我國上市公司營利事業所得稅負擔之研究

• Main Authors: 楊朝旭, Yang, Chaur Shiuh
• Language: 中文
• Published: 國立政治大學
• Subjects: 政治成本假說; 營利事業所得稅; 有效稅率; 縱查資料; political cost hypothesis; business income tax; effective tax rate; longitudinal data
• Online Access: http://thesis.lib.nccu.edu.tw/cgi-bin/cdrfb3/gsweb.cgi?o=dstdcdr&i=sid=%22B2002003753%22.

　　美國的租稅正義公民會社(Citizens for Tax Jusitce)在一九八五年發表一篇震撼美國政府、企業界與學術界的研究報告，文章中指出美國最大約兩百五十家大企業中，有半數以上，在一九八一年到一九八五年當中 ，至少有一年未納一毛錢的聯邦所得稅，或者甚至收到聯邦的退稅款。這篇報告引起美國財經學者與會計學者對公司租稅負擔的熱烈探討，亦對美國國會通過一九八六年租稅改革法案(the Tax Reform Act of 1986)取消投資扣抵、擴大稅基並加強最低稅負(alternative minimum tax)之稽徵有重要的影響。 　　本文之主要研究目的在於探討我國上市公司之營利事業所得稅負擔，是否受到企業規模、產業別以及稅制變化之影響。本研究使用我國教育部電子計算機中心提供之「台灣股票上市公司財務報表資料庫」，取出自民國七十三年到民國八十一年連續九年間，相同八十家上市公司所構成之縱查資料（longitudinal data或panel data），利用二元固定影響模型(two-way fixed effects model)等方法進行分析，發現上市公司之規模越大有效稅率越低（租稅負擔越低），顯示較大的上市公司可能運用較大的政治力量來減輕其租稅負擔，此結果代表用租稅來反映政治成本的政治成本假說(political cost hypothesis)不適用於我國的上市公司。在財務會計的研究上，經常以政治成本假說為基礎來解釋經理人員為何採用可降低當期盈餘的會計方法，亦即假設政治成本假說為正確，則企業規模愈大，愈容易受到政治上的監督（例如稅捐單位的嚴格審查），其政治成本愈大，為了避免這種政治成本，規模愈大約企業愈可能採用降低當期盈餘的會計方法。但是由本文的研究結果顯示，同租稅來表示政治成本的政治成本假說在我國並不成立，因此，未來有關之研究，在解釋我國公司規模對會計方法之選擇的影響時，不能直接從政治成本假說的觀點來解釋。 　　本文亦參考國外有關文獻，將我國與美國上市公司同期間(1986~1989)的租稅負擔加以比較，發現我國上市公司的有效稅率低於美國，但其主要原因為我國上市公司有虧損所引起。 　　本文另外亦使用產業的虛擬變數來分析不同產業是否有不同的租稅負擔，檢定結果顯示產業別對租稅負擔有顯著的影響。最後，本文比較民國七十五年營所稅自30%降為25%與實施加值稅前後，公司租稅負擔的變化情形，發現七十五年稅制變動後，規模愈大之上市公司須繳愈多的營利事業所得稅。 === 　　A 1985 study by Citizens for Tax Justice examined the corpor-ate income tax burden of 250 largest U.S. corporations. They found that almost half of the largest corporations paid absolutely nothing in federal income taxes, or received tax rebates, in at least one of the four years from 1981 to 1984.This thesis intends to examine the income tax burden of public companies in Taiwan. In particular, the thesis investigates whether firm size, specific industry and tax system change will affect the income tax burden of public (listed) corporations. Financial statement data of the public corporations were obtained from the Econometric Programm- ing System (EPS) data base provided by the Ministry of Education. The main sample consists of eighty corporations that appear in all nine years over the 1984-1992 period. 　　To control for individual company and time heterogeneity and reduce omitted variable bias, fixed effects models are employed in this study. Using a two-way fixed effects model, the result shows that the larger the corporation's size the lower the effective tax rate (the lower the corporation's tax burden). It indicates that larger corporations may have more political power to reduce their tax burden. The finding suggests that political cost hypothesis is not justified in Taiwan's public corporations from the standpoint of effective tax rates. 　　This thesis also analyzes whether tax burden varies across industries. The analysis shows that a corporation's tax burden is related to the industry in which the corporation operates. Finally, this study examines the changes of corporation's tax burden pre and post 1986 when the highest statutory business income tax rate was reduced from 30 percent to 25 percent and the value-added tax was adopted in the same year. The analysis indicates that after the tax system change in 1986, large corporations paid more business income taxes.