SEASONAL ADJUSTMENT IN A MARKET FOR FEMALE AGRICULTURAL WORKERS

Traditionally, the analysis of labor market adjustment refers to the interaction between the demand and the supply for labor, e.g., between employers and job seekers. The labor market is said to 'clear' when the wage and labor force participation adjust so that supply and demand are equal. However, it has been observed that the adjustment process in the agricultural casual labor market is often uneven and incomplete. This phenomenon has challenged economists to search for additional explanation of the workings of these markets (see for example Rosenzweig, 1986; Binswanger and Rosenzweig, 1981). Our study contributes to the literature by examining this issue using a Chilean data set collected by one of the authors in which pronounced seasonality is evident in both labor demand and labor supply, resulting in large changes in wages, participation and unemployment. The principal objective of this paper is to analyze the socio-economic and demographic factors that determine seasonal labor force participation, as well as the events that appear to be associated with labor market entry/exit of seasonal workers in order to enable us to understand the welfare implications of changes in economic behavior. We will analyze these phenomena in considerable detail and with additional econometric techniques attempting to shed additional light on seasonal adjustment in agriculture. A major strength of our study resides in the data set available. From January to March 1992, data were collected from 599 workers in three growing regions of Chile, who were then working in table grape processing sheds. For each worker, information was collected on measured productivity in a piece rate task in 1992, personal and family characteristics (e.g., age, sex, education, work experience, marital status, family composition, and family income), as well as on labor force participation, employment, type of work, incentive mechanism (wage or piece rate) and earnings for every day of calendar year 1991. Many of the workers surveyed held more than one job from more than one employer during the year. This rich longitudinal data set, including male and female workers, offers a new dimension to existing research on seasonal labor market analysis. We have data on a large number of factors that determine participation, employment, and earnings, and having a large number of observations throughout the entire year allows improved treatment of seasonal dynamics. We use econometric models for panel data to analyze the labor participation of individual seasonal workers, allowing for unobserved heterogeneity and consequently endogeneity/selection bias (Vella and Verbeek, 1999). In the process, we determine the expected wage of each worker for each day, based on that worker's human capital characteristics, and include the expected wage as a determinant of labor force participation. Preliminary results indicate that real wages fluctuate considerably over the year and that seasonal wage variation is an important aspect of labor market adjustment, contributing to a large change in labor force participation. The labor force participation rate of women is significantly more elastic to changes in their expected wage than is the labor force participation rate for men, though much of the difference in response appears to be due to the household roles that males and females play rather than gender differences per se. The labor force participation behavior of women is similar to that of men when the former are the heads of household. Although the labor force participation rates for seasonal workers vary greatly across seasons, we still find evidence of substantial open unemployment during the slack season. The analysis will also provide estimates of the premium earned per day for piece rate as opposed to wage work, and of differences in pay for male and female workers. The study will contribute longitudinal findings that will enhance our understanding of the dynamics of casual employment in agriculture