AbstractA credit rating compresses a large variety of information that needs to be known about thecreditworthiness of the issuer of bonds and certain other financial instruments. Credit ratingagencies provide the borrower’s ability to repay the debt obligation (creditworthiness)information to investors thereby serving as information intermediary between lenders andborrowers.The information contained in the credit rating is debated worldwide. One school of thoughtbelieves that change in credit rating does not carry any new information to the market. The otherschool of people believes that change in credit ratings do carry valuable information which isobserved in the abnormal returns of the stock prices.In general, there is considerable evidence that downgraded rating announcementsprovide new information seen through statistically significant abnormal returns, while theupgraded rating announcements do not provide any new information and is already embedded inthe stock prices.
In this paper, we examine whether rating change announcements signal new informationto Indian stock market in general and banking sector in particular using event studymethodologyKey Words: Credit rating changes, Abnormal Returns, Event Study1. INTRODUCTIONThe countries have opened up ; the world has become small. Localization is replaced byglobalization, the horizons of markets are expanding. Investments are crossing the boundaries,Collaboration at international level has become common and the very business thinking isundergoing metamorphosis. The Growth of any country to a greater extent depends upon theefficient ; vibrant economic system. The financial system which facilitates the transformation offunds from Surplus spending units to Deficit spending units started growing in geometricprogression from 1970?s (The era of LPG) at the global level and from 1990?s onwards at thenational level. All the four major components of Financial System – Financial Institutions,Financial Instruments, Financial Markets, and Financial Services – have been growing rapidly.Financial institutions create Financial Services as their output for consumers, they aremotivated to serve needs in society and earn profit just as the manufacturing firms that producetangible goods.
Manufacturing firms use physical assets to produce tangible products, where asfinancial firms use financial instruments to offer their services. Thus Financial Instrumentsrepresent paper wealth that substitutes for real wealth. To cater to the needs of diverse categoriesof investors, various tailor made financial instruments were created ; introduced.
Increasingvarieties of instruments and the growth of financial market created not only the opportunities forthe investors but also complexities in the market, the reputation of the issuer Co. is no longer aguarantee to the quality of Financial instrument which the Co. issues.
Hence, general caution tothe customers. Caveat emptor (buyer beware), is also applicable to financial market investors.To demystify the complexity of financial instruments in the financial market, to facilitatethe investors to take informed investment decisions and to enable the issuer companies /Institutions to mobilize the required funds. Various credit rating agencies emerged across theglobe.
Credit RatingA credit rating compresses a large variety of information that needs to be known aboutthe creditworthiness of the issuer of bonds and certain other financial instruments. Credit ratingagencies (CRAs) provide the borrower?s ability to repay the debt obligation (credit worthiness)information to investors thereby serving as information intermediary between lenders andborrowers. The CRAs thus contribute to solving principal agent problems by helping lenders”pierce the fog of asymmetric information that surrounds lending relationships and helpborrowers emerge from that same fog” (White, 2001).
While, on the other hand, CRAs areheavily criticized for not providing proactive, reliable and credible information in the form ofratings during the Asian Crisis, Subprime crisis, Satyam, Lehman brothers etc.The information contained in the credit rating is debated worldwide. Studies by Weinstein(1977), Wakeman (1978, 1990), Zaima and McCarthy (1988), Pinches and Singleton (1978);Creighton, Gower and Richards (2007) ; Mohindroo (2008) shows that CRAs have no specialinformation, while they summarize publicly available information and rating changeannouncements convey no new information to the market. On the other hand, studies by Ingramet al., (1983), Hand, Holthausen, ; Leftwich, (1992), Dichev and Piotroski (2001), observedabnormal returns in stock prices owing to rating change announcements, and concluded thatCRAs deliver valuable information to the market. Also, the rating agencies declare that theyreceive inside information and rating is a means of communicating significant facets of suchinformation to the stock holders, without exposing detrimental details to the opponents (Rao ;Sreejith, 2013). In general, there is considerable evidence that downgraded rating announcementsprovide new information seen through statistically significant abnormal returns, while theupgraded rating announcements do not provide any new information and is already embedded inthe stock prices.
