Beyond the LTV Ratio: Lending Standards, Regulatory Arbitrage, and Mortgage Default

Abstract

Booming house prices are historically correlated with loose lending standards. Nonetheless, in Spain the loan‐to‐value (LTV) ratio failed to capture imbalances during the last housing boom. Using loan‐level data from millions of mortgages we show that inflated collateral valuations, used by banks as a mechanism to circumvent regulation, distorted the informative value of LTV, and masked the accumulation of risk. We identify that regulation relying upon a single measure is more prone to suffer from regulatory arbitrage, and that the optimal policy mix varies over the financial cycle. Overall, our study provides useful insights for the implementation of borrower‐based measures.

Micro-Assessment of Macroprudential Borrower-Based Measures in Lithuania

“…The insight that secondary mortgages are riskier than single loans is also in line with those of Kelly and O'Toole (2018) and Galán and Lamas (2019), who find multi-loan borrowers and second-home properties to have a higher chance of default. Our policy message on restricting LTVs for secondary mortgage takers is compatible with the agent-based model conclusion of Tarne and others (2022), which states that differentiation of BBM limits to different classes of borrowers may be welfare-improving.…”

“…This is also supported by other research papers, including that of Kelly and O'Toole (2018), who show that multi-loan borrowers have a higher default risk, even after controlling for the DSTI ratio and other borrowerloan characteristics. Adding to this, Galán and Lamas (2019) find that mortgages on second-home properties have a higher PD. Further, secondary mortgages are riskier, because they are not taken out by first-time buyers, who are known to be low-risk borrowers, as found by Kelly and others (2015), Mihai andothers (2018), andothers (2019), among others.…”

“…However, since micro level data availability and often quality are major obstacles for researchers, the use of such method is not as prevalent as might be expected. Authors that use this approach include Kelly and others (2015), Kelly and O'Toole (2018), Mihai and others (2018), Galán and Lamas (2019), Nier and others (2019), de Haan and Mastrogiacomo (2020), Andries andothers (2021), andKukk (2021). They typically find that BBMs, or indicators that are restricted by BBM parameters, significantly and often nonlinearly reduce the probability of default.…”

“…Conversely, if the DSTI option had been chosen, it would have affected around 15 percent of borrowers -roughly the size of mortgage overflow -whose DSTI is on the righthand side of the distribution (see Figure 4b). Stricter regulation of DSTI (*) -headline DSTI or the stressed DSTI * -would be more targeted to containment of risk, since the ratios are strongly related to the customer's PD, as shown by multiple authors, including Mihai and others (2018), Galán and Lamas (2019), and Nier and others (2019). Using a comprehensive framework, Gross and Población (2017) find that "DSTI limit is more effective than LTV cap from the perspective of reducing household risk parameters while implying less pronounced macro feedback effects.…”

Mortgage Lending Valuation Bias Under Housing Price Changes and Loan-to-Value Regulations

“…Earlier studies have found appraisals to be upwardly biased (Calem et al, 2021) and that a need to refine costly unsecured debt (Reite et al, 2022), a steady increase in real estate prices (Galán and Lamas, 2023), and competition between banks (Conklin et al, 2020) can contribute to such bias. We hypothesize that government-mandated LTV policies may increase incentives to upwardly bias valuations by systematically choosing the less conservative valuations available.…”