(Japan Society of Family Sociology, 7th annual meeting at Tokyo, 1997.7.24.)
- TANAKA Sigeto (Faculty of Human Sciences, Osaka University)
Women's employment has been rising these years. To evaluate such movement, we face a question. How does related changes in women's employment status to changes in the sexual division of labor? To answer the question, it is important to distinguish two types of women's employment.
The first type is to be called a serious challenge to the sexual division of labor. If a woman perticipate into employment in such manner, she is no longer able to undertake all of the housework. Thus such type of women's employment cause the collapse of the sexual division of labor.
In the second type, women's employment is only secondary to housework. A woman engages paid work only if her housework does not hinder it. This type of women's employment is more friendry to the sexual division of labor.
From the empirical viewpoint, we consider that these two types of women's employment correspond to full-time (and long-term) employment and part-time (or short-term) one. Though this correspondence is incomplete in practice, for which we will look out later, we accept this simple correspondence for a while.
Over the changes in these two types of women's employment, two schools, Marxist-feminist theory (cf. Ueno ) and rationalistic economics (cf. Higuchi ), have argued with each other. In this paper, we will specify their argument with a mathematical model. Then we will test the two theories with empirical evidences.
Her employment behavior depends on the relations among x and two thresholds P, F (P<F):
In regards to the employment behavior of all women in a society, consider that x is distributed according to a probability density function Z(x). Assume that Z(x)>0 for all x.
- x < P
- ... not employed
- P <= x < F
- ... part-time or short-term employment
- F <= x
- ... full-time and long-term employment .
Assume that all women in the society have the same value of the two thresholds P and F.
Both Rationalistic Economints and Marxist-Feminists concur in the view that the insentive pulling out women into employment has been increasing. Consider that the distribution of x is according to Z0(x), Z1(x), at two moment 0, 1. The increace in the insentive pulling women can be translated as follows:
Z1(x) = Z0( x - k ) . (1)Equation (1) means that the distribution of x has shifted right by a constant k (>0).
The question where the two schools are antagonistic is about changes in the two threshold P and F with the shift of Z(x) stated in equation (1). Let P0, P1, F0, and F1 denote the values of P and F at the two moment 0,1. Then we go through our argument.
P1 = P0, F1 = F0 . (2)Thus, whenever Z(x) shifts right, full-time and long-term employment of women must increase.
P1 = P0 , F1 = F0 + k . (3)Thus, as F shifts right as rapid as Z(x), the decreace in the number of women who do not employed all pours into part-time or short-term employment.
Tanaka [3: 157] comfirmed the followings, with the personal history data from the 1985 SSM (Social Stratification and Social Mobility) Survey :
Table 1. shows the results of an analysis of 1995 SSM data (A division, women only), with the same method as Tanaka . This table indicates the following:
Those indexes might be disturbed by the changes in the system of support for child care. If support system is enough developed to lessen the burden of child care, even the complete sexual division of labor does not hinder women from continuous full-time employment. Now we make attempt to elaborate such disturbance.
Table 1. Change in women's employment status (1995 SSM Survey)------------------------------------- Age C (N) | E R (N) ------------------------------------- 60's .216 ( 88) | .094 .820 (128) 50's .230 (126) | .147 .691 (136) 40's .214 (224) | .176 .449 (136) 30's .251 (167) | 20's .128 ( 47) | ------------------------------------- Total .221 (652) | .140 .650 (400) V .072(p>.10) | .241 (p<.01) -------------------------------------
- C: Rate of continuous full-time employment in the childrearing stage [3: 154].
- E: Rate of new entry into full-time employment after the childrearing stage (only those who does not employed in the childrearing stage)
- R: rate of remaining not employed (rate of housewife) after the childrearing stage (only those who does not employed in the childrearing stage)
- V: Cramer's association coefficient (parenthesized are the results of chi-square tests of statistical independence).
First, other women in an extended family may take over the childrearing. We estimate such effect from the Japan Population Census data (Statistics Bureau, Management and Coordination Agency of Japan) [6: 254] [7: 308]. According to the result, the effect of such change on C is very small and negregible (-.007 in 20 years).
Second, the externalization of childrearing, such as baby shitting, and legal institutionalization of child-care leave should have been being developed. We conclude that the increase in the rate of full-time and long-term employment is small, as the development of such institution.
(Women in agriculture, self-employed, and in family enterprise are excluded from all analyses in this chapter.)
In the Marxist-feminist critics, this is called as the second and third compromise between capitalism and patriarchy [ Ueno : 1: 40f.] But such a expression can be only a ad hoc description. It should be our main issue to clarify the background mechanism.
(C) Tanaka Sigeto
This paper will be presented at Japan Society of Family Sociology, 7th annual meeting at Tokyo: Waseda University, 1997.7.24.
A direct correspondence with the author is welcome.
Duplication in any form without the author's permission is forbidden.
Created at 1997-06-20. Last modified at 2002-04-22.