본 연구는 금융위기 기간의 주식수익률을 이용하여 한국 기업들이 위기경험으로부터 학습효과를 얻어 위기대응력을 제고하였는지, 아니면 위기경험에도 불구하고 위험문화나 영업방식을 유지하였는지를 검증하였다. 선행위기 기간의 주식수익률은 후행위기 기간의 주식수익률에 대한 예측력을 보유하여 위험문화가설의 타당성을 확인하였다. 위기수익률 간의 양의 상관관계는 대주주가 존재하는 기업의 경우 유의하게 약해지고, 업력에 따른 차이는 발견하지 못하였다. 기업특성변수를 사용하여 금융위기 기간 저조한 주가수익률을 야기한 요인을 살펴본 결과, 수익성(총자산수익률)이 낮고 재무적 불안정성(차입금의존도)이 높은 기업은 지속적으로 저조한 위기수익률을 시현하였다. 1997년 금융위기 기간 중 외국인 투자자의 매도거래가 많았던 기업일수록 더욱 심각한 주가하락을 경험하였으며, 이는 자본자유화 이후 외국인 투자자의 거래 행태가 국내 주식시장에 미치는 영향이 증대되었음을 의미한다. 일반적으로 주식수익률에 영향을 준다고 알려진 베타, 시장가치 및 장부가대시가비율은 위기기간에는 설명력이 제한되거나 상이한 영향력을 갖는다는 점을 확인하였다.

Are some firms have something unique to make them prone to perform poorly during crises? The purpose of this study is to answer the above question. In this paper, we show that a firm`s stock return during the preceding crisis (stock return during the 1997 crisis, 1997 crisis return) has predictive power to the posterior crisis return (stock return during the 2008 crisis, 2008 crisis return). Analyses in this paper are related to those presented in the previous literatures (Fahlenbrach et al., 2012; Kim & Yang, 2013; and Kim, 2014). This paper analyzes whether Korean non-financial firms raised the confrontational ability or didn`t change risk culture (or business model) after crisis experience. To this end, we test the two hypotheses through regression analysis. The main explanatory variable is 1997 crisis return, and the dependent variable is 2008 crisis return. Risk culture hypothesis refers to the situation that firms with poor performance during the preceding crisis would also perform poorly during the posterior crisis. Difficulty to change risk culture or business model that make firms sensitive to crises is the basis of an argument to this hypothesis. Learning hypothesis refers to the situation that firms with low stock return in 1997 would perform well in 2008 through crisis learning. Firms could change its risk culture or business model after crisis experience to raise the confrontational ability to the crisis. Our empirical findings are as follows. First, 2008 crisis return is positively associated with 1997 crisis return. This positive association is outstanding among firms which performed poorly during the 2008 crisis. This result lead us to conclude that the risk culture hypothesis is more appropriate for Korean non-financial firms than the learning hypothesis. Second, the above mentioned predictive power is weakened among firms with major shareholder. Major shareholder refers to the shareholder with ratio of share more than 2% in KOSPI (Korea Composite Stock Price Index) and 4% in KOSDAQ (Korea Securities Dealers Automated Quotations), or market capitalization more than 5 billion Korean won in KOSPI and 4 billion Korean won in KOSDAQ according to the Korean tax law. Corporate age - relating to the concept of organizational learning, organizational life cycle, and organizational inertia - is regarded to favor the risk culture hypothesis. But we can`t find any significant difference according to the corporate age. Third, we examine the effects of firm characteristics and foreign investors` trading trend on the predictability of poor performers during the crisis period by probit analysis. Firms with higher borrowing to asset ratio and low turnover ratio of total liability and net worth had poor performance during the both crises. These results are consistent with the previous literatures which insist funding fragility is the main weakness during the crisis period. Kim and Yang (2013) found borrowing to asset ratio is the key factor that makes crisis return of Korean financial institutions low. High net sales growth rate had negative effect to the crisis return only in 1997. And firms with low foreign investors` net purchase ratio had poor performance during the 1997 crisis. This result is in line with the arguments that capital decontrol may be the cause of financial crisis. Forth, these results are robust to the change of the calculation method and period of the crisis return. We found statistically significant positive coefficient of regression as ever from the robustness check with two newly calculated main explanatory variables. Overall, the findings are consistent with the risk culture hypothesis. It means that some firms has vulnerability to the crisis. And this fragility would come from the difficulty in the change of the risk culture or the business model of them. Previous study with the sample of Korean financial institutions is in favor with the learning hypothesis (Kim & Yang, 2013; and Kim, 2014). It may be a big difference between financial institutions and non-financial firms in organizational learning after crisis experience, and of the persistence in the risk culture or the business model. This is considered from the dissimilarity of corporate governance. It is because the existence of major shareholder has significant effect to the firm`s behavior after crisis experience. Our study is the first analysis over the Korean non-financial firms` behavior after crisis experience. This paper contributes to the literature on the financial crisis, the crisis return, and also on, in part, the organizational learning and the capital decontrol.