自第三次经济全球化浪潮以来，全球关税税率呈现出长期普遍下降的趋势，区域贸易协定的签署数量也急剧上升。一国对外贸易的开放程度根植于其国内政治经济因素，在经典的关税内生化理论框架下，学者们刻画行政部门、选民以及利益集团之间的偏好和行为，重点关注利益集团之间的游说竞争，最终在均衡分析中得到政策结果。已有文献在刻画利益集团如何影响贸易政策的同时，过度简化了行政部门的相关设定，但行政部门恰恰是无法回避政策制定主体，其内部不同主体的竞争程度也会通过影响行政部门偏好，最终对政策结果产生重大影响。本文的分析包括理论分析和实证分析两部分。理论分析部分，构建了带有政治竞争因子的共同代理人模型，从形式模型的角度刻画了国内政治竞争如何影响对外贸易政策结果。实证分析部分，基于自主搭建的1988-2019年共125国的政治经济数据库，使用非平衡面板数据建立固定效应模型，验证国内政治竞争对关税税率水平、区域贸易协定签署数量的影响：在国内政治竞争与关税税率方面，本研究考虑了国内政治竞争与关税税率之间的反向因果问题，选取工具变量和使用两阶段最小二乘法（2SLS）回归克服其内生性；在国内政治竞争与贸易协定方面，本研究提出了有关利益集团俘获的竞争性解释。此外，实证研究还开展了一系列异质性分析和稳健性检验。本文理论分析表明，一国国内政治竞争程度越高，行政部门更加重视社会总体福利且更加忽视来自利益集团的补偿性捐献，使得均衡关税税率越低、签订的贸易协定数量越少，本文的实证结果与理论分析相一致。在解释现实中贸易协定的大幅扩散时，本文认为，给定国内政治竞争激烈，但利益集团的独占度较高时，行政部门仍可以被利益集团俘获并签署贸易协定，该竞争性解释通过了经验数据的检验。基于以上结论，本研究提出了针对性的政策建议，在进行双边以及区域贸易关系协调处理时，需要在传统的多边协调机制之下，更加注重各经济体的国内政治竞争、利益集团间竞争。

Since the third wave of economic globalization, global tariff rates have shown a long-term general downward trend and the number of Regional Trade Agreements (RTAs) signed has risen sharply. The degree of a country‘s foreign trade openness is deeply rooted in its domestic political and economic factors. Under the classic Tariff Endogenous Theory framework, scholars have portrayed the preferences and behaviors among the government, voters, and Special Interest Groups (SIGs), focusing on the lobbying competition among SIGs, and ultimately obtaining policy outcomes in equilibrium analysis. However, existing literature over-simplifies the setting related to the government when characterizing how SIGs influence trade policy. The government is the unavoidable policy-making subject, and the degree of competition among its different components can also have a significant impact on the policy outcome eventually by influencing the government’s preferences.The analysis of this paper consists of two parts: theoretical analysis and empirical analysis. For the theoretical analysis, a Common Agent Model with a political competition factor is constructed to describe how domestic political competition affects foreign trade policy outcomes from the perspective of a formal model. For the empirical analysis, based on an independently constructed political economy database for a total of 125 countries from 1988-2019, a fixed effects model is built using unbalanced panel data to verify the effects of domestic political competition on the level of tariff rates and the number of RTAs signed: in terms of domestic political competition and tariff rates, this study addresses the issue of reverse causality between domestic political competition and tariff rates by selecting instrumental variables and using Two-stage Least Squares (2SLS) method to overcome the endogeneity; in the case of domestic political competition and RTAs, this study proposes a competitive explanation regarding interest group capture. In addition, a series of heterogeneity analysis and robustness tests are conducted in the empirical study.The theoretical analysis of this paper suggests that the higher the level of domestic political competition in a country is, the more the government values the general welfare of society and ignores compensating contributions from SIGs, resulting in lower equilibrium tariff rates and fewer RTAs. This conclusion is consistent with the empirical results of this paper. In explaining the surge of RTAs in reality, this paper argues that given a high level of domestic political competition but a high degree of dominance by the privileged SIGs, the government can still be captured by SIGs and sign the agreements. This competitive explanation passes the test of empirical data.Based on the above findings, this study proposes policy recommendations that each economy should pay more attention to its domestic political competition and competition among SIGs under the traditional multilateral coordination mechanism when coordinating bilateral and regional trade relations.