부동산PF 금융의 위험과 보상 관계 연구- 담보인정가액 비율(LTV)과 총 대출비용(all-in-cost) 스프레드를 중심으로 -

Banks and thrifts in private sector take up majority share in real estate development finance that decision making process and practices need to be studied and revealed to set up more effective housing policy. Nonetheless, real estate finance is an area that has not been fully studied in empirical manner in Korea as statistical data are not available with confidential information policy of financial institutions. That's why we focus on building up dataset of and analyze decision practices in empirical manner in this area. In this paper, we reviewed 130 real estate development finance deals of a national bank in Korea, from 2011 to 2017, and build up statistical regression models to see whether loan exposure risk represented with loan-to-value ratio(LTV) is related to loan pricing represented with all-in-cost spread. In the whole dataset, It is revealed that the former affects the latter, precedented in previous studies. Second-step analysis for 8 sub-groups gives more detailed picture of the impact of LTV on all-in-cost spread. Seemingly, LTV affects individual differential of the spread rather than inter-group differential. In only 4 of the 8 sub-groups, LTV gives significant impact on the spread, with same sign as in the whole dataset, which implies that LTV can be more significant on the spread under a certain condition.