Skip to main content

Searching, Matching, Sorting, and Separating in the Labour Market: Costs, Structure and Consequences of Employment Adjustment in Dynamic Frictional Economies

Final Report Summary - SMSS (Searching, Matching, Sorting, and Separating in the Labour Market: Costs, Structure and Consequences of Employment Adjustment in Dynamic Frictional Economies)

How do recessions affect the age-related structure of the firms’ workforce adjustment? How are employment adjustment costs distributed among workers searching for jobs and firms paying hiring and firing costs? How does labor adjustment influence sorting patterns between employees and employers? These three interrelated questions sum up the research objectives of Dr. Sebastian Buhai’s Marie Curie IOF research project “Searching, Matching, Sorting, and Separating in the Labour Market: Costs, Structure and Consequences of Employment Adjustment in Dynamic Frictional Economies”. The project ran from May 1st 2010 to 30th of April 2013, with Northwestern University, USA the outgoing host and Aarhus University, Denmark, as the EU host. The mentors of the project are Professors Dale Mortensen at Northwestern University and Michael Rosholm at Aarhus University.

The first theme addresses the impact of the business cycle on employment adjustment patterns. In the paper “Firm downsizing, public policy, and the age structure of employment adjustments”, co-authored with Dr. Hans-Martin von Gaudecker from the University of Bonn, Dr. Sebastian Buhai is investigating the structure of workforce adjustments when firms facing adverse demand conditions are offered public financial incentives for downsizing. Specifically, the authors focus in how the age composition of such employee outflows is shaped by corresponding age-dependent institutional arrangements. Their labour demand framework, characterized by stochastic product demand and firing costs heterogenous in workers' early retirement eligibility, has as core prediction that distressed firms will dismiss with predilection those employees eligible to retire early. The model’s implications are then empirically tested on the entire set of mass layoff events in larger Danish private firms over 1980-2001, period covering several reforms to the early pension system. Inter alia, it is found that firms behave as predicted with regards to their lower-skilled workforce, but not vis-a-vis their high-skilled employees. An extension of the firm-level aggregate model to narrow within-firm employee categories