MERROUCHE Ouarda

The purpose of this thesis is to examine the effectiveness of Loan-to-Value (LTV) and Debt-to-Income (DTI) regulations, as well as the regulatory arbitrage that results from the implementation of these two measures. This study is motivated by the fact that our understanding of regulation is still in its infancy in this area, despite the non-negligible importance of the role of these devices in terms of financial stability. To this end, our study deals with the Korean case, which presents an appropriate framework from an empirical point of view, on the one hand, because of the variation of these regulations according to the geographical region, and on the other hand, because of the availability of a relatively longer history of the implementation of these measures. In this study, we find that regulations are effective in curbing mortgage growth. However, their impact could be significantly destroyed by regulatory arbitrage. Specifically, two possible leakages are explored: i) the migration of mortgages to less regulated areas and ii) the compensation of the mortgage shortage by unregulated loans. In addition, we show that regulation could increase the credit risk of household loans instead of reducing it. This unexpected result could be due to their impact extending even to unregulated loans: regulation could have negative, unintended effects on the credit risk of unregulated loans, which is transmitted through the evolution of the housing auction market . moreover, these negative effects seem to dominate the expected positive outcome from regulated loans. Moreover, this study confirms that nonbank lending could be a loophole due to a regulatory gap between the banking and nonbank sectors. Thus, closing this gap should be done to combat the regulatory arbitrage that exists there, although it is certainly not the final solution in the sense that the "cat-and-mouse game" between regulators and the regulated can emerge in a different form at any time.