Últimas Publicaciones

Revisiting the Omitted Price Bias in the Estimation of Production Functions (joint with Xulia González and Saul Lach). Journal of Industrial Economics, forthcoming.

Impact of Spanish renewable support scheme reforms on the revenues of photovoltaic power plants (joint with Fidel C. Rodriguez) Utilities Policy, 80.

Gender imbalance in housework allocation: a question of time? (joint with Begoña Alvarez) Review of Economics of the Household, 17.

See Google-Scholar

Working papers

Parameter Stability as a Diagnostic Tool for Production Function Specification.

The control function approach remains the dominant method for estimating production functions. In this framework, unobserved productivity is recovered by inverting a proxy function that relates it to observed state variables and flexible inputs. A key specification test proposed by Levinsohn and Petrín (2003) relies on the invariance of parameter estimates across different nonparametric proxies for flexible inputs. Given the widespread use of Cobb-Douglas production functions in empirical industrial organization, we propose a related test within the parametric inversion framework of Doraszelski and Jaumandreu (2013, hereafter DJ). Specifically, we demonstrate that under the identification assumptions of DJ, the recovered productivity process—and hence the structural parameter estimates—should be invariant to the choice of flexible input used for inversion. Consequently, systematic discrepancies across inversion strategies can serve as a diagnostic tool for detecting violations of the model’s underlying assumptions. Link

Billing protocols and the Identification of Price Effects in Consumer Expenditure Data

Consumer expenditure surveys collect the information on recurrent household expenditures, such as electricity, natural gas, and telecommunications services, from the most recent bill available at the time of the interview. As a result, the data reported by these surveys depend on the rules governing how consumption, expenditure, and prices are recorded on invoices. Focusing on electricity, we show that billing protocols may undermine the identification of price effects in empirical demand analysis because of non-classical measurement errors and temporal misalignment between prices and consumption. As an illustration, we revisit Labandeira, Labeaga, and Teixid ´o (2022), who evaluate an electricity price reform using consumer expenditure survey data. Our findings suggest that, once billing protocols are taken into account, the variation exploited by their design no longer identifies the causal effect of prices on consumption. For example, a placebo regression using their specification returns a significant ”price effect” on consumption for a billing-protocol change that altered no electricity prices. More broadly, we emphasize that recurrent expenditure data cannot be interpreted in isolation from the billing systems that generate them.
Link.

Comment on Majors Reforms in Electricity Pricing

In a paper published in the May 2022 issue of The Economic Journal, Labandeira, Labeaga, and Teixidó assess the impact of a Spanish electricity price reform implemented in August 2013 on household electricity consumption. Their difference-in-differences identification strategy rests on two assertions: (i) that, between September 2011 and March 2014, no other relevant events affected electricity consumption or prices in Spain; and (ii) that the reform increased the average electricity price while simultaneously reducing the marginal price. In this comment, we provide evidence that both assertions are not consistent with contemporaneous developments in Spain’s electricity sector or the institutional design of the 2013 reform. Furthermore, we show that the price of potatoes is not a valid instrument for the price of electricity.
Omitted regulatory changes.
Online Appendix.
STATA Appendix.

Work in progress

VAT pass-through

This study examines the extent to which changes in VAT—both increases and decreases—are passed through to cinema ticket prices using a database of more than 9,000,000 observations.

Unequal Burdens: Gender Differences in Mental Load

We examine gender asymmetries in mental load using as an approximation daylight saving time changes.