Inefficient Regulation: Mortgages versus Total Credit
Artashes Karapetyan, Jens Soerlie Kvaerner, Maximilian Rohrer
Review of Finance, Volume 28, Issue 1, January 2024, Pages 311–351, https://doi.org/10.1093/rof/rfad023
Regulators have implemented macro-prudential policies in response to rising household debt and real estate prices. One common borrower-based instrument is a loan-to-value (LTV) cap. We quantify the willingness-to-pay (WTP) for unregulated debt available for bypassing the LTV cap.
Identifying how an LTV cap affects the demand for unregulated debt used for equity downpayment is challenging. First, the available datasets lack information on how people finance the equity proportion of a home purchase. Even administrative datasets with a complete breakdown of all sources of credit do not contain information about how people spend their credit. Second, macro-prudential regulations often come during a volatile period, with overheated debt markets and strong house price development. Aggregate economic conditions, or other non-observable factors, may change concurrently, driving both house prices and credit demand simultaneously.
We solve these challenges by combining a unique regulatory change with Swedish real estate data. To illustrate the empirical strategy, imagine a homebuyer with insufficient equity after the LTV cap to buy a particular home. Assume an identical home exists; the only exception is that it comes with building-level unregulated debt taken by the cooperative (so-called co-op debt). As a result of the co-op debt, this home has a lower transaction price and hence requires a lower equity downpayment. Because co-op debt is not part of the LTV calculations, the homebuyer can circumvent the LTV cap by buying the dwelling with co-op debt.
We find that an LTV cap can increase housing costs for many homebuyers as they switch to unregulated debt to circumvent the LTV cap. Our estimates suggest that the average homebuyer pays 7.3 SEK to avoid 1 SEK of equity down payment. Additional analyses reveal that financially weaker households drive the willingness-to-pay estimate.Using a two-period model, before and after the LTV cap, we quantify the shadow cost of the LTV regulation. Our focus is on affected individuals, those who, after the LTC cap, only can afford a unit with a high-debt co-op. The shadow cost is the percentage change in their value function due to the price increase of the high-debt co-op unit. We derive a simple analytical expression for the lower bound of the shadow cost, estimated at 0.5%. By comparing localities by financial resources, we find that the shadow cost of the LTV regulation was several times greater in poorer areas.