Prior research on the determinants of credit ratings has focused on rating agencies’ use of quantitative accounting information, but the there is scant evidence on the impact of textual attributes. This study examines the impact of financial disclosure narrative on bond market outcomes. We find that less readable financial disclosures are associated with less favorable ratings, greater bond rating agency disagreement, and a higher cost of debt. We improve causal identification by exploiting the 1998 Plain English Mandate, which required a subset of firms to exogenously improve the readability of their filings. Using a difference-in-differences design, we find that the firms required to improve the readability of their filings experience more favorable ratings, lower bond rating disagreement, and lower cost of debt. Collectively, our evidence suggests that textual financial disclosure attributes appear to not only influence bond market intermediaries’ opinions but also firms’ cost of debt.
This is a preview of subscription content, log in to check access.
Buy single article
Instant unlimited access to the full article PDF.
Price includes VAT for USA
Subscribe to journal
Immediate online access to all issues from 2019. Subscription will auto renew annually.
This is the net price. Taxes to be calculated in checkout.
As discussed in more detail in Section II, deterioration in disclosure readability is likely driven by a combination of increases in regulatory requirements during our sample period (Radin 2007), lack of managerial ability, or managerial obfuscation (Bloomfield 2002). Standard agency theory predicts that managers are likely to put their own interests above the interests of the corporation. As such, in many cases, managers may have incentives to obfuscate their financial disclosures (e.g., hide poor performance as documented in Li (2008)) and forgo potential benefits to the corporation (e.g., the potential for lower cost of debt).
Unlike equity investors, bond rating agencies have access to material nonpublic information even after Regulation Fair Disclosure (FD). Although this access may provide rating agencies the opportunity to clarify certain aspects of the report, it is unlikely that it would eliminate the role of the public disclosure in forming rating opinions. Specifically, studies by Kim and Verrecchia (1994) and Barron et al. (1998) model public information as truth plus noise and private information as a signal that reduces the noise in the public signal. As such, although it is plausible that access to management will reduce the association between less readable reports and raters’ opinions, the frictions due to factors such as limited rating agency resources are likely large enough that the noise in the public signal cannot be eliminated. Theoretically, the Dodd-Frank Act of 2010 limited raters’ access to nonpublic information, but in practice, rating agencies were not affected by the regulation, as they re-classified themselves to no longer be investment advisers. Furthermore, rating agencies also sign express confidentiality agreements, enabling them to continue to receive material nonpublic information. We confirm this notion that access to private information has been unaffected by the Dodd-Frank Act directly with the rating agencies themselves.
This disagreement could manifest itself in different ways. First, less readable filings may lead rating agencies to process the same disclosure differently (Karpoff 1986; Kandel and Pearson 1995). Second, higher processing costs could lead rating agencies to focus on different factors within a filing because processing the entire report may be too costly (Merton 1987; Hirshleifer and Teoh 2003). Given the unobservable nature of the mechanism through which higher processing costs arise from less readable financial reports, we focus more generally on the impact of processing costs stemming from disclosure readability.
The focus of our examination is on the impact of writing clarity in firms’ disclosures incremental to the content of textual disclosures. We control for and separately examine the debt market implications of several content based textual measures of disclosure (e.g., tone, uncertainty, risk disclosure, and the extent of forward-looking statements). A recent working paper by Bozanic and Kraft (2015) focuses on Moody’s hard and soft adjustments related to the qualitative content of firms’ annual reports.
Work by Ederington (1986) examines these mappings and finds no evidence of a difference in ratings standards between the two agencies. However, to provide assurance that our results are not affected by potential differences in rating mappings into default risk, we follow Akins (2013) and compute the historical default rate for each rating category and agency as of the beginning of a calendar year. We employ the S&P RatingsXpress and Moody’s Default and Recovery databases to gather rating and default information for this test. We adopt this rolling computation approach to mitigate the effects of look-ahead bias on our measure of default risk difference. For each bond in our sample, we compute the absolute value of the difference between the historical default rate for Moody’s and S&P corresponding to the categorical ratings assigned to the bond issue. We use this measure as an alternative proxy for rating agency uncertainty in a Tobit model. The Tobit model generates positive coefficients for all of our readability proxies at the same statistical significance levels as in our binary or ordered logit tests.
