The Backward-bending Supply Curve of Labor in Africa: Models, Evidence, and Interpretation- and Why It Makes a Difference
1981; Western Illinois University; Volume: 15; Issue: 2 Linguagem: Inglês
ISSN
1548-2278
Autores Tópico(s)Agriculture and Rural Development Research
Resumoed model, see Stephen Hymer and Resnick, Model of an Agrarian Economy with Nonagricultural Activities, 'American Economic Review 59 (September 1969): 493-506. This content downloaded from 157.55.39.112 on Wed, 07 Sep 2016 05:26:55 UTC All use subject to http://about.jstor.org/terms The Backward-bending Supply Curve 261 leisure minus his marginal utility from work (i.e., his net utility from an additional increment of leisure). Although to noneconomists the jargon may appear esoteric, the model allows us to focus on four key variables, knowledge or assumptions about which are necessary to predict the farmer's behavior: his marginal utility of income, his marginal product, his marginal desire for leisure, and the marginal disutility of the work he must perform. Economists, applying the concept of diminishing returns, go on to make assumptions about the behavior and value of the variables, e.g., they assume that the marginal product of labor and the marginal utilities of income and leisure are all greater than zero but diminishing and that the marginal utility of work is negative and growing more so.43 A few observations are in order: 1. If the marginal utility of income is held constant, then an increase in wage would call forth an increased effort in hours worked. The lefthand side of the equation would be greater than the right side, and for the right to increase in value, the farmer would have to reduce his leisure (with the marginal unit therefore more valuable to him) and increase his labor (with the marginal unit more disagreeable to him). The amount of responsiveness would depend upon the farmer's valuations, but, assuming work grows more disagreeable per unit as more is performed, and leisure less pleasurable per unit as more is obtained, in no case would the response be to work less as wages increased. In short, increased marginal costs will not serve to explain a backward-sloping supply curve of labor.4 2. If leisure is a normal good (i.e., if, as income rises, the farmer uses some of his income to purchase more leisure), then less time will be available for working, and the supply curve of labor for the individual may be backward bending.45 The normal expectation would be for some of the income to be used to purchase leisure and most to be used in purchasing other goods (and thus reducing the marginal utility of additional increments of income) and for less time to be worked on balance as wages and income rise. While this framework is overly simple, its implications have led economists to label as rational the behavior signified by a backward-bending 13 Winkelman, Traditwnal Fanner, p. 8. 44 Viewed from another perspective, one might ask why, if supply of labor is dependent only upon costs, one level of price brings forth (if the curve slopes backwards) two levels of supply. It instead we consider the backward bend to have been produced by shifts in the supply function over time (assuming the demand function to have been steady or rising), then the bend, to fit historical descriptions (in which a raise in wages elicited a decrease in laboring) would have to have been caused by a shift uptard in costs over time (rather than the downward trend historically observed and noted by Miracle and Fetter). One must question whether such shifts would have been probable in the short run or whether the above explanation sounds more reasonable. 45 It is, of course, possible that, because leisure has now become more expensive than other goods purchased (the wage that could be earned having risen), individuals will switch from leisure to other goods to such an extent as to lead to less leisure being purchased at bigher incomes. See below. This content downloaded from 157.55.39.112 on Wed, 07 Sep 2016 05:26:55 UTC All use subject to http://about.jstor.org/terms
Referência(s)