隨著人類的經濟活動發展造成環境的污染，近年來，人們開始注重環境的保護，並試著在經濟發展和環境保護之間取得平衡，本文便以此想法結合地區之間的租稅競爭，進而研究地方政府之間在環境政策的競爭是否會造成環境品質的向下沉淪 (race to the bottom)。本文中的地方政府利用對製造污染的廠商課徵資本稅(本文以從量稅和從價稅為例，但資本稅稅率為外生條件)以做為公共財的財源，同時管制廠商所排放的污染量。 本文主要得到了三個結論：(1) 地方政府之間競爭的結果將會造成環境政策的向下沉淪；(2) 在單位稅額相同的情況下，地方政府選擇以從量稅或從價稅課徵資本稅，並不影響環境政策向下沉淪的程度，以及 (3) 影響環境政策向下沉淪的因素主要取決於該地方政府面臨的競爭對手數量，競爭的轄區愈多，環境政策向下沉淪的程度愈甚。 / As the development of economy, the environment becomes much polluted. In recent years, environment protection has become an important issue, and people have tried to strike a balance between the development of economy and environment protection. In this thesis, we combine this idea and tax competition among jurisdictions, and discuss the issue that whether the tax competition among jurisdictions leads to “the race of bottom.” In this environmental quality, the local governments levy tax on the capital of polluting industries, with either unit tax or ad valorem tax and with fixed tax rate, and use the tax revenues to finance public goods. Each government also imposes a cap on firms’ emissions. I obtain three main results: (1) The competition between local governments will lead to the race to the bottom in environment policy. (2) Given the same revenue per unit of tax, both the unit tax and the ad valorem tax lead to the same environmental policy. (3) The level of race to the bottom in environment policy depends on the number of the competitors to the local government. As the number of jurisdictions increases, the more the level of race to the bottom in the environment policy. Key words: environment policy, capital tax, tax competition

本文以Tiebout（1956）及Zodrow與Mieszkowski（1986）為基礎，利用地方政府間之租稅競爭行為，結合租稅負擔與支出性質，探討地方公共財最適提供條件與資本稅歸宿條件。 　租稅競爭行為通常只以課徵資本稅作為分析對象，但資本稅因存在扭曲性質，故政府提供之公共財數量必然低於最適水準。而本文引入政府提供不同類型公共財搭配課徵資本稅之觀點，利用要素彈性解釋地方公共財最適提供條件。本文發現在基本模型（僅考慮資本要素）中，消費性公共財提供數量會低於最適水準，生產性公共財則可能為提供不足或過度提供，主要原因取決於生產性公共財對生產動力之影響是否足以彌補資本稅帶來之扭曲性，而在延伸模型（同時考慮資本及勞動要素）中，消費性及生產性公共財雖仍提供不足，但消費性公共財較基本模型接近最適數量，而生產性公共財則較遠離最適數量。 　　由於廠商及居民在租稅制度之機制下皆有可能承擔稅負，新觀點之歸宿理論認為課徵單一資本稅因不存在替代效果，故稅負全部由資本所有者承擔，而本文則以政府提供不同類型公共財之觀點，分析資本稅競爭環境之最後稅負承擔者，本文發現地方政府間租稅競爭行為會使資本稅發生轉嫁現象，且稅負由資本所有者及使用者共同承擔。 當政府可透過支出政策，藉由提供吸引消費者遷入之消費性公共財或增加廠商生產動力之生產性公共財時，本文認為因公共支出成本係由使用之居民或廠商負擔，故資本稅具有使用者付費之受益性質。另本文發現在均衡時，於同轄區內定居之消費者及投資之廠商對政府公共支出品質與稅負具有相同偏好，此係政府、廠商與消費者運用自我選擇機制而使彼此間行為產生相互配合之現象，因而可滿足Tiebout模型強調之區分機制。

近期的租稅文獻認為影響一國公司所得稅稅率政策制定的國際因素最主要為國際間的租稅競爭，政治經濟文獻則強調政治才是最後決定租稅的關鍵因素；然而過去的實證文獻幾乎並未針對兩者綜合討論。因此，不同於過去的租稅及政治經濟文獻，本文除探討來自鄰國的稅率競爭及經濟因素對一國之法定公司所得稅稅率的影響外，同時亦分析法定公司稅率是否受到政治運作如政府體制與貪腐等因素之干擾。本文的樣本跨越全球五大洲，涵蓋不同開發程度之50個國家於1985-2007年間的追蹤資料，採用固定與隨機效果模型來估計，結果得到法定公司稅率同時受到國際租稅競爭與政治因素的顯著影響，但國際租稅競爭所給予的影響力較強；而且政府品質越差、政體為議會制的國家之法定公司稅率會較高。另外，開發程度越高的國家，對於稅率壓力的反應越敏感。

