本研究以2001年至2005年為研究期間，從股權結構及董事會監督功能分別探討公股角色對投資標的公司經營績效之影響。本文之研究結果顯示：(1)股權結構變數中，公股投資標的公司之經營績效較無公股投資之公司為佳。(2)公股持股比率愈高，公股投資標的公司之經營績效則愈差。(3)董事會組成變數中，由公股代表出任董監之公股投資標的公司之經營績效較未派任公股代表之公司為佳。(4)當公股代表的席次比率愈高時，公股之投資標的公司之經營績效愈差。(5)公股持股比率偏離公股代表席次比率之程度愈高，公股投資標的公司之經營績效愈差。 / This research from 2001 to 2005 as research period, aim to explore the role of government ownership in the relationship between board composition and firm performance in the enterprises with government ownership. The empirical results document that from ownership structure variables, the effect of government shareholding on firm performance is found to be positive. Furthermore, firm performance decreases with an increased size of government shareholding. From board composition variables, the effect of the board with the government shareholding represents on firm performance is found to be positive. Furthermore, firm performance is worse with an increased size of government shareholding represents in the corporate board composition. Finally, when the government shareholding and the government shareholding represents’ seat ratio deviation degree is bigger, the firm performance would be worse.

我國公股銀行民營化後，對於經營績效的影響目前未有定論，然而我國政府對於已完成民營化公股銀行更衍生出後續之政府持股管理問題。因此本研究整理相關文獻，希望可以透過資料包絡分析法，客觀評估銀行經營績效，並且研究政府持股如何影響公司經營效率。並且考慮資產品質會對銀行經營造成壓力，因此將備抵呆帳是為銀行投出的一部分，希望可以更廣泛的討論銀行經營績效，並且可以對政府持股管理問題提出可能觀點。主要研究結果如下： 一、 公股銀行平均技術效率値顯著優於民營銀行，主要來自於純粹技術效率的優勢。而資產品質主要對民營銀行產生規模效率之減損。 二、 100%公股銀行在純粹技術效率與技術效率，無論是否有考慮備抵呆帳皆顯著最佳，民營化銀行次之，民營銀行最差。然民營化銀行在無論有無考慮備抵呆帳的情形下，規模效率皆不如100%公股銀行或民營銀行。 三、 我國政府不同的持股比例對於公司經營效率之影響，可能並非單純線性關係。若政府低於33%，則持續降低政府對於經營之影響力，可能有助於公司經營效率。 四、 Malmquist生產力指數顯示100%持股公股銀行生產力因為生產技術的提升，累積後較民營化銀行為大，此與三類銀行間t-test之檢定結果相同。顯示政府持股較多的銀行可能有助於提升技術效率與純粹技術效率。

本論文主要目的係探討政府持股與企業避稅間之關係。本文以2005年至2014年台灣公開發行以上公司為樣本，分析政府持股比例、代表政府董監事席次比例對租稅規避的影響，並檢視公股主導企業之避稅程度。 實證結果顯示，政府持股比例對企業避稅無顯著影響，然而，當代表政府的董監事席次比例愈高，企業會減少避稅行為。此外，研究結果亦指出公股主導之企業相對於家族企業，其避稅程度較低。進一步分析公股主導之企業和雙權偏離互動對企業避稅之影響。實證分析發現，當股份盈餘偏離程度愈大，公股主導之企業相對於家族企業，其避稅程度較高，隱含公股主導之企業會透過避稅行為將財富移轉至企業本身；反之，當席次盈餘偏離程度愈大，公股主導之企業相對於家族企業，其避稅程度較低，表示政府掌握董事會席次時，能有效監督企業遵循稅法規定，採用較保守的租稅政策。 / The purpose of this study is to examine the association between government shareholding and tax avoidance. This study uses the data of Taiwan public companies from 2005 to 2014 to analyze whether government shareholding percentage and the ratio of directors held by the government have impact on tax avoidance, and investigate the level of tax avoidance of government-controlled companies. The empirical results reveal that government shareholding does not significantly affect corporate tax avoidance. However, when the ratio of directors held by the government increases, companies would reduce the level of tax avoidance. In addition, the results also indicate that government-controlled companies are less tax avoidance compared with family-controlled companies. This study conducts further analysis on the influence of the interaction of government-controlled companies and the deviation between control rights and cash flow rights on tax avoidance. The results show that the higher the level of the deviation between voting rights and cash flow rights, the higher the level of tax avoidance of government-controlled companies than family-controlled companies. It implies that government-controlled companies might transfer fund to themselves through tax avoidance. On the contrary, when the deviation between the ratio of controlled board seats and cash flow rights increases, the government-controlled companies engage in less tax avoidance than family-controlled companies. It represents that when the board of directors is controlled by the government, it could effectively supervise the company to comply with tax regulations, and the company would adopt conservative tax strategies.