ScopeThe study focuses on Indian Credit rating agencies (CRISIL, CARE, ICRA, FITCH,BRICKWORKS and SMERA) and it is restricted to study the bond rating announcements ofbanking sector.2. LITERATURE REVIEW2.1 Literature on the Impact of credit rating changes on stock prices in Indian ContextRao and Sreejith (2013) examined the impact of credit ratings by all five credit rating agencies(CRISIL, ICRA, CARE, Fitch and Brickwork) on equity returns in India during the period 1stJanuary, 1999 to 31st March, 2013.
The authors employ event study methodology. Abnormalreturns were computed using Mean Adjusted Model, Market Adjusted Model andConditional Risk Adjusted Model (Standard Market Model) and yielded similar results. T testis used to test the significance of the abnormal returns.
The study revealed that downgrades had aconsiderable negative impact and upgrades had negligible positive impact.Chandrashekar and Mallikarjunappa (2013) studies the impact of bond rating on Indian stockmarket for the period 1998 to 2005. The results show statistically insignificant abnormal returnassociated with the bond down grades, small but insignificant positive abnormal returns forupgrades and concludes that bond upgrades and downgrades do not convey any importantinformation to the market.
Chandrashekar and Mallikarjunappa (2013) examines the reaction of stock returns to theinitial bond rating and concludes that returns associated with the rating events are insignificant,unlike the prior studies which showed stock prices react negatively to the announcement ofdowngrades of bond ratings, while weaker positive excess bond and stock returns are found forupgrades.Lal and Mitra (2011) examines the effects of rating changes announcements on share prices inIndia using event study methodology during the time period 1 April 2002 to 31 March 2008. Thestudy found that rating upgrade or downgrade does not come as a surprise to the investors so asto impact the pricing significantly.
But at the same time investors react moderately for upgradesand downgrades are received more negatively by investors with significant negative abnormalreturns.Rao and Ramachandra (2004) evaluate the response of stock prices and volumes to bond ratingchanges in Indian capital market. They found that stock price incorporates the factors that lead torating revisions. They also report that upgrades are received cautiously by the investors with nosignificant abnormal returns where as downgrades are perceived as bad news by investors withsignificant negative abnormal returns.2.
2 Literature on the impact of credit rating changes on stock prices in Global contextHui, Nuttawat and Puspakaran (2004) studied the effects of Credit Rating Announcements onShares in the Swedish Stock Market for the period February, 1992 to February, 2003. Theauthors employ event study methodology using EVENTUS package and cumulative averageabnormal returns (CAAR) were computed based on a GARCH (1,1) model. The study showedthat there is no significant share price reaction for rating assignments, positive outlooks andaffirmations announcements following credit rating announcements in both the long-term andshort-term. However, there is significantly positive (negative) market reaction to the upgrade(downgrade) announcements.Taib, Iorio, Hallahan and Bissoondoyal-Bheenick studies the impact of corporate bond ratingannouncements on stock prices of companies in UK and Australia.
The authors use daily datafrom 1st Jan 1997 to 31st Dec 2006 and find a significant effect to downgrades both in the UKand Australia markets. Also, finds evidence that stock price reaction during upgradeannouncements is weak.Taib, Iorio, Hallahan and Bissoondoyal-Bheenick examines the impact of corporate bond ratingrevisions on UK stock market during the period January 1997 to December 2006 and alsocompares alternative techniques for estimating abnormal returns. The authors employ four returngenerating models (the conventional market model, the quadratic market model, the downsidemarket model and higher order downside market model) to analyze the stock price impact. Theresults show that there is an announcement effect for bond downgrades, but not upgrades.Despite differences in approach, all four return generating models produce similar results whenused in the event studies.Barron, M.J.
, Clare, A.D, and Thomas, S.H. (1997) examines the impact of new credit ratings,credit rating changes and Credit Watch announcements during the period 1984 to 1992 on UKcommon stock returns. We find significant negative excess returns around the date of adowngrade and positive returns close to the date of a positive CreditWatch announcement. Newratings, whether short or long-term, have no significant impact on returns.Creighton, Gower and Richards (2007) study the response of bond yield spreads and equityprices to credit rating changes in the Australian financial markets between January 1990 and July2003. The empirical evidence reveals that bond spreads appear to widen in response to ratingsdowngrades and contract with upgrades and equity prices tend to fall on days of downgrades andrise on days of upgrades.
