台灣補習產業競爭程度對於家戶補習支出之實證分析

Abstract

在臺灣如此重視「學歷」的觀念下，補習早已成為臺灣社會的主流價值，即使政府近年來施行一連串教育改革，卻絲毫沒有降低補習的風潮，各類補習班朝連鎖化發展，且科目也因應多元的升學方式擴及至備審資料的撰寫、各類語言口說訓練與特殊技能的培養等。過去文獻對於補習支出方面的研究，大多著重於探討個人或家庭特徵對於補習支出的影響，並未有研究將補教產業競爭度與家戶補習支出兩者結合，並以實證方法進行分析，因此本研究將利用短期補習班資訊管理系統資料計算各地區補教產業HHI指數做為產業集中度指標，並結合民國104年家戶收支調查資料選取家戶補習支出與家戶背景資料；在計算HHI時，依招生對象的不同，界定不同相關市場，將樣本分為混合家戶、只存在國小生家戶、只存在國中生家戶與只存在高中生家戶，以明確估計補教產業集中度對於補習支出之影響效果。 本研究主要使用Tobit模型進行實證分析，但由於補習產業集中度與家戶補習支出具有互為因果的內生性問題，因此本研究除了利用Tobit模型估計外，尚利用工具變數分析法（Instrumental Variable Method）解決內生性存在之問題。主要實證研究結果顯示，補習產業集中度與家戶每學童補習支出確實存在負向關係，亦即補習班競爭程度越高，家戶每學童補習支出越多。此外，統整每學童補習支出與補習班集中度關係可發現混合家戶估計數值最小，亦即補習班當補習班集中度上升或下降0.1（補習班競爭程度降低或增高），混合家戶每學童補習支出減少或增加金額最多，推估其原因為計算混合補習班產業集中度時將所有補習班納入考量，因此各縣市混合補習班集中度的提升或降低相較國小、國中、高中補習班不易，因此在上升或下降同幅度之HHI時，所造成家戶補習支出波動幅度較大；各縣市高中補習班相較國中小補習班較少，補習產業集中度的提高或降低較容易，因此下降或上升同幅度HHI，造成家戶補習支出波動的幅度最小。

Supplementary tutoring has been the mainstream value of Taiwanese society. Even if the government had implemented a series of educational reforms in recent years, it has not reduced the trend of tutoring. All kinds of tutoring classes have been developed in a chain, and the subjects have been expanded to the writing of interview materials, the training of various language speaking skills and the training of special skills. Most of the researches on private tutoring expenditures in the past literature focused on the impact of individual or family characteristics. There is no research that combines the competition of supplementary tutoring industry, household tutoring expenditures and analyzes them by empirical methods. This study will use the Short-term Supplementary Tutoring Information Management System data to calculate the HHI index of the supplementary tutoring industry as the industrial concentration index, and combine The Survey of Family Income and Expenditure of Taiwan in 2015 to select household tutoring expenditure and household background information. In the calculation of HHI, according to the different enrollment targets this study defines different relevant markets; divide the sample into mixed households, only elementary student households, only junior high school student households and only senior high school student households, to explicitly estimate the effect of the supplementary tutoring industrial concentration on the expenditure of tutoring. This study mainly uses the Tobit model for empirical analysis. However, because the tutoring industrial concentration and the household tutoring expenditure have interrelated and endogenous problems, this study uses the Instrumental Variable Method to solve the problem. The main empirical results show that there is a negative relationship between the HHI of tutoring industry and the expenditure of tutoring for each student. In other words, the higher level of competition in the tutoring industry, the more tutoring expenditure will be spent per student. In addition, it can be found that the mixed households have the smallest estimated value; when the HHI increase by 0.1, the mixed households have the largest reduction in the amount of tutoring expenditure per student. The reason is that all the tutoring classes are taken into consideration when calculating the mixed HHI. Therefore, the increase in mixed HHI in counties is more difficult than others. When the HHI increase the same magnitude, the mixed households tutoring expenditures fluctuated significantly. The number of senior high school tutoring classes in all counties is the least so it is easier to increase the HHI. Therefore, the same magnitude of HHI is increased, the fluctuation of household tutoring expenditure is the smallest.