本文以 Oates and Schwab (1988) 及 Lockwood (2004) 兩篇文章為基礎。採用 Oates and Schwab (1988) 中各地方政府同時選擇資本稅率及環境標準的模型為主要架構，並結合 Lockwood (2004) 中從量及從價資本稅之相關設定，將公共財及環境品質納入代表性個人的效用函數中，求得對稱納許均衡解 (symmetric Nash equilibrium) 。我們發現在考慮地區間策略性行為 (strategic behavior) 時，不論地區採用從量或從價資本稅以融通地區性公共財，均衡稅率及環境品質皆會低於最適水準，其中又以從價資本稅的扭曲更為嚴重。而中央集權時，無論採用從量或從價稅制，均衡稅率及環境品質皆為最適水準；完全地方分權時，從量稅和從價稅兩者等價，各區間彼此競爭，不論在公共財或環境水準， 都有向下沉淪 (race to the bottom) ，偏離最適解的結果。 / The thesis is based on the setting of Oates and Schwab (1988) and Lockwood (2004). I consider that each local government has two policy instruments available: the capital tax and the environmental standard. Local governments levy tax on the capital of polluting industries, with either unit taxes or ad valorem taxes.All capital tax revenues are used to provide local public goods. The aim of the thesis is to examine whether the two distinct capital tax systems lead to different outcomes. I find that when strategic behavior among jurisdictions is considered, both the tax rates and the environmental standards are set below the optimal levels under the two tax systems. Compared to the unit tax case, it's even more inefficient with the ad valorem tax. Another finding is that the government would set both tax rate and environmental standard at the optimal level under the centralized case. Contrarily, under the decentralized cases, the competition among jurisdictions will lead to a race to the bottom in both local public goods provision and environment policy. Unit taxes and ad valorem taxes are equivalent in these two special cases.

傳統探討利潤移轉效果的租稅競爭文獻中，忽略廠商租稅規避效果對公共財的影響。然而，在貿易自由化與國際化興盛的時代下，廠商對於國與國之間的租稅安排是複雜且不可避免，所以在探究租稅競爭時，租稅規避也是一項重要的考量因素。因此本文以Lai (2006) 一文為基礎，建構一個具有利潤移轉效果的國際租稅競爭模型，分別為Cournot數量競爭與Bertrand價格競爭，來探討廠商租稅規避效果對國際租稅競爭導致公共財提供量不足的影響問題。 本文章有幾點發現：Cournot 數量競爭下，避稅效果會強化稅基、稅收效果，和利潤移轉效果，加劇公共財提供數量不足的問題。而Bertrand價格競爭下，避稅效果同樣會使公共財提供量下降，但此模型下的利潤移轉效果對公共財影響為正向，使得公共財數量皆大於Cournot模型之下。此外，本文也探討眾多參數對政府最適決策的影響。 / The traditional literatures of tax competition, which include the rent shifting effect, ignore the impact of tax avoidance effect on public goods. However, the taxation arrangements between countries which firms face are complex and inevitable in this era of trade liberalization and internationalization. Therefore, when investigating tax competition, we should consider an important factor- tax avoidance. So, this paper is based on Lai (2006) to build an international tax competition model containing rent shifting effect. We will discuss the problem of insufficient public goods and how tax avoidance effect influence this question in Cournot competition and Bertrand competition. This conclusion shows that tax avoidance effect enhances tax-base effect, revenue effect and rent shifting effect in Cournot competition. In this condition, the issue of insufficient public goods will worsen. In Bertrand competition, tax avoidance effect will also make the quantity of public goods fall. But rent shifting effect in this model is so good for public goods that the quantity of public goods is larger than Cournot competition. Furthermore, this paper also uses software to test many parameters and finds the parameters how to influence government's optimal decision.

臺灣現有移轉計價之規範未有明確的罰則(Penalty)，因而衍生許多稅負規避的問題。本文採用動態最適化(Dynamic Optimization)的模型來觀察跨國企業移轉計價的行為，在面臨懲罰與兩國稅差時企業會如何利用移轉價格及數量來進行獲利移轉以規避稅負，進而分析政府調降稅率以降低稅差並吸引獲利移轉的稅率政策對企業移轉計價的影響，最後再探討罰則在法規制定上的必要性。結果顯示，預料到的稅率政策在長期能有效減少企業從事移轉價格操弄(Transfer Price Manipulation)，但在短期﹝除了宣告那一刻之外﹞反而更助長移轉價格操弄的發生，特別是當政策宣告至執行之期間過長時更為嚴重。此外，由先前的文獻可知無罰則下的最適移轉價格為一邊界解(Boundary Solution)，本文也證明了此邊界解亦可能出現於有罰則的情況下。然而，罰則的存在創造了內部解(Interior Solution)的可能性，此內部解較邊界解更趨近於常規交易價格，因此我們仍建議政府制定罰則。 / This paper employs a dynamic optimization model to determine the equilibrium price and quantity in a multinational firm (MNF) faced with a threat of a penalty. We analyze the impact on transfer pricing that arises from the unanticipated and anticipated permanent taxation policy of home country and host country. Anticipated taxation policy for reducing tax differentials can reduce transfer price manipulation in the long term. However, except for the moment of announcement, such reduction of transfer price manipulation does not occur in the short term, especially in the case of a large time lag of policy. We also show that the boundary solution is possible even though transfer price penalty exists and suggest that governments impose penalty which creates the possibility of interior solution.