影響我國府際間補助分配之因素探討

Determinants of the Allocation of Intergovernmental Grants in Taiwan

陳曉鈴(Hsiao-Ling Chen)

指導教授 ： 何志欽 ； 李顯峰

國立臺灣大學/社會科學院/經濟學研究所/碩士(2007年)

https://doi.org/10.6342/NTU.2007.02088

摘要

我國地方財政困窘問題由來已久，而在民眾普遍存有「財政幻覺」，對公共財需求過多，與政治人物過度承諾下，地方政府支出持續擴張。中央政府雖自民國87年進行簡化政府層級的精省工程後，又擴大中央統籌分配稅款額度，並建立該分配稅款與補助款公式化的分配原則等改善地方財政措施，然其收支差絀卻日益擴增，而地方政府卻致力爭取中央更多補助，而非努力開闢自有財源。
本研究探討影響上級政府對地方補助分配的決定因素，期釐清我國補助制度運作係基於公平與效率考量達成財政政策目標，抑或受制於選票、民代勢力及遊說活動等政治力牽制？藉由對台灣地區21縣（市）於民國82年至93年間的Panel Data進行迴歸分析，經檢定後採用OLS Without Group Dummy Variables模型分析，其結果歸納如下：
（一）財政需求與能力方面：縣（市）的實質收支比率、平均每人可支配所得及民國88年財劃法修正後都對平均每人獲得補助有負向影響；老年人口比例、前期平均每人補助及民國90年當年度的影響則為正向。
（二）外溢效果方面：犯罪率對平均每人獲得補助有正向影響；都市化程度及縣（市）立學校學生比例的影響則不顯著。
（三）遊說成本及其他政治因素方面：反映遊說成本的各縣（市）至首都交通時間與平均每人獲得補助為負向關係；至於縣（市）長黨籍、立委席次及是否為縣市長選舉年度的影響則不顯著。
上述結果顯示補助分配對均等化的重視，而外溢效果方面可能因本研究取得的資料係一般與計畫型補助總數，致該因素的影響降低；另遊說成本以外的其他政治面因素影響也都不顯著，可能因補助制度法制化、未區分一般補助與計畫型補助，以及本研究納入多個財政面與外溢效果的變數致降低其影響力。但欲瞭解補助制度，政治因素仍不容忽視。

關鍵字

財政幻覺 ； 補助款 ； 財政需求 ； 外溢效果 ； 遊說成本

並列摘要

Many counties and cities have met with financial shortage for a long time in Taiwan. The general public’s fiscal illusion, the excess dependency for public goods and the politicians’ excess promises to voters have all contributed to increased local spending. To remedy the issue, the central government began implementing several measures since 1998 that included the simplification of the government bureaucracy, expansion of central fiscal planning process that weighted in local tax revenues and fair allocation of resources according to pre-set formulas. Despite all effort, local governments continue to operate in unprecedented deficits. They incessantly look to the central government for relief, not to mention becoming self reliant.
This analysis will closely examine the substantial determinants that affect the allocation of grants to the local governments to address the questions of equity and efficiency. Or, is the process impeded by election votes, congressional influences and the lobbying efforts of local government? The samples are taken from 21 counties/cities between 1993 and 2004 by means of panel data. The empirical results of OLS without Group Dummy Variables regression model are as follow：
1. Fiscal Need and Capability
Per capita grant is negatively affected by local government’s revenue/spending ratio, per-capita income and the 1999 fiscal policy, but positively affected by elderly population ratio, previous per capita grant and 2001 fiscal policy.
2. Spillover Effect
Crime rate has positive impact on the per capita grant, whereas the impact is insignificant in urbanization and the percentage of county’s/city’s students.
3. The lobbying costs of local government and other political variables
Commute time increases lobbying costs, and thus reduces the per capita grant. The partisanship of magistrate/mayor, the number of legislators and whether or not it was in the election year did not have a significant effect on per capita grant.
In conclusion, the per capita grant is helpful in equalizing horizontal fiscal capability among counties and cities. Spillover effects and other political variables, excluding the local lobbying costs, are not sufficient to explain the design of these transfers. However, political factors should not be ignored in the process.

並列關鍵字

fiscal illusion ； grants ； fiscal need ； spillover effect ； lobbing costs

參考文獻

江豐富、董安琪、劉克智（2004），「立法院老人議題的質詢趨勢與模式」，《臺灣經濟預測與政策》，中央研究院經濟研究所，34:2，1-29。

羅清俊、萬榮水（2000），「選舉與補助款的分配：綁樁？還是平衡地方財政？《選舉研究》，6:2，121-161。

謝文盛、歐俊男（2003），「台灣地區政黨政治對地方政府補助收入影響之研究」，《問題與研究》，42:6，97-111。

Aderson, G. M., and Tollison, R. D. (1991), “Congressional Influence and Patterns of New Deal Spending, 1933-1939, ” Journal of Law and Economics, 34, 161-175.

Ahmad, E and R. Thomas (1997), “Type of Transfer-A General Formulation,” Financing Decentralized Expenditures: An International Comparison of Grants, Brookfield, Vermont: Edward Elgar.