Mutual Fund Brokerage Commissions and Trading Costs
John P. Hussman, Ph.D.
President, Hussman Investment Trust
Originally published on October 1, 2003
(Updated December 31, 2012)
How much do mutual funds pay in commissions and trading costs? This is a question that the mutual fund industry seems reluctant to answer. In an industry already complicated by fees - sales loads, soft dollars, trailing fees, 12b-1 marketing fees - asking mutual funds to prominently disclose trading costs is unpopular among fund companies.
Since few mutual funds offer prominent, detailed information regarding trading costs (in fact, none that we know of), taking the lead is risky. After all, if one fund company publishes its costs, there is a risk that shareholders will simply add these costs to the expense ratio for that company's funds, without doing the same for other companies that don't disclose this information. For that reason, even funds with low trading costs (which we believe include the Hussman Funds) are reluctant to go first.
We've decided to take the lead anyway. It's the right thing to do.
The following data is for the Hussman Strategic Growth Fund (HSGFX). For the fiscal year ended June 30, 2012, the Fund's gross expense ratio was 1.05%, compared with an average expense ratio of 2.05% among the limited group of mutual funds pursuing similar strategies and classified as "long-short" funds by Morningstar. The Fund has no sales load (a charge for purchasing the fund), no soft-dollar arrangements (where fund managers receive research, data terminals and other benefits in return for paying higher commissions to brokers), no trailing fees (where funds pay brokerages an ongoing percentage of assets in order to bring business to the fund), and no 12b-1 marketing fees (where shareholders pay an amount over and above management and operating expenses, so that funds can advertise and attract new shareholders).
While it is possible to attain even lower expense ratios through passive investment strategies that neither evaluate securities nor manage market risk, our investment objectives are more demanding. For the Strategic Growth Fund, the objective is long-term capital appreciation, with added emphasis on protection of capital during unfavorable market conditions. This is an approach that requires a much different set of skills than passive strategies do. Our expenses should be evaluated based on the extent to which we satisfy our investment objectives, and whether we deliver incremental long-term return and investment stability that is worth the management expense.
According to data from Greenwich Associates presented in testimony to the House Committee on Financial Services (Harold Bradley of American Century Management, March 12, 2003), mutual funds pay an average of between 5.1 and 5.5 cents per share in commissions to make securities transactions - a rate that has not changed significantly in the past decade.
As of December 31, 2012, the transactions for the Hussman Strategic Growth Fund were subject to commissions of 1.3 cents per share and $1.30 per option contract. Given the Fund's approach to trade execution, our transactions are handled using verbal orders through traders with direct access to the trading floors of major exchanges. While it would be possible to obtain somewhat lower commissions using strictly electronic trading platforms, we believe that until liquidity on these platforms becomes substantially deeper, we would lose more on execution than we would gain through modest commission savings. Still, we do intend to seek further commission reductions as net assets grow over time.
The following table presents the brokerage costs of the Hussman Strategic Growth Fund from its inception on July 24, 2000 through December 31, 2012. When evaluating commission costs, it is important to remember that taking advantage of opportunities in the financial markets requires trading. Commission costs (as well as other costs) should be judged in the context of whether these transactions generate value for shareholders over time. Note also that while the assets of the Fund have grown substantially since inception, this growth began from an extremely small base. As noted later in this report, there is no evidence that the Fund's asset growth has resulted in substantial market impact or reduction in Fund performance.
* Initial period from 7/24/2000 - 9/30/2000
A Note on Investment Style
We evaluate securities on the basis of their valuation and market action. When we trade, we generally attempt to buy highly ranked candidates on short-term weakness, and to sell lower ranked holdings on short-term strength. In our view, foregoing these opportunities in order to avoid transaction costs would substantially reduce our returns and increase our risk.
In evaluating how the stocks held by the Strategic Growth Fund have performed, we estimate a positive correlation between the brokerage costs we incurred in any given quarter and the performance of our stock holdings relative to the S&P 500. We would not draw too strong a conclusion from this. It is certainly not true that we could increase Fund returns simply by trading more actively. Rather, the Fund's turnover is a measure of how frequently the market gives us opportunities to add desirable candidates on short-term weakness, or to replace less attractive holdings on short-term strength.
