Skip to main content

The roles of rating outlooks: the predictor of creditworthiness and the monitor of recovery efforts

Review of Quantitative Finance and Accounting volume 55, pages 1063–1091 (2020)Cite this article

Abstract

Using a comprehensive U.S. rating sample from S&P between 1981 and 2015, we examine the information content, responsiveness to credit risk and recovery efforts associated with rating outlooks. We find that rating outlooks (and credit watches) have important information content and are significantly associated with creditworthiness, measured by expected default frequency. More importantly, we show that by assigning negative outlooks, credit rating agencies induce some issuers to exert recovery efforts to prevent subsequent downgrades. The findings support the theoretical prediction of Boot et al. (Rev Financ Stud 19(1):81–118, 2006) that credit rating actions serve as a coordination mechanism between rating agencies and issuers.

This is a preview of subscription content, access via your institution.

Access options

Buy single article

Instant access to the full article PDF.

USD 39.95

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Fig. 1
Fig. 2
Fig. 3
Fig. 4

Notes

  1. S&P uses the term “credit watch” where Moody’s adopts the term “rating under review” for issuers/ratings that are placed on the watchlist. As both terms essentially mean the same rating procedure, we simply use the term “credit watch” (CW) to refer to both.

  2. Boot et al. (2006) conjecture that credit watch procedures can induce firms to exert recovery efforts to avoid downgrades and rating outlooks should be considered as a refinement of firms’ ratings (see Footnote 15 in their paper). In our empirical study, we examine whether both CWs and OLs are associated with future default risk (the role of refinements of credit ratings) and whether negative OLs are associated with subsequent improvements in financial strength of the firm (the role of inducing recovery efforts).

  3. CRAs are criticized for assigning unreasonably high ratings to structured products before the financial crisis. With respect to corporate bonds, Hung et al. (2017) show that, because of slow revisions of credit ratings by CRAs, the firms facing downgrades issue more debt in order to take advantage of current higher ratings.

  4. Hill and Faff (2010), and Alsakka and ap Gwilym (2012) analyze OLs and CWs of sovereigns, not corporate issuers. Finnerty et al. (2013) focus on the impact of credit rating announcements on credit default swap spreads.

  5. According to Manso (2013), “Rating agencies are supposed to provide an independent opinion on the credit quality of issuers. However, if market participants rely on credit ratings for investment decisions, then credit ratings themselves affect the credit quality of issuers”. This is called “feedback effects” of credit ratings.

  6. Starting from Holthausen and Leftwich (1986), the studies in credit ratings exclude the credit events overlapped with corporate news surrounding the event window (contaminated events). The relevant corporate news is normally selected from some sources like The Wall Street Journal. However, Galil and Soffer (2011) argue that the practice of excluding contaminated events leads to selection-bias, and that market responses by the uncontaminated sample are underestimated. Given that the credit events of OLs and CWs may not have strong market reaction, we do not exclude the contaminated events so as to better estimate the impacts of OLs and CWs.

  7. There are several steps used to calculate the expected default frequency according to the Merton model. The first step is to estimate the volatility of equity return from historical stock prices, and to calculate the face value of the debt in a firm as the sum of current liabilities and 50% of the long-term debt. The key step is to estimate the volatility of the firm’s asset value and the market asset value based on Merton model from the equity volatility, the face value of debt, risk-free rate and time to maturity [Equation (2) and (5) in Bharath and Shumway (2008)]. The expected default frequency is one minus the cumulative probability that the firm value is higher than the face debt value [Equation (7) in Bharath and Shumway (2008)]. More details can be found in Bharath and Shumway (2008).

  8. We appreciate the referees’ suggestions to adopt the change of EDF as the dependent variable to test H2. In our earlier version, EDF was used as the dependent variable and we found that OL or CW assignment significantly affects the level of expected default risk.

  9. Liu and Sun (2017) find that the firms receive negative watches and subsequent downgrades have better improvements in the financial strength than the firms with direct downgrades. However, the improvements of the firms with negative credit watches in their paper are measured after they are downgraded. Hence, the improvements may not be attributed to the recovery efforts in response to credit watch assignments. Our Hypothesis H3 explores the recovery efforts that the issuers undertake to avoid potential downgrades after negative OLs, which is more relevant to Boot et al. (2006).

