The labor supply of Thailand : an empirical analysis of the determinants of participation rates

Abstract

This study analyzes the variation in labor force participation rates of several age-sex groups of the Thai population in 1970 in terms of an econometric model incorporating economic as well as demographic variables. The study is a cross-sectional one based mainly on census data for 71 provinces of Thailand. The econometric model incorporates eclectic theories of labor supply with human capital theory as well as economic/demographic theory. It consists of three structural equations and an identity equation reflecting the joint dependency among activity rates, wage rates, and family income. In the participation rate equation, age-sex specific activity rates are functions of wage rates, family income, unemployment rates, industrialization, education, population-density, and family size. Parameters of the model are estimated by two-stage least squares technique. The results of the statistical estimation are generally consistent with the underlying theoretical concepts. The participation rates of prime-age males were found less sensitive to socioeconomic variations compared to those of other groups of the population. The effect of the wage variable is positive and significant for females of every age group, but it was negative and not significant for males of every age group except the 20-29 age group (where it was negative and significant). The coefficients of the family income variable are negative and significant for females of all ages and young men (aged 11-19). The income effect is positive and significant for males aged 40 and above. Unemployment rates were found to have negative effects upon most workers except for young women aged 11 to 14 whose coefficient is insignificant statistically. In terms of elasticity participation rates are generally less sensitive to the unemployment rates than to family income and wage rates. Education has a relatively large positive effect on participation rates of the young of both sexes while it has a large negative effect on participation rates of women 40 to 59 years of age. A milder positive effect of education occurs in females aged 20-39. The coefficient of industrialization which captures the effects of difference in economic structure, rural-urban residence, and occupational composition among areas is positive for secondary workers. Family size is positively associated with participation rates of the young aged 11-19 and women aged 40 to 60 and above. Final1y population density was found negatively related to participation rates of secondary workers and positively related to participation rates of primary workers. In conclusion, this study can serve as a guide to the effects of various variables on different measures of labor supply. Provided the parameters of the model remained constant, a planner could use the model to test the effects of various manpower policies of labor supply if he could deduce how a policy change would affect the variables in the model.