This interview
was first published by the Global Association for Risk Professionals on April 14th
2016
Lending programs best help fund
families weather crises when the funds are well governed, Prof. Vikas Agarwal
says
Liquidity dries up during periods of crisis when there
is flight to quality in the markets. The credit crisis of 2008 was exacerbated
by illiquidity of assets. The Southeast Asian currency crisis of 1997 and the
collapse of Long Term Capital Management in 1998 also brought attention to
liquidity as a factor in systemic and market risks. In late 2015, it was an
issue in the closing of the Third Avenue Focused Credit fund, which faced
redemption pressures on its holdings of distressed-debt assets.
A March 21,
2016 U.S. Treasury Department press release (https://www.treasury.gov/press-center/press-releases/Pages/jl0393.aspx) regarding a Financial Stability Oversight Council
meeting, stated that “the Council discussed its ongoing assessment of potential
risks to U.S. financial stability from asset management products and activities,
including a discussion regarding potential financial stability risks related to
liquidity and redemption risks and risks associated with the use of leverage by
asset management vehicles.”
Vikas Agarwal, H. Talmage Dobbs Jr. Chair and
Professor of Finance, J. Mack Robinson College of Business, Georgia State
University, says that open-end mutual funds bear significant costs because of
their unique obligation to provide continuous, sufficient liquidity to their
investors. “One potential channel for these costs is the fire sale of assets by
fund managers to meet investor redemptions,” he says, citing Joshua Coval and
Erik Stafford, “Asset fire sales (and purchases) in equity markets,” Journal of
Financial Economics 86, 479—512 (2007). “An investor in an open-ended fund can
exit anytime. Investors especially withdraw their money when the market is
down, when it is the worst time for the fund manager to sell the assets.”
A London
Business School (University of London) PhD in finance who has served as a
distinguished visiting scholar in the Securities and Exchange Commission’s
Division of Risk and Economic Analysis, Prof. Agarwal (http://www2.gsu.edu/~fncvaa/vikasgsucv3.pdf) has published extensively on hedge fund and mutual fund subjects.
Interviewed recently by Dr. Nupur Pavan Bang of the Indian School of Business,
Hyderabad, the professor discussed the provision and impact of interfund
lending within mutual fund families, once they have obtained permission from
regulators to set up interfund lending programs (ILP).
Would it be
less expensive if the funds maintained enough cash or borrowed from banks to
deal with redemption pressures?
Funds can easily hold the cash to deal with the
problem, but cash does not earn any returns. If the fund has a liquidity
buffer, then it can lead to a drag on performance. Funds have to pay fees to
maintain any committed lines of credit, and interest expenses when borrowing
takes place. If the fund has an uncommitted line of credit, it is possible that
when things go bad, even the banks may refuse to lend. There can also be some
restrictions imposed by shareholders on external borrowings in order to avoid
leverage in funds.
What kinds of
costs do funds incur in the event of a fire sale of assets?
A fire sale forces unwinding of illiquid positions in
a short period of time at disadvantageous prices. In addition, there can be
costs related to predatory trading. If a mutual fund experiences outflows,
other market participants such as hedge funds can try to take advantage of
“distressed” funds by front-running (e.g., short selling the securities that
the mutual funds are expected to sell in a fire sale) or by buying the
securities sold in a fire sale at cheap prices.
How can funds
minimize some of these costs? Can they borrow money when facing redemption
pressures?
The 1940 [Investment Company] Act prevents affiliated
funds (those belonging to the same fund family) from engaging in any kind of
transactions — for example, borrowing and lending — with each other. The
rationale is that such interfund transactions should not result in one set of
investors being worse off than others who might benefit. There is some evidence
in the literature that in order to create star funds, fund families engage in
cross-fund subsidization. For example, families can allocate the hot IPOs to their
better-performing funds and, as a result, other funds in the family are worse
off. [Jośe-Miguel Gaspar, Massimo Massa and Pedro Matos, “Favoritism in mutual
fund families? Evidence on strategic cross-fund subsidization,” Journal of
Finance 61, 73—104 (2006)]
But, in the 1940 Act there is a provision that allows
the fund families to obtain an exemption from the Securities and Exchange
Commission to engage in interfund lending if they can fulfill certain
requirements. The underlying idea is that if there are controls in place to
make sure that there is no cross-fund subsidization, then it seems okay for the
funds within a family to borrow/lend amongst them. This is something that the
fund family needs to convince the SEC about when applying for the ILP.
How would an
interfund lending program work?
