近年來整體生育率持續下跌，少子化問題已成為社會關注的焦點，也是諸多社會安全相關政策改革時的重要變數。針對少子化問題，直至2012 年止，政府尚無提出完整的應對及配套方案。在中央政府方面，主要的作法有2011 年啟動之「5 歲幼兒免學費教育計畫」以及2012 年起適用之「幼兒學前特別扣除額」。各地方政府雖然陸續提出相對較多元的生育政策，主要的作法是以一次性之「生育津貼」現金補助，期能刺激民眾生育意願、提升各地生育率。然而，生育津貼是否真能提高民眾的生育意願，達成提升生育率的政策目標，並沒有相關研究成果可供參考。本研究根據Panel 迴歸模型的計量方法，探討1998-2010 年間各縣市政府生育津貼的發放與育齡婦女總生育率之關聯性。實證估計上最大的問題在於處理縮減式（reducedform）迴歸估計所面臨的內生性問題。由於縣市政府會因其財源決定是否發放生育津貼，但縣市財政狀況一般而言與生育率無關。故本研究選取歲入作為工具變數，來處理以生育津貼發放作為解釋變數所存在的內生性問題。研究結果發現，生育津貼與生育率之間雖是正向關係，但是效果並不大。對於本來就想生育的育齡婦女來說，地方政府每發放生育津貼新台幣1,000 元，將有提前懷孕的刺激效果，但也僅增加整體生育率0.0057。此外，其他國家的經驗也鮮少僅靠生育津貼就能提升生育率之案例；因此，臺灣若要有效解決少子化問題，各級政府應積極思考更完整的應對方案。

The decreasing total fertility rate has been a major public concern in Taiwan. Except the “5-year-old Children Education Subsidy” (starting from 2011) and the “Special Deduction for Preschool Children” (starting from 2012), the central government has failed to provide a complete and effective measure to boost the dropping birth numbers. On the other hand, local governments, however, promote child births with policies of a wider variety. Among them, the most popular and welcome policy is to distribute cash maternity allowance to child births in the jurisdictional area. Nonetheless, the effectiveness of using one-time-allowance to promote birth rate has never been seriously studied. With data from 23 jurisdictional area of Taiwan in the period from 1998 to 2010, panel regressions are used to investigate the relationship between total fertility rate and the distribution of cash maternity allowance by local governments.In order to resolve the potential endogeneity problem in the regressions, local government revenue is used as an instrument. The regression results suggest that there is positive correlation between total fertility rate and the amount of cash maternity allowance, but the effect is small. For women who want to have children in the first place, each NT$1,000 cash allowance raises total fertility rate by 0.0057.