Choy, E. Y. W, Gray, S. F, and Ragunathan, V. (2006) examines the impact of rating changesdone by two agencies Moody?s and S;P on the Australian stock market between 1989 and 2003.
The results indicated a significant and negative impact for downgrades that are anticipated andunanticipated, and an insignificant impact for upgrades.Abner and Andrea investigates the Latin American stock price reaction for a rating change orCredit Watch announcement. The authors employ event study methodology to analyze stockmarket reaction to rating change news in four major Latin American economies: Argentina,Brazil, Chile and Mexico. The results show similar results to those previously observed in theliterature, where in the impact is quite significant for rating downgrades but less relevant forrating upgrades and Credit Watches.
Cross section regressions indicate absolute change in thenumber of notches for downgrades is a significant variable that best explains the impact ratingchanges announcements have on stock prices in these countries.Jorion and Zhang (2007) examined the impact of rating changes on senior unsecured corporatebonds of US issued during 1996 Jan to 2002 May. The results support the previous findings thatthe downgrades have a greater impact than upgrades. Their results also showed that thedowngrades of speculative grade bonds increase the default probability and cost of capital tocompany, while downgrades on investment grade bonds create ripple like fluctuations in defaultprobability and cost of capital. Therefore downgrades in speculative issues more heavily impactprice changes than downgrades in investment grade issues. They also find a significant averageCAR for upgrades of speculative grade issues.The review highlights that there is limited studies in Indian context and there is no clearevidence pertaining to the impact of rating change announcements of bank?s stock prices. Basedon the above context, the following objectives, hypothesis is developed.
3. OBJECTIVES AND HYPOTHESIS OF THE STUDY1. To study the impact of credit rating changes (Upgrades and Downgrades) on the bank?sstock prices2. To investigate whether there are any significant abnormal returns (whether positive or negative)related to the credit rating change announcements.HYPOTHESISH0: Credit rating announcements has no impact on bank?s stock pricesCredit rating announcements has an impact on bank?s stock pricesH0: There is no significant abnormal return associated with credit rating announcementsCAARt = 0There is significant abnormal return associated with credit rating announcementsCAARt ? 04. SAMPLE AND DATAThe rating changes by CRISIL, CARE, ICRA, FITCH, BRICKWORKS, SMERA are extracted from2007 to 2015 April using Ace equity database.
Our initial sample consisted of 28 events (17 upgrades and11 downgrades). The sample was checked for other major events (such as merger or acquisition,divestment, buyback of shares, stock split etc) during the period, if found, the event is said to becontaminated. After applying the above criteria, the final sample consisted of 26 events (15 upgrades and11 downgrades).
Daily stock prices are taken from BSE historical prices and Yahoo finance portals foreach of the event from day – 280 to + 30. The Benchmark Index considered for the study is BSESENSEX.5. METHODOLOGYThe methodology used here is event study. The basic idea is to find the abnormal return attributableto the event being studied by adjusting for the return that stems from the price fluctuation of the market asa whole.
(Ronald and Bernard 1995).5.1 Event WindowThe literature about market reaction to rating announcement does not have a consensus in theevent window definition. Dichev and Piotroski (2001) check different event windows: 0 (date of theannouncement) to 3 months, to 6 months, to 1 year, to 2 years and to 3 years after the announcement.Jorion and Zhang (2007) checked the event window of 1 year before to 1 year after the announcement.
Ee(2008) tested different windows: 1 day before to 1 day after, 3 days before to 3 days after, 50 days beforeto 26 days before, 25 days before to one day before. (Abner de Pinho ; Andrea Maria, 2013). However,the choice of the window is arbitrary and “should not be too long, because it would be encompassingother events, generating biases, nor too small, because it would be failing to fully capture the abnormalityin prices” (Camargos ; Barbosa, 2003).
Similarly, Brown and Warner (1985) uses eleven day event period (– 5 to + 5) to analyse daily stockreturns. Wansley. J. w.
, Lane. W. R and Yang H.
