255 that for infrequent originators, the subordination level of securitization with the OBS method is slightly lower than for the FLT method. There are no significant differences between the FLE and FLT method for infrequent originators. 6.5 Empirical Results The results of our first regression model, Equation (6.3), are shown in Table 6.5. In Table 6.5, we report the results of the ordered logit regressions with Credit Rating as the dependent variable and Risk Retention Methods as key independent vi na rTi aa bb ll ee s. T6h. 6e ar en sdu 6l t .s7 .o If no ut hr esseec ot anbdl erse, gwr ees sr ei opno rmt ot hd ee l r, eEsquul tast i oo fn t (h6e. 4o)r, daer ree sdh loowg int regressions with Rating Discrepancy as the dependent variable and Risk Retention Methods as key independent variable. The results of our third regression model, Ethqeuaistsiounan(c6e.5), are provided in Tables 6.8 and 6.9. In Tables 6.8 and 6.9, we use Spread as the dependent variable, and Risk Retention Methods as the key independent variable. In all our models, the independent variable Risk Retention Methods is of categorical nature. Where we report the values for only VES, OBS and FLE, as the FLT is the omitted variable. 79 6.5.1 Risk retention methods and credit ratings We first analyze if the credit ratings of CRAs differ for securitization tranches with dpirfofevridenedt riinskTarbelteen6t.5io,nwmheertehods. The results of our ordered logit regressions are Credit Rating is our dependent variable and the Risk Retention Methods our key independent variable. Panel A presents the results ot rfaonuc rh ems owd ei tlhu fsri enqg utehnet f u( cl lo tl ur amnnc h2e)- l ea vnedl si na fmr epql eu e(nc ot l (ucmo lnu m1 )n a 3n )d ot hr ieg irneas tuol rt ss . f oI nr P( caonl eulmBn, w2 )e, Ds hBoRwS (tchoel ur me snu l3t s) af onrd tFhiet cchr e( dc oi tl urma t ni n4g )s soefpMa roaot de lyy’.s ( c o l u m n 1 ) , S & P 80 79 We use the first loss tranche method as the baseline as this method is used most in our sample. 80 The number of observations for KBRA are too limited to do statistical analyses. Chapter 6 - Risk Retention in the European Securitization Market
RkJQdWJsaXNoZXIy MTk4NDMw