There can be saving on the transaction costs when
funds engage in borrowing and lending to each other within a fund family. Let’s
say that the borrowing fund can borrow outside at 5%, and the lending fund can
lend outside at 3%. In such a case, the funds can settle down at a rate between
3% and 5%, say 4%. So the lender is actually able to earn a return which is
greater than what they would earn outside 4% rather than 3%). The borrower’s
rate of 4% is cheaper than the 5% rate outside.
Effectively, ILP creates an “internal capital market”
for fund families and has become more popular over time. In 1987, Fidelity
Investments was the first to apply to the SEC for the ILP. By 2013, funds with
almost 40% of the equity holdings of all mutual funds had applied for the ILP.
What kind of
compliance and controls does the SEC look for?
The onus is on the fund family to convince the SEC
about the proper administration and implementation of the program. The board of
directors is obliged to monitor and review the fund’s participation in the ILP
for compliance. If the fund is not well governed, it can be expensive in the
long run to ensure compliance and to make sure that the program is implemented
in the right way. It takes about a year or so for the SEC to finish the process
of reviewing an application and deciding whether to grant permission to the
fund families for interfund borrowing and lending. In addition, the funds are
required to obtain shareholder approvals and fully disclose material
information about ILP to engage in interfund lending.
If the lending
fund has cash, why would it not invest directly in risky assets rather than
lending to the other fund?
The lending funds are typically money-market funds,
which are not really trying to do fancy stuff. The borrowing funds are mostly
equity funds. The idea here is that liquidity shocks to funds within the fund
family are not perfectly correlated. The ILP essentially relies on the
heterogeneity in the assets and liquidity characteristics of the funds within
the family. So if one fund is experiencing outflows while the other is not, then
one fund can demand liquidity while the other can supply it.
Can you
elaborate on how governance of the fund has an impact on the performance of the
fund?
In my working paper with Haibei Zhao, “Interfund
lending in mutual fund families: Role of internal capital markets” [version:
March 8, 2016], we show that families which are well governed perform better
than those that are not. A better governed fund indicates that the ILP will be
implemented in the right way.
Some of the proxies of fund governance are the size of
the fund family and the fund itself, and the characteristics of the managers.
Larger fund families and larger funds are likely to be better governed. Funds
with fewer managers are less likely to have the free rider problem and hence
should be better governed. Younger managers with career concerns are likely to
be more disciplined, and hence associated with better governance.
Then there is certainly a reputational cost if the
program is not well administered. When investors realize this to be the case,
the fund will face outflows, and such cost is greater for larger funds and
families.
How do the
managers’ characteristics play out?
We find that there is a decline in the sensitivity of
the managerial turnover to past performance after the funds apply for the ILP.
This, in turn, implies that there will be less pressure on the managers in
terms of outflows after they have performed poorly. As a result, bad managers
will continue to stay in the funds, which can actually hurt future performance.
So this is a cost of applying for the ILP for funds that are not well governed.
Note that in well-governed funds, such costs will not be borne by the funds
since the managers would be fired subsequent to poor performance.
Do fund
families that have ILP perform better than those that do not?
We find that funds belonging to families that have ILP
show better performance — if the funds are well governed. Moreover, as
mentioned earlier, the sensitivity of flows to past performance goes down after
the ILP.
In the mayhem that followed the September 11, 2001
attacks on the World Trade Center, a lot of funds experienced outflows. The
funds which had the ILP actually suffered less, because ILP helped them absorb
liquidity shocks much better. In fact, there was a release from the SEC [No.
25156, September 14, 2001] which stated that those funds which have the ILP
could borrow from other funds within their families, and the borrowing limit
was relaxed for a period of five business days after the markets reopened.
Incidentally, we also notice that the number of
applications to the SEC for ILP approval is more around crisis periods such as
1998-1999 (LTCM), 2001-2002 (dot-com bubble) and 2008.
What is the
impact on funds’ liquidity after their families participate in ILP?
Illiquidity of portfolios goes up, and there is an
increase in the portfolio concentration of borrowing funds. Specifically, our
analysis shows that there is a drop in the overall cash holdings of the equity
funds, as they do not need to keep as much cash to meet investor redemptions.
Overall, how
would you rate the ILP as a tool for liquidity management?
In principle, the ILP is a good idea, but it has costs
associated with it if the funds are not well governed. In other words, the
benefits of ILP accrue to well-governed funds. We also show that the funds use
the ILP for the right reasons. Only about 7% of the funds which have ILP used
it during the period of our study. We find that funds borrow when they
experience outflows and poor performance. Also, we observe that well-governed
funds are more likely to borrow, consistent with the benefits accruing to such
funds.