  10. We choose S&P’s data for several reasons. First, S&P has the longest history of OL and CW credit actions. The OLs and CWs sample by S&P is more comprehensive than the sample from Moody’s or Fitch. Second, Hill and Faff’s (2010) study of sovereign OLs and CWs shows that S&P tends to be more active, provide more timely rating assessments, and offer more new information than Fitch and Moody’s. Also, other existing related studies have used mainly Moody’s data. Our study can complement the current credit rating literature, especially the information content of OLs and CWs. We believe that our major conclusions still apply to the samples from Moody’s or Fitch.

  11. In addition to negative and positive views of OLs and CWs, CRAs also give stable outlooks (11,391 actions), developing outlooks (330 actions) and developing watches (642 actions). We do not report the frequencies of these actions in the table as CRAs argue that these actions do not indicate specific directions of future rating changes.

  12. The number of negative outlooks (5651) for the test of the recovery effort hypothesis is less than the total number of negative outlooks (6336) reported in Table 1. The reason is that some negative outlooks are resolved immediately in the quarter that the issuers are put on the OL list. These outlooks are deleted as the recovery efforts cannot be detected from the changes of quarterly financial statement variables.

  13. The improvements include the increases of interest coverage ratio and ROA, as well as the decreases of the leverage ratio, short-term debt to total debt and capital expenditure.

  14. It is worth noting that our CW sample is much larger than the previous studies. The sample in Chung et al. (2012) has totally 1911 negative watches and 963 positive watches; Kiesel and Kolaric (2018) analyze 1526 watchlist placement announcements; the number of negative watches is 611 in Chan et al. (2011); and the numbers of firms with negative watches and positive watches are 104 and 23 in Hand et al. (1992).

  15. As the dummy variables downgrade (upgrade), negative (positive) OL and negative (positive) CW are included in the same regression model, one potential concern is that some of these variables may be highly correlated. Therefore, we have computed the correlations among rating change, credit watch and rating outlook in the sample. The correlation matrix of these variables shows that the correlations between each pair of these variables are less than 6%. Also, the variance inflation factors (VIFs) of the model with all these variables are less than 1.5. Therefore, the collinearity is not a serious problem in Eq. (1). The results remain similar if each variable of rating actions is included in the model one at a time.

  16. We thank the referees for their suggestions on robustness tests.

References

  • Alsakka R, ap Gwilym O (2012) Rating agencies’ credit signals: an analysis of sovereign watch and outlook. Int Rev Financ Anal 21:45–55

    Google Scholar 

  • Altman EI, Rijken HA (2004) How rating agencies achieve rating stability. J Bank Finance 28(11):2679–2714

    Google Scholar 

  • Altman EI, Rijken HA (2007) The added value of rating outlooks and rating reviews to corporate bond ratings. Paper presented at the financial management association meeting, Barcelona

  • Bannier CE, Hirsch CW (2010) The economic function of credit rating agencies: what does the watchlist tell us? J Bank Finance 34(12):3037–3049

    Google Scholar 

  • Bao MX, Liu Y (2018) Level 3 assets and credit risk. Rev Pac Basin Financ Mark Policies 21(01):1850003

    Google Scholar 

  • Bharath ST, Shumway T (2008) Forecasting default with the Merton distance to default model. Rev Financ Stud 21(3):1339–1369

    Google Scholar 

  • Bonsall S, Koharki K, Neamtiu M (2015) The effectiveness of credit rating agency monitoring: evidence from asset securitizations. Account Rev 90(5):1779–1810

    Google Scholar 

  • Boot AW, Milbourn TT, Schmeits A (2006) Credit ratings as coordination mechanisms. Rev Financ Stud 19(1):81–118

    Google Scholar 

  • Campbell JY, Hilscher J, Szilagyi J (2008) In search of distress risk. J Finance 63(6):2899–2939

    Google Scholar 

  • Chan H, Faff R, Hill P, Scheule H (2011) Are watch procedures a critical informational event in the credit ratings process? An empirical investigation. J Financ Res 34(4):617–640