Our primary results related to rating disagreement are based on rating differences of a single “notch” (e.g., Moody’s = Aa2 and S&P = AA-). Despite the fact that single notch disagreements capture observable differences of opinion between Moody’s and S&P, one could raise the concern that single notch disagreements are less likely to capture rating agency uncertainty than more pronounced disagreements. To address this concern, we compress the rating scale to nine categories where, consistent with the rows in Appendix B, we group all ratings with the same letter portion of the rating together (e.g., the S&P ratings of A+, A, and A-). We then recalculate our SPLIT and SPLIT_GAP in the same manner as under the full rating scale and find that our results are unaffected by measuring the disagreement using this broader scale.
For ease of presentation and comparison, this table presents raw writing style components, instead of the transformed variables used to create the Bog Index (e.g., the Bog Index incorporates a squared version of the average sentence length).
We also control for several other earnings properties that Akins (2013) finds associated with rating agency disagreement. Specifically, we control for accruals quality, as defined by Francis et al. (2005), the debt contracting value of accounting information as defined by Ball et al. (2008), and the asymmetric timeliness coefficient from the Basu (1997) model and find that our results are unaffected by the inclusion of these alternative controls.
The inclusion of multiple fixed effects can induce bias in coefficient estimates in logit models (Chamberlain 1980). Thus, as a robustness check, we re-estimate our regressions using a linear probability model (OLS) with both firm and year fixed effects and find that the significance levels are unaffected by the inclusion of both fixed effects in the same model.
We limit our sample to initial bond issuances of fixed rate nonconvertible bonds. This restriction not only ensures that both issuers provide ratings at the same time but also creates homogeneity among the types of debt instruments within our sample and thus alleviates the possibility that other characteristics are influencing results. This restriction does not appear to impact generalizability, as over 70% of corporate bonds in the Mergent FISD universe of new bond issues meet our sample selection criterion. From this sample of available bond issuances, we randomly select one bond per firm-year to reduce the possibility our test statistics are inflated by the inclusion of multiple issuances for the same firm. Results are unaffected if we use the entire sample of available bond issuances and continue to cluster standard errors by firm and year.
The average (median) firm in our sample issues 2.47 (one) separate debt instruments per fiscal year. Thus, there are a small number of firms that issue a larger number of debt instruments per year.
Kappa is calculated as [ po – pe ] / [ 100 – pe ], where po is the percentage of bonds with the same rating from both Moody’s and S&P observed and pe is the percentage expected, given the distribution of ratings.
In untabulated tests, we provide more direct evidence that changes in readability around the SEC’s Plain English Mandate were directly associated with various bond outcomes. To calculate our changes in measures around the mandate, we use the last bond issued by treated firms before the regulation and the first bond issued following the regulation and exploit variation in readability changes to examine the impact of readability on bond outcome variables. Despite the fact we are left with only 108 bond changes, we find evidence that both the change in the 10-K Bog Index and the change in the prospectus Bog Index are associated with favorable changes in bond ratings, less pronounced rating disagreement, and reductions in the cost of debt capital.
In the tabulated tests reported in Panel B of Table 7, we use issuances made by firms that filed prospectuses in 1999–2000 and examine the periods before the regulation (POST_PSEUDO). In untabulated tests, we also classify firms that filed prospectuses in 1996–1997 and find similar insignificant results on the interaction term to those reported in Panel B of Table 7.
In untabulated tests, we observe that nonprospectus filing firms resemble prospectus filers, and, as expected, they are larger, have more analyst following, and slightly lower readability than prospectus filing firms. To provide greater assurance that the differences between these two groups of firms are not driving our results, we implement an entropy balancing technique that achieves covariate balance across first, second, and third moments of the respective variable distributions (Hainmueller 2012; McMullin and Schonberger 2015). After employing the entropy balancing procedure, we not only find that the mean, variance, and skewness of all covariates are nearly identical across the treated and control firms but also that our results from Panel B of Table 7 continue to be significant after employing the weighted regression approach described by Hainmueller (2012).
