Part of a series of ICI Viewpoints about problems in the IMF’s analysis of the asset management industry.
In April, the International Monetary Fund (IMF) released its most recent Global Financial Stability Report (GFSR), including a chapter on “The Asset Management Industry and Financial Stability.” On the basis of this chapter, the IMF has declared that “Even simple investment funds such as mutual funds can pose financial stability risks.”
We have heard more than one person suggest that here at last is strong analysis—chock full of data, charts, and tables—that regulators may rely on when it comes to discussing whether regulated funds or assets managers could pose systemic risks.
In contrast, in an ICI Viewpoints titled “The IMF Is Entitled to Its Opinion, but Not to Its Own Facts,” we pointed out a significant error in the IMF’s analysis. The April GFSR considerably overstated regulated funds’ holdings of emerging market (EM) bonds, and erroneously claimed that such holdings had surged by more than $900 billion in the past two years. In fact, as we pointed out, regulated funds’ holdings of EM debt have been steady over the past two years. Subsequently, the IMF retracted and replaced its data and charts on its website.
Not an Isolated Error
One might wonder whether we were nit-picking—perhaps this was just one chart and the IMF’s error doesn’t detract from its analysis as a whole. Closer examination, however, reveals that this is not the lone error in an otherwise-solid analysis. The report contains numerous data errors, misinterpretations, and misleading charts. By and large, these issues arise because the IMF lacks expertise in, and institutional knowledge of, regulated funds.
Over a series of posts, we’ll lay out several areas where the IMF’s inexperience with regulated funds has led to data errors, inconsistencies, results that don’t bear statistical scrutiny, and misleading claims—all of which seriously undermine the report’s validity as a foundation for policy changes that could end up harming funds, their managers, and their investors.
Let’s start with one striking example—the IMF’s analysis of redemption fees. The IMF tries to analyze whether redemption fees (charged to an investor and paid back into the fund) help stem investor redemptions. It suggests that average redemption fees have fallen substantially—from more than 2.5 percent in December 2001 to about 1.0 percent in August 2014 (see the red lines in the figure below). The IMF attributes this fall to “competitive pressures.”
Trend of Mutual Fund Redemption Fees
Simple average, percent; monthly, December 2001–August 2014
Source: Calculated based on data from the survivor-bias-free U.S. mutual fund database
©2015 Center for Research in Security Prices (CRSP), The University of Chicago Booth School of Business
But the IMF’s figures are incorrect. Since 1979, Securities and Exchange Commission (SEC) staff has taken the view that no open-end mutual fund can have a redemption fee greater than 2 percent. In 2005, the SEC codified this in a rule. Thus, redemption fees simply cannot average more than 2 percent.
The correct figures for redemption fees are shown in the chart as blue lines. The number of funds charging redemption fees rose in 2004, as asset managers put such fees in place to discourage investors who were “market timing”—trading in and out of funds to arbitrage predictable price movements—and to otherwise address costs to long-term shareholders from other shareholders’ short-term trading. Since 2008, redemption fees have trended down modestly, in large part because funds and pricing vendors have worked together to correct these pricing issues and because many funds have found more effective ways to deter short-term trading.
A Case of Confusion
How did the IMF come to misinterpret the data? The answer is that the IMF confused redemption fees with contingent deferred sales charges (CDSCs). To build its chart, the IMF relies on mutual fund data from the Center for Research in Security Prices (CRSP). The CRSP database labels the levels of both CDSCs and redemptions fees as “rear load” and “fees charged when withdrawing funds.” Analysts who are unfamiliar with the CRSP database and the details of fund regulations could easily miss this, leading to a misinterpretation of CDSCs as redemption fees.
Redemption fees and CDSCs serve two completely different purposes. Redemption fees are designed to compensate the fund for costs imposed by short-term traders, and may have the practical effect of deterring redemptions. In contrast, a CDSC is a sales load—a charge paid to compensate a broker or financial adviser for advising and servicing the investor’s account. (CDSCs are also known as back-end loads.)
One should not assume that CDSCs deter redemptions. Here’s an example that shows why.
If a bond fund sets a redemption fee of 1 percent for shares held less than 90 days, an investor who redeems or exchanges out of that fund in less than 90 days must pay a 1 percent fee into the fund. This is a cost the investor likely will consider when deciding whether to redeem out of the fund in an effort to reduce his or her exposure to the bond market and move to another type of exposure, such as cash.
However, funds typically allow investors to exchange into another fund within the same family without paying a CDSC. Suppose a bond fund imposes a 5 percent CDSC if an investor redeems his or her shares within the first year. Normally, such a fund would allow the investor to exchange out of that specific fund into any other fund in the same fund complex (e.g., into a money market fund offered by the same adviser) without incurring the 5 percent CDSC. Consequently, the 5 percent CDSC isn’t likely to deter an investor from reducing his or her exposure to bonds in favor of cash. For precisely this reason, some funds—actually, some fund share classes—that have CDSCs also have redemption fees.
The downward trend in the IMF calculation for “redemption fees” arises primarily because CDSCs (which can be as high as 4 to 5 percent) were more common in the early 2000s, spuriously boosting the “average redemption fee” reported by the IMF. As CDSCs have become increasingly less common among funds (and, hence, in the CRSP database), the IMF’s average has increasingly been driven by actual redemption fees—capped at 2 percent—and thus falls below 2 percent.
So what the IMF presents as a significant trend is, in fact, an artifact of a database that IMF analysts have little experience with and have misinterpreted.
Some may view these details as arcane fund minutiae. But the consequences of misinterpreting these data—and drawing the wrong policy prescriptions from them—are anything but trivial. In the United States, more than 90 million investors hold roughly $18 trillion in regulated funds. These investors will be affected by public policies that alter the structure of funds. Consequently, public policy intended to influence the structure of regulated funds and their investors should be based on expert, careful analysis of the data at hand.
Next, we’ll consider how the IMF’s emphasis on “herding”—its belief that fund managers and investors move in packs—has led it astray.
- The IMF Is Entitled to Its Opinion, but Not to Its Own Facts
- The IMF Quietly Changes Its Data, but Not Its Views
- The IMF on Asset Management: The Perils of Inexperience
- The IMF on Asset Management: Which Herd to Follow?
- The IMF on Asset Management: Sorting the Retail and Institutional Investor “Herds”
- The IMF on Asset Management: Handle Empirical Results with Care
Sean Collins is chief economist at ICI.