    Google Scholar 

  • Chou T (2013) Information content of credit ratings in pricing of future earnings. Rev Quant Finance Account 40(2):217–250

    Google Scholar 

  • Chung KH, Frost CA, Kim M (2012) Characteristics and information value of credit watches. Financ Manag 41(1):119–158

    Google Scholar 

  • Driss H, Massoud N, Roberts GS (2019) Are credit rating agencies still relevant? Evidence on certification from Moody’s credit watches. J Corp Finance 59:119–141

    Google Scholar 

  • Finnerty JD, Miller CD, Chen R (2013) The impact of credit rating announcements on credit default swap spreads. J Bank Finance 37(6):2011–2030

    Google Scholar 

  • Fitch Ratings Ltd. (Fitch) (2005) Special report: Fitch sovereign rating outlook and watch study. https://www.fitchratings.com/site/search?content=research&filter=REPORT%20TYPE%5EHeadlines%5ERating%20Action%20Commentary. Accessed 30 Nov 2016

  • Galil K, Soffer G (2011) Good news, bad news and rating announcements: an empirical investigation. J Bank Finance 35(11):3101–3119

    Google Scholar 

  • Hamilton DT, Cantor R (2004) Rating transition and default rates conditioned on outlooks. J Fixed Income 14(2):54–70

    Google Scholar 

  • Hand JR, Holthausen RW, Leftwich RW (1992) The effect of bond rating agency announcements on bond and stock prices. J Finance 47(2):733–752

    Google Scholar 

  • Hill P, Faff R (2010) The market impact of relative agency activity in the sovereign ratings market. J Bus Finance Account 37(9–10):1309–1347

    Google Scholar 

  • Hill P, Brooks R, Faff R (2010) Variations in sovereign credit quality assessments across rating agencies. J Bank Finance 34(6):1327–1343

    Google Scholar 

  • Hirsch CW, Krahnen JP (2007) A primer on rating agencies as monitors: an analysis of the watchlist period. http://dx.doi.org/10.2139/ssrn.967463. Accessed 30 Nov 2016

  • Holthausen RW, Leftwich RW (1986) The effect of bond rating changes on common stock prices. J Financ Econ 17(1):57–89

    Google Scholar 

  • Hull J, Predescu M, White A (2004) The relationship between credit default swap spreads, bond yields, and credit rating announcements. J Bank Finance 28(11):2789–2811

    Google Scholar 

  • Hung CD, Banerjee A, Meng Q (2017) Corporate financing and anticipated credit rating changes. Rev Quant Financ Account 48(4):893–915

    Google Scholar 

  • Kedia S, Rajgopal S, Zhou X (2014) Did going public impair Moody׳s credit ratings? J Financ Econ 114(2):293–315

    Google Scholar 

  • Kiesel F, Kolaric S (2018) Measuring the effect of watch-preceded and direct rating changes: a note on credit markets. Rev Quant Financ Account 50(2):653–672

    Google Scholar 

  • Liu AZ, Sun L (2017) Revisiting post-downgrade stock underperformance: the impact of credit watch placements on downgraded firms’ long-term recovery. J Account Audit Finance 32(2):271–299

    Google Scholar 

  • Liu AZ, Subramanyam K, Zhang J, Shi C (2017) Do firms manage earnings to influence credit ratings? Evidence from negative credit watch resolutions. Account Rev 93(3):267–298

    Google Scholar 

  • Liu L, Luo D, Han L (2019) Default risk, state ownership and the cross-section of stock returns: evidence from China. Rev Quant Finance Account 53(4):933–966

    Google Scholar 

  • Löffler G (2013) Can rating agencies look through the cycle? Rev Quant Finance Account 40(4):623–646

    Google Scholar 

  • Luo H, Chen L (2019) Bond yield and credit rating: evidence of Chinese local government financing vehicles. Rev Quant Finance Account 52(3):737–758

    Google Scholar 

  • Manso G (2013) Feedback effects of credit ratings. J Financ Econ 109(2):535–548

    Google Scholar 

  • Merton RC (1974) On the pricing of corporate debt: the risk structure of interest rates. J Finance 29(2):449–470

