This dissertation studies different Unemployment Insurance (UI) systems and how critical design and institutional features affect individual behavior. In the first chapter, I analyze the UI system in Chile, whose benefits are primarily financed by work- ers’ own savings in Individual Savings Accounts (ISA). In 2002 Chile implemented a mandatory savings system for the formal labor sector. Every worker accumulates funds to be withdrawn in future unemployment spells, accompanied by a Solidarity Fund (SF) that may provide benefit extensions in case of low savings balances. Using administrative records of the Chilean formal labor market, I exploit the SF extension’s eligibility conditions to identify its effect on nonemployment duration. I estimate a Regression Discontinuity Design (RDD) that uses the eligibility for SF extension as an instrument for its take-up. Results show that an additional potential monthly installment financed by the SF extension increases nonemployment durations by 11.4 days, similar to what the literature documents for other developing economies.
The second chapter builds on the previous one by empirically examining the im- pact of an additional monthly installment of UI benefits financed by ISAs. I implement a dynamic panel data model with individual fixed effects controlling for time-invariant unobserved heterogeneity. Using this empirical strategy, I find that nonemployment durations, on average, increase by 5.1 days when the unemployed have an additional monthly payment financed by their savings. Finally, I combine these two reduced-form outcomes to perform a decomposition of the disincentive effect of a more generous UI system funded by general taxes. Job search effort is depressed via a substitution effect when UI benefits increase (moral hazard). Still, there is also an incentive to exert less effort because the household’s consumption is sensitive to cash-on-hand (liquidity effect). I extend the existing literature by proposing a ratio pertinent to a system based on ISA that measures the liquidity effect over the total disincentive effect. Results show that around one-third to one-half of the negative impact of a UI extension on job search effort can be attributed to moral hazard, which this design of UI system eliminates.
In the 1990s, Germany saw a massive rise in unemployment of workers in their late 50s compared to younger workers. In the third chapter (joint with Matthew Gudgeon, Johannes Schmieder, Simon Trenkle, and Han Ye), we show that a large share of this increase can be explained by the interaction of UI and the retirement system, where UI benefits affect labor supply by inducing individuals to leave employment. We show considerable bunching in UI inflows at age thresholds that allow for using UI as a bridge to early retirement. The bunching mass moves as the age threshold moves due to reforms of the UI system. To quantify the impact of this channel on labor supply, we use our reduced-form evidence to estimate a dynamic lifecycle model of labor supply that endogenizes unemployment and retirement transitions. We show that the increase in potential benefit durations in the late 1980s increased unemployment rates for workers aged 55-59 by around three percentage points (or about a 30 percent increase), playing a significant role in the large increases in unemployment rates for older workers.
| Identifer | oai:union.ndltd.org:bu.edu/oai:open.bu.edu:2144/45277 |
| Date | 28 October 2022 |
| Creators | Guzman Pinto, Pablo Ignacio |
| Contributors | Schmieder, Johannes F. |
| Source Sets | Boston University |
| Language | en_US |
| Detected Language | English |
| Type | Thesis/Dissertation |
Page generated in 0.0014 seconds