刑事政策與犯罪研究論文集(16)-「影響賄選行賄罪量刑之因素」

本文採內容分析法進行，排除部分賄選類型後，以 2008 年 1 月 1 日至 2012年 12 月 31 日近 5 年臺灣各地一審賄選案行賄者有罪之判決案件共 157 件為樣本。分析結果發現，在主刑及從刑方面，有期徒刑宣告刑遠低於法定刑、併科罰金比例偏低及緩刑比例偏高。至於顯著影響量刑結果之因素，包括：行賄人數、查扣金額、法定因素中《刑法》第 57 條酌量刑度、59 條其情可憫及《選罷法》第 99 條自首自白減刑、非法定因素中被告身分等。參選層級、政黨、被告性別及當選與否等因素不會影響法官量刑。本文提出四點建議如下：（一）訂定明確之賄選行賄罪緩刑標準、（二）量刑理由應明確記載、（三）扮演角色、受賄人數、被告身分、查扣金額、每票行賄價值可納入賄選行賄罪量刑考量因素。

This study employs content analysis. After excluding certain types of vote-buying cases, a total of 157 guilty judgment samples of bribe-givers in first-instance vote-buying cases across Taiwan from January 1, 2008 to December 31, 2012 (a nearly 5-year period) were selected as the research sample.

The analysis results show that, regarding principal and supplementary punishments, the announced sentences of fixed-term imprisonment are far lower than the statutory penalties, the proportion of concurrent fines is relatively low, and the proportion of suspended sentences is relatively high.

As for the factors that significantly affect sentencing outcomes, they include the number of bribe-givers, the seized amount, statutory factors such as Article 57 of the Criminal Code (discretionary sentencing considerations), Article 59 of the Criminal Code (compassionate mitigating circumstances), and Article 99 of the Election and Recall Act (sentence reduction for surrender and voluntary confession), as well as non-statutory factors such as the defendant's status.

Factors such as the level of electoral participation, political party affiliation, the defendant's gender, and whether the defendant was elected do not significantly affect judges' sentencing decisions.

This study puts forward four recommendations as follows: 1. Formulate clear standards for suspended sentences for the crime of vote-buying bribery; 2. Clearly record the reasons for sentencing; 3. Include factors such as the role played, the number of bribe recipients, the defendant's status, the seized amount, and the value of each bribed vote as sentencing considerations for vote-buying bribery offenses.