≡ Menu

On Pigou on Externalities

Many years ago I read the sections of the fourth and final edition of A.C. Pigou‘s most influential book, The Economics of Welfare (1932), that deal with a distinction made famous (among economists) by Pigou: the difference between private cost and social cost.  Pigou’s formulation was in terms of output – what he called the difference between “marginal social net product” and “marginal private net product.”

I re-read those sections today.  (They are Chapter IX [‘Divergences Between Marginal Social Net Product and Marginal Private Net Product’].)  I’m embarrassed to realize how few of the details of Pigou’s discussion I remember from my first reading of it.  One fact that strikes me today as being especially interesting is that the first kinds of cases that Pigou identifies and discusses as ones that typically give rise to a difference between the private costs born by economic decision-makers and the social costs of those decisions are contractual arrangements between owners of agricultural land and tenants.  (Pigou introduces this class of cases more generally as ones involving “owners of durable instruments of production, of which the investor is a tenant” [p. 174].)  Pigou goes on at some length explaining how tenants who hold agricultural lands under leases from landlords do not have incentives toward the land that are exactly and naturally in line with the interests of the landlords.  (Pigou mostly assumes that the landlords’ interests in the the land correspond to the ‘social’ interest.  Tenants’ incentives to treat lands differently than lands would be treated were there no divergence between the interests of tenants and that of landlords is what, in Pigou’s telling, gives rise to socially inefficient use and improvement in leased agricultural lands.)

Pigou reviews several different sorts of contracts, legal rules, customs, and statutes that deal with this problem of divergent interests.  Some perform better than others, but none perfectly aligns the interests of tenants with those of landlords.

Two things are surprising about this treatment by Pigou.  The first is Pigou’s suggestion that this divergence is notable.  As I read page after page of Pigou’s discussion, I kept asking myself: Why did Pigou suppose that this fact of reality is an economic-policy problem?   Of course, I write several generations after Pigou, but as I see matters, the fact that contracting parties do not naturally have perfectly aligned interests is simply a reflection of the reality that contracting parties are different people in different circumstances.  Were it otherwise – if the parties’ interests were naturally and perfectly aligned – there would be no need in the first place for contracting (or at least not for contracts that specify any matters other than price and quantity).  The very point of contracts is to create binding law between the contracting parties that brings their interests into closer alignment than would exist in the absence of contracts.  And the fact that no contract achieves perfect alignment of interests – the fact that in all contracting situations there comes a point where the further effort and time necessary to bring the parties’ interests into even closer alignment are too great compared to the expected benefit of closer alignment – is no sign of inefficiency or market failure.  It’s simply a reflection of cost.  Contracting isn’t free, and perfection in a contract is a luxury that few can afford and that none find worthwhile to purchase.

Moreover – and more importantly – the costs of these contractual ‘imperfections’ fall on the contracting parties themselves.  They’re not “social costs.”  A landlord who fails to write a contract with his tenants in ways that prevent his tenants from mistreating his land privately incurs the cost of that failure.  It’s not a social cost.  A tenant who agrees to a contractual term under which she is required to exert “too much” effort to maintain the land privately incurs the cost of that poor agreement.  It’s not a social cost.

The second surprising thing about Pigou’s treatment is that all of the problems that he identifies as emerging because of the failure to perfectly align the incentives of tenants with the interests of landlords would disappear if the tenants owned the land.  Pigou himself recognized that tenants generally act more in line with the interests of landlords the longer and more secure are the tenants’ leaseholds.  The reason this fact is interesting is that Pigou himself – without, it seems, realizing it – stumbled long before Ronald Coase did upon the realization that secure private property rights are an effective means of internalizing externalities.  The structure of property rights matters.