Estimating Market Impact
Market impact or "slippage" measures the amount by which a fund's transactions affect securities prices. For example, suppose that the last sale on a particular stock is $30.50 a share, the current bid is $30.48 and the ask is $30.52. Suppose that a mutual fund enters the market with a large purchase and executes it at an average price of $30.65. Though many factors could have contributed to the price change, it is likely that the fund's large purchase has contributed to this "market impact." Measured from the last sale, the impact is 15 cents, or 0.49%. Measured from the ask, the market impact is 13 cents, or 0.43%.
In contrast, suppose that a fund enters the market with large sell order and executes it at an average price of $30.35. Though the order has driven the price lower, the market impact or "cost of slippage" is still 15 cents, or 0.49% as measured from the last sale, and 13 cents or 0.43% measured from the bid.
There have been a few questionable attempts at estimating slippage without using actual trading information. For example, a 1998 study by Barra estimated that a 10,000 share order on a $40 large-cap stock in the S&P 500 would have a total market impact of 20 cents per share, or 0.50%. This figure was estimated by examining the typical number of shares offered at various prices, and estimating how deep "into the book" a fund would have to go in order to execute 10,000 shares. In our view, it is unrealistic to believe that funds would execute their transactions using unrestricted market orders of this sort. In our experience, we are generally able to trade blocks of 100,000 shares or more in large-capitalization stocks with little or no market impact, by using limit orders.
Another frequently quoted but statistically careless study estimates market impact as high as 2.55% for a single turnover of assets (i.e. 1.275% for each buy and sell). The 2002 study related the 10-year returns of mutual funds with two variables: their turnover ratio and their overall expense ratio. For the average mid-cap fund, the "regression coefficient" on turnover was -2.55%, so the study reports this as the estimated market impact from trading. Unfortunately, the company (which we won't legitimize by name) leaps from bad statistics to outright deceit by using these estimates to fabricate bogus "cost reports" for specific funds.
Statistically, the problem with this study is that it ignores the effect of outside factors that affect both turnover and returns, and therefore profoundly mis-measures the relationship between the two. (Our technical criticism is that the regression was not properly instrumented, as both the dependent and independent variables are endogenous).
We believe that the main factor being ignored is investment style. Consider a momentum fund, which tends to chase rising stocks and dump losing ones. For this sort of fund, a higher frequency of trading means that the fund is repeatedly buying strength and selling weakness. In itself, this is likely to result in lower returns over time, even if the trades could be executed without any price impact at all. Without taking this into account, the statistical methodology loads the entire burden of a fund's poor performance onto its turnover ratio.
In contrast, consider an approach that generally uses limit orders, attempting to buy attractively valued stocks on short-term weakness and to replace less attractive holdings on short-term strength. For this approach, one would expect (and indeed, we find in practice) that higher turnover is associated with better performance, since portfolio turnover in this case measures the frequency of favorable trading opportunities.
In the financial markets, part of the return earned by investors represents compensation for providing liquidity to other investors. Part of the return given up by investors is payment for drawing liquidity away from other investors. To the extent that an investor or mutual fund generally makes trades that provide liquidity to other investors (providing buying support for attractively valued stocks under short term selling pressure, or providing supply for overvalued stocks under short term buying pressure), it does not follow that these transactions are costly at all. Indeed, transactions like these can be a source of profit.
Market Impact Based on Actual Trading Data
To illustrate the foregoing points, we measured the actual market impact of every transaction in the Strategic Growth Fund between June 17, 2003 and December 31, 2012. If we paid a purchase price for a security higher than the last trade price (as of the time the order was placed), the difference was added to our "slippage cost" or market impact. Similarly, if the sale price of a security was lower than the last trade price as of the time the order was placed, the difference also increased our slippage cost. Due to the width of bid-ask spreads in the options market, the last sale price for options was generally taken to be the average of the bid and ask prices.
The total value of equity security transactions during this period was $67,441,967,188. The total market impact of these transactions, measured as the difference between execution price and last sale, was 0.26% (e.g. price slippage of 26 cents per share on a $100 share price). Measured from the bid/ask and initial limit price, the market impact for equity trades has been 0.25% and 0.26%, respectively.
