Mutual Funds that Borrow

• Date: 01 December 2019
• DOI: http://doi.org/10.1111/jels.12232
• Published date: 01 December 2019
• Author: A. Joseph Warburton,Michael Simkovic

Journal of Empirical Legal Studies

Volume 16, Issue 4, 767–806, December 2019

Mutual Funds that Borrow

A. Joseph Warburton*and Michael Simkovic

Securities laws prohibit open-end mutual funds from borrowing more than one-third of

their capital structure because of concerns that too much borrowing may harm investors.

This is the first article to examine the performance of open-end funds that borrow money

within these permissible limits. We construct a database from annual filings of open-end

domestic equity mutual funds covering 17 years from 2000 to 2016. Eighteen percent of

funds borrowed money for leverage within that time. We find that borrowing funds

underperform their nonborrowing peers by 62 basis points per year on a total return basis,

while also incurring greater risk. After accommodating for the greater risk taking, we find

that borrowers underperform by 48 to 72 basis points annually. By contrast, funds that use

derivatives and other financial instruments perform about as well as unlevered mutual

funds, before and after adjusting for risk, and with less volatility. This suggests that many

mutual funds use derivatives to hedge risk rather than as a substitute for leverage through

the capital structure. Thus borrowing may present a greater risk than derivatives, which

have received more attention than borrowing. Fund investors and regulators would benefit

from greater transparency into mutual fund capital structure.

I. Introduction

Leverage can amplify returns, both positive and negative. Although financial economists

often assume that open-end mutual funds donot borrow money at all for leverage,

1

mutual

funds are permitted to have a capital structure that is up to one-third debt.

2

This might

provide enough leverage to significantly alter the risk-return profile of a mutual fund.

Leverage might be detrimental to fund investors, as leverage has the potential to

magnify fund losses. To protect investors against the adverse effect of leverage, the

*Address correspondence to A. Joseph Warburton, Professor of Law at the College of Law and Professor of

Finance at the Whitman School of Management, Syracuse University; email: warburto@syr.edu. Simkovic is Profes-

sor of Law & Accounting, University of Southern California.

We thank participants at the 13th Conference on Empirical Legal Studies, the International Conference on

Economic Modeling, and the National Business Law Scholars Conference as well as Michael Guttentag, Kevin

Haeberle, RichardKaplan, Monty Mansur, Darius Palia, Adriana Robertson, Gregory Shill, ClemensSialm, and three

anonymous referees.Thanks also to Courtney Deklotz, Shelby Lindholm,and Justin Lee for research assistance.

1

See e.g., Elton et al. (2013) (stating that open-end funds “cannot use leverage”); Natter et al. (2016) (“Mutual

funds are generally prohibited from using any type of leverage.”).

2

In other words, a mutual fund that raised $100 from equity investors could borrow another $50, and so increase

its assets from $100 to $150. See also Morley (2013), noting that mutual funds can borrow money from banks, but

that data on the extent of such borrowing were not readily available.

767

Investment Company Act of 1940 prohibits open-end mutual funds from borrowing more

than one-third of their capital structure. Notably, the statute restricts borrowing but does

not prohibit it. Hence, it is conceivable that borrowing, within permissible limits, may

benefit investors. As one borrowing fund declares: “The fund may use leverage and is

especially likely to do so when we believe prospects for businesses are favorable and stock

prices of those businesses do not reflect those prospects.”

3

While Congress worried that

borrowing would harm investors, funds that borrow maintain that they do so to help

them. Another possibility is that borrowing has no impact on fund performance. Borrow-

ing might be so constrained by the statute that the effects are trivial. If so, the literature

is justified in having overlooked the borrowing activity of open-end mutual funds.

This raises the question: Is mutual fund borrowing (within the legally permissi-

ble limits) irrelevant, beneficial, or harmful for fund investors? This article is the first

to study empirically the performance of open-end mutual funds that invoke their statu-

tory borrowing authority. We find meaningful differences in the performance of bor-

rowers and nonborrowers.

Despiteits importance, capitalstructure has received little attention in theacademic lit-

erature onopen-end funds. Previous studieshave tended to assumethat borrowing is a substi-

tute for and equivalent to leverage that can be obtained through other methods such as

options,futures, margin, and shorting. However, we find evidence that —whereas borrowing

is used to amplify risk — derivatives and other financialinstruments are used to mitigate risk.

