10%

0.20%

2.00%

0.12%

0.20%

0.40%

5.00%

0.30%

0.50%

1.00%

 

 

Short Proceeds Retained By Broker

130/30

 

30%

50%

100%

Spread

1.00%

0.09%

0.15%

0.30%

2.00%

0.18%

0.30%

0.60%

5.00%

0.45%

0.75%

1.50%

 

 

Short Proceeds Retained By Broker

150/50

 

30%

50%

100%

Spread

1.00%

0.15%

0.25%

0.50%

2.00%

0.30%

0.50%

1.00%

5.00%

0.75%

1.25%

2.50%

(There is nothing special about Russell’s 130/30 fund, it just happens to be one we came across.)

So what can an investor do to evaluate the impact of negative alpha on the actual return of a short extension mutual fund? Unfortunately, very few disclosures are available in prospectuses and they are often useless. In the annual report of the above quoted Russell Investment Company Quantitative Equity Fund, deep in the footnotes of the financial statements, you find this useful sentence in the connection with short sales:

As of October 31, 2006, $512,688,999 was held as collateral.

You will then notice that the Quantitative Equity Fund has only $160 million in short sales. In the case of this fund, we can not determine how much of that deposit represents collateral against the, because some of the collateral is held against written options, swaps and other derivatives for all of the 20 funds within the family of the trust. It is impossible to allocate the collateral back to the 130/30 Quantitative Equity Fund. In short, you know you lose something to the spread on the broker deposit, but you have no way of knowing how much.

Investors in 130/30 funds should be wary of funds offered by large financial services institutions with affiliated brokerage and lending operations. The temptation of squeezing extra margin out of a 130/30 fund through low short rebates and high lending rates could be too great for bonus hungry executives to resist. In the absence of clear disclosures, investors have no way to distinguish between poor investment performance and being short changed on the short rebate. We would look at short extension funds for the next big mutual fund scandal.

Thomas Kirchner manages the Pennsylvania Avenue Event-Driven Fund (PAEDX), a mutual fund that practices short selling.

add to del.icio.us :: Add to Blinkslist :: add to furl :: Digg it :: add to ma.gnolia :: Stumble It! :: add to simpy :: seed the vine :: :: :: TailRank

About these ads

4 Responses to Negative Alpha Is Built Into 130/30 Funds

  1. David Harper says:

    Thomas,

    That’s great, really helpful. btw, your exhibit looks correct but the 18bps should match the 12bps (=20%*30%*2% = 12bps).

    The FAJ just published a study that incorporates this cost (the borrowing spread) into the IR/IC metric (it’s about time that got adjusted for costs!). They went even further than you on costs. They add the above “Explicit” costs plus they add “implicit” operational cost of adding the extension strategy.

    David Harper

  2. David has an excellent analysis of the FAJ article (Long-Short Extensions: How Much Is Enough? by Clarke, De Silva, Sapra and Thorley) on his website linking the FAJ article to our negative alpha:
    http://www.bionicturtle.com/learn/article/how_to_include_incremental_costs_in_extension_strategies_like_130_30/

    Another interesting follow-up is Greg Boop’s post on his HingeFire blog: http://hingefire.blogspot.com/2008/02/will-13030-funds-hold-water.html

    Finally, HedgeWorld summarized the problem in a great article by Emma Trincal (by way of AllAboutAlpha):
    http://allaboutalpha.com/blog/2008/03/13/manager-we-would-look-at-short-extension-funds-for-the-next-big-mutual-fund-scandal/

  3. [...] drag of reduced short interest rebates on 130/30 funds. (Deal Sleuth via Kirk [...]

  4. Tristan says:

    Excellent article. FYI I calculated something similar when using index futures which allow 5% performance margin. Here is my analysis:

    I have $100m of assets
    I think that Russell 2000 will outperform the S&P 500
    I buy $130m worth of Russell 2000 index futures
    I short sell $30m worth of S&P 500 index futures
    Performance margin requirements = 5% x 160m = $8m
    Lost interest on $8m @ 4% (just an average) is $320,000 or 32bp of the original amount

    And that analysis suggests that in a higher interest rate climate, the impact of the performance margin (or in your analysis, the collateral) is also higher.

    Thanks,
    Tristan Yates
    Author: Enhanced Indexing Strategies

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Follow

Get every new post delivered to your Inbox.

%d bloggers like this: