由於良好的人口控制與一國的經濟發展以及社會資源的分配有著極為密切的關係，而許多文獻也指出教育程度的提升有助於降低育齡婦女的生育率，以減緩過快成長的人口數，因此本文試圖以台灣為研究對象，從控制人口成長的面向去討論女性教育程度的提升對其生育行為的影響效果。本文主要利用台灣在1968年時所實施的義務教育改革做為女性教育程度的外生變異來源，並以兩階段最小平方法估計教育程度對於小孩個數的影響效果。本文的資料來源為1990年的台閩地區戶口及住宅普查資料，其中年齡介於31–39歲（不包含35歲）的已婚女性為我們主要的研究對象。本文利用受教育改革政策影響指標，即31–34歲的世代，做為女性教育程度的工具變數之一，另外再以受教育改革政策影響指標與女性年齡的交叉項做為其他的工具變數，試圖檢定教育改革政策是否造成女性在教育程度上的差異，最後再將此差異與女性的小孩個數做連結。我們發現相對於36–39歲的世代，受教育改革政策影響的31–34歲的世代，其教育程度明顯較高，且此政策效果會隨著世代年輕化而增強。另外，教育程度越高的女性，其生所生的小孩個數會越少，並進一步發現女性的教育程度與生一個小孩以上或是兩個小孩以上的情況並無顯著的負向關係，但此教育效果會顯著降低女性生三個小孩以上的機率。最後本文進一步將樣本縮小為34–37歲（不包含35歲）的女性，結果發現教育程度於小孩個數所產生的負向效果變得更大且更為顯著。The literature generally points to a strong negative relationship between female education and fertility. In view of the above, many policymakers have adopted policies designed to reduce rapid population growth. However, due to the endogeneity of schooling, the causality between female education and fertility remains open. In this paper, I use 1990 Population and Housing Census and exploit the reform in compulsory education in Taiwan to estimate the effect of female education on fertility accounting for the endogeneity of schooling. First, we study women aged between 31 and 39 in 1990, excluding aged 35, and we form a reform indicator of women between 31 and 34 to reflect an individual who was affected while the education reform was implemented. Thus we exploit the reform indicator and the interaction between the reform indicator and age as instruments for female education, and employ the two-stage least squares to estimate the effect of female education on her number of children. Our results suggest that women with additional one year of schooling would have on average 0.04–0.05 less children, and women with education attainment above junior high school have on average 0.18–0.21 less children. Moreover, our study suggests that the fertility-reducing effect of education attainment for women with more than 3 children is the most evident. We further limit our sample to women aged between 34 and 37 in 1990. Our further results indicate that the fertility-reducing effect of education attainment gets larger and much more evident. These results imply that rising level of education attainment can account for a sizable fraction of the decline in fertility rates, especially for women with education attainment above junior high school.
