Firing the wrong workers

Choosing which worker to remove

During the 2007-2014 period, youth unemployment increased substantially in many European countries. In Spain, the increase was from around 20% to more than 53%, and was still as high as 35% at the end of 2018. In part, this large increase in unemployment was caused by firms that were forced to fire workers because of their inability to obtain credit during the financial crisis.

Youth unemployment is very costly both for individuals and for society, as it reduces the ability of young people to acquire skills and experience, and it can have long-term negative effects on their career paths.

In their Barcelona GSE Working Paper (No. 1058) “Firing the Wrong Workers: Financing Constraints and Labor Misallocation,” Andrea Caggese, Vicente Cuñat and Daniel Metzger study how financing constraints affect the firing decisions of firms.

They show that when firms are forced to lay off workers because of financial problems, they are more likely to “wrongly” fire recently hired and less experienced workers than more tenured workers. This is the case even when young workers are highly skilled and with more growth potential. Conversely, when firms fire workers for reasons unrelated to financial problems, they are more likely to retain these promising recently hired workers and fire more mature ones instead.

These decisions are “wrong” from the society standpoint, but firms with financial difficulties have the incentive to take actions which save costs in the short term even when they damage the long run productivity gains for both the firm and the worker. The short-term costs of keeping a young worker include the severance pay of firing and older worker or the cost of waiting for the young worker to fully develop its potential.

The model

The authors illustrate these ideas by developing a stylized partial equilibrium model of a firm that makes hiring and firing decisions regarding heterogeneous workers: short-tenured and long-tenured workers. Financial frictions make credit-constrained firms discount future cash flows more severely than unconstrained firms. The model is built on the following three key features:

  • Wages are rigid and do not fully adjust to compensate for fluctuations in the productivity of workers
  • Recently hired workers have more upside potential than long-tenured ones
  • Firing costs increase with worker’s tenure in the firm

In this context, financial frictions affect not only the overall level of firm employment but also the optimal mix of short-tenured and long-tenured workers.

The model provides four implications regarding employment and firing decisions conditioning on the financing capacity of the firm. First, the more financially constrained a firm is, the more likely it fires a short-tenured worker, and the less likely it fires a long-tenured one. Second, the more financially constrained a firm is, the younger the tenure profile of its labor force is. Third, given a negative demand shock, a financially unconstrained firm reduces employment by firing relatively more long-tenured workers than it would during normal times. Fourth, given a negative demand shock, a more financially constrained firm fires workers with shorter tenures.

Empirical test

The authors test those predictions using matched employer-employee data about the whole active population of Sweden between 1990 and 2010 from the Swedish Companies Registration Office (Bolagsverket). The exceptional quantity and quality of information available in this data set makes it ideal for the analysis. To measure firm’s financing constraints the authors use three discrete rating categories provided by the main rating company in Sweden. To identify the causal effect of an exogenous change in financing constraints (i.e., independent of firms’ characteristics) the authors use two further identification strategies: a regression discontinuity design (RDD) and a within-firm-year estimator. RDD is an econometric technique that consist of comparing individuals just above and just below a particular threshold. In this case, rating agencies have a continuous measure of firm’s risk but the rating is discrete. Therefore, firms near the thresholds with almost equal risk measure might be assigned in different credit ratings. Finally, the authors use shocks to the exchange rate of a firm-specific basket of currencies to identify negative demand shocks.

Results

Using the empirical strategy described above, they find results in line with the model predictions. In normal times, financially constrained firms have on average between 1.5% and 4.3% more short-tenured workers than non-financially constrained ones, and they also tend to fire these workers more frequently than the other firms. Moreover, after an exogenous demand shock, non-financially constrained firms fire more long-tenured workers and fewer short-tenured ones relative to normal times. Taken together, the results show that in normal times, the firing rate of short-tenured workers is three times higher than that of long-tenured workers in unconstrained firms but five times higher in constrained firms. During negative-shock times, the firing differential narrows down to a factor below two for unconstrained firms, but the convergence of firing rates is absent for constrained firms.

The analysis of the expected productivity of retained and fired workers confirms that while unconstrained firms are more able to retain good and promising workers and use periods of negative shocks to fire the least productive ones, constrained firms are not able to do so. As a result, financing constraints increase labor duality within firms with highly protected long-tenured workers and more exposed short-tenured ones. This has important negative consequences for the misallocation of resources across firms, the process of human capital accumulation, and the career paths of young workers.

This publication was part of the research project “Financial Constraints, Hiring and Firing, and the Careers of Young Workers” (2015-2018), funded by the La Caixa Foundation and is forthcoming in the Journal of Financial Economics.