政治競爭對軍事支出的影響：一個跨國的實證研究

本研究旨在以實證途徑探討競爭性選舉這一民主制度的核心要素對軍事支出的影響。本研究建立在推選人團理論與中位選民理論的理論基礎上，假設競爭性具有降低軍事支出的效果，並透過使用161 個國家，1990 至2017 年的時間序列橫斷面資料驗證該假設。實證分析上以V-Dem 資料庫的選舉要素指數作為主要解釋變數，以線性迴歸並控制地理區域和時間固定效果作為主要的估計方法。實證分析的結果顯示競爭性選舉不僅對軍事支出具有顯著的負向影響，而且亦具有降低軍隊規模的效果。競爭性選舉對軍事支出的負向影響在經過增加控制變數、改變資料結構、替換解釋變數和使用不同估計方法，例如平行校正標準誤和廣義動差法等一系列的穩健性測試後依然獲得實證上的支持，且該影響在中等收入和低收入國家更為強烈。為了處理解釋變數和依變數之間的潛在內生性，本研究使用工具變數對模型的內生變數進行校正以確認兩者的因果關係。本研究根據探討氣候條件對政治制度長期影響的文獻，使用降雨量扣除全球平均值作為預測競爭性選舉的工具變數。二階最小平方法的估計結果顯示競爭性選舉對軍事支出的負向影響依然維持統計上的顯著性且競爭性選舉的迴歸係數增加，顯示競爭性選舉對軍事支出的負向影響可能受到模型內生性低估。本研究進一步將政治競爭區分為獲勝聯盟和中位選民兩個次要素，並比較兩者對軍事支出的影響。分析結果顯示獲勝聯盟的規模比起政治結社的自由度對於減少軍事支出具有更強的效果，暗示了推選人團理論相對於中位選民理論具有更強的解釋力。

關鍵字

V-Dem ；競爭性選舉；軍事支出；推選人團理論；中位選民理論；工具變數；二階最小平方法

並列摘要

This thesis aims to investigate the causal effect of electoral competition on military expenditure empirically. Based on the Selectorate Theory and Median Voter Theorem, we hypothesize that competitive elections have a negative effect on military expenditure. We test this hypothesis by using panel data of 161 countries from 1990 to 2017 and the Electoral Component Index from V-Dem database as the main explanatory variable. We use OLS regression and control for geographic region and time fixed effect as our main estimation strategy, empirical results show that competitive elections have a negative effect on military expenditure, either measured by military spending as the share of GDP or government expenditure. Our empirical analysis also demonstrated that competitive elections has an effect of reducing scale of army. Our empirical result is supported by a series of robustness checks, including adding further controls, changing data structure, switching explanatory variable and using other panel techniques like panel-corrected standard error or system-GMM estimator. To address the potential endogeneity between competitive elections and military expenditure, we conduct an instrumental variable approach. Based on past literature investigating the long-run effect of weather conditions on political institutions, we use precipitation per year minus the global average to instrument our main explanatory variable, and the result of two-stage least square (2SLS) estimation shows that competitive elections remain a negative effect on military expenditure at conventional level of statistical significance, and the coefficient of competitive elections increases, implying that the negative effect of competitive elections could be underestimated by endogeneity. Furthermore, we separate political competition into two subcomponents, winning coalition and freedom of association, and compare their effects on military expenditure. Our results of analysis demonstrate that the effect of coalition size is larger than that of freedom of association, suggesting the Selectorate Model has greater explanatory power than Median Voter Theorem.

並列關鍵字

V-Dem ； Competitive Elections ； Military Expenditure ； Selectorate Theory ； Median Voter Theorem ； Instrumental Variable ； Two-stage Least Square Estimator

參考文獻

Acemoglu, Daron and Johnson, Simon. 2005. “Unbundling Institutions.” Journal of Political Economy 113:949-995

Google Scholar

Acemoglu, Daron, Johnson Simon and Robinson, James. 2001. “The Colonial Origins of Comparative Development: An Empirical Investigation.” The American Economic Review 91(5): 1369-1401

Google Scholar

Acemoglu, Daron and Robinson, James. 2000. “Why did the West Extend the Franchise? Democracy, Inequality and Growth in Historical Perspective.” The Quarterly Journal of Economics 115(4):1167-1199

Google Scholar

Acemoglu, Daron., Ticchi. Davide and Vindigni, Andrea. 2010. “A Theory of Military Dictatorships.” American Economic Journal: Macroeconomics. 2(1):1-42

Google Scholar

Adser`a, Alicia., Boix, Carles and Payne, Mark. 2003. “Are You Being Served? Political Accountability and Quality of Government.” Journal of Law, Economics, and Organization. 19(2):445–90

Google Scholar

國際替代計量

政治競爭對軍事支出的影響：一個跨國的實證研究

全文下載

主題瀏覽