We find that there is some correlation between BOG and the alternative measures of readability that we consider in this section. Specifically, across our sample, firms BOG is significantly positively correlated with PASSIVE (0.3531), LOG_WORDS (0.5870), FOG (0.5217), and LOG_FILESIZE (0.4469).
These weaker results on file size are consistent with the work of Bonsall, et al. (2016), who point out that the file size measure contains more than just the textual information (e.g., pictures, XML, HTML, PDF’s etc.). Given that these nontextual components are unrelated to readability, this measure is likely a noisy proxy for readability.
Akins, B. (2013). Financial reporting quality and uncertainty about credit risk among the ratings agencies. Working Paper, Rice University.
Atiase, R. K. (1985). Predisclosure information, firm capitalization and security price behavior around earnings announcements. Journal of Accounting Research, 23(1), 21–36.
Ball, R., Bushman, R., & Vasvari, F. (2008). The debt-contracting value of accounting information and loan syndicate structure. Journal of Accounting Research, 46(2), 247–287.
Bamber, L. S. (1987). Unexpected earnings, firm size, and trading volume around quarterly earnings announcements. The Accounting Review, 62(3), 510–532.
Barron, O. E., Kim, O., Lim, S. C., & Stevens, D. E. (1998). Using analysts’ forecasts to measure properties of analysts’ information environment. The Accounting Review, 73(4), 421–433.
Basu, S. (1997). The conservatism principle and the asymmetric timeliness of earnings. Journal of Accounting and Economics, 24(1), 3–37.
Beatty, A., Weber, J., & Yu, J. (2008). Conservatism and debt. Journal of Accounting and Economics, 45(2–3), 154–174.
Beyer, A., Cohen, D., Lys, T., & Walther, B. (2010). The financial reporting environment: Review of the recent literature. Journal of Accounting and Economics, 50(2–3), 296–343.
Biddle, G. C., Hilary, G., & Verdi, R. S. (2009). How does financial reporting quality relate to investment efficiency? Journal of Accounting and Economics, 48(2–3), 112–131.
Bloomfield, R. J. (2002). The “incomplete revelation hypothesis” and financial reporting. Accounting Horizons, 16(3), 233–243.
Bloomfield, R. J., & Fischer, P. E. (2011). Disagreement and the cost of capital. Journal of Accounting Research, 49(1), 41–68.
Blume, M. E., Lim, F., & Mackinlay, A. C. (1998). The declining credit quality of U.S. corporate debt: Myth or reality? Journal of Finance, 53(4), 1389–1413.
Bonsall, S. B., Holzman, E., & Miller, B. P. (2016). Managerial ability and credit risk assessment. Management Science. doi:10.1287/mnsc.2015.2403.
Bonsall, S. B., Leone, A. J., Miller, B. P., & Rennekamp, K. (2016). A plain English measure of financial reporting readability. Working paper, the Ohio State University, the University of Miami, Indiana University, and Cornell University.
Bozanic, Z., & Kraft, P. (2015). Qualitative corporate disclosure and credit analysts' soft rating adjustments. Working paper, the Ohio State University and New York University.
Bozanic, Z., Roulstone, D., & Van Buskirk, A. (2015). Management earnings forecast and other forward looking statements. Working paper, the Ohio State University.
Campbell, J. L., Chen, H., Dhaliwal, D. S., & Steele, L. B. (2014). The information content of mandatory risk factor disclosures in corporate filings. Review of Accounting Studies, 19(1), 396–455.
Chamberlain, G. (1980). Analysis of covariance with qualitative data. Review of Economic Studies, 47(1), 225–238.
Core, J. (2001). A review of the empirical disclosure literature: A discussion. Journal of Accounting and Economics, 31(1–3), 441–456.
Cox, C. (2007). Closing remarks to the second annual corporate governance summit. Speech delivered at the USC Marshall School of Business, Los Angeles, CA, March 23, 2007. U.S. Securities and Exchange Commission. http://www.sec.gov/news/speech/2007/spch032307cc.htm. Accessed 3 October 2013.