    Google Scholar 

  • Moody’s Investors Service (Moody’s) (1998) Special comment: an historical analysis of Moody‘s watchlist. https://www.moodys.com/creditfoundations/Default-Trends-and-Rating-Transitions-05E002. Accessed 30 Nov 2016

  • Moody’s Investors Service (Moody’s) (2004) Special comment: rating transitions and defaults conditional on watchlist, outlook and rating history. https://www.moodys.com/creditfoundations/Default-Trends-and-Rating-Transitions-05E002. Accessed 30 Nov 2016

  • Moody’s Investors Service (Moody’s) (2005) Special Comment: Rating transitions and defaults conditional on rating outlooks revisited: 1995–2005. https://www.moodys.com/sites/products/DefaultResearch/2004700000425139.pdf. Accessed 30 Nov 2016

  • Moody’s Investors Service (Moody’s) (2008) Special comment: Moody’s credit transition model: a summary of the watchlist/outlook extension. https://www.moodys.com/creditfoundations/Default-Trends-and-Rating-Transitions-05E002. Accessed 30 Nov 2016

  • Moody’s Investors Service (Moody’s) (2016) Rating action: Moody’s confirms Microsoft’s Aaa senior unsecured rating; outlook changed to negative. https://www.moodys.com/credit-ratings/Microsoft-Corporation-credit-rating-698200. Accessed 20 Sept 2018

  • Moody’s Investors Service (Moody’s) (2017) Rating action: Moody’s affirms Microsoft’s Aaa senior unsecured rating; outlook changed to stable. https://www.moodys.com/credit-ratings/Microsoft-Corporation-credit-rating-698200. Accessed 20 Sept 2018

  • Norden L, Weber M (2004) Informational efficiency of credit default swap and stock markets: the impact of credit rating announcements. J Bank Finance 28(11):2813–2843

    Google Scholar 

  • Salvadè F (2018) Is less information better information? Evidence from the credit rating withdrawal. Rev Quant Finance Account 51(1):139–157

    Google Scholar 

  • Standard & Poor’s Financial Services LLC (S&P’s FS) (2005) Credit trends: creditwatch and ratings outlooks: Valuable predictors of ratings behavior. https://www.spglobal.com/ratingdirect. Accessed 30 Nov 2016

  • Standard & Poor’s Financial Services LLC (S&P’s FS) (2011) Guide to credit rating essentials: what are credit ratings and how do they work? https://www.spglobal.com/ratings/en/about/understanding-ratings. Accessed 30 Nov 2016

  • Standard & Poor’s Financial Services LLC (S&P’s FS) (2016) RatingsDirect global credit portal. https://www.spratings.com/en_US/login. Accessed 30 Nov 2017

  • Standard & Poor’s Financial Services LLC (S&P’s FS) (2017) RatingsDirect: S&P global ratings definitions. https://www.spglobal.com/ratingdirect. Accessed 30 June 2018

  • Vassalou M, Xing Y (2004) Default risk in equity returns. J Finance 59(2):831–868

    Google Scholar 

  • Xia H (2014) Can investor-paid credit rating agencies improve the information quality of issuer-paid rating agencies? J Financ Econ 111(2):450–468

    Google Scholar 

Download references

Acknowledgements

Poon acknowledges a research Grant (LU13501214) from the General Research Fund (GRF), Research Grants Council, Hong Kong. Poon and Shen acknowledge a Business Faculty Research Grant (DB14B2) from Lingnan University, Hong Kong, and a research Grant (UGC/FDS14/B20/16) from the Faculty Development Scheme (FDS), Research Grants Council, Hong Kong. Shen acknowledges a research Grant (P0030199) from the Hong Kong Polytechnic University. The authors thank Dorla Evans for her editorial work and Cheung Chun-Kit for his dependable research assistance.

