This thesis is composed of three chapters on different topics of macro-finance. Thematically, they are linked by their focus on household credit constraints - be they exogenous regulatory constraints, as in Chapter 1, or endogenous constraints as in Chapters 2 and 3.

The first chapter is joint work with Juan Castellanos Silvan and Gonzalo Paz-Pardo. We propose a joint model of the aggregate housing and rental markets in which both house prices and rents are determined endogenously. The key part of the model is that households can choose their housing tenure status (renters, homeowners, or landlords) depending on their age, wealth, and income. We show how the reliance on heterogeneous household landlords generates an upward sloping supply curve for rented accommodation. With the model in hand, it can be used to study the introduction in Ireland in 2015 of macroprudential policies that limited loan-to-value (LTV) and loan-to-income (LTI) ratios of newly originated mortgages. The introduction of stringent LTV and LTI ratios mitigates house price growth, but increases rents and reduces homeownership rates. As a result, middle-income households are negatively affected.

The second chapter stays with the theme of housing markets but introduces endogenous default. The main innovation is to consider delinquency as an important stage of default. In some European countries, mortgage delinquency rates are much higher than foreclosure rates. The stock of delinquent mortgages peaked at 9% of GDP in the Eurozone periphery and the average length of a delinquency spell was over 10 months. This fact has been largely neglected in the macro-finance literature. This chapter provides a framework for understanding why high levels of persistent mortgage delinquency can emerge as an equilibrium outcome during a housing market crisis. Banks tolerate delinquency because the gain to foreclosing is less than the option value of continuing with the delinquent loan. By nesting a straightforward game between debt-distressed households and banks within a quantitative macro-housing model, the option to enter delinquency is shown to significantly attenuate (by roughtly half) the consumption drop during a crisis. Importantly, I show that the ability of households to gain insurance through delinquency is significantly impacted by the degree of recourse available to banks upon foreclosure. The model features realistic lifecycle dynamics, tenure choice between renting and owning, endogenous liquidity in the housing market and defaultable, long-term debt.

The third and final chapter leaves housing markets behind but sticks with the theme of default, credit risk, and interest rates. Using the Brazilian administrative credit registry data with the universe of all consumer loans originated by banks in the country from 2013 to 2019, the chapter documents high borrowing interest rates, which vary systematically with individual characteristics. In particular, even after controlling for several observable individual attributes - such as income, debt, occupation, and default probabilities, low-income individuals pay higher interest rates than high-income borrowers. We quantitatively analyze a consumer credit market with these characteristics observed in Brazil in a model with endogenous default, and where consumers face idiosyncratic income and expenditure shocks. We perform counterfactual analyses to assess the impact of different financial reforms on borrowing rates, consumption inequality, consumption insurance, and welfare. We find that reforms aiming to reduce intermediation costs and bank market power have sizeable average and distributional welfare implications.