1、中文 2194 字 外 文 翻 译 原文： Disclosure Quality and Cash Flow We study a relatively recent change in voluntary disclosure practices by management, namely the issuance of management cash flow forecasts. While we find some management cash flow forecasts in the early 1980s, the incidence of such disclosures i
2、s low until recent years. However, since 2000, there has been a dramatic increase in the issuance of management cash flow forecasts and the number of such forecasts has more than tripled from pre-2000 levels. One potential explanation for this trend is Regulation FD, which went into effect in 2000.
3、For example, if managers were disclosing cash flow forecasts privately to analysts prior to Regulation FD, they would have to publicly disclose such forecasts to all parties after Regulation FD or curtail their management cash flow forecasts completely. Regulation FD potentially increases company di
4、sclosure of a wide array of financial information, including cash flow information. Consistent with this, we also document an increasing frequency of management earnings forecasts in recent years. Another potential explanation for the recent trend of more management cash flow forecasts is investors
5、and analysts paying more attention to cash flow information than before. Recent corporate scandals involving Enron, WorldCom and others, have heightened investor concern over potential accounting earnings manipulations. Such concerns were recently noted in a Business Week article entitled “Fuzzy Num
6、bers”. Consistent with an increase in the demand for cash flow information by investors, we also find analyst forecasts of cash flow during the 2000-2003 period more than doubled from pre-2000 levels. We study voluntary management cash flow forecasts and test hypotheses on managers incentives to iss
7、ue these forecasts. Prior voluntary disclosure literature offers varying predictions for managements incentives to provide disclosure as well as the nature of information conveyed in management disclosures. For example, theoretical models demonstrate that when there are proprietary costs of disclosu
8、re or when investors are uncertain about the information management has, firms will voluntarily disclose good news and withhold less favorable news. Early empirical studies on management earnings forecasts provide evidence consistent with this prediction. More recent empirical work, however, suggest
9、s the importance of litigation risk in affecting management earnings forecasts. Skinner1994 and Kasznik and Lev1995 document that management earnings forecasts are more likely to convey bad news, consistent with managers are concerned with the risk of litigation and issue preemptive earnings forecas
10、ts to adjust downward investor expectations. Earnings forecasts likely play a special role in reducing the risk of litigation, and a more important role than management cash flow forecasts. For firms with bad news and thus concerned about potential litigation, earnings-related disclosures are likely
11、 to be more effective in conveying that bad news to investors than disclosures of other financial information such as cash flows because, in general, earnings is the most informative summary performance measure. Thus, earnings disclosure is more likely to bring about the needed adjustment to investo
12、r expectations. As a result, the propensity for earnings forecasts to reflect bad news as documented in some of the prior studies may not apply to other types of management forecasts such as cash flow forecasts. Consistent with this, using data since the 1980s, researchers find that better performan
13、ce is associated with higher overall disclosure levels. We predict that management issues cash flow forecasts to signal good news in cash flow, to meet investor demand for cash flow information, and to pre-commit to a certain composition of earnings in terms of cash flow versus accruals, thus reduci
14、ng the degree of freedom in earnings management. Our findings are consistent with these predictions. We find that the likelihood of management cash flow forecasts increases in periods when there is a large increase in operating cash flow, when analysts are forecasting an earnings loss, when manageme
15、nt specifically reveals in their press releases that earnings will be either below or above expectations and when firm is young. In addition, we find that the likelihood of management cash flow forecasts decreases in periods with extreme positive discretionary accruals. We document that firms issuin
16、g management cash flow forecasts tend to have better cash flow information than those without a forecast and that management cash flow forecasts tend to beat existing expectations. This applies to situations where there is very bad or very good news in earnings and when the firm is young, suggesting
17、 that management uses cash flow forecasts to mitigate the negative impact of bad news in earnings, to lend credibility to good news in earnings and to signal economic viability in young firms. We provide evidence on how management reports actual cash flow information in (subsequent) press releases.
18、We document that managers use discretion in reporting realized cash flows in earnings announcement-related press releases and adopt alternative definitions of cash flows (vis-vis GAAP definitions of cash flow per the statement of cash flows). Specifically, management-announced actual cash flows tend
19、 to exceed actual GAAP cash flows. As a result, management-announced actual cash flows meet or beat management cash flow projections more often than actual GAAP cash flows meet or beat management projections, generating significantly positive cash flow forecast errors. This finding suggests that a s
20、imilar practice to managements announcement of pro forma earnings also exists for management announcement of cash flows. Sloan (1996) provides evidence that investors overestimate the persistence of accruals and underestimate the persistence of cash flow. This results in mispricing labeled the accru
21、al anomaly where a trading strategy designed to exploit investors misunderstanding of the persistence of earnings components earns significant abnormal returns. Literature extending Sloan (1996) primarily focuses on accrual overvaluation (e.g., Xie 2001； Thomas & Zhang 2002； Collins, Gong, & Hribar
22、2003). However, recent research suggests that research focusing on accrual mispricing without also considering cash flow mispricing is incomplete (Desai, Rajgopal, & Venkatachalam 2004； Yu 2007； Barone & Magilke 2008). Thus, in this paper, we investigate the role that disclosure quality plays in the accurate
