Passive Exit

Economist Albert O. Hirschman (1970) classically set out the two alternatives facing dissatisfied members of an organization: They can voice displeasure or exit for greener pastures. Hirschman’s model has long explained the tradeoff facing shareholders of a poorly governed firm: Agitate for change or take the “Wall Street Walk” by selling shares.[1] Professor John C. Coffee (1991) showed that large institutional investors have little incentive to voice their concerns to monitor portfolio firms,[2] a trend exacerbated by low-fee “passive”[3] portfolio management (Bebchuk et al., 2017).[4]

While voice is often too costly for passive investors, exit is downright impossible. An index fund is contractually bound to replicate its underlying index, which implies buying and holding the securities of every constituent firm regardless of how poorly it is governed. Whether the unavailability of exit hampers monitoring is hotly contested (Fisch et al., 2019).[5] But precisely because they are unable to sell their shares, index funds are well-positioned to lend those shares to short sellers (Palia and Sokolinski, 2019), who place a negative bet that yields a profit if a stock price declines.[6] By lending shares to be sold first and bought later, passive investors delegate monitoring of portfolio firms to short sellers.

In a new paper, I use share lending data to study the exit contract between passive investors and short sellers. Like ordinary exit, passive exit is informationally sensitive: While index funds and ETFs cannot sell shares of underperforming firms directly, they can lend to short sellers borrowing shares of those firms’ stock.

To be sure, passive investors can share in the surplus from shorting on negative information simply by lending their shares. Increased demand will drive up share lending rates, leading to de facto profit sharing between shareholder lenders and informed short-seller borrowers. Making shares available for borrowing reflects a kind of institutional delegation of exit from passive investors to short sellers, even if lenders are uninformed. But in my article, I examine whether there might also be a supply-side exploitation of information and market power held by “passive” share lenders. When lenders possess negative information about a firm and sufficient market power to price discriminate against borrowers, do they constrict supply and raise lending rates above and beyond the demand-side increase?

I take advantage of an institutional characteristic of mutual funds, namely that the same fund family sometimes manages both active and passive funds. For example, when a portfolio manager drops a firm from an actively managed Vanguard fund, that information is shared with Vanguard’s portfolio review department and other Vanguard “access persons” who receive portfolio updates as frequently as every day. While access persons are prohibited from personally profiting from nonpublic information about active funds’ trading, there is no publicly disclosed policy prohibiting that information from being shared with Vanguard index funds and ETFs lending their shares to short sellers. When Vanguard controls a large volume of shares in the securities lending market for that firm, Vanguard is theoretically able to exploit that informational advantage by raising lending rates.

Of course, it is difficult empirically to determine whether an increase in share lending rates correlated with exit by an active fund is driven by supply-side information exploitation as opposed to increased demand from short sellers. Negative information could reach fund managers, passive lenders, and borrowers at once, simultaneously increasing the price of share lending.

I overcome this identification challenge by exploiting an exogenous supply-side shift in the elasticity of share lending rates to negative information: a prior change in the identity of the portfolio management team of the active fund. The intuition underlying this sort of longevity effect is that a stable portfolio management team likely has better relationships with the securities lending business than one that has experienced turnover. A stable team could also be more effective at digesting information due to the “tournament” nature of fund management (Qiu, 2003).[7]  My identifying assumption is simply that instability in a portfolio management team is a supply-side shock affecting the future ability of a share lender to exploit information about a change in an affiliated active fund’s portfolio.

My study also highlights the link between passive exit and market power in the share lending market, which is a product of the concentration of share ownership in the hands of large institutional investors. A growing literature examines the anticompetitive effects of common ownership (e.g., Azar et al. (2018)),[8] but the effects of concentration on share lending have not been considered. I find that the link between portfolio exit by active managers and higher share lending rates is strongest when affiliated passive index and ETF funds hold a large fraction of a firm’s shares. The exercise of market power is concentrated in intrinsic value lending programs targeting hard-to-borrow securities.

I find that passive lenders with market power in the share lending market capture most, but not all, of the surplus accruing to short sellers by engaging in this form of price discrimination: from 62 percent to as much as 87 percent, depending on the specification.  The risk to short selling is low over this period, explaining why short sellers are willing to open a position even though the share lender has extracted a large fraction of the proceeds.

Finally, I present evidence suggesting that the exercise of market power in the share lending market screens for higher-quality short sellers. Mitts (2019) finds that pseudonymous short campaigns may facilitate profitable market manipulation as measured, in part, by systematic price reversals.[9] Here, I find that lending by stable managers is followed by fewer price reversals as compared with unstable managers, suggesting that the flow of information to passive funds not only captures rents but also deters less-informed short selling. By screening for more informed short sellers, the exercise of market power by share lenders imposes a positive externality in the form of greater price accuracy (Fox, 1999).[10]


[1] A. O. Hirschman. Exit, voice, and loyalty: Responses to decline in firms, organizations, and states, volume 25. Harvard University Press, 1970.

[2] J. C. Coffee. Liquidity versus control: The institutional investor as corporate monitor.

Columbia Law Review, 91(6):1277–1368, 1991.

[3] Robertson (2019) shows that “passive” indexes are often anything but truly passive, as index committees exercise significant discretion over which firms are included in the index.

[4] L. A. Bebchuk, A. Cohen, and S. Hirst. The agency problems of institutional investors.

Journal of Economic Perspectives, 31(3):89–102, 2017.

[5] J. E. Fisch, A. Hamdani, and S. Davidoff Solomon. The new titans of wall street: A theoretical framework for passive investors. University of Pennsylvania Law Review, Forthcoming, pages 18–12, 2019.

[6] D. Palia and S. Sokolinski. Does passive investing help relax short-sale constraints? Available at SSRN 3335283, 2019.

[7] J. Qiu. Termination risk, multiple managers and mutual fund tournaments. Review of Finance, 7(2):161–190, 2003.

[8] J. Azar, M. C. Schmalz, and I. Tecu. Anticompetitive effects of common ownership. The Journal of Finance, 73(4):1513–1565, 2018.

[9] J. Mitts. Short and distort. Journal of Legal Studies (forthcoming), (592), 2020.

[10] M. B. Fox. Retaining mandatory securities disclosure: Why issuer choice is not investor empowerment. Virginia Law Review, pages 1335–1419, 1999.

This post comes to us from Joshua Mitts, associate professor of law and Milton Handler Fellow at Columbia Law School. It is based on his recent article, “Passive Exit,” available here.