[[alternative]]運用隨機共同生產邊界探討多國會計師事務所之效率與生產力

博士[[abstract]]本研究運用Huang et al. (2014) 所提出之隨機共同生產邊界模型探討臺灣、大陸與美國會計師事務所之效率與生產力。研究對象以三個國家前百大會計師事務所之排名資料，針對2007至2009年相同的投入產出要素進行靜態與動態之比較分析。隨機共同生產邊界與先前所提出之確定共同邊界模型最大的不同處在於其第二階段仍採用隨機邊界法進行估計。此外，本研究亦將一般化Metafrontier Malmquist生產力指數的概念從距離函數延伸至生產函數。主要實證結果如下：首先，臺灣會計師事務所的平均總技術效率最高，然平均技術差距比率則以美國會計師事務所為最佳，顯示美國會計師事務所整體所採用的生產技術居三國之首。然而，兩者間並未呈顯著差異。臺灣屬於小型開放經濟體，雖然其事務所的規模明顯小於美國與大陸，但其專業人員因擁有傑出的生產力使其整體的總技術效率優於其他兩國。其次，在樣本期間大陸會計師事務所的總技術效率與技術差距比率雖居三者之末，但觀其每年技術差距比率的成長幅度卻顯著優於臺灣與美國，顯示大陸會計師事務所的生產邊界正逐漸趕上共同生產邊界。最後，大陸會計師事務所的平均生產力 (gMMPI) 成長幅度亦顯著高於美國與臺灣，主要原因來自其本身的技術進步與潛在技術變動的追趕。推此原因應與近年來大陸政府大力推行會計師事務所做大做強的政策有關。[[abstract]]This study employs the newly developed stochastic metafrontier production function by Huang et al. (2014) to compare the technical efficiencies and productivity changes of public accounting firms among Taiwan, China, and the U.S., operating under different technologies. Although the public accounting firms play an important role in a nation’s capital market, the accounting firm has not attracted much attention to academic researchers. The main difference between the stochastic metafrontier production function and the one proposed by Battese et al. (2004) and O’Donnell et al. (2008) lies in the second step, where the stochastic frontier approach (SFA) is recommended instead of programming techniques. Furthermore, this study extends the concept of the generalized metafrontier Malmquist productivity index (gMMPI), which is originally based on the distance function framework (Orea 2002; Chen and Yang 2011), to production function. In this research, Taiwan’s accounting firms are found to have the highest average metafrontier technical efficiency (MTE) and accounting firms in the U.S. have the highest technology gap ratio (TGR). Nonetheless, the average TGR and MTE values of American firms are closer to those of Taiwan. Taiwan is a small open economy and although its accounting firms are smaller in scale compared to the U.S. and China, its overall production efficiency is superior. This may be attributed to the outstanding productivity of professional employee in Taiwan’s accounting firms, indicating that the professionalism of employee plays a vital role in the production process of firms. Although Chinese accounting firms have the lowest average values for TGR and MTE, the mean TGR of Chinese firms grows more rapidly than that of firms in the other countries, indicating that the gap between the metafrontier and Chinese frontier shrinks more quickly than do the gaps in the U.S. and Taiwan. In addition, their average gMMPI surpasses the other two countries. This is mainly incurred by the fastest technical progress and catch-up with the potential technology in Chinese firms. This may be associated with Chinese governmental policies that encourage fast expansion in the scale of accounting firms.[[tableofcontents]]Table of Content Chapter 1 Introduction 1 1.1. Research Background and Motivation 1 1.2. Research Purpose 5 1.3. Research Contribution 5 1.4. Flow Chart of Research 6 1.5. Glossary 8 Chapter 2 A Comparison of the Technical Efficiency of Public Accounting Firms across Countries 11 2.1. Introduction 11 2.2. Literature Review 14 2.2.1. Efficiencies of the Accounting Firms 14 2.2.2. Literature Review of Metafrontier Approach 17 2.3. Methodology 18 2.3.1. Stochastic Metafrontier Approach 18 2.3.2. Empirical Model 24 2.4. Variable Definition and Data Sources 25 2.4.1. Data Sources and Sample Selection 25 2.4.2. The Output 26 2.4.3. Input Variables 27 2.4.4. Descriptive Statistics 28 2.5. Empirical Results 31 2.5.1. Results of Group Frontier Estimation 31 2.5.2. Results of Metafrontier Estimation 35 2.5.3. Various Efficiency Measures 37 2.5.4. Hypothesis Testing 40 2.6. Conclusion 41 Chapter 3 A Comparison of the Metafrontier Productivity Changes of Public Accounting Firms across Countries 43 3.1. Introduction 43 3.2. Literature Review 47 3.2.1. Productivity Growths of the Public Accounting Firms 47 3.2.2. Metafrontier Approach and gMMPI 49 3.3. Methodology 51 3.3.1. Technical Efficiency and TGR 51 3.3.2. Metafrontier Malmquist Productivity Index 53 3.3.3. Stochastic metafrontier approach 59 3.4. Hypothesis Development 63 3.5. The Data 65 3.5.1. Data Sources and Variable Definition 65 3.5.2. Descriptive Statistics 66 3.6. Empirical Results 69 3.6.1. Results of Group Frontier Estimation 69 3.6.2. Results of the Metafrontier Model 72 3.6.3. Results of the gMMPI 76 3.6.4. Significance Tests of the Difference in the Components of the gMMPI 83 3.7. Conclusion 84 Chapter 4 Conclusions and Recommendations 88 4.1. Concluding Remarks 88 4.2. Limitations and Recommendations 90 References 91 Appendix A 97 List of Tables Table 2-1 Descriptive statistics 30 Table 2-2 Joint tests for equality of mean inputs and outputs among sample countries 31 Table 2-3 Parameter estimates of group-specific production frontiers 34 Table 2-4 Parameter estimates of the metafrontier production function 36 Table 2-5 Summary statistics of various measures of efficiency in accounting firms 38 Table 2-6 Significance tests for measures of TGR and MTE across countries 40 Table 3-1 Descriptive statistics 67 Table 3-2 Joint tests for equality of mean inputs and outputs among countries 68 Table 3-3 Parameter estimates of group-specific production frontiers 71 Table 3-4 Parameter estimates of the metafrontier production function 73 Table 3-5 Summary statistics of various measures of efficiency in the accounting industry 75 Table 3-6 Significance Tests for Measures of TGR and MTE across Countries 76 Table 3-7 Decomposition of the gMMPI 85 Table 3-8 Cross-country Significance test for elements of the gMMPI 86 Table 3-9 Cross-period significance test for elements of the gMMPI 87 List of Figures Figure 1-1 Flow chart of research 7 Figure 2-1 SMPF Approach-GTE 39 Figure 2-2 SMPF Approach-TGR 39 Figure 2-3 SMPF Approach-MTE 39 Figure 2-4 QP Approach-GTE 39 Figure 2-5 QP Approach-TGR 39 Figure 2-6 QP Approach-MTE 39 Figure 3-1 GMPI across countries 79 Figure 3-2 gMMPI across countries 79 Figure 3-3 CUT across countries 81 Figure 3-4 PTC across countries 81 Figure 3-5 CUS across countries 82 Figure 3-6 CUE across countries 82 Figure A-1 A Production Frontier 97 Figure A-2 Metafrontier Production Model 101[[note]]學號: 898620496, 學年度: 10