This thesis is a collection of three essays which study how unemployment insurance (UI) can be provided in the most beneficial way for society. In particular, a great deal of this work aims to advance the scientific understanding regarding the following question: how generous unemployment benefits should be in order to maximize social welfare? In other words, how much unemployment insurance do we need?

Chapter 1 and 2 are introductory to the thesis and to the specific topic. The first of them introduces the specific questions addressed in this thesis and presents the main results. Chapter 2 provides a more throughout review of the related literature and highlights how each of this thesis’ essays contribute to advancing the scientific understanding on the topic.

Chapter 3 presents the first of the three essays. It studies the existence of a causal link between the availability of potential unemployment benefits for employed workers and the duration of their employment spells. After discussing few straightforward reasons why and how UI may affect employment duration, I apply a regression kink design to address this question using linked employer-employee data from the Brazilian labor market. Exploiting kinks in the Brazilian UI schedule, I find a statistically and economically significant effect of benefit level on the duration of employment spells at the lower end of the skill distribution. Surprisingly, the results for these workers indicate that the elasticity of employment duration to benefit level is positive and as large as 0.5. To assess the economic relevance of this result, I generalize the reduced welfare formula from Chetty (2008) to deal with this effect on employment duration and show that this elasticity is as relevant for welfare as the elasticity of unemployment duration to benefit level.

Chapter 4 contains the second thesis’ essay. It first exploits a “bonus” policy providing low-income workers with cash grants in Brazil to study the effect of liquidity provision on unemployment outcomes. Based on a RDD, I find that granting unemployed workers with a bonus equal to half of their previous monthly earnings decreases the probability of exiting unemployment within 8 weeks by around 0.65%. Second, by exploiting the UI potential duration schedule, I find that granting workers with an extra month of unemployment benefits decreases the same outcome by 1.9%. Then, theoretical results from Landais (2014) are used to combine these estimates and disentangle liquidity and moral hazard effects of UI. Based on these, I estimate the liquidity-to-moral hazard ratio in Brazil to be as large as 98%, similarly to values previously found in the US. It suggests that, contrary to common belief, providing UI in developing countries with large informal labor markets may be welfare increasing.

Chapter 5 is composed by the third and last essay. This work investigates how unemployment insurance (UI) affects unemployment inflow. By using administrative data from the Brazilian Labor Market and applying a Regression Discontinuity Design, I show that UI significantly increases the lay-off hazard rate at the minimum eligibility requirement for benefits. Then, I provide a learning model with work effort which is able to explain this finding and the hazard rate profile over time by relating unemployment benefits to work effort and lay-off hazard rates. The model supports the hypothesis that UI may increase unemployment inflow because it undermines work effort. Then, personnel data on absenteeism supporting this prediction is provided.

The main conclusion from these three essay are summarized and related to each other in Chapter 6.

Additional Metadata
Promotor A.M. Pacces (Alessio) , G. Zanella (Giulio)
Publisher Erasmus University Rotterdam
Persistent URL
Series EDLE - The European Doctorate in Law and Economics programme
de Britto, D.G.C. (2015, December 8). Essays on Unemployment Insurance. EDLE - The European Doctorate in Law and Economics programme. Erasmus University Rotterdam. Retrieved from

Additional Files
cover-image.PNG Cover Image , 10kb