260 of skin-in-the-game. This might incentivize the originators to do ‘do their jobs bo re itgt ei nr a’, tfoo rr s eaxraema pl l loew, be yd itmo ap pr opvl yi nmg ut hl t ei pi rl es mc r ee tehnoi nd gs taon rde mt a oi nn si tkoi rni -ni ng -st thaen- dg aa mr des, . wA es sWe ee kd ot os ion bv ey sat ni gaal tyez ii nf gi nivf ei ns tvoersst opresr cdeeivviea tt eh ei ns et hme ei rt hpor di csi nt og (bme ee aqsuuarl el yd rbi sykiys sourannoc te. spread) between the different risk retention methods. In Table 6.8, we report the results of the ordinary least square regressions with Spread at issuance as our dependent variable and the Dummy Risk Methods as the ki nevyeisnt do re sp teenndde nt ot rveadr iuacbel et h. Oe iur ri sr seus ua nl tcses shporwe at dh ac to fmo rp tarraendcthoetshwe iFt LhTa mV eEtShmo de, twh oi tdh, a coefficient of -27.45 (t-stat= -2.40), significant at the 5% level. Hence, tranches wl o iwt he rt, hwe i Vt hE 2S 7m. 4e5t hboads irsepc oe ii vnet so, nt haavne trha og es ea ws pi trhe tahdeaFt Li Ts smu aent hc eo dt h. Sa itmi si lraerml y,afrokratbhl ye OBS risk retention method, compared to the FLT method, investors reduce the spread at issuance, with a coefficient of -36.08 (t-stat= -4.61), column (5). Thus, investors also demand a significantly lower issuance spread, with 36.08 basis points on average, for tranches with the OBS method, relative to the FLT method. Wr e emfoi nv de coounr sci sotnetnrto rl se si un l tcso sl ui gmn ni fsi c (a1n)t taot t(h4e) . 1 %F o lre Fv eLlEf owr eb odtoh nvoatr ifai nb dl e sc ownhsei snt ewnet (highly) significant results throughout columns (1) to (5). Hence, it appears that it nh ve e sotroi gr si nlaotookr baepypol ni eds t thhee c rVeEdSi t arna dt i nOg BaSn dmaedt hj uosdt ttho e irre ti as si nu asnkcien -si pn r- teha ed- gwa hmeen. Wa phpiel ea rws et hf iant dwnhoi l he i cgohnl yt rsoi gl l ni ni fgi cfaonr t t ihme pcarcetdoi tn rpa triinc ign, gi nf voer stthoer sF fLi En dmtehteh oVdE. SS oa n, idt OorBigSinmaettohroadnldesinsvreissktoyr.and potentially better aligning the interests between the 84 84 In column (3) of Table I, Appendix II, we repeat the analysis of Table 6.8 by including the following additional control variables for robustness purposes; STS Compliant, Single Originator, GDP Growth Rate and Country of Risk c m o o n d tr e o l. ls. We show that our results of Table 6.8 remain robust when including several additional controls to our
RkJQdWJsaXNoZXIy MTk4NDMw