男性婚姻溢酬之研究: 傾向分數配對法之應用 / Marital wage premium for male: an application of propensity score matching

在勞動市場裡可以觀察到,每個人的工資都不相同。在過去的研究裡,學者嘗試以各種人力資本變數來解釋工資率的差異。其中,在男性的工資方程式中,婚姻狀態特別受到關注,且在過去研究皆發現男性存在著婚姻溢酬。文獻上解釋婚姻溢酬的來源,主要分成三個假說,分別是家庭分工假說、選擇性假說以及雇主的偏好。
本文和過去研究不同的地方在於消除選擇性偏誤的方法。我利用傾向分數配對法使婚姻狀態變成一個隨機決定的過程。配對完,並將資料整理為追蹤資料的型態後,再估計婚姻溢酬。最後結果顯示,在利用配對方法把單一年度資料轉換為兩期的追蹤資料後,婚姻溢酬仍然相當顯著,因此本文不支持選擇性假說。但是因為追蹤資料涵蓋期間過短,因此我無法利用現有的資料區分出家庭分工和雇主的偏好這兩種效果。 / The study of determinants of individual wages is important for exploring why some individuals earn more than other individuals. In the past, numerous efforts have been devoted to explain wage differentials by various human capital variables. Marital status is also used as a control and is of interest in estimating the male’s wage regression. It is commonly acknowledged that in cross-sectional analyses, married men have higher wages than single men, other things being equal. The higher wages paid to married men compared with their single counterparts is defined as marital wage premium. There are three leading hypotheses: family role specialization, selection hypothesis and employer favoritism.
This paper uses a new method called propensity score matching to estimate the male marital wage premium. Propensity score matching is used to create a quasi-experiment that makes the decision to get married become a random treatment. The principle is to match a married man to a comparable single man who is closest in terms of important characteristics. And then, I transform the matched sample into panel data where units are observed at two periods, having the unmarried man as a proxy of married man, as if he was single. Therefore, I can apply fixed effect and random effect models to the transformed data.
The results show no evidence to support the selection hypothesis since after transforming the data into panel data set and applying the fixed effect model, the marital wage premium is still significant. But being limited by the short period of the quasi-panel data set, I cannot distinguish between family role specialization and employer preference.

Identiferoai:union.ndltd.org:CHENGCHI/G0099258007
Creators林瑋
Publisher國立政治大學
Source SetsNational Chengchi University Libraries
Language英文
Detected LanguageEnglish
Typetext
RightsCopyright © nccu library on behalf of the copyright holders

Page generated in 0.002 seconds