Investments in physical capital, relationship-specificity, and the property rights approach

Highlights

• Reconsiders the property rights approach to the theory of the firm.

• Investments are in physical capital (instead of human capital).

• Joint ownership can be optimal only if exogenous relationship-specificity is low.

• Parties prefer strictly positive level of relationship-specificity.

• Ownership by the more productive party is optimal if specificity is endogenous.

Abstract

We reconsider the property rights approach to the theory of the firm based on incomplete contracts. We explore the implications of different degrees of relationship-specificity when there are two parties, $A$ and $B$, who can make investments in physical capital (instead of human capital). If relationship-specificity is exogenously given, it turns out that joint asset ownership can be optimal only if the degree of relationship-specificity is sufficiently small. If relationship-specificity can be freely chosen and if party $A$’s investments are more productive, then the parties deliberately choose a strictly positive level of relationship-specificity and they always agree on sole ownership by party $A$.

Introduction

The property rights approach to the theory of the firm (Grossman and Hart, 1986, Hart and Moore, 1990, Hart, 1995) is one of the major achievements in microeconomic research in the past three decades, as it provides a formal framework to analyze basic questions about economic institutions such as firms that were first raised by Coase (1937).¹ In a nutshell, given that contracts are incomplete, a party’s incentives to make relationship-specific investments depend on the fraction of the investments’ returns that the party can capture in future negotiations. Asset ownership matters, because ownership improves a party’s position in the case that future negotiations fail, and hence ownership increases the fraction of the investments’ returns that a party will be able to capture in the negotiations.

The standard model of the property rights approach considers a party’s investments in its human capital only (see Hart, 1995). In this case, it turns out that joint ownership of an asset by two parties cannot be optimal. Under joint ownership, each party has veto power over the use of the asset. Instead, making one party the sole owner of the asset improves this party’s incentives to invest in its human capital, while the other party’s investment incentives are not changed. However, Hart and Moore (1990, pp. 1132–1133) and Hart (1995, pp. 68–69) briefly point out that joint ownership can be optimal if the parties invest in physical capital, so that both parties’ investments can be used by a single asset owner, even in the case that negotiations fail. Joint ownership can be optimal in the presence of physical capital investments, because under sole ownership the non-owner improves the owner’s bargaining position by investing, so that under joint ownership one of the two parties has stronger investment incentives.

In the present paper, we take a closer look at investments in physical capital, which have been largely neglected in the literature on the property rights approach. In particular, we analyze the impact of the investments’ relationship-specificity on the optimality of joint ownership, an issue that to the best of my knowledge has been unexplored so far.

In the first step, we assume that the degree of relationship-specificity is exogenously given. It turns out that joint ownership can be optimal only if the investments are not too relationship-specific. Otherwise, the party whose investments are more productive should be the owner (just as in the standard case where investments are in human capital).

In the second step, we endogenize the degree of relationship-specificity. Suppose that party $A$’s investments are more productive than party $B$’s investments. It turns out that if the degree of relationship-specificity can be freely chosen, then joint ownership cannot be optimal, even when investments are in physical capital. Instead, the parties will agree on $A$-ownership. Moreover, while in the case of investments in human capital the parties would prefer to completely remove any relationship-specificity, in the case of investments in physical capital the parties deliberately choose a positive level of relationship-specificity.