How Regulators Can Improve the Quality of Mutual Fund Data

Information is the lifeblood of smart investors. To make good decisions, you need good timely information. With that in mind, I have some suggestions for how regulators can make existing disclosure more valuable to fund investors. We've seen a number of new initiatives in recent years to help fund investors make better decisions, but the end result isn't always satisfying. Were these data points for corporate filings, they would probably have been cleaned up right away, but an overly paternal view of fund investors leads regulators to settle for less.

Turnover
Turnover is the percent of a fund's assets that get traded during the year, right? Wrong. To calculate turnover, fund companies take the lesser of purchases and sales and divide by assets. That means turnover is really understating fund trading. Moreover, if there's a big difference between purchase and sales, it can vastly understate trading. Say a fund gets huge inflows, and management simply buys more of the stocks it likes most at the moment and lets less-promising stocks dwindle in position size rather than actually sell the stocks. In such a case, turnover would be zero because the manager didn't sell anything, but he clearly did some trading. Fixing this issue to provide more clarity for investors would be a piece of cake: Just require that purchases and sales be added together and then divided by assets.

Tax-Advantaged Dividends
Congress and fund companies have done the heavy lifting to reduce the tax you pay on qualified stock dividends to 15%, and fund companies have also done their part to accommodate the new rules. They've rolled out new dividend-focused funds, and 1099 forms now also distinguish between tax-advantaged dividends and regular-income dividends.

Unfortunately, however, investors looking to compare funds' aftertax returns in SEC documents are out of luck. That's because funds report their distributions through Nasdaq, but Nasdaq isn't equipped to break out tax-advantaged dividends from regular-income dividends. Thus, a useful data point, aftertax returns, isn't nearly as meaningful as it should be. This can be easily remedied by Nasdaq simply breaking out income distributions into those taxable at the lower dividend rate and those taxable as ordinary income.

Manager Investment Values
Do you think stock investors would be satisfied to know that Microsoft CEO Steve Ballmer owns more than $1 million worth of shares of his company? His ownership stake might be $1.1 million or it might be $1 billion, but you couldn't actually know how much skin he had in the game. I doubt stock investors would settle for that level of disclosure, but that's what we get from fund managers. The SEC created broad ranges for manager ownership: $0; $1-$10,000; $10,000-$50,000; $50,000-$100,000; $100,000-$500,000; $500,000-$1 million; and more than $1 million. This framework has three key flaws. First, a range can change without a manager changing his investment level. Say a fund loses 10% of its value and a manager's investment range falls from $500,000-$1 million to $100,000-$500,000. Maybe something bad is going on and the manager has slashed his investment. Then again he might not have changed a thing but his investment shrunk to the lower level. So, you can't be sure if the manger has bought, sold, or held.

Second, managers can be weasels and buy in at the lowest end of the range while claiming to be at the high end. There's a big difference between $100,000 and $500,000, but you can't know where a manager fits in. They could even invest $10 in their fund and they'd be in a range where the midpoint is $5,000.

Finally, the bands simply don't go high enough to draw meaningful conclusions about whether a manager believes in his fund or not. A $1 million investment is a big deal for some managers. The best-paid managers, however, earn more than $10 million a year and may have been investing for 30 years, so their $1 million investment isn't quite as meaningful. All this can be solved by simply requiring the actual dollar amount, just like you see for executives at publicly traded companies.

Manager Compensation
The SEC requires fund companies to disclose the factors on which a manager's bonus is based, but fund companies are allowed to be extremely vague. The descriptions typically mention the benchmark and time periods, but they don't say how the levers work. Is the manager rewarded for top-decile one-year returns? If so, she has an incentive to gamble because one has to take big risks to have top-decile returns for a short time period when the market is rallying. I'm not talking hypothetically here. We saw a number of fund managers, at Janus and many other firms, make huge bets in 1999 and 2000 because managers had huge bonuses on the line in exchange for hot single-year performance. In addition, Janus' managers didn't have much downside if they had poor performance, so it was house money. Janus has since adopted a much healthier focus on long-term results, but the compensation rules wouldn't have been much help to shareholders in 1999. And there are still plenty of firms with poorly structured bonuses out there.

Consider the disclosure for the largest mutual fund,  American Funds Growth Fund of America (AGTHX): "In order to encourage a long-term focus, bonuses based on investment results are calculated by comparing pretax total returns to relevant benchmarks over both the most recent year and a four-year rolling average, with the greater weight placed on the four-year rolling average. For portfolio counselors, benchmarks may include measures of the marketplaces in which the relevant fund invests and measures of the results of comparable mutual funds. For investment analysts, benchmarks may include relevant market measures and appropriate industry or sector indexes reflecting their areas of expertise. Capital Research and Management Company also separately compensates analysts for the quality of their research efforts. The benchmarks against which The Growth Fund of America portfolio counselors are measured include: S&P 500 and Lipper Growth Funds Index (customized to remove The Growth Fund of America and index funds)."

So, you get some useful information about time periods and benchmarks but you don't know how outperformance is defined. Nor do you know if the pendulum swings equally so that underperforming hurts as much as outperforming is rewarded. In this case, we know American's long track record, low turnover, and stable management confirm that American really is focused on the long term and limiting risk in the short term but that's not the case at many other fund companies. The SEC should require that fund companies explain the exact structure of the bonus. What are the weights, and what percent of base salary do they get for the different performance possibilities?

How Much Does Your Manager Run?
Another useful bit of information the SEC recently required was disclosure of the total amount run by a manager. One of the biggest threats to repeating strong performance is asset bloat, so it's crucial that investors know how much a manager is running. For the most part this works well, but some fund companies have found a way to thwart the SEC's intentions. For managers who run portions of multiple funds, they have included all the assets run by all the fund managers on all the funds. Thus, a fund manager who is really running $10 billion is listed as running a ridiculous sum like $190 billion. The SEC should ensure that the sum listed only refers to that amount run by the manager across funds. Otherwise, they've left a loophole wide enough to drive a $100 billion fund through.

Poll Results: Fund Company Turnarounds
Last week I asked "Of the bottom three fund companies, which has the best chance of a turnaround?" Here's how you voted:

52% - American Century

18% - Putnam

30% - Van Kampen