I Don’t Recall: Information and the Governance Tradeoffs in Securities Lending

Institutional investors often lend shares to earn extra revenue on their portfolio holdings. But from a governance perspective, this lending comes at a cost: the votes attached to lent shares belong to whoever holds them on the record date. Lending and voting are therefore commonly viewed as substitutes, with lenders trading off voting power against revenue. There is considerable debate over how to weigh this tradeoff. The SEC recommends that funds recall lent shares ahead of material votes, while BlackRock states that lending revenue outweighs the value of the forgone votes in almost all instances. In a new paper, I argue that both sides miss an important consideration: lending can also make the lender’s remaining votes more informed.

Consider a firm that is underperforming, perhaps because its managers are shirking or incompetent. Short sellers specialize in uncovering exactly this sort of bad news, which enables them to earn trading profits by shorting overvalued stocks. But to short the stock, a short seller must first borrow shares from a lender who, by definition, is long the stock (e.g., an index fund). Her very request to borrow shares conveys some of this bad news to the lender, who can use it to vote against management and thereby punish the managers for the underperformance. Lending shares therefore makes the lender’s vote more informed.

Securities lenders can also recall lent shares at any time, in which case the short seller must return them within the normal settlement period, restoring the lender’s voting power. Lending shares and then recalling them before the record date would seem to offer the best of both worlds: the lender acquires the information without losing any votes. However, being recalled is costly to the short seller, since it forces her to close out her position prematurely or borrow replacement shares. A short seller who anticipates being recalled will therefore not borrow in the first place, curtailing the flow of information in the lending market. If the lender wishes to learn from the short seller, he must therefore commit not to recall shares. Higher lending fees supply exactly this commitment: the higher the fee, the more revenue the lender forgoes by recalling, and the more secure the short seller feels borrowing. The lender cannot simply lower the fee to attract more borrowing, since a low fee is cheap to walk away from. In short, there is no free lunch: the only way the lender can acquire lending-market information is by committing not to recall the shares.

Recalling shares to vote is usually seen as “good governance,” since it returns votes to a long-term owner who is more likely to cast them responsibly. However, the logic above shows that the picture is more nuanced and depends on what other sources of information the lender has. Compare firm value when the lender does not recall shares with a counterfactual in which he always recalls them: without recall, the lender casts fewer votes, but those votes are more informed. That is, the key tradeoff is between a smaller number of more informed votes and a larger number of less informed ones. I show that not recalling shares improves firm value whenever the lender lacks other high-quality sources of information. Intuitively, the extra recalled votes do little to improve firm value when the lender is otherwise uninformed—they are just noise. By not recalling shares, the lender ensures that at least some of his votes carry information. On the other hand, when the lender is already quite informed, the extra information from the lending market has only a small impact on the informativeness of his vote, so recalling shares is better for firm value. Nevertheless, the lender may privately prefer not to recall shares in this situation: if the lending fees are high enough, they more than compensate him for his share of the destroyed firm value.

This insight is particularly important for the relationship between index funds and corporate governance. Index funds are often criticized for having weak incentives to acquire firm-specific voting information because they own small, diversified stakes in many firms. However, index funds are major lenders of securities, so they passively receive a steady flow of such information via their lending operations. Index funds’ direct incentives to acquire information therefore understate how informed they are as voters. Moreover, index funds resemble the otherwise-uninformed lender in the previous paragraph, so not recalling shares may actually improve firm value in their case. Share recall frequency is therefore a poor proxy for how much index funds value governance.

My analysis shows that lenders gain more than revenue when lending shares—they also acquire information that makes their remaining votes more informed. Policies that push funds to recall shares ahead of every material vote can then damage firm value by curtailing the flow of information between short sellers and institutional investors. Hence, lending and voting are not necessarily substitutes but can be complements through informational spillovers.

George Nurisso is an Assistant Professor at the University of Florida's Warrington College of Business.

This blog is based on a paper presented at the Corporate Governance Academic Forum at the University of Toronto with ECGI, held on 2 October 2026. Visit the event page to explore more conference-related blogs.

The ECGI does not, consistent with its constitutional purpose, have a view or opinion. If you wish to respond to this article, you can submit a blog article or 'letter to the editor' by clicking here.