Toward a More General Model of Land Tenancy and Reform: Comment

In a recent article A. Y. C. Koo [1973] sought to advance a "more general theory of land tenancy" and draw implications for land reform. While his contention that different models of share-tenancy (with conflicting efficiency properties) have different implications for land reform is correct, unfortunately, his mathematical analysis of oligopoly land market is both misleading and erroneous. It is misleading in the sense that it is not (as a close look at the model would verify) an oligopoly model of share-rental determination, which it purports to be. Thus, on this account, it is not really a critique of either Bardhan and Srinivasan [1971] or of Cheung [1974], which it claims to be.1 Rather it is an oligopoly model of determination of fixed-rental in the landlease market. If so, it loses much of its importance, at least empirically, since the fixed-rental system is of limited empirical significance. And the efficiency arguments for and against the share-rental lease, which he advances in the beginning, and also the land-reform question of reducing the share-rental become irrelevant for the analysis that follows. Second, even if we consider his mathematical argument on land renting on its own grounds, it is unfortunately marred by a mathematical flaw that prevails throughout his analysis and leads him to counterintuitive results. Section I gives a restatement of the Koo model and points to its errors and the resulting counterintuitive results. Section II reformulates the model in correct terms and shows how intuitively and empirically plausible results may be derived therefrom.