Unemployment Insurance, Strategic Unemployment, and Firm-Worker Collusion

Heterogeneous Impacts on Layoffs of Changes in Brazilian Unemployment Insurance Eligibility Rules

“…There is no evidence of employer-worker collusion among moderate-tenure workers in their second UI request (data subset 4), suggesting that the decrease in the probability of unjustified dismissal might not be associated with the direct collusion channel between workers and firms. 25 This suggests that changes in legislation might affect collusion more when workers have a shorter 24 Estimations considering workers reemployed by the same firm four to 10 months after being laid off as in Van Doornik et al (2018) show that the probability of being rehired by the same employer is about 0.9 pp lower after PM 665 for workers in the treatment group. This is half of the effect estimated by Van Doornik et al (2018), indicating that they overestimate the collusion effect.…”

“…25 This suggests that changes in legislation might affect collusion more when workers have a shorter 24 Estimations considering workers reemployed by the same firm four to 10 months after being laid off as in Van Doornik et al (2018) show that the probability of being rehired by the same employer is about 0.9 pp lower after PM 665 for workers in the treatment group. This is half of the effect estimated by Van Doornik et al (2018), indicating that they overestimate the collusion effect.…”

Unemployment Insurance and Labor Turnover: Evidence From Brazil

“…12 This finding represents a rise in the unemployment inflow caused by non-market conditions and is, therefore, likely to be related to inefficiencies in the labor market, such as: (i) higher spending on redundancy costs; (ii) greater expenditure on vacancy-posting in equilibrium; (iii) less investment in training by employers; (iv) lower firm-specific human capital (FSHC) accumulation by workers. This paper sheds light on the role of UI as a determinant of worker turnover by developing a framework where workers may quit their jobs in 1 Carvalho et al, 2018 andDoornik et al, 2018 exploit an unanticipated change in the Brazilian UI program eligibility criteria implemented in 2015. They use diff-in-diff to estimate the effect of eligibility for UI on the probability of unjustified dismissal using the same data source over similar sample periods [Carvalho et al, 2018uses 2012-2015and Doornik et al, 2018uses 2013-2015.…”

“…This paper sheds light on the role of UI as a determinant of worker turnover by developing a framework where workers may quit their jobs in 1 Carvalho et al, 2018 andDoornik et al, 2018 exploit an unanticipated change in the Brazilian UI program eligibility criteria implemented in 2015. They use diff-in-diff to estimate the effect of eligibility for UI on the probability of unjustified dismissal using the same data source over similar sample periods [Carvalho et al, 2018uses 2012-2015and Doornik et al, 2018uses 2013-2015. Though Carvalho et al, 2018 use a 10% random sample of workers, the results are quite similar, .55-.65pp increase in layoff probability, both consistent with strategic unemployment.…”

“…4 Adding to the list of works that document increases in separation hazard due to UI eligibility, Christofides andMcKenna, 1996 andBaker andRea, 1998 study different changes in the minimum number of weeks worked required to collect UI benefits in Canada, the former in 1986-87 and the latter in 1990, and both conclude that separation is significantly more likely after UI eligibility is met. Also, papers like Fuller et al, 2015 andDoornik et al, 2018 regarding frauds in the UI system make more appealing the reading of these evidence as not a consequence solely of firms behavior. Further evidence on workers behavior responding to the incentives on UI system is provided in the next section by showing the relation of the hazard function to the replacement ratio.…”