Author information

Authors and Affiliations

  1. Department of Finance and Insurance, Lingnan University, 8 Castle Peak Road, Tuen Mun, Hong Kong

    Winnie P. H. Poon

  2. Department of Building and Real Estate, The Hong Kong Polytechnic University, Hung Hom, Kowloon, Hong Kong

    Jianfu Shen

Authors
  1. Winnie P. H. Poon
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Jianfu Shen
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Jianfu Shen.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Appendices

Appendix 1: Variable definitions

Variable code Variable name and brief explanation
CAR Cumulative abnormal return over 3-day event window
BHAR Buy-and-hold abnormal return over 3-day event window
EDF Expected default frequency in a month of a firm, calculated from Merton’s model
ΔEDF Change of EDF from previous month to current month of a firm
CONFIRM Dummy variable; it equals to 1 if a firm received rating confirmation after negative OL assignment
DOWNGRADE Dummy variable; it equals to 1 if a firm was downgraded in the month
UPGRADE Dummy variable; it equals to 1 if a firm was upgraded in the month
NEGOL Dummy variable; it equals to 1 if a firm was placed on negative OL list in the month
POSOL Dummy variable; it equals to 1 if a firm was placed on positive OL list in the month
NEGCW Dummy variable; it equals to 1 if a firm was placed on negative CW list in the month
POSCW Dummy variable; it equals to 1 if a firm was placed on positive CW list in the month
RATING Numerical value of credit rating at the end of the month or quarter
INVESTGRADE Dummy variable; it equals to 1 if a firm’s rating is above BB + 
INTCOV Interest coverage in a quarter; = EBITDA/interest expense
LEV Leverage ratio in a quarter; = total debt/total assets
STDTTD Short-term debt to total debt ratio in a quarter; = short-term debt/total debt
ROA Return on assets in a quarter; = net income/total assets
CAPEX Capital expense in a quarter; = capital expenditures/total assets
RECOVERY A set of recovery effort variables including ΔINTCOV, ΔLEV, ΔSTDTTD, ΔROA and ΔCAPEX
DINTCOV Recovery effort variable; it is the increase of average interest coverage from pre-OL to post-OL assignment period
DLEV Recovery effort variable; it is the decrease of average leverage from pre-OL to post-OL assignment period
DSTDTTD Recovery effort variable; it is the decrease of average short-term debt to total debt from pre-OL to post-OL assignment period
DROA Recovery effort variable; it is the increase of average ROA from pre-OL to post-OL assignment period
DCAPEX Recovery effort variable; it is the decrease of average capital expenditure from pre-OL to post-OL assignment period
OPROFIT Operating profitability in a quarter; = operating income before depreciation/total assets
MTB Market to book ratio in a quarter; = market value of assets/total book value of assets; market value of assets is the sum of market equity and total debt
TANG Tangibility in a quarter; = net property, plant, and equipment/total assets
SALES The natural logarithm of sales in a quarter
SIZE The natural logarithm of total assets in a quarter
CASH Cash ratio in a quarter; = cash and marketable securities/total assets
LEVGVOL Volatility of leverage during the past eight quarters
OPROFITVOL Volatility of operating profitability during the past eight quarters

Appendix 2: Standard and Poor’s long-term issuer credit ratings and their assigned numeric values

According to S&P, “An S&P Global Ratings issuer credit rating is a forward-looking opinion about an obligor’s overall creditworthiness. This opinion focuses on the obligor’s capacity and willingness to meet its financial commitments as they come due.” (S&P’s FS 2017, pp. 6–7).

Ordinal/numeric value assigned to each rating category S&P’s long-term issuer credit ratings  
22 AAA Investment grade
21 AA+
20 AA
19 AA−
18 A+
17 A
16 A−
15 BBB+
14 BBB
13 BBB−
12 BB+ Speculative grade or non-investment grade
11 BB
10 BB−
9 B+
8 B
7 B−
6 CCC+
5 CCC
4 CCC−
3 CC
2 SD
1 D

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Poon, W.P.H., Shen, J. The roles of rating outlooks: the predictor of creditworthiness and the monitor of recovery efforts. Rev Quant Finan Acc 55, 1063–1091 (2020). https://doi.org/10.1007/s11156-019-00868-7

Download citation

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11156-019-00868-7

Keywords

  • Rating outlook
  • Information content
  • Credit risk
  • Recovery effort

JEL Classification

  • G14
  • G20
  • G24