Previous research has shown that the obese are more likely to be unemployed than the non-obese. This tendency is strongly evident for females. Being obese has a significant effect on marriage also, especially for women. Several studies have found that marriage is less likely for heavier women.
The aim of this paper is to assess the prevalence of overweight among Japanese and to examine the effects of body weight on employment and marriage. First, we analyze the effects of body weight on the choice of regular work, non-regular work, self employment job and no work. Second, we investigate the impact the body weight has on employment and marriage, taking account of a speculation that employment and marriage are linked together and determined in mutually dependent way.
Data/Methods
The data used in this study is taken from Keio Household Panel Survey (hereafter KHPS). The first wave of KHPS was conducted in January 2004 and since then data has been collected annually. People on whom the survey is targetted are 4,005 of males and females who were 20-69 years old in January 31, 2004, selected by two-stage random sampling. We use the second wave of KHPS in 2005.
The multinominal logit model by sex is used for the first analysis mentioned above. Dependent variable takes 0 to 3 for the 3 types of employment. Independent variables include BMI, age, education, drinking and smoking, etc. For the second analysis above, it is appropriate to specify a bivariate probit model in which decisions of employment and marriage are simultaneously made. Here, we use two dependent variables. One of them takes 1 if married and 0 if not married. The other is divided into 4 cases: ‡@takes 1 if regular-worker and 0 if not regular worker, ‡Atakes 1 if non-regular worker and 0 if not non-regular worker, ‡Btakes 1 if self-employed and 0 if not self-employed, and ‡Ctakes 1 if no-work and 0 if not no-work. The independent variables are the same as the first analysis in the logit model. We estimate this bivariate probit model by sex.
Results/Expected Results
This study provides some evidence on the marriage and employment for the overweight: (1) the probability for females to become self-employed increases as the body weight increases. (2)Males tend to become non-regular workers as their body weight increases. (3) For females, the increase of BMI by 10 raises the probability of becoming self-employed and remaining single by 3 %. (4) For males, when BMI increases by 10, the probability of getting married and at the same time working as a regular worker will decrease by 15 %, the probability of remaining single and working as a regular worker will increase by 11 %, and the probability of becoming single and working as a non-regular worker will increase by 4%.