Hedge Funds, Side Pockets, And Gates

Discretionary Liquidity: Hedge Funds, Side Pockets, And Gates by Warrington College Of Business

Adam L. Aiken

Quinnipiac University

Christopher P. Clifford

University of Kentucky

Jesse Ellis

University of Alabama

First draft: November 2012

This draft: November 2012

Abstract:

During the recent financial crisis, more than 30% of hedge fund managers used their discretion to restrict investor liquidity through the use of “gates” or “side pockets.” Using a database of hedge fund investor interests, this paper is the first to empirically examine the determinants of these discretionary liquidity restrictions (DLRs) and their consequences for hedge fund investors. We find that funds enacted DLRs following poor performance and when their portfolio assets were more illiquid. However, despite claims from managers that DLRs protected investor interests by preventing fire-sales, funds that enacted DLRs continued to underperform comparable funds. Consistent with DLRs reflecting agency problems, we find that restricting investor liquidity during the crisis had a negative impact on fund reputation that spilled over across the hedge fund family. DLR funds and their family affiliates had a more difficult time raising capital and were more likely to cut their fees in the post crisis era.

Discretionary Liquidity: Hedge Funds, Side Pockets, And Gates – Introduction

Hedge funds invest in complex and illiquid assets. Because they offer redeemable claims to investors, however, their strategies and survival are constrained by their ability to retain outside financing (Shleifer and Vishny, 1997). Hedge funds, therefore, typically maintain withdrawal restrictions, such as lockups, to slow down the flow of capital from their funds. These contractual restrictions ensure that, under normal market conditions, the funds can invest in illiquid assets and have the flexibility to meet redemptions without resorting to selling illiquid portfolio assets at “fire-sale” prices. However, market conditions during the financial crisis of 2007-2009 were anything but normal. As market liquidity dried up and performance suffered, many funds found themselves subject to substantial withdrawal requests that overwhelmed ordinary redemption restrictions, creating the potential for the funding liquidity spiral discussed in Brunnermeier and Pedersen (2009). In order to combat the run on their assets, some hedge fund managers enacted “gates” or “side pockets” which served to prevent investor withdrawals from illiquid hedge fund portfolios. These restrictions were imposed ex-post at the discretion of fund managers and were in addition to the ordinary withdrawal restrictions (e.g. lockups and redemption notice periods)present ex-ante in the partnership agreements. In this paper, we study the causes and consequences of these discretionary liquidity restrictions (DLRs) for hedge fund managers andtheir investors.

The majority of DLRs can be broadly classified as either gates or side pockets. When invoking a gate, the manager temporarily suspends redemptions from the fund. When a side pocket is created, the fund segregates a portion of its assets into a separate illiquid investment vehicle. DLRs can also be viewed as a liquidity restriction option from the viewpoint of the hedge fund manager (Ang and Bollen, 2010). Most hedge fund agreements afford the manager the option to restrict investor liquidity by invoking DLRs when redemptions would force the fund to liquidate assets too quickly in illiquid markets at unfavorable prices.

The value of DLRs for hedge funds and investors remains a contentious issue, especially given their prevalence in the recent financial crisis. Managers argue that by initiating DLRs the fund can protect investors from themselves and ensure that assets can be sold at fair values after asset markets stabilize. However, restricting redemptions is costly for investors as it removes their option to “vote with their feet” by removing capital from poorly performing funds (Fama and Jensen, 1983). In addition, because of their discretionary nature, it could be tempting for some hedge fund managers to abuse side pockets and gates as a means of preserving fund capital and earning excess fees. The notoriety of DLRs in the press is partly due to investor outrage over being unable to access their capital.2 Despite the attention these DLRs received from the media, there have been no empirical studies of DLRs, and thus we know little about their economic importance or relevance to investors.

In this paper, we utilize a hand-collected dataset of hedge fund holdings from a sample of institutional investors to provide the first, large-scale empirical study of discretionary liquidity restrictions (DLRs) and their impact on investors. We begin by documenting the incidence of DLRs over the period 2006-2011 and find them to be especially prevalent during the crisis period. Though most partnership agreements allow for managers to enact DLRs in extreme circumstances, few hedge funds had ever exercised that right prior to the financial crisis. For instance, in 2006 only 4% of the funds in our sample had enacted a gate or side pocket. Strikingly, by the end of 2009 over 30% of funds in our sample had enacted some form of DLR, restricting investor redemptions beyond their ordinary contract terms. This is consistent with reports in the popular press that the widespread use of DLRs came as a surprise to many investors. We find these restrictions tend to last several quarters after being enacted, and were created by some of the largest and most well-known hedge funds. Motivated by the prevalence of DLRs during the financial crisis, we seek to understand their economic determinants, costs, and benefits for hedge funds and their investors. The primary question we ask is whether DLRs ultimately served the interests of investors by protecting them from themselves, or instead served the interest of managers by protecting them from investors voting with their feet.