Bebchuk, Lucian A. and Cohen, Alma and Hirst, Scott, The Agency Problems of Institutional Investors (June 1, 2017). Available at SSRN: https://ssrn.com/abstract=2982617:
We analyze how the rise of institutional investors has transformed the governance landscape. While corporate ownership is now concentrated in the hands of institutional investors that can exercise stewardship of those corporations that would be impossible for dispersed shareholders, the investment managers of these institutional investors have agency problems vis-à-vis their own investors. We develop an analytical framework for examining these agency problems and apply it to study several key types of investment managers.
We analyze how the investment managers of mutual funds - both index funds and actively managed funds - have incentives to under-spend on stewardship and to side excessively with managers of corporations. We show that these incentives are especially acute for managers of index funds, and that the rise of such funds has system-wide adverse consequences for corporate governance. Activist hedge funds have substantially better incentives than managers of index funds or active mutual funds, but their activities do not provide a complete solution for the agency problems of institutional investors.
Our analysis provides a framework for future work on institutional investors and their agency problems, and generates insights on a wide range of policy questions. We discuss implications for disclosure by institutional investors; regulation of their fees; stewardship codes; the rise of index investing; proxy advisors; hedge funds; wolf pack activism; and the allocation of power between corporate managers and shareholders.
The problem I have with this analysis (besides my usual disagreement with the assumption that we want to goad investors to be more active) is that "stewardship" is just one form of agency problem in the institutional investor context. And maybe not even one of the most serious.
Stewardship is basically concerned with the extent to which institutional investors pester corporate managers, which may or may not redound to the benefit of the hedge fund's investors.
In contrast, there are serious conflicts of interest that directly affect the relationship between funds and their investors:
- Side letters that give some investors in hedge funds a better deal than others.
- Side pockets that make investor funds harder to withdraw funds and "give hedge funds other ways to cheat, in part because managers often don't tell investors how those holdings are doing on a regular basis." (Link)
- "A hedge fund could exile poor investments into a side pocket, allowing the fund to post jazzy returns on its stronger-performing investments, while keeping sour deals in a black hole. If the side-pocket investments later are sold at a loss, investors take the hit -- without dinging managers' performance fees." Id.
- Simultaneous management: "The simultaneous management of hedge funds and private equity funds is rife with potential conflicts of interest, including issues relating to allocation of investment opportunities between the funds, possession of material nonpublic information, valuation and allocation of expenses."
- Personal account dealing: Many are hedge fund managers are extremely high net-worth individuals with substantial portfolios. They can have incentives to "spend more time worrying about their own positions rather than the state of the fund (particularly during periods of market stress)." Outside business interests that can create conflicts of interest are also "quite common among senior hedge fund personnel."
The point that institutional investor stewardship conflicts exist is not new, although bebchuk et al. have done a service by detailing the point at length. In my book Corporate Governance after the Financial Crisis
Even where gains might arise from activism, only a portion of the gains would accrue to the activist institutions. Suppose that the troubled company has 110 outstanding shares, currently trading at $10 per share, of which the potential activist institution owns ten. The institution correctly believes that the firm’s shares would rise in value to $20 if the firm’s problems were solved. If the institution is able to effect a change in corporate policy, its ten shares will produce a $100 paper gain when the stock price rises to reflect the company’s new value. All the other shareholders, however, will also automatically receive a pro rata share of the gains. As a result, the activist institution confers a gratuitous $1,000 benefit on the other shareholders.
Put another way, the gains resulting from institutional activism are a species of public goods. They are costly to produce, but because other shareholders cannot be excluded from taking a pro rata share, they are subject to non rivalrous consumption. As with any other public good, the temptation arises for shareholders to free ride on the efforts of those who produce the good.
Granted, if stock continues to concentrate in the hands of large institutional investors, there will be marginal increases in the gains to be had from activism and a marginal decrease in its costs. A substantial increase in activism seems unlikely to result, however. Most institutional investors compete to attract either the savings of small investors or the patronage of large sponsors, such as corporate pension plans. In this competition, the winners generally are those with the best relative performance rates, which makes institutions highly cost conscious. The problem is exacerbated by the performance metrics applied to many asset managers. Investment managers commonly are assessed on the basis of short-term—mostly quarterly—returns as compared to the returns earned by other managers. This makes fund managers even more cost conscious than otherwise would be the case.
Given that activism will only rarely produce gains, and that when such gains occur they will be dispensed upon both the active and the passive, it makes little sense for cost-conscious money managers to incur the expense entailed in shareholder activism. Instead, they will remain passive in hopes of free riding on someone else’s activism. As in other free riding situations, because everyone is subject to and likely to yield to this temptation, the probability is that the good in question—here shareholder activism—will be under-produced.
This is especially true of the ever growing number of passively managed index funds. Index funds tend to be especially cost conscious, because they compete almost exclusively on keeping fees low and tracking their index as closely as possible. Investing resources in governance activism makes no sense for such funds given their business model.
Even if activism made sense from a cost-benefit perspective, corporate managers are well-positioned to buy off most institutional investors that attempt to act as monitors. Bank trust departments are an important class of institutional investors, but are unlikely to emerge as activists because their parent banks often have or anticipate commercial lending relationships with the firms they will purportedly monitor. Similarly, insurers “as purveyors of insurance products, pension plans, and other financial services to corporations, have reason to mute their corporate governance activities and be bought off.”
Mutual fund families whose business includes managing private pension funds for corporations are subject to the same concern. A 2010 study examined the relationship between how mutual funds voted on shareholder proposals relating to executive compensation and pension-management business relationships between the funds’ families and the targeted firms. The authors concluded that such ties influence fund managers to vote with corporate managers rather than shareholder activists at both client and non-client portfolio companies. Voting with management at non-client firms presumably is motivated by a desire to attract new business and send signals of loyalty to existing clients.
With many categories of institutional investors thus eliminated as potential activists, we are left mainly with union and state and local pension funds, which in fact generally have been the most active institutions with respect to corporate governance issues. Unfortunately for the proponents of institutional investor activism, however, these are precisely the institutions most likely to use their position to self-deal or to otherwise reap private benefits not shared with other investors. With respect to union and public pension fund sponsorship of Rule 14a-8 proposals, for example, Roberta Romano observes that:
It is quite probable that private benefits accrue to some investors from sponsoring at least some shareholder proposals. The disparity in identity of sponsors—the predominance of public and union funds, which, in contrast to private sector funds, are not in competition for investor dollars—is strongly suggestive of their presence. Examples of potential benefits which would be disproportionately of interest to proposal sponsors are progress on labor rights desired by union fund managers and enhanced political reputations for public pension fund managers, as well as advancements in personal employment. … Because such career concerns—enhancement of political reputations or subsequent employment opportunities—do not provide a commensurate benefit to private fund managers, we do not find them engaging in investor activism.
There have been several troubling incidents of the sort to which Romano refers. In 2004, for example, CalPERS organized a proxy campaign to remove Safeway’s CEO and chairman of the Board Steven Burd. At that time, Safeway was in the middle of acrimonious collective bargaining negotiations with the United Food & Commercial Workers Union. It turned out that CalPERS had acted at the behest of Sean Harrigan, Executive Director of the United Food and Commercial Workers Union, who was then also serving as president of CalPERS board. Nor is this an isolated example. Union pension funds tried to remove directors or top managers, or otherwise affect corporate policy, at over 200 corporations in 2004 alone. Union pension funds reportedly have also tried shareholder proposals to obtain employee benefits they could not get through bargaining.
Public employee pension funds are vulnerable to being used as a vehicle for advancing political/social goals unrelated to shareholder interests generally. Mid-decade activism by CalPERS, for example, reportedly was “fueled partly by the political ambitions of Phil Angelides, California’s state treasurer and a CalPERS board member,” who planned on running for governor of California in 2006. In other words, Angelides allegedly used the retirement savings of California’s public employees to further his own political ends.
To be sure, like any other agency cost, the risk that management will be willing to pay private benefits to an institutional investor is a necessary consequence of vesting discretionary authority in the board and the officers. It does not compel the conclusion that we ought to limit the board’s power. It does suggest, however, that we ought not to give investors even greater leverage to extract such benefits by further empowering them.
I also noted that:
The analysis to this point suggests that the costs of institutional investor activism likely outweigh any benefits such activism may confer with respect to redressing the principal-agent problem. Even if one assumes that the cost-benefit analysis comes out the other way around, however, institutional investor activism does not solve the principal-agent problem but rather merely relocates its locus.
The vast majority of large institutional investors manage the pooled savings of small individual investors. From a governance perspective, there is little to distinguish such institutions from corporations. The holders of investment company shares, for example, have no more control over the election of company trustees than do retail investors over the election of corporate directors. Accordingly, fund shareholders exhibit the same rational apathy as corporate shareholders.
