Why Trading Costs Matter

Domestic stock funds spend an average of 1.44% annually on buying and selling stocks -- more than they charge in annual expenses, according to a new study. But the results of this rigorous study are much more fascinating than that one number indicates. The study found that how much a fund's trading actually costs investors varies dramatically depending on what kind of fund it is -- and why the fund managers buy and sell stocks.

Despite the costs, a lot of trading ends up boosting performance. When a manager buys or sells a stock because he or she thinks the price is right, on average, the trades add value to the fund even after subtracting all the trade-related costs.

Aren't all trades made because managers think they're a good idea? Unfortunately, no. About 30% of all trades are entered into because investors pour money into a fund or pull money out. Those cash flows force managers to either buy or sell shares of stocks. And such trades reduce performance by, on average, 1.4 percentage points per year. Managers also trade to rebalance their portfolios and to ensure they don't drift out of their designated investment style box.

Subscribe to Kiplinger’s Personal Finance

Be a smarter, better informed investor.

Save up to 74%

Sign up for Kiplinger’s Free E-Newsletters

Profit and prosper with the best of expert advice on investing, taxes, retirement, personal finance and more - straight to your e-mail.

Profit and prosper with the best of expert advice - straight to your e-mail.

Sign up

The study's bottom line: On average, every dollar spent on trading costs investors 42 cents in performance. Multiply the 1.44% spent annually on trading by $0.42, and you conclude that trading in the average fund costs investors 0.60% of performance annually.

The study's results can help you do a better job of deciding which funds to buy and sell. So please read on. But be aware throughout that two different types of numbers are tossed around in the paper and consequently in this article: First, there's the total cost of trading -- think of it as the gross cost. Second, there's the net amount that that trading costs the fund in performance.

The nitty-gritty

No one has ever questioned that trading costs are a hidden expense of mutual funds. But no one that I know of has ever done the laborious work necessary to determine with any accuracy how much those costs really are.

Brokerage commissions have long been public, albeit buried in fund prospectuses and documents called Statements of Additional Information. But commissions account for only about 20% of the true costs of trading, according to the study.

What's the rest? Every stock has a bid-ask spread -- the difference between what you can buy the stock for and what you can sell it for. On heavily traded stocks, the bid-ask spread can be as a little as a penny; on thinly traded stocks it be ten cents or more.

Much more important for funds, though, is the impact that their trades have on stock prices. It's the law of supply and demand: If you want to sell three million shares, especially of a thinly traded stock, you're going to push the price down before you fill your order.

Consequently, most funds delay selling big positions in a stock all at once. Instead, they may ease out of a stock over days, weeks or even months. But if the manager is right that the stock should be sold, delaying can cost, too. Over the course of an afternoon some ten years ago, I watched a trader in Fidelity's trading room whittle down a position in the stock of a midsize company. When I asked him how long it would take to complete selling the stock, he answered "three or four months." Trading stocks in a fund with a lot of assets can be like turning around a tanker.

I'm not going to bore you with the details of how the professors -- Gregory Kadlec of Virginia Tech, Roger Edelen of Boston College, and Richard Evans of the University of Virginia -- conducted the study. If you're a fund and statistics geek, you can read the study, "Scale Effects in Mutual Fund Performance: The Role of Trading Costs" at papers.ssrn.com/sol3/papers.cfm?abstract_id=951367). But I think they did the job right. In a nutshell, they looked at the quarterly holdings of 1,706 domestic funds from 1995 through 2006. When a stock was sold during a quarter, they determined the average bid-ask spread over that period and used sophisticated methods to determine how big an impact each trade had on the stock's price.

The results

The biggest predictor of total trading costs is the size of the companies in which a fund specializes. Annual trading costs for small-cap funds are a staggering 2.85%, compared with 1.73% for mid-cap funds and 0.77% for large-cap funds.

Value funds spend less on trading (1.21%) than do growth funds (1.84%) -- but the gap is much smaller than I would have expected. After all, value funds try to sell stocks that other investors want while growth managers often buy what's popular.

And bloated funds -- those with a lot of assets -- are the big losers from fund trades. On average, $1 in total trading costs a big fund 88 cents. Conversely, $1 in trading costs in a small fund improves that fund's performance by an average of 38 cents, the study found. The study bolsters the case for investing in funds with small asset bases.

Not surprisingly, trading costs correlate with fund performance. The higher the trading costs, the lower the performance. That's common sense.

But the professors dug deeper into trading costs. It turns out that the reason a fund makes a trade makes a huge difference. When a fund manager sees a promising stock and snaps it up, or dumps one he's soured on, on average, for each dollar in trading costs the trade adds 22 cents to a fund's value -- after subtracting the trading costs. Fund managers aren't as dumb as we may have thought.

The worst trades are those made to accommodate inflows and outflows by investors. "When fund flows force a trade, watch out," Kadlec says. For every dollar in trading costs, the fund's value falls by 53 cents. The bottom line: Be careful of both hot funds and cold funds. When money is pouring into a fund, the manager may be forced to make poor trading decisions. Ditto when investors are fleeing a fund.

Large trades are much more costly than small trades to performance -- no surprise here.

Trading costs have long been acknowledged to be a hidden cost of mutual funds. The significance of this study is that it brings to light how much trading costs really matter -- and when they matter.

Steven T. Goldberg (bio) is an investment adviser and freelance writer.