The total value of option transactions during this same period was $142,713,607,437. On average, the Fund executed option trades at 0.18% through the mid-market price (the average of the bid and ask prices). Measured from the ask price (for purchases) and the bid price (for sales), the market impact for option trades has been (2.00)%. The negative impact of 2.00% versus bid/ask prices implies that, on average, the Fund executes its option transactions within the quoted bid/ask spread.
Trading impact for options is substantially greater, largely because bid/ask spreads tend to be a substantial percentage of premium value, and also because of the very large notional value which these options represent. For example, an S&P 100 option at a strike price of 500 might be quoted at a bid of 14 and an ask of 15. The spread alone represents 6.9% of the mid-market price, but only 0.20% of the underlying notional value of the option.
Using the previous example, the Fund's actual average trading impact of 0.18% versus mid-market prices would be about 3 cents on an option valued at 14.50, and would represent about 0.005% based on the notional value of that underlying index.
In short, regardless of how one measures these costs, it is clear that the Fund's transactions did not have significant impact on the securities traded by the Fund. Also, while it would be ideal to avoid these costs altogether, they are necessary in order to take the opportunities that the market provides.
Importantly, investors should not assume that the transaction costs for the Strategic Growth Fund are typical for all mutual funds. We believe that our transaction costs are partly the result of a very specific investment discipline that seeks to purchase stocks on short-term weakness and replace them on short-term strength. The market impact of trading may be much higher for momentum-based approaches that pursue stocks during advances and liquidate them during declines.
Putting all of this together, we estimate that the total cost of transactions in the Strategic Growth Fund for the calendar year ended December 31, 2012, including both brokerage commissions and market impact, was approximately 0.56%. We calculate this figure as follows. During the past year, the Fund paid 0.13% of net assets in commissions, with a turnover rate of 63%, and an average price impact of 0.26%. Given that each turnover involves a purchase and a sale, we estimate market impact of turnover at 2 x 0.63 x 0.26% = 0.33%. Net shareholder flows during the year represented about 37.6% of the Fund's average daily assets, requiring security transactions with an estimated market impact of 0.376 x 0.26% = 0.10%. The total of commission costs and estimated market impact comes to 0.56%. In a world of frictionless, costless trading, this might be the additional amount that could be earned in returns. In the real world, it is the cost of taking investment opportunities that otherwise could not have been taken. This cost should be evaluated in context of how well we achieve our long-term investment objectives.
Frankly, we don't know what these figures are for other fund companies. If transaction costs are high and a fund company believes they are reasonable given their investment approach, management can always explain why. Though it would be interesting for funds to report standardized calculations of market impact, a regulatory requirement to do so would be onerous. In contrast, tracking and reporting brokerage costs should be straightforward.
Since commission costs are directly associated with investments made by funds, and are fully reflected in total returns, we don't think it makes accounting sense to include commissions as part of the reported expense ratio of mutual funds. However, we do believe that annual reports should include a breakout of total commissions paid, average commissions per share, and disclosure about soft-dollar arrangements, trailing fees and other costs borne by shareholders (at present, this is considered "non-standard information"). Some kinds of information are reasonable to protect, such as proprietary methods or real-time portfolio holdings, because disclosing them could ultimately be detrimental to shareholders. Brokerage cost and fee arrangements do not have these characteristics. Shareholders are perfectly within their rights to know.
The foregoing comments represent the general investment analysis and economic views of the Advisor, and are provided solely for the purpose of information, instruction and discourse.
Prospectuses for the Hussman Strategic Growth Fund, the Hussman Strategic Total Return Fund, the Hussman Strategic International Fund, and the Hussman Strategic Dividend Value Fund, as well as Fund reports and other information, are available by clicking "The Funds" menu button from any page of this website.
For more information about investing in the Hussman Funds, please call us at
513-326-3551 outside the United States
Site and site contents © copyright Hussman Funds. Brief quotations including attribution and a direct link to this site (www.hussmanfunds.com) are authorized. All other rights reserved and actively enforced. Extensive or unattributed reproduction of text or research findings are violations of copyright law.
Site design by 1WebsiteDesigners.