C., (1987) and Dodd Peter (1980) used – 50 to +50event period to examine the effect of merger announcement on stock return. Chandrashekhar R andMallikarjunappa T (2013) use 61 day event period (-30 to + 30), Vaithanomsat (2001) uses -10 to + 10,Sehgal (2013) uses -20 to + 20, Goh and Ederington(1999) uses -60 to + 60, Lal and Mitra (2011) uses -30 days to + 30 days.In this study, we have used 61 day event window, 30 days before (-30) and thirty days after (+30) the dateof rating change announcement (0).5.2 Calculating expected returns and Abnormal returnsMarket adjusted model developed and suggested by Sharpe (1963) is used to calculate theexpected return. The prior studies use extensively the market model to determine the expected return onspecific asset, given the return on market and the two parameters of the market model (alpha and beta ofthe security). Market model is based on the fact that the most important factor affecting stock returns ismarket factor and it is captured in the market model in the form of the parameters.
The market model for calculating expected return is given by the following regression equation:E (Rjt) = ?j + ?j RmWhere,E (Rjt) is the expected return on security j,?j is intercept. (Mean return over the period not explained by the market).Rm is the expected market return,?j is the slope of the regressionDaily returns/actual returns are calculated as below:-Rjt = ln (Pjt/ Pij-1)WhereRjt is the daily return on security „j? on day „t?.Pit is the daily adjusted price of the security „i? at the end of period„t?.Pit-1 is the daily adjusted price of the security „i? at the end of period„t-1?.Rmt = ln(I.
t/ I t-1)Where,Rmt is the daily return on market index on day „t?. I.t and I t-1 is the closing index value on day „t? and „t-1?, respectively.The abnormal return is the difference between the actual return on day t and the expected return i.e.,ARjt = Rjt – E(Rjt)Where,ARjt is the abnormal returnAbnormal returns represents that part of the return which is not predicted and is, therefore, an estimate ofthe change in firms share price on that day which is caused by the announcement of credit rating.
Abnormal returns are averaged across firms to produce AARt for day „t? using the following formula,Where, N is the number of firms in the sample. Finally we calculate the cumulative average abnormalreturn (CAAR) for the event period. The cumulative average abnormal return represents the average totaleffect of the event across all firms. Where,5.3 Parametric Significance testParametric t-statistic is used to examine the statistical significance of AARs and CAARs. It is tested at 5percent level of significance and appropriate degree of freedom. It is given by5.
3.1 The t Test Statistic for AARsThe statistic is given byt = AARt/ ?AARt (Standard error of AAR)Where AAR =average abnormal return, ?AARt = standard error of average abnormal return.The standard error is calculated by using following formula.SE = ?/?nWhere, S.
E = standard error, ? = standard deviation, n = number of observation.5.3.2 The t Test Statistic for CAARsThe statistic is given byt = CAARt/ ?CAARt (Standard error of CAAR)SE = ?/?nWhere, S.E = standard error, ? = standard deviation, n = number of observationChart1: Average Abnormal Returns of event window for upgradesAARs are negative for 19 days and positive for 11 days before the Announcement of the eventand negative for 15 days after the announcement and positive for 16 days after the Announcement of theevent.
During the whole event period for upgrades, AARs are negative for 34 days and positive for 27days. AARs are significant for only 1 day before the event in the event window. This shows that the stockmovement persists even after the change in credit rating signaling that credit rating upgrade has notincreased the bank?s stock price.. AARs are negative for 14 days and positive for 16 days before the Announcement of the eventand negative for 8 days after the announcement and positive for 23 days after the Announcement of theevent.
During the whole event period for downgrades, AARs are negative for 22 days and positive for 39days. AARs are positive for majority of the days in the event window and statistically insignificant formajority of the days (56 of 61 days). Hence, we accept the null hypothesis which states that change incredit rating has no impact on bank?s stock prices. Also, abnormal returns are found which are statisticallyinsignificant.7.
CONCLUSIONThe study examines the impact of bond rating changes on bank?s stock prices. Analysis revealsthat AARs are negative and statistically insignificant for majority of the days in case of ratingupgrades and AARs are positive and statistically insignificant for majority of the days for ratingdowngrade announcements. The study also reveals that credit rating announcements have nospecial information, while they summarize publicly available information and rating changeBITM-DMS. A Conference on: The Future of Management. 29 th Aug, 2015www.shreeprakashan.com [email protected],Vol-IV,Issue –VIII, Aug,2015.
Page 269announcements convey no new surprises to the market and hence, we conclude that there are nosignificant abnormal returns associated with rating change announcements. However, our resultscannot be generalized as it based on a small sample
