Major league baseball’s Moneyball at age 15: a re-appraisal

In the 15 years since publication of the book Moneyball: The Art of Winning an Unfair Game in 2003, major-league baseball (MLB) has seen the embrace of analytics by both fans and teams. Using 45 seasons of MLB data, since the introduction of the designated-hitter in the American League in 1973, the present study validates the central premise in Moneyball regarding the importance of certain performance metrics, such as on-base percentage (OBP). The terms in our empirical model are constructed such that our results permit a straightforward comparison of the relative contribution of each factor towards MLB teams’ success in winning games. We also provide evidence indicating that MLB teams have moved to align their payrolls with Moneyball analytics in the years following the book’s publication.     

Constrained output maximization and US agriculture

This paper presents a framework for estimating an indirect production function and then applies it to aggregate US agriculture. Issues addressed include tests of standard hypotheses about the underlying technology (homotheticity, neutral technical change, etc.) as well as examination of consistency in aggregation and the effects of changes in the level of aggregate expenditure on farm output. Since there already exist several studies empirically describing agricultural production technologies with indirect objective functions, one might fairly ask whether this paper really breaks any new ground? We feel that it does. One reason is that existing studies (e.g. Binswanger, 1974; Lopez, 1980; Ball and Chambers, 1982; Ray, 1982; and Weaver, 1983) all rely on either profit or cost functions. To date there appears to be only one study using an indirect production (Appelbaum, 1979), and it is not in agriculture. This seems unfortunate because there are many instances in agriculture for which an indirect production function (or revenue function approach) seems more appropriate. The reason why relates directly to one's belief about the objective function and constraints farmers face. Ultimately, it seems plausible in a certainty framework that producers maximize profits. And since cost minimization and output maximization are just constrained versions of profit maximization, both cost and indirect production functions are more appropriately viewed as restricted profit functions. What differs is the constraint. Assuming cost minimization implies that farmers are constrained by a fixed output which they must produce; in most instances, this is implausible. More likely, the level of output is itself a choice variable. Output maximization, on the other hand, suggests that the main constraint is the amount of money that producers can muster to hire resources. In other words, farmers may face binding constraints in obtaining the profit maximizing level on expenditure on input utilization. One might think that expenditure may not be a binding constraint because there exists a relatively active US credit market. However, the US credit market is far from being perfectly competitive in a stylized sense, and thus, the possibility of a farmer being able to finance a given level of expenditure may be limited.

Readers familiar with the dual approach to production problems may already be asking why these ideas rule out cost functions since it is well known that the cost function is the distance function of the indirect production function (Blackorby et al., 1978)? In principle, therefore, one could always estimate a cost function and then invert it to obtain the indirect production function. There are two problems with this approach: the first is that if output maximization is truly the objective, output is not predetermined and cannot be treated as exogenous in cost function estimation. Second, estimation of direct and indirect functions using the same data set does not always yield comparable results. Estimation of a transcendental logarithmic (translog) indirect production function and a translog cost function, separately, does not generally yield identical estimates of the Allen elasticities of substitutions. Solving the first problem in simple, but the second suggests that estimation should proceed on the basis of the objective function the researcher deems most appropriate. Therefore, if output maximization is more plausible than cost minimization an indirect production function should be used.

The plan of the paper is as follows. We first outline the theoretical developments necessary to our approach (included are conditions necessary for consistent aggregation over firms). We then present estimates of an indirect production function and use these estimates to investigate the plausibility of various restrictions on the technology, consistency in aggregation, substitution possibilities in agricultural production, and the effect of expenditure on agricultural input utilization. The paper closes with a discussion of the results and suggestions for future research.     

Costs and benefits of housing subsidies to tenants from voluntary and involuntary rent control: a comparison between tenures and income groups

This paper estimates the economic subsides to housing for different groups using the 1988 Joseph Rowntree Memorial Trust sample survey of households.An Olsen type model is used to estimate consumber surplus to households from subsidized housing.These estimates are compared to costs, and housing services in the owner occupier market. Poorer families receive larger benefits than richer families. Transfer efficiency is not high; but the local authority sector receives the greatest benefits from controlled rents.     

A stochastic production function for machine repair

Although work by Levhari and Sheshinski (1970) and Syrquin (1972) suggests that general functional forms, such as the Cobb–Douglas and CES, accurately approximate a machine-repair stochastic production process, this paper shows that these authors have not adequately captured the stochastic nature of the production process. More importantly, while general forms are convenient and widely employed, this paper argues that there is a more direct method for determining the properties of the production process.     

Cross-country growth regressions: problems of heterogeneity,stability and interpretation

The paper discusses the issues of heterogeneity and stability of cross-country growth regressions that have been used to study the problem of convergence. Almost all studies use pooled regressions. The paper considers the issue of pooling under heterogeneity using a hierarchical Bayesian method and estimates growth regressions for different panels studied in earlier papers, and different regimes. The conclusion is that the convergence rates are higher than those obtained from pooled regressions under the assumption of homogeneity and that there is instability over time in the relationships.     

Coffee export booms and monetary disequilibrium: some evidence for Colombia

The theoretical models that analyse the monetary consequences of export booms show that under a regime of fixed exchange rates, they affect not only the demand for money, via real income, but also the money supply via foreign exchange accumulation. Within this theoretical framework, this study proposes an empirical approach to determine whether the coffee booms of the second half of the 1970s and mid–1980s led to excess money supply in the Colombian economy. The findings provide evidence in favour of a direct association between coffee export booms and excess money supply, implying that external disturbances jeopardize the ability of the economic authorities to carry out successful monetary policy.