"The Price of Leverage: Learning from the Effect of LTV Constraints on Job Search and Wages" , with Gazi Kabas, March 2023 [Link to SSRN]

Does household leverage matter for workers’ job search, matching in the labor market, and wages? Theoretically, household leverage has been shown to have opposing effects on the labor market through, among others, a debt-overhang and a liquidity constraint channel. To test which channels dominate empirically, we exploit the introduction of a macroprudential borrowing restriction that exogenously reduces household leverage in Norway. We study home-owners who lose their job and find that a reduction in leverage raises wages by 3.3 percentage points after unemployment. The mandated restriction of leverage enables workers to search longer for jobs, and thereby find positions in firms that pay higher wage premia and switch to new occupations and industries. We observe no evidence that a greater use of credit during unemployment drives the extended job search. The positive effect on wages is persistent and more pronounced for workers who are more likely to benefit from improved job search, such as young people. Our findings contribute to the debate on the costs and benefits of policies that constrain household leverage and show that such policies, while primarily aiming at enhancing financial stability, have other positive effects such as improved labor market outcomes.

An early version of this paper circulated as "Household Leverage and Labor Market Outcomes Evidence from a Macroprudential Mortgage Restriction".

Executive Labor Market Frictions, Corporate Bankruptcy and CEO Careers, with Morten Grindaker and Andreas Kostøl [paper at SSRN]

Featured on Harvard Law School bankruptcy roundtable 

CEOs of large firms filing for bankruptcy are more likely to exit the executive labor market after bankruptcy and experience substantial compensation losses (Eckbo et al., 2016). While the fear of reputational scarring can lead to lower risk-taking and manifest itself as lower rates of entrepreneurship and job growth, the mechanisms through which bankruptcy affects CEO careers are not well understood. In this paper, we examine the effect of "random bankruptcy" decisions on small and medium-sized business CEOs’ careers. By random, we mean job separation for reasons unrelated to a firm or CEO quality but rather through a court’s bankruptcy decision. We control for the unobserved ability of bankrupt and non-bankrupt CEOs by using randomly assigned judges’ propensity to liquidate firms as an instrument. We then combine our sample of CEOs with administrative records containing granular information on income, wealth, new employers and job titles. Our results show that bankrupt CEOs find new employment quickly, but that a large share exits the executive labor force. On average, bankruptcy reduces CEOs’ variable income components. While the net present value of CEOs’ loss of future capital income equals more than 60 percent of annual pre-bankruptcy income, we observe no effect on wage income. We find that displaced CEOs are more likely to reallocate to new industries and new geographic areas, suggesting that managerial skills are portable. We explore how the income and employment effects of bankruptcy vary with industry conditions. Consistent with the executive labor market using bankruptcy as a noisy signal of managerial ability, we find the displacement effect is stronger when industry conditions are good. Our evidence is consistent with the presence of information frictions that could entail important social costs. 

What Do12 Billion Card Transactions Say About House Prices and Consumption?, with Knut Are Aastveit,  Jesper Böjeryd, Magnus Gulbrandsen and Ragnar Juelsrud , draft coming soon 

Rising house prices can drive up consumption and investment while falling house prices can depress the labor market and the economy at large. Using close to the universe of debit card payment data linked with relevant administrative background data for the full Norwegian population, we investigate how changes in house prices affect household consumption. We exploit the oil price shock of 2014-2015 as an instrument and compare the impact of changing house prices between Norwegian government workers, who face essentially no change in unemployment or income risk, and private sector workers in the regions of Norway that are highly dependent on oil production. Durable goods such as cars and furnishing have the strongest negative response to falling house prices while semi-durables such as clothes are less affected. Essential goods such as food and beverages have an opposite response, suggesting a substitution away from luxury consumption. The effects we find are highly concentrated among wealthier households. Older, retired, households do not respond to the house price shock.  

Credit Information, Discipline and Strategic Behavior by Firms, with Marieke Bos and Paola Morales, 2018 

Credit bureaus in many countries restrict the length of time that negative credit information of firms can be retained. The large variance in retention times across countries illustrates the lack of consensus on the optimal memory of negative information. By exploiting a variation in retention time of negative information for firms, provided by the introduction of the Habeas Data law in Colombia, we are able to analyze the causal link between the length of credit bureaus’ retention time and the subsequent behavior of lenders and borrowers. The law was ratified in 2009 and prohibited institutions in Colombia from accessing the entire credit history of borrowers. Since then, negative credit information is observably only for periods determined by the length of the delinquency period. Our results suggest that after the introduction of the Habeas Data law: i) the duration of loan delinquency periods is longer, ii) firms seem to strategically wait until their negative records disappear from the credit bureaus, before switching banks, iii) banks grant loans with higher interest rate spreads, lower collateral requirements, larger loan amounts and longer maturities. In addition, we find empirical evidence of both ex ante and ex post theories of collateral.

An earlier version of this paper circulated as "The Impact of Sharing Credit Information, Evidence from a Quasi-Experimental Variation"

Cross-Border Bank Flows and Bank Risk-Taking: Evidence from European Households, with Dominik Boddin and Daniel te Kaat 

In this paper, we study the consequences of the increase in bank inflows after the ECB's implementation of non-conventional monetary policy in 2014/15 on household lending. For this purpose, we employ granular household survey data matched with supervisory bank data from Germany and estimate difference-in-differences regressions around this increase in bank inflows. We show that the inflow of liquidity from abroad induced more exposed banks, i.e., those with greater initial non-core funding ratios, to increase their consumer loan supply to low-income households. Mortgage lending is largely unaffected by bank flows. We also document that consumer expenditures by more exposed households rise simultaneously. our findings  speak to the capital flow-bank risk-taking nexus by providing new evidence at the household level.