Vivian van Breemen

100 Iimn Tpaabctleo3f .d4e, walecsohmopwletxhietyocrdhearreadctleorgisittitcesstosnof model (2), where we measure the Rating Discrepancy between credit ratings provided by Moody’s and S&P on the same CLO, including controls. In column (3), we find that a one standard deviation change in Capital Allocation increases the odds of experiencing an increase in notches difference by 200%, which is significant at a 5% level (Ɀ-stat=2.04). This result suggests that in the precwrhi sei ns pt he reiroed i as ha i ghhi ge hr erra tlienvge ld iosfc rceappai tnacl y ubnedt ewrel yeinn gM tohoed yC’ Ls Oa ntdr aSn&c hP ei .s Hr eopwoervt ee dr, the sign of the coefficient changes with the inclusion of the post crisis dummy (l ickoelluymt on r6e)p. To rhtotswe oC Lr aOtsi nt hg sa twr ei tphoar trhe il ga ht i ev re lcya ph ii tgahl lneuvme l bs earf toe fr nt hoet cchrei ssi sd wi f feerree nl ecses, with the odds of 443%, statistically significant at a 5% level (Ɀ-stat=-2.06). We fdui sr ct hr eepr afni nc dy, bt huat to nt rl ya nact hae1 0c o%u ns itg ni si f inceagnactei vl ee vaenl . dWsei gf innidf i cnaonstilgyn ri fei lcaatnetdr et os url tast ifnogr Log Tranche Size. For completeness, the Appendix32 presents additional ordered logit regressions to study the impact of business cycles on rating discrepancy and complexity in more detail. These findings confirm our results in Table 3.4 that ms pol irtercaot imn gpsl ebxeCt wL Oe se nw Si t&h Pa ahni gdh Me rocoadpyi t’ sa ldaul lroi nc ag t pi oenr iaordesl eosf sal irkeecleystsoi or ne p, aonr td l amr og er er likely during the period after the Global Recession. Overall, the results of Table 3.4 show that of the three complexity characteristics, cr aa tpi int ga l dai lslcorceapt ai onnc yi sdti hs cel oms eods t asti gi snsi uf iac na nc et . f Wa c ht oi lrs tt ht ha te sdee tf ei nr dmi ni ngess vtahl ei d saitzee coafp ti thael aHl yl opcoatthi oe ns i sa 3s , ai . em. , eo ausruorrei goi nf ac lotmh op ul egxhi tt yt, hma toar eh i igmh ep ro Cr tLaOn tcl oy ,mt hp el eyx idt yo wn oo tu lsdu rpepsourl tt in more rating discrepancy must be rejected for the period after the crisis. 32 The table is available in Appendix I. Supplementary Table I for Chapter 3.

RkJQdWJsaXNoZXIy MTk4NDMw