Cross-Sectional Variation in Stock Price Reaction to Bond Rating Changes: Evidence from India (Empirical Results-7)

Cross-Sectional Variation in Stock Price Reaction to Bond Rating Changes: Evidence from India (Empirical Results-7)Thus, firm characteristics do impact the relationship between rating changes and stock returns. In case of less informationally efficient firms, wealth redistribution effect dominates in downgrades while signalling effect generally dominates in upgrades. For more informationally efficient firms, the results are not so clear. Further, stock price reaction is stronger for low P/B and low profitability firms (which are expected to be informationally less efficient) viz a viz their counterparts for both rating upgrades as well as downgrades. Similar conclusions however, cannot be drawn while classifying firms on other characteristics.
Table 3 shows the results of regression between the post – event CAR and the independent variables both in case of downgrades and upgrades.
In case of downgrades, two of the variables namely the pre – event CAR (CARi) and magnitude of rating change (NUM_GRADE) are found to significantly affect the post -event stock returns. The pre – event CAR negatively affects the post – event CAR. The negative sign implies that post – event abnormal returns are high in cases where pre -event abnormal returns are low and vice – versa. Thus, where the downgrade is anticipated by the investors, the post – event abnormal returns are low whereas, the abnormal returns are larger where the downgrade is a surprise.
The results also show that the post – event CAR is negatively affected by the magnitude of rating change. This means that the more the magnitude of rating change the lower the abnormal returns after announcement. This is not in line with expectations. Usually studies find a direct relationship between the two variables. A possible explanation may be that a large magnitude of rating change is easier to anticipate as compared to a small rating change. It seems that in case of downgrades, the signalling effect tends to become stronger for a larger magnitude of rating change which relatively offsets the wealth redistribution effect. This may be the reason why more the number of grades changed, the less the impact on post – event returns. Other variables (BUSINESS_CYCLE, SPEC, FALLEN) were not found to be significant.
In case of upgrades, none of the variables was found to be significant.

Table 3. Results of Regression Between Post – event CAR and Pre – event CAR, Magnitude of Rating Change, Business Cycle Dummy, Movement within Speculative Grade dummy and Movement from Investment to Speculative Grade Dummy

Downgrades Upgrades**
Beta t Beta t
(Constant) 0.025 0.517 0.008 0.188
CARi -0.326* -2.162 0.646 1.677
NUM GRADE -0.081* -3.392 -0.121 -1.072
BUSINESS CYCLE -0.031 -0.491 0.024 0.438
SPEC -0.007 -0.056 -0.074 -0.634
FALLEN 0.075 0.648
Adjusted R Square 0.337 0.138

Tags: , , , ,