公司經營危機預警模型‐以財務比率與公司治理指標為例

The Warning Models of a Corporate Operational Crisis‐Using the Indexes of Financial Ratio and Corporate Governance

林芳利(Fang-Li Lin)

指導教授 ： 王親仁

國立臺中科技大學/商學院/企業管理系碩士班/碩士(2016年)

摘要

在經營全球化與國際市場競爭的環境下，公司經營面臨許多的挑戰，導致每年皆有許多上市櫃公司毫無預警的發生經營危機。有鑑於此，有效預測公司經營危機的發生，有助於確保公司永續經營能力以及降低投資人財務損失的風險。因此，如何建構一個符合公司經營環境之有效危機預警模型乃重要的議題。
本研究選取2010年至2014年245家(68家危機公司與177家正常公司)台灣上市上櫃公司為樣本，利用SPSS 18.0統計軟體，以敘述性統計、相關分析、獨立樣本檢定、卡方檢定、逐步選取迴歸模型與羅吉斯迴歸模型等，分析如何以財務比率與公司治理指標有效預測公司經營危機。
實證結果發現，無論是全體產業或電子產業，危機公司之負債比率平均較正常公司高，而正常公司之總資產報酬率、淨值報酬率、稅後淨利率及每股盈餘相對較危機公司高，表示公司經營負債比率越高，以及總資產報酬率、淨值報酬率、稅後淨利率及每股盈餘越低，則越容易發生經營危機。此外，全體產業與電子產業獨立樣本檢定結果顯示，預測前一年至前三年之「負債比率」、「總資產報酬率」、「淨值報酬率」、「每股盈餘」，以及預測前二年至前三年之「經理人持股比率」等指標為統計上顯著的重要變數。
以二元羅吉斯迴歸模型預測全體產業之準確率前三年分別為71%、79%、65%，而預測電子產業之準確率前三年分別為78%、86%、67%。由此可見，預警模型之預測準確率，無論是全體產業或電子產業皆以危機發生前二年為最高，表示所建構之公司經營危機預警模型較適合預測危機發生前二年，而且依產業別所建構之預警模型其預測準確率會提高。

關鍵字

經營危機 ； 財務比率 ； 公司治理 ； 羅吉斯迴歸分析

並列摘要

Under the situation of operation globalization and competitive international market, the company faces many challenges, resulting in occurrence of company’s management crisis without warning every year. Therefore, an effectively forecasting the occurrence of company operational crisis will be helpful to ensure sustainable operation capacity and reduce the risk of financial losses to investors. Furthermore, how to construct an effectively crisis warning model to fit the company operational environment is a critical issue.
This study collected 245 Taiwaneselisted companies(68 Crisis and 177 normal companies) that encountered financialdistress between 2010 and 2014. SPSS 18.0 statistical software will be used for the analysis of descriptive statistics, correlation analysis, independent sample test, chi-square test, stepwise regression model and logistic regression model to analyze how financial ratios and corporate governance indicators can effectively predict the company management crisis.
Empirical results show that normal company has the current ratio higher than 260, the debt ratio less than 45, and dependence on loanless than 70. Furthermore, total return on assets, return on equity, net profit margin and earnings per share in a normal company are relatively higher than those in a crisis company. It implies that the company management with the higher debt ratio, and the lower total return on assets, return on equity, net profit margin and earnings per share will occur easily a management crisis. In addition, the variables, the 「debt ratio」, 「total return on assets」, 「return on equity」, and 「earnings per share」 which are 1-3 years prior the prediction and the indicator,「managers shareholding ratio」which is 2-3 year prior prediction are all statistically significant factor based on the results of independent samples test no matter all industries or electronic industries.
The accuracy rate of prediction by the Logistic regression model was 71%, 79%, and 65%for prior Year 1, Year 2and Year 3 in all industries respectively.While the accuracy rate of them was78%, 86% and 67%for prior Year1, Year2andYear3 inelectronics industry respectively. Therefore, the prediction accuracy rate warning model in two year prior the occurrence of operational crisis for all industries and electronics industry. It implies the construction of a company management crisis warning model by the study is comparatively more suited to predict the first two years of the crisis occurrence. Furthermore, the prediction accuracy of the warning model will be increased if you construct a model with the type of industry.

並列關鍵字

Operational Crisis ； Financial Indexes ； Corporate Governance ； Logistic Regression Model

參考文獻

1. 丁秀儀、張邦茹、林倍勤(2011)。危機過後─從公司治理角度探討。證券市場發展季刊，23(4)，63-110。

2. 余惠芳、王永昌(2011)。財務預警與公司治理─台灣傳統產業之實證研究。應用經濟論叢，90，209-241。

5. 林郁翎、黃建華 (2009)。考慮公司治理之企業財務危機預警模型。東吳經濟商學學報，64，23-56。

6. 林淑萍、黃劭彥、蔡昆霖(2007)。企業危機預警模式之研究─DEA-DA、邏輯斯迴歸與類神經網路之應用。會計與公司治理，4(1)，35-56。

9. 邱瓊玲(2011)。關係人交易對財務危機預測模式影響之探討。育達科大學報，26，31-48。