本論文主要由三大研究主題組合而成：第一個主題是依照中國金融體系的現況，針對國有銀行、政策性銀行、股份制商業銀行及城市商業銀行等四種不同類型銀行的獲利績效進行比較實證研究。其中的股份制商銀與城市銀行，其資本組成除少數仍有國務院持股（即中央政府持股）外，股東成員擴及地方政府、國有企業及民間企業，股本結構明顯較國有商業銀行及政策銀行多元化。不同的股本結構，尤其是官股比例（政府持股比例）是否對各家銀行的獲利能力產生影響，是本文主要研究興趣所在。本文採用Bankscope資料庫，涵蓋1997至2004年中國大陸49家銀行的財務及股東資料。結果發現：銀行股權結構中，中央政府持股比例越高，銀行獲利表現越差，地方政府持股的影響則較不明顯；而四類銀行中，股份制商銀與城市商銀的表現遠優於國有銀行及政策銀行。但1999年以後中國國有銀行的財務整頓措施的確使得國有銀行的相對表現有所提升。 鑑於外商銀行大舉進入中國金融市場，勢必對中國銀行帶來更多的競爭壓力，本論文第二個研究主題進一步將研究範圍擴及至世界資產排名前一百大的銀行，企圖比較中國銀行與世界百大銀行經營績效的差異。在這個主題中，延續前一主題「股權結構」的重要性，除了官股比例，同時考慮外資持股及本國人持股，分析三類持股對績效的影響，又鑑於總體面的政府治理因素可能對股權與銀行績效的關係產生變化，本研究也將政府治理變數，包括文獻中經常提到的證券市場內線交易情況、政府政策透明度及政府貪污情況等因素納入考量，比較這些因素對中國銀行與百大銀行影響有何差異。實證結果如下：對中國銀行及世界百大銀行而言，政府持股對銀行獲利的影響是負向的，而外資持股的影響則恰相反，開放程度愈高、銀行自由度愈高越有利於銀行績效提升。中國政府越無貪污情況時，越有助於外資持股對銀行績效的正向影響。對世界百大銀行而言，政府越無貪污情況時，越有助於外資持股對銀行績效的正向影響。而政府政策透明度越高，越有助於政府持股與本國人持股的正向影響。 第三個主題探討我國上市櫃公司不同的對外與大陸投資決策對母公司獲利性的影響。本文將投資決策對公司獲利的影響視為內生，也就是投資決策受公司特性因素的影響。同時，我們進一步假設公司的投資決策是分兩階段完成，第一步會先決定是否進行海外投資，一旦確定後，再決定是否赴大陸投資。因為有兩層的投資決策，本文以延伸的Heckman’s Two Stage Method進行估計。 實證結果顯示，在投資決策的影響因素方面，對第一層赴海外投資而言，規模愈大、國際化程度愈高、愈勞力密集的公司愈傾向赴海外投資；對第二層赴大陸投資的決策，影響因素也類似，只是公司規模不再是重要影響因素。至於獲利的影響因素方面，第一類「赴大陸投資的公司」與第二類「未赴大陸但赴其他國家投資的公司」的利潤型態較類似，長期負債對獲利都有負向顯著影響，總資產也傾向負向影響，至於第三類「未從事任何海外與大陸投資公司」的利潤型態則與前兩類不同，總資產與研發支出都呈現顯著負向影響。而三類公司的績效比較部分，第一類公司的平均預期資產報酬率雖若高於第二類，第二類公司又高於第三類公司，但三類公司兩兩之間平均預期資產報酬率的差距並未達統計上的顯著性。 / This dissertation contains three main parts. The first part studies the effects of government owned share on Chinese banks’ performance. After a series of financial reforms in the 1990s, joint stock commercial banks and city commercial banks started to boom and play an increasingly important role in China’s banking industry which had previously been monopolized by four state-owned commercial banks. These two new bank-types are considerably more diversified in that the primary shareholders include the central government, local governments, state-owned enterprises and private enterprises, not just the central government. Using 49 Chinese banks’ financial data retrieved from Bankscope, this part examines the effect of different ownership structures, in general, and government-owned-shares, central vs. local, in particular, affect the profitability and risk of banks in China. It also compares the profitability of four types of banks, namely state-owned banks, policy banks, joint-stock commercial banks and city commercial banks. We conclude that, without equivocation, the higher the ratio of state-owned shares is, the worse is the profitability of the bank. And we also note the profitability of joint-stock commercial banks and city commercial banks is much better than that of state-owned and policy banks. In light of the ever-growing foreign competition facing Chinese banking industry, the second part expands the sample banks to World’s top 100 banks and studies the differences between Chinese banks and foreign banks from the respects of ownership structure, government governance and bank regulations. Our empirical results reveal that for both Chinese banks and top 100 banks, government owned share has a negative effect on banks’ profitability, while foreign owned share has an opposite effect. For both of them, the more the foreign institutions have access to the banking market and the more freedom the banking industry enjoys, the better the bank performs. For Chinese banks, lesser corruption helps strengthen foreign share’s positive effect on banks’ performance. Furthermore, higher transparency would increase the positive effect of government owned share and domestic owned share. The third part studies the profitability of Taiwan’s listed companies with and without investment in China. Unlike past studies which deal with the decision of investment in overseas and China as exogenous, the decision is endogenous in this paper. That is, the decision making is based on the firm’s characteristic factors. We further assume that there are two hierarchy decisions made by firms when they decide in investing in overseas and China. The first decision is whether it should invest overseas. Once the first decision is made, the next one is whether a firm should invest in China. This two decision model, which allows us to extend Heckman’s two-step method, is referred to as an extended-Heckman method in this paper. Our empirical results regarding the determinants of investing overseas and China are as follows. With respect to the investment in overseas, firm size and export ratio show positive influence on the decision of investment, while capital labor ratio shows negative effect. Next, with respect to the investment in China, determinants are similar except that firm size is no longer significant. Our results regarding the profit performance reveal that for ‘the firms with investment in China’ and ‘the firms without investing in China, only investing other countries’, long term liability ratio has a significantly negative impact on profitability, whereas total asset, R&D expenditure show a negative effect on profitability for ‘the firms without investing overseas, nor investing in China’. As for the performance comparison among firms of these three investment types, the differences are insignificant. This result implies that there are still considerable individual differences among the firms of the same investment type.