Demerjian, P., Lev, B., Lewis, M., & McVay, S. (2012). Managerial ability and earnings quality. The Accounting Review, 88(2), 463–498.
Demerjian, P., Lev, B., & McVay, S. (2012). Quantifying managerial ability: A new measure and validity tests. Management Science, 58(7), 1229–1248.
Ederington, L. H. (1986). Why splits ratings occur. Financial Management, 15(1), 37–47.
Francis, J., LaFond, R., Olsson, P., & Schipper, K. (2005). The market pricing of accruals quality. Journal of Accounting and Economics, 39(2), 295–327.
Ganguin, B., & Bilardello, J. (2005). Fundamentals of corporate credit analysis. New York: The McGraw-Hill Companies.
Guay, W., Samuels, D., & Taylor, D. (2016). Guiding through the fog: Financial statement complexity and voluntary disclosure. Journal of Accounting and Economics, 62(2–3), 234–269.
Gunning, R. (1952). The technique of clear writing. New York: McGraw-Hill International Book Co.
Hainmueller, J. (2012). Entropy balancing for causal effects: A multivariate reweighting method to produce balanced samples in observational studies. Political Analysis, 20(1), 25–46.
Hand, J. R. M., Holthausen, R. W., & Leftwich, R. W. (1992). The effect of bond rating agency announcements on bond and stock prices. The Journal of Finance, 47(2), 733–752.
Hirshleifer, D., & Teoh, S. H. (2003). Limited attention, information disclosure, and financial reporting. Journal of Accounting and Economics, 36(1–3), 337–386.
Kandel, E., & Pearson, N. (1995). Differential interpretation of public signals and trade in speculative markets. Journal of Political Economy, 103(4), 831–872.
Kaplan, R., & Urwitz, G. (1979). Statistical models of bond ratings: A methodological inquiry. Journal of Business, 52(2), 231–261.
Karpoff, M. (1986). A theory of trading volume. Journal of Finance, 41(5), 1069–1087.
Khan, M., & Watts, R. (2009). Estimation and empirical properties of a firm-year measure of accounting conservatism. Journal of Accounting and Economics, 48(2–3), 132–150.
Kim, O., & Verrecchia, R. E. (1994). Market liquidity and volume around earnings announcements. Journal of Accounting and Economics, 17(1–2), 41–67.
Kothari, S. P., Li, X., & Short, J. (2009). The effect of disclosures by management, analysts, and business press on cost of capital, return volatility, and analyst forecasts: A study using content analysis. The Accounting Review, 84(5), 1639–1670.
Kraft, P. (2015). Rating agency adjustments to GAAP financial statements and their effect on ratings and credit spreads. The Accounting Review, 90(2), 641–674.
Kravet, T., & Muslu, V. (2013). Textual risk disclosures and investors’ risk perceptions. Review of Accounting Studies, 18(4), 1088–1122.
Lawrence, A. (2013). Individual investors and financial disclosure. Journal of Accounting and Economics, 56(1), 130–147.
Lehavy, R., Li, F., & Merkley, K. (2011). The effect of annual report readability on analyst following and the properties of their earnings forecasts. The Accounting Review, 85(6), 2107–2143.
Li, F. (2008). Annual report readability, current earnings, and earnings persistence. Journal of Accounting and Economics, 45(2–3), 221–247.
Livingston, M., Naranjo, A., & Zhou, L. (2007). Asset opaqueness and split bond ratings. Financial Management, 36(3), 49–62.
Livingston, M., & Zhou, L. (2010). Split bond ratings and information opacity premiums. Financial Management, 39(2), 515–532.
Loughran, T., & McDonald, B. (2011). When is a liability not a liability? Textual analysis, dictionaries, and 10-Ks. Journal of Finance, 66(1), 35–65.
Loughran, T., & McDonald, B. (2014a). Measuring readability in financial disclosures. Journal of Finance, 69(4), 1643–1671.
Loughran, T., & McDonald, B. (2014b). Regulation and financial disclosure: The impact of plain English. Journal of Regulatory Economics, 45(1), 94–113.