In other words, broadly inclusive definitions of “leverage” commonly used in the literature

may not be as useful as the isolated measureof borrowing developedin this article.

This analysis is based on a comprehensive sample of open-end funds contained in

the Center for Research in Security Prices (CRSP) mutual fund database. Information

on mutual funds’ use of borrowing and other financial instruments is extracted from

Form N-SAR, which each mutual fund is required to file semi-annually with the Securi-

ties and Exchange Commission (SEC). Mutual funds are required to disclose on Form

N-SAR whether they used any long-term or short-term debt during the reporting period

(typically six months or a year).

4

We collect data from each filing over a 17-year period,

and combine them with the CRSP data.

Eighteen percent of open-end domestic equity mutual funds borrowed money for

leverage within the sample period. These funds in our sample are “plain-vanilla” domestic

equity funds. Figure 1 shows that, as expected, funds that borrow money appear to out-

perform nonborrowing peers during years when stock returns are high, but underperform

when returns are low or negative.

However, we find in our 17-year panel that funds that borrow money underperform

nonborrowers in meaningful ways. After controlling for other differences, borrowing

funds significantly underperform nonborrowers while also engaging in greater risk

3

Fact Sheet of Baron Partners Fund (Mar. 31, 2018).

4

Disclosures of mutual fund borrowing are relatively opaque, as funds are required to disclose whether they

“engaged in” borrowing but not how much debt they used or for how much of the reporting period it was

outstanding.

768 Warburton and Simkovic

taking. A borrowing fund suffers a 62 basis point reduction in annual performance com-

pared to funds that do not borrow (or use other financial instruments), all else equal.

Not only do borrowing funds underperform, they also expose investors to greater risk tak-

ing (increased volatility), thus imposing an additional cost on investors in borrowing

funds. After accommodating for the difference in risk taking, we estimate that borrowers

underperform nonborrowers by 48 to 72 basis points per year on a risk-adjusted basis. It

is not the interest expense on the debt that causes borrowers to underperform. We find

evidence that funds use debt opportunistically within the year in an attempt to catch up

to their better-performing peers by year end. This tournament-driven behavior is harmful

to the fund’s investors, as such risk-shifting funds typically fall further behind their peers.

Funds that borrow more often do not employ debt with greater skill but instead generate

greater value destruction.

By way of contrast, we also examine funds that use financial instruments,

broadly defined to include not just borrowing but also derivatives such as options and

futures as well as margin and shorting. Funds that use financial instruments perform

about the same as funds that do not, both on a total return and risk-adjusted basis.

We also find consistent evidence that funds using instruments are typically less risky

than other funds. These findings suggest that funds tend to use instruments (of the

nonborrowing kind) in order to hedge risk more than they use instruments for

leveraging purposes. Figure 1 shows that the only year in which funds using instru-

ments were able to outperform was the financial crisis year of 2008. However, inves-

tors pay for this hedging activity, not in risk-adjusted performance but in out-of-

pocket costs. Expense ratios are significantly higher for funds that use these instru-

ments (though not for funds that borrow).

Because borrowing predicts meaningfully different risk and returns for inves-

tors, our analysis suggests that investors would benefit from greater transparency into

mutual fund capital structure. The approach pioneered in this study could contribute

to further research on mutual fund borrowing, particularly with more detailed disclo-

sures of borrowing activity. The SEC recognizes the importance of disclosure of

mutual fund leverage. The SEC is currently implementing a major change in mutual

fund reporting requirements that focuses on detailed monthly disclosure of a fund’s

use of derivatives.

5

However, the new disclosures do not require similar detailed

reporting of borrowings, which also create leverage. Hence, risks from borrowing may

not be apparent to investors as borrowing remains confined to disclosures that are

relatively opaque.

The rest of the article is organized as follows: Section II contains a review of the

related literature on the use of leverage and financial instruments by mutual funds.

Section III briefly discusses the regulation of mutual fund leverage. Section IV describes

5

The SEC is introducing a new monthly reporting requirement (Form N-PORT) that will require detailed disclo-

sures of fund derivatives usage. Investment Company Reporting Modernization, Release Nos. 33-10231; 34-79095;

IC-32314; (Dec. 8, 2017) (Final Rule). In addition, the current Form N-SAR would be discontinued, replaced by

Form N-PORT and a new Form N-CEN.

Mutual Funds that Borrow 769