Print
E-MailThe following post comes to us from Reena Aggarwal, Professor of Finance at Georgetown University; Pedro Saffi of the Finance Department at the University of Navarra; and Jason Sturgess, Assistant Professor of Finance at Georgetown University.
In the paper Does Proxy Voting Affect the Supply and/or Demand for Securities Lending? which was recently made publicly available on SSRN, we provide a comprehensive analysis of the securities lending market during a period of tremendous growth in the market. In the past, understanding the securities lending market has been limited partly due to the lack of transparency in this fragmented market. To study the securities lending market in the United States during the period 2005-2009, we use a proprietary data set consisting of shares available to lend (supply), shares borrowed (demand), and loan fees. The data covers most of the securities lending activity in the United States.
We find that on average, 19.57% of a firm’s market capitalization is available for lending, 3.3% is actually borrowed, and the annualized loan fee is 42 basis points. The variation in the supply of lendable shares shows great variation with minimum and maximum values of 0.01% and 74.38% of market capitalization. There is considerable interest in some stocks and almost 100% of the available supply of such stocks actually gets borrowed and is on loan. For these high-utilization stocks, the annual fee can be quite high, with the maximum being 1926 bps. Fee is negative in some cases, implying the lender pays the borrower. The negative fee can happen in fixed contracts in which the rebate is fixed in advance but interest rates are volatile. During 2005-2009, 10% of the stocks were very expensive to borrow and had a fee greater than 100 basis points. 2007 was the peak year for the securities lending market with activity dropping off after the financial crisis.
We examine the role of institutional investors in the voting process by analyzing the supply of lendable shares around the time of a proxy vote. At the time of a proxy vote, there is a significant reduction in the supply of shares available to lend because institutions restrict or call back their loaned shares prior to a vote. We find that there is a marked reduction in supply of lendable shares around the time of a proxy vote. The reduction in supply of more than 1.64% of market capitalization on the record date is economically significant. Our results imply that institutions take seriously their responsibility to vote, and that they are even willing to give up revenue from lending securities when they see benefits from voting. The reduction in securities lending activity by institutions around the time of a vote is direct evidence of institutions playing a role in the voting process in order to bring about changes at companies. The reduction in the supply of lendable shares is most pronounced in cases for which ISS recommends voting against the proposal. Institutions restrict the supply and recall shares already on loan in order to vote on important proxy agenda items. These findings are consistent with institutions responsibly restricting and recalling shares, hence reducing supply of lendable shares ahead of voting record dates.
To address concerns related to empty voting, we also examine the changes in borrowing demand around the time of a vote. There is some evidence of increased demand around the time of the record date. However, the increase in demand is economically small compared to the sharp reduction in supply. Most of the increase in loan fee around the time of a vote is associated with reduced supply, which is related to the desire of institutions to vote their shares.
Focusing on the outcome of proposal votes, we show that votes cast in favor of a proposal are negatively related to lending supply but are not related to demand. There are fewer favorable votes cast if either management or ISS oppose the proposal, and for firms with larger institutional ownership.
In contrast to the activity around the record date of a proxy vote, we find that the large increase in loan fee around the time of the ex-dividend record date is primarily driven by an increase in borrowing demand for cash flow reasons. However, there is little change in the supply of lendable shares. During the financial crisis of 2008, activity in the securities lending market decreased as institutions cut back on their lending programs. The demand to borrow stocks and the fee also experience large reductions during the financial crisis.
Our results suggest that to obtain a complete understanding of the dynamics, it is important to examine both the supply and demand side of the market. Supply is not a constraint for most stocks but supply does get restricted for certain stocks around important events. Policy makers should address several issues related to proxy voting including the need for investors to learn about proxy items before the record date so that they can decide whether to lend their shares. In a similar context, we also believe it is important to address issues of over- and undervoting. However, our results suggest that the use of borrowed shares for voting purposes is limited. It is quite possible that this activity has been reduced in recent years because of the publicity related to empty voting.
The full paper is available for download here.