Lundholm, R., Rogo, R., & Zhang, J. (2014). Restoring the tower of babel: How foreign firms communicate with U.S. investors. The Accounting Review, 89(4), 1453–1485.
Mansi, S. A., Maxwell, W. F., & Miller, D. P. (2011). Analyst forecast characteristics and the cost of debt. Review of Accounting Studies, 16(1), 116–142.
McMullin, J. L., & Schonberger, B. (2015). Entropy-balanced discretionary accruals. Working paper, Indiana University and University of Rochester.
Merton, R. C. (1987). A simple model of capital market equilibrium with incomplete information. Journal of Finance, 42(3), 483–510.
Miller, B. P. (2010). The effects of disclosure complexity on small and large investor trading. The Accounting Review, 85(6), 2107–2143.
Morgan, D. P. (2002). Rating risk and uncertainty in an opaque industry. American Economic Review, 92(4), 874–888.
Radin, A. (2007). Have we created financial statement disclosure overload? New York Society of CPAs. http://www.nysscpa.org/cpajournal/2007/1107/perspectives/p6.htm. Accessed 13 September 2013.
Rennekamp, K. (2012). Processing fluency and investors’ reactions to disclosure readability. Journal of Accounting Research, 50(5), 1319–1354.
Schroeder, M. (2002). SEC proposes rules to improve disclosure by public companies. Wall Street Journal, (May 1), C5.
Securities and Exchange Commission. (1998). A plain English handbook: How to create clear SEC disclosure. SEC Office of Investor Education and Assistance. http://www.sec.gov/pdf/handbook.pdf. Accessed 07 September 2013.
Securities and Exchange Commission. (2013). The path forward on disclosure. Speech given by SEC Chair Mary Jo White to the National Association of Corporate Directors. http://www.sec.gov/News/Speech/Detail/Speech/1370539878806#.UrBX57Rojvc. Accessed 21 March 2014.
Securities and Exchange Commission. (2014). Disclosure effectiveness: Remarks before the American Bar Association Business Law Section Spring Meeting. http://www.sec.gov/News/Speech/Detail/Speech/1370541479332#.U4Zb-SjrB3s. Accessed 15 June 2014.
Securities Industry and Financial Markets Association. (2016). Statistics and data pertaining to financial markets and the economy. http://www.sifma.org/research/statistics.aspx. Accessed 18 August 2016.
Sengupta, P. (1998). Corporate disclosure quality and the cost of debt. The Accounting Review, 73(4), 459–474.
Standard and Poor’s (2008). Corporate ratings criteria. http://www.nafoa.org/pdf/CorprateCriteriaBook-2008.pdf. Accessed 03 November 2013.
You, H., & Zhang, X. (2009). Financial disclosure complexity and investor underreaction to 10-K information. Review of Accounting Studies, 14(4), 559–586.
Yu, F. (2005). Accounting transparency and the term structure of credit spreads. Journal of Financial Economics, 75(1), 53–84.
Ziebart, D., & Reiter, S. (1992). Bond ratings, bond yields and financial information. Contemporary Accounting Research, 9(1), 252–282.
We are grateful for helpful comments from Brian Burnett, Christine Cuny (FARS Discussant), Paul Fischer, Eric Holzman, Jeff McMullin, Kenneth Merkley, Patrick Hopkins, Brady Twedt, Hal White, and workshop participants at Brigham Young University, Florida State University, Massachusetts Institute of Technology, University of Connecticut, University of Miami, University of Missouri, and the Ohio State University, and the 2014 Financial Accounting and Reporting Mid-Year Meeting. We also thank Feng Li for his generosity in providing PERL code for the textual analysis measures used in this project and Bonsall et al. (2016) for Bog readability scores. Finally, we thank Bradley Carrico, Adam Hadley, Eric Holzman, Jessie Watkins, Anna Wellman and Barrett Wheeler for research assistance.
About this article
Cite this article
Bonsall, S.B., Miller, B.P. The impact of narrative disclosure readability on bond ratings and the cost of debt. Rev Account Stud 22, 608–643 (2017) doi:10.1007/s11142-017-9388-0
- Narrative disclosure
- Bond ratings
- Cost of debt capital
- Plain english