公司財務危機預測-考慮公司治理及關係人交易影響之實證研究

An Empirical Study of The Predictive Financial Distress of Corporate—Considering The Impact of Corporate Governance and The Related Party Transaction

陸怡伸(Yi-Sheng Lu)

指導教授 ： 林江亮

中原大學/商學院/會計研究所/碩士(2003年)

https://doi.org/10.6840/cycu200300319

摘要

摘 要

過去許多財務危機研究指出以傳統財務性指標及公司治理因素建立財務危機模式，可獲得良好之危機預警模式，但並未有研究考量將關係人交易資訊納入危機模型。近年來爆發許多掏空公司資產、違約交割及跳票的情況，這些事件大都與關係人交易有關連性，關係人可經由非常規交易影響公司經營狀況，例如，較低之進貨價格，較高之售價間接影響公司損益；關係人資金融通將可能造成公司大量呆帳。本研究將我國財務會計準則公報第六號規定揭露之關係人交易資訊加入危機預警模型中，觀察關係人交易資訊是否可提昇財務危機預測之能力。

本研究依據Lee and Yeh（2001）之研究方法，採用Foster’s（1997）對財務危機的定義選取危機公司，並以「1：2配對樣本法」採取相同產業及相當規模之正常公司為配對樣本。利用採用因素分析法及不採用因素分析法兩種方式分別建構危機發生當年、前一年及前二年之財務危機預警模型，並考量加入財務性解釋變數、公司治理解釋變數及關係人交易解釋變數，是否能提高模式之正確區別率。

研究結果顯示：
一、 運用Logit模式以1997-2000年之樣本公司為估計樣本，採用因素分析法預測2001年驗証樣本公司之預測正確區別率為0.8485，最佳臨異點為0.4。不採用因素分析法最佳臨異點為0.4，正確區別率為0.6667。故採用因素分析法所建構之模式較不採用因素分析法佳。
二、 不論採用因素分析法或不採因素分析法，加入公司治理解釋變數所建構之財務危機預警模型之預測能力較僅含財務性解釋變數之財務危機預警模型為佳。
三、 不採用因素分析法下，加入關係人交易解釋變數所建構之財務危機預警模型之預測能力較僅含財務性解釋變數之財務危機預警模型為佳。採用因素分析之財務危機預警模型其正確區別率在危機發生當年及前二年並無提昇現象。
四、 不論採用因素分析法或不採因素分析法，同時加入公司治理及關係人交易解釋變數之財務危機預警模型其正確區別率均高於僅加入財務性解釋變數或公司治理及關係人交易解釋變數之模型，且隨著危機時點接近，模型之區別能力會越來越高。

關鍵字

財務危機預測 ； 公司治理 ； 關係人交易

並列摘要

Abstract

According to previous studies on financial distress of corporate pointed out that financial ratios and corporate governance factors made the construction of financial distress model could obtain excellent financial distress predicting model, but there were no study considering the impact of the related party transaction. It happened many excavation of the company’s assets, check bounce situations, those events most had correlations with the related transactions, the related party could influence on condition of corporate operating, for example, the lower purse price and the higher sales price will indirectly affect corporate income：pur-chases from the related parties will make corporate have many bad accounts. This study will add most existing related-party transaction disclosures required by Taiwanese Financial Accounting Standard Board Statement N0.6 in model of financial distress and observe whether related-party transaction information makes the construction of financial distress model better accordingly.

This study depends on Lee and Yeh（2001）research method, and adopts Foster’s（1997）defined a state of financial crisis to chose distress firms, and use 「1：2 matched pairs sample method」chose the same industry and the same size non-distress company as matched sample. Using factor analysis method and non-factor analysis method to establish the period between current years to 2 years before financial distress the financial distress model, and considering whether financial explanation variables, corporate governance explanation variables and related party explanation variables could increase classify correct ratio.

The results of empirical studies shows：
1. Using 1997-2000 years estimate samples applying Logit regression model, Using factor analysis method and non-factor analysis method to accurate 2001 years holdout samples’ predicting classify correct ratio is 0.8485 and the best cut off point is 0.4；non-factor analysis holdout samples’ predicting classify correct ratio is 0.6667 and the best cut off point is 0.4. Using factor analysis method to establish the financial distress is more better than non-factor analysis method.
2. Whether using factor analysis method or non-factor analysis method, add corporate governance explanation variables in model of financial distress is provided with more predicting ability than only including financial explanation variables financial distress model.
3. In non-factor analysis method, add related party explanation variables in model of financial distress is provided with more predicting ability than only including financial explanation variables financial distress model. Using factor analysis method establish financial distress modeol, during the period between current year and 2 years before financial distress classify correct ratio is not increasing.
4. Whether using factor analysis method or non-factor analysis method, add corporate governance explanation variables and related party explanation variables in the same time classify correct ratios is higher than only including financial explanation variables financial distress model.

並列關鍵字

financial distress ； the related party transaction ； corporate governance

參考文獻

葉銀華、李存修和劉容慈,2002,「整合公司治理、會計資訊與總體經濟敏感度之財務危機模型」，行政院國科會專題研究計畫成果報告。

Hsu, S. C. 1997. Familism, Professionalism and Entrepreneurship - In the Perspective of Chinese Business Firms. Management Review 16：1-9.

Yeh, Y. H., S. B. Chiu, and H. C. Ho. 1997. Wealth Exploitation and Ownership Structure: A Study of Agency Theory on Taiwan Stock Market. Journal of Financial Studies 4：47-73.

Altman, E. I. 1968. Financial Ratios, Discriminate Analysis and the Prediction of CorporateBankruptcy. Journal of Finance 23：589-609.

Aziz, A., and G. H. Lawson. 1989. Cash Flow Reporting and Financial Distress Models: Testing of Hypotheses. Financial Management 18：55-63.