Matching Capital and Labor

• Author: BINYING LIU,JONATHAN B. BERK,JULES H. BINSBERGEN
• Published date: 01 December 2017
• DOI: http://doi.org/10.1111/jofi.12542
• Date: 01 December 2017

THE JOURNAL OF FINANCE •VOL. LXXII, NO. 6 •DECEMBER 2017

Matching Capital and Labor

JONATHAN B. BERK, JULES H. van BINSBERGEN, and BINYING LIU∗

ABSTRACT

We establish an important role for the firm by studying capital reallocation decisions

of mutual fund firms. The firm’s decision to reallocate capital among its mutual fund

managers adds at least $474,000 a month, which amounts to over 30% of the total

value added of the industry. We provide evidence that this additional value added

results from the firm’s private information about the skill of its managers. The firm

captures this value because investors reward the firm following a capital reallocation

decision by allocating additional capital to the firm’s funds.

WE DEMONSTRATE THAT AN IMPORTANT ROLE of a firm in the mutual fund sector is

to efficiently match capital to skill. In a world with perfectly rational players,

no information asymmetries, and no other frictions, the role of a mutual fund

firm would be irrelevant because investors themselves would efficiently allocate

their own capital among managers. In reality, mutual fund executives play a

very important role in capital allocation. We find that a decision to increase a

portfolio manager’s assets under management (AUM) leads to an increase in

the manager’s productivity as measured by value added. Similarly, a decision

to reduce a manager’s responsibilities by taking away assets also leads to an

increase in value added. Indeed, at a minimum, the decision to reallocate capital

to a manager adds, on average, $474,000 per manager per month, implying

that the firm is responsible for at least 30% of the total value added of the

average manager. Mutual fund firms therefore appear to add substantial value

by intermediating between investors and managers and thereby efficiently

matching capital to skill.

We further find that investors are unable to match the firm’s capital realloca-

tion decisions themselves. We hypothesize that this is due to the significant in-

formational advantage that mutual fund executives have relative to investors.

∗Jonathan Berk is with Stanford University and the NBER, Jules van Binsbergen is with the

Wharton School at the University of Pennsylvania and the NBER, and Binying Liu is with the

Hong Kong University of Science and Technology. None of the authors have any relevant or ma-

terial financial interests that relate to the research described in this paper. We thank the Acting

Editor, David Robinson; two anonymous referees; Anat Admati; Peter DeMarzo; Daniel Ferreira;

Ravi Jagannathan; Dirk Jenter; Francisco Perez-Gonzalez; Paul Pfleiderer; Anamaria Piescha-

con; Jonathan Reuter; Michael Roberts; Paulo Volpin; and seminar participants at Georgia State

University, the University of Chicago, Oxford University, University of Warwick, Baruch College,

Stanford University,UCLA, the University of Maryland, Wharton, University of New South Wales,

University of Sydney, University of Melbourne, the Adam Smith Workshop, the SFS Cavalcade,

and the NBER Corporate Finance Meetings, for helpful comments and suggestions.

DOI: 10.1111/jofi.12542

2467

2468 The Journal of Finance R

In particular, firm executives know every trade a manager makes, as well as

trades that the manager chooses not to make. Executives can use this infor-

mation to direct capital away from overfunded managers toward underfunded

managers.

We find support for the informational advantage hypothesis: (1) external

hires that involve a change in AUM do not lead to a detectable change in

future value added, (2) while past performance does explain investor flows,

it does not have much power to explain firm capital reallocation decisions,

and (3) investors respond to these capital reallocation decisions by investing

additional capital in the firm’s funds. We also find that the capital realloca-

tion decision adds more value for young managers, in line with the hypothesis

that the firm’s advantage derives from its access to superior information about

managerial ability. These results are consistent with the view that firm ex-

ecutives use factors not easily observable to people outside the firm to make

personnel decisions (Alchian and Demsetz (1972)). They are also consistent

with investors trusting mutual fund firms to make sound financial decisions

on their behalf (Gennaioli, Shleifer, and Vishny (2015)), and so when they see

these firms making a managerial change, they respond by investing additional

capital.

Although a large body of evidence demonstrates the importance of the firm’s

informational advantage in assigning workers to jobs and (thereby) determin-

ing compensation, there is little evidence on whether firms also use their su-

perior information to determine an employee’s job scope. Our results imply

that, by correctly determining how much responsibility to give an employee in

a particular job, firms add considerable value. Consequently, the documented

wage gains that result from internal job assignments likely not only result from

the productivity gains from correctly matching jobs to workers, but also from

correctly assigning worker responsibility within a job.

Many questions pertaining to the economics of organizations are difficult

to address because it is often hard to measure employee output directly. An

advantage of our study is that, because the performance of a mutual fund

is public information, employee output is directly observable. A mutual fund

manager has one task—to invest capital on behalf of investors. The return he

generates as well as the amount of capital invested are public information.

The investor’s next-best alternative investment opportunity, an investment in

passively managed index funds, is also observable. By comparing the manager’s

performance against this alternative, we can directly calculate an individual

manager’s productivity, that is, value added. A second advantage of our study

is that mutual fund firms own little physical capital.1Consequently, ownership

rights to capital cannot play an important role in why mutual fund companies

exist or are valuable. Our results therefore imply that other factors, such as

the informational role of the firm, are also important.

1Although the industry is capital-intensive, firms do not own their own capital. Instead, firms

manage capital on behalf of outside investors. That is, capital providers retain full ownership

rights to their capital and can call it back at any time.

Matching Capital and Labor 2469

Although we do not find strong evidence for a behavioral explanation of our

findings, we cannot definitively rule out the possibility that our results derive

at least in part from suboptimal investor behavior. For example, part of the

value that the firm adds could result from investor inattention—firms add

value by paying attention on investors’ behalf. We leave it to future research

to determine the possible importance of this explanation for our results.

In Section I, we describe how our paper relates to the existing literature.

We develop a simple model that formalizes our hypothesis for why firms add

value in Section II. In Section III, we define the value added of firms, funds,

and managers. We describe the data set and provide summary statistics in

Section IV. In Section V, we show that we can reject our null hypothesis that

firm capital reallocation decisions do not add value and, more importantly,

provide a lower bound on the value firms add. In Section VI, we show that

investors respond to these capital allocation decisions by investing additional

capital with the firm, indicating that investors are aware of the value these de-

cisions add. Section VII presents evidence consistent with the hypothesis that

the value the firm adds by reallocating capital derives from the firm’s informa-

tional advantage from observing the quality of its employees. In Section VIII,

we discuss the importance of using value added instead of gross alpha as our

main variable of interest. Finally,in Section IX we provide concluding remarks.

I. Background

The literature on the economics of organizations raises several important

questions about the role of firms. What makes a firm successful? Is it an inher-

ent characteristic of the firm itself, or is it that a successful firm is simply a

collection of particularly talented employees? Why do people choose to work for

firms rather than for themselves? And do personnel decisions within the firm

add to overall firm value?

A large theoretical literature attempts to shed light on these questions (see,

for example, Holmstrom and Tirole (1989), Hart and Moore (1990), Hart (1995),

and Rajan and Zingales (1998)). A key component of modern theories of the firm

is ownership. In a world with incomplete contracting, incomplete information,

and bounded rationality, ex post bargaining power is affected by ownership.

In particular, because they retain control rights, asset owners have inherently

more bargaining power. An important insight of this literature is that firms

exist to ensure that ex ante ownership is concentrated to allow for efficient ex

post outcomes. However, while these theories undoubtedly help explain why

modern firms exist, they cannot explain the existence of an increasingly im-

portant type of firm, namely, a firm that consists almost exclusively of human

capital. These firms have little physical capital other than perhaps some in-

tangible capital such as the firm’s brand name. As a result, a primary reason

for the existence of these types of firms cannot be the assignment of ex post

bargaining rights through asset ownership.

Our paper relates to a growing literature pioneered by Bertrand and Schoar

(2003) that seeks to isolate the productive role of the worker (manager) from