본 연구에서는 기업의 정보위험과 관련된 회계공시의 질이 부채조달비용을 줄이는지 여부를 분석하였다. 2003년부터 2008년까지 일반 회사채를 발행한 12월 결산 비금융 상장기업 가운데, 분석에 필요한 재무자료를 구할 수 있는 기업을 표본으로 선정하였다. 공시의 질을 설명하는 변수로 외국인지분율, 재무분석가의 수, 주주의 수, 대형회계법인여부 그리고 등록시장구분 등을 선정하여 요인분석을 실시한 다음, 요인분석의 결과로 산출된 요인점수를 공시의 질로 측정하였다. 부채조달비용은 한신평의 채권평가등급과 부채이자비용을 사용하였다. 한신평의 채권평가등급이 최우량 채권등급인 경우에 부채조달비용에 1의 값을 부여하고, 상환불능채권등급인 경우에는 20의 값을 부여하였다. 또한 부채이자비용은 이자비용을 기초부채와 기말부채의 평균값으로 나누어 계산하였다. 실증분석 결과, 선행연구에서 부채조달비용에 유의적인 영향을 미치는 것으로 판명된 변수의 영향을 통제한 후에도, 기업의 정보위험 대용치인 공시의 질은 부채조달비용에 유의적인 음의 영향을 미치는 것으로 나타났다. 공시의 질과 부채조달비용 측정의 타당성을 검증하고 연구결과의 강건성을 높이기 위한 추가적인 분석에서도 동일한 결과를 보였다. 이러한 결과는 회계공시가 확대되어 공시의 질이 증가되면 외부투자자들이 겪는 정보비대칭이 감소될 수 있기 때문에, 공시의 질이 높은 기업에게 상대적으로 낮은 위험프리미엄을 적용할 것이며, 이는 부채조달비용의 감소로 이어질 것이라는 사실로 해석할 수 있다.
This paper examines the effect of firms` disclosure quality on cost of debt financing. We conjecture that poor quality of disclosure decrease information asymmetry between managers and market participants, thereby reducing cost of debt capital. Francis et al.(2005) argue that accruals quality (i.e., disclosure quality) tells investors about mapping of accounting earnings into cash flow. Relatively poor disclosure quality weakens this mapping, and therefore, increases information risk. Accordingly, firms with more information risk will have higher costs of capital. Disclosure quality is difficult to measure because it is not directly observable. Disclosure measures used in prior research include analysts` evaluations of firms` disclosure practices by the Association for Investment Management and Research (AIMR) (e.g., Lang and Lundholm 1996; Botosan Plumlee 2002), a disclosure index created by the researcher (e.g., Singhvi and Desai 1971; Botosan 1997), and an indirect measure of disclosure quality combining three correlated variables of institutional holdings, low bid-ask spread, and high analysts following that are associated high disclosure quality (Lee et al. 2006). Among these approaches, we use the Lee et al.(2006)`s measure to estimate a proxy for disclosure quality. In doing so, we perform a factor analysis on foreign investors` holdings, analysts following, audit quality, number of shareholders, and listing status to create one variable (i.e., common factor) for disclosure quality. The major benefit of using a common factor, as maintained by Lee et al. (2006), is that it will be less subject to random measurement errors. Our measure of firms` cost of debt is their senior debt ratings assigned by Korea Investors Service (KIS), one of the Korean major bond-rating firms. KIS assigns bond ratings of Korean firms on a scale of twenty grades from AAA to D. Following Ahmed et al.(2002), we code these ratings such that AAA corresponds with 20 (the best rating) and D (the worst rating) with 1 to facilitate our empirical analyses. We also use the ratio of interest to interest-bearing debt for a robustness check of main results. Prior studies has found that firm`s debt ratings are closely associated with its eventual payoff of interest and principal obligations. We draw an initial sample of firms listed on the Korea Stock Exchange for five years from 2003 to 2007. We obtain data on bonds by searching through FnGuide and KIS-Value databases. Among these, non-financial firms that satisfy all of the following criteria are selected: (1) fiscal-year ending December 31, (2) listed since 5-year before each sample year, and (3) availability of relevant financial data including daily and monthly stock returns. This sample selection procedure yields an final sample of 589 firm-year observations. To test our hypothesis of the effect of firms` disclosure quality on cost of debt financing, we carry out OLS regressions and an ordered-logit regression analyses. We find that firms with poorer disclosure quality have lower debt ratings and higher ratios of interest expense to interest-bearing debt. This results are consistent with several alternative specifications of the disclosure quality and cost of debt metrics. Our empirical findings confirm the notion that greater disclosure reduces information risk arising from investors` estimates of the parameters of an asset (i.e., bonds)`s return or payoff distribution, where information risk is captured by disclosure quality. Our paper makes two contributions. First, consistent with theories that demonstrate a role of information risk in asset pricing, we show that firms with poor disclosure quality have higher costs of debt capital than do firms with good quality. Second, we provide support for using a parsimonious variable measured by factor analysis on a set of variables that are correlated each other but are significantly associated with disclosure quality.