Long‐term value versus short‐term profits: When do index funds recall loaned shares for voting?

• Published date: 01 September 2024
• Author: Haoyi (Leslie) Luo,Zijin (Vivian) Xu
• Date: 01 September 2024
• DOI: http://doi.org/10.1111/corg.12576

SPECIAL ISSUE ARTICLE

Long-term value versus short-term profits: When do index

funds recall loaned shares for voting?

Haoyi (Leslie) Luo | Zijin (Vivian) Xu

Finance Department, Faculty of Business and

Economics, The University of Melbourne,

Melbourne, Victoria, Australia

Correspondence

Haoyi (Leslie) Luo, Finance Department,

Faculty of Business and Economics, The

University of Melbourne, Melbourne, Victoria,

Australia.

Email: haoyil3@student.unimelb.edu.au

Funding information

The authors acknowledge fundings from the

Faculty of Business and Economics Graduate

Research Travelling Scholarship provided by

the University of Melbourne.

Abstract

Research Question/Issue: In this paper, we examine the effects of share lending or

recall on proxy voting, with a particular focus on the role of index funds.

Research Findings/Insights: Our study reveals that higher index ownership in a firm

is associated with an increased likelihood of share recall, particularly in the presence

of higher institutional ownership, lower past return performance, smaller firm size,

and when more shares are held by younger fund families with higher turnover ratios

or higher management fees. Using the Russell 1000/2000 Index reconstitution as an

exogenous shock, we establish a causal relationship between index ownership and

share recall through instrumental variable (IV) analysis. Furthermore, we find a posi-

tive correlation between index ownership and share recall for proxy voting proposals

related to compensation, director election, and those sponsored by management. In

subsequent proxy votes, shareholder-sponsored proposals and environmental, social,

and governance (ESG) proposals receive more support in firms with higher index

ownership, especially when share recall is more prevalent. Our analysis does not pro-

vide evidence to support the conjecture that firms with higher index ownership are

more vulnerable to empty voting issues.

Theoretical/Academic Implications: Our study enhances the understanding of how

index funds recall shares during proxy voting and the impact of index ownership on

voting outcomes. The findings support the practice of index funds recalling shares

to actively engage in proxy voting, effectively addressing the conflict between

short-term profit-seeking through securities lending and long-term governance

responsibilities.

Practitioner/Policy Implications: We contribute to a better understanding of the role

of index funds in corporate governance and shed light on the consequences of secu-

rities lending in proxy votes. These findings have important implications for investors,

policymakers, and market participants in managing the potential conflicts arising from

securities lending activities and promoting effective corporate governance practices.

KEYWORDS

corporate governance, index fund ownership, proxy voting, securities lending

Received: 18 September 2022 Revised: 15 December 2023 Accepted: 4 January 2024

DOI: 10.1111/corg.12576

This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium,

provided the original work is properly cited.

© 2024 The Authors. Corporate Governance: An International Review published by John Wiley & Sons Ltd.

856 Corp Govern Int Rev. 2024;32:856–889.

wileyonlinelibrary.com/journal/corg

1|INTRODUCTION

There is growing concern over the extensive engagement of index

funds in the securities lending market and its implication on the cor-

porate governance. One concern stems from the fact that index funds

often engage in the practice of securities lending to maximize their

return on investment. This process involves lending stocks to market

participants in return for a nominal fee, thereby supplementing the

fund's income and enhancing investor returns. However, this practice

entails a transfer of not only the securities but also the voting rights

and entitlements associated with those securities.

When a security is lent, the voting rights and entitlements associ-

ated with the security are transferred to the borrower. This gives rise

to the issue of “empty voting,”whereby the investors accumulate vot-

ing power in excess of their economic share of ownership by borrow-

ing stock shares prior to or on the record dates. Anecdotal evidence

has shown that share borrowers may vote against the proposals and

profit from their short positions when the share price moves in the

direction favorable to them (Hu & Black, 2006,2007). This opportu-

nistic voting could pose a risk to long-term firm value as borrowers

may vote against critical investments or strategic decisions that are

projected to boost the company's long-term performance but may

negatively impact short-term profitability.

Thus, index funds encounter a palpable trade-off. Securities lend-

ing facilitates the augmentation of short-term profits and can provide

enhanced returns to investors. However, this practice could poten-

tially harm the long-term value of the firms in which the funds hold

shares because of the phenomenon of “empty voting.”Should the

strategic actions of these voters lead to a decline in the firm's long-

term value, the returns of the index fund could be adversely affected.

In this study, we take a close look at this trade-off facing index funds

by investigating their share recall decision around proxy voting.

These governance concerns may have become even more signifi-

cant given that the proportion of equities held by index funds has

risen dramatically over the last two decades and is foreseeable to con-

tinue growing strongly.

i

According to a study by Bebchuk and Hirst

(2019), the Big Three—BlackRock, State Street Global Advisors, and

Vanguard—collectively vote about 25% of the shares in all S&P

500 companies. However, these investing giants may give away their

voting power ahead of shareholder fights by choosing to loan out

shares to maintain the “rich stream of (lending) fees,”as The Wall

Street Journal (Jun, 2020) uncovered.

Prior studies (e.g., Aggarwal et al., 2015) have found that institu-

tional investors often limit the supply of lendable shares or recall

loaned shares before a vote to exercise their voting rights. The likeli-

hood of recall increases when investors have stronger incentives to

monitor corporate activities, in cases of underperforming or weakly

governed firms, and when proposals are made where governance

returns are anticipated to be high. However, the role of index funds in

share lending and corporate governance has largely escaped from aca-

demic attention. Index funds are distinctive among other shareholders

(e.g., long-term blockholders) because of several key features. First

and foremost, their extensive engagement in securities lending

activities creates a unique landscape where short-term gains and

long-term investment goals might conflict. This dynamic offers an

intriguing area of study for their behavior, particularly within the con-

text of active voting. Their immense size and extensive market cover-

age allow index funds to wield considerable influence in the securities

lending market, unmatched by active funds or other blockholders.

Two primary factors contribute to their lending prowess. First, their

size and diversified pool of securities enable large-scale lending activi-

ties without significantly disrupting their investment strategy. Second,

their passive investing approach, aimed at tracking the market,

ensures a stable supply of securities for lending, as it involves less fre-

quent trading compared to active funds.

Furthermore, index funds face intense competition on lowering

fund fees which strengthens their motivation to generate profits

through security lending to gain comparative advantage in market

competition, especially because peer index funds carry identical bas-

kets of stocks. Therefore, the revenue an index fund earns from lend-

ing bears higher marginal value for attracting investment flows than it

would for active funds. As an example, income from securities lending

substantially reduced fund expenses for TIAA, as reported by McCol-

lough (2018). Unlike the “free riding”issue associated with passive

funds' monitoring and other initiatives—where a rise in the funds'

holdings value also directly benefits competing funds holding the

same securities—lending income can boost performance relative to

peer funds, particularly if these peers are less active in the lending

market.

As a response to the clients' growing awareness of responsible

investment in recent years, the leading index fund management com-

panies have been emphasizing the importance of corporate gover-

nance and their strong commitment to it. For instance, the BlackRock

CEO Larry Fink stated that “our responsibility to engage and vote is

more important than ever”as “in managing index funds, BlackRock

cannot express its disapproval by selling the company's securities as

long as that company remains in the relevant index.”

ii

These compa-

nies particularly address that investors are protected in many ways

with regard to the issue of borrowing to vote. For example, in a 2019

report, State Street demonstrates their attention to this concern by

citing an announcement from one of the world's largest pension funds

that “decided to suspend stock lending until further notice.”As keen

as they are to fulfill their governance responsibilities, however, it is

not a legal obligation for these asset managers in the United States to

cast votes by recalling shares and giving up lending income.

iii

In other

words, whether and when to recall loaned shares ultimately fall into

the discretion of lenders. Therefore, it remains unclear to what extent

that index funds, as significant shareholder and securities lender, com-

mit to their fiduciary duty particularly in terms of informed voting.

In this paper, we examine share recall activities associated with

index fund ownership around proxy voting. Specifically, we aim to

answer the following research questions: (1) Do firms with higher

index ownership have a lower likelihood of recalling loaned shares for

voting? (2) Are there significant differences in voting outcomes

between firms with higher and lower index ownership? We also

attempt to explore the answer to the question whether firms with

LUO and XU 857

higher index ownership are more likely to experience potential short

sales and “empty voting”than firms with lower index ownership.

We construct a sample of the US public firms using the proxy vot-

ing data from Institutional Shareholder Services (ISS) and equity lend-

ing data from Markit for the period of 2005–2018. Following the

approach of Christoffersen et al. (2007), we develop measures to cap-

ture abnormal securities recall and lending activities around record

dates.

iv

Consistent with prior studies (Aggarwal et al., 2015), we find

evidence that shareholders recall shares for voting purposes. To

answer the first question, we then examine the relationship between

securities recall/lending and index ownership. Our findings indicate

that firms with higher index holdings are more likely to engage in

share recall rather than lending out. Additionally, we consider the

ownership of two other investor types, dedicated investors and tran-

sient investors (Bushee, 1998,2001). Interestingly, we observe that

higher transient ownership is associated with less share recall, con-

trary to the pattern observed for index ownership. These results high-

light the varied governance engagement of institutional investors

(Duan & Jiao, 2016; Matvos & Ostrovsky, 2010) and underscore the

value index funds place on their voting rights as a means to actively

participate in firm governance.

To address endogeneity concerns, we utilize the Russell

1000/2000 indexreconstitutions as exogenousshocks to index owner-

ship. Implementing a difference-in-differences specification in an instru-

mental variable (IV) analysis, our results provide strong support for the

positive association between index ownership and share recall. This IV

analysis helps address endogeneityissues and establishes a causal rela-

tionship. To explore the heterogeneity in the relationship between

index ownershipand share recall, we divide the sample intotwo groups

(below and above median) based on a battery of firm characteristics.

We find that higher index ownership is more strongly associated with

share recall in firms with higher institutional ownership, lower past

return performance, smaller size, and moreshares held by younger fund

familieswith higher turnover ratios or highermanagement fees.

We further explore the relationship between index ownership

and share recall in the context of proxy voting proposals and out-

comes. By categorizing the proposals into different types, such as

management compensation, director election, corporate control, envi-

ronmental, social, and governance (ESG), and others, we observe a

positive correlation between index ownership and share recall for pro-

posals related to compensation, director election, and those spon-

sored by management. This analysis provides insights into the specific

areas where index ownership has a significant impact on share recall

behavior during proxy voting.

To address the second research question regarding the influence

of index investors' share recall on proxy voting outcomes, we analyze

the impact of share recall on the percentage of votes in favor of a pro-

posal in subsequent proxy votes. We consider the interaction

between share recall and proposal item indicators, comparing the

results between firms with below- and above-median index owner-

ship. Our findings indicate that higher share recall is associated with

increased support for shareholder-sponsored proposals and ESG pro-

posals, while facing greater opposition if opposed by ISS. These

results provide insights into how index investors' share recall

behavior can shape proxy voting outcomes, particularly in relation to

shareholder-sponsored proposals and ESG issues.

As an attempt to detect the potential presence of empty voting in

the context of higher index ownership and securities lending supply,

we analyze trading activities around record dates and meeting dates.

Following established methodologies (e.g., Chang et al., 2017;

Lakonishok & Vermaelen, 1986), we construct measures abnormal

lending activities and abnormal trading. Contrary to expectations, our

results do not support the notion that firms with higher index owner-

ship are more susceptible to empty voting issues. However, we

acknowledge the need for further research to fully explore the impact

of index holdings and share recall on empty voting concerns.

Overall, our findings suggest that index funds strive to balance

short-term profit-seeking through share lending with their long-term

stewardship responsibilities by actively recalling shares for voting.

The remainder of this paper is structured as follows. Section 2

provides a brief overview of the securities lending market and proxy

voting in the United States. In Section 3, we review literature and

develop our hypotheses. Section 4outlines the data sources, vari-

ables, and sample construction. Empirical results pertaining to the

relationship between index ownership and share recall for proxy vot-

ing, as well as the variation across different voting proposals and out-

comes, are presented in Section 5. Section 6concludes the paper.

2|INSTITUTIONAL BACKGROUND

2.1 |Securities lending by mutual funds

Securities lending is a transaction where securities are loaned by their

owners to borrowers in exchange for collateral, typically cash or other

securities. In the United States, the collateral for domestic securities is

required to be at least 102% of the market value, while international

securities require 105%. During the loan period, the borrower tempo-

rarily gains ownership and is responsible for any dividend payments to

the lender. Securities lending serves various purposes, including short

selling, arbitrage strategies, and potentially empty voting (e.g., Hu &

Black, 2006). Lending agents play a crucial role in facilitating these

transactions, and the lending fee paid by borrowers is typically split

between the lender and the agent.

Under the Investment Company Act of 1940, US mutual funds

are obligated to disclose their lending activities in the semi-annual and

annual N-SAR filings with the Securities and Exchange Commission

(SEC). This information includes whether they are permitted to lend

securities and whether they actually engage in lending. Additionally,

mutual funds are required to disclose security lending income and col-

lateral details in their annual SEC N-CSR filings.

v

Mutual funds have three general approaches to managing their

lending programs. First, they can hire an independent third-partyagent.

Second, they can use an agent affiliated with the same company that

manages the funds. Last, they can act as their own internal lending

agent. Larger asset management companies often choose to operate

858 LUO and XU