Essays in household finance

Carella, Agnese (2021) Essays in household finance. PhD thesis, London School of Economics and Political Science.

Text - Submitted Version Download (4MB)

Abstract

In the first chapter, I exploit the rebranding of a mortgage lender, under a more salient name and in some Italian provinces, to empirically analyze households’ choice behaviour in response to brand popularity. Loan-level data on both the universe of newly originated mortgages and the offer rates suggest that (1) brand awareness reduces the equilibrium price of residential mortgage contracts and (2) the reduction mainly reflects consumers’ selection into cheaper products. Comparing contracted rates with concurrent market offers from the main online mortgage broker in Italy, I show that households’ reallocation towards less expensive choices is unlikely to reflect pure substitution behaviours induced by brand persuasion. In fact, my findings support the informative view that brand awareness improves consumers’ search and allows them to obtain more convenient deals, with an overall decrease in price dispersion. In the second chapter, we back empirical findings with theoretical foundations, and quantify the impact of brand name on consumers’ search costs and borrowers’ transition across lenders within a life-cycle model. The model is well calibrated to replicate main features of the Italian household sector and to match the level of dispersion in the price of mortgage products encountered in the data. Model calibrations imply a 330 euro reduction in consumers’ search costs due to brand popularity, and roughly a 10 percentage points increase in the share of households that move to cheaper lenders. The treatment effect of brand name on price dispersion is in line with the empirical evidence in chapter one. In the third chapter, we use information on mortgage supply available from the online broker to assess trends in lending strategies of Italian banks. We document that (1) riskier mortgages (high loan-to-value, low borrower’s income, and long maturity) are offered by fewer banks that charge higher rates; (2) keeping the level of risk constant, online banks offer better price conditions than traditional ones. We then use online offer rates to nowcast bank-level official rates (MIR). By relying on both regression analysis and machine learning algorithms (random forest), we show that online prices have a high predictive content for the equilibrium price of fixed-rate mortgages, and allow for a very timely assessment of changes in household financing conditions.

Item Type:	Thesis (PhD)
Additional Information:	© 2021 Agnese Carella
Library of Congress subject classification:	H Social Sciences > H Social Sciences (General) H Social Sciences > HC Economic History and Conditions H Social Sciences > HG Finance
Sets:	Departments > Finance
Supervisor:	Paravisini, Daniel
URI:	http://etheses.lse.ac.uk/id/eprint/4278

Actions (login required)

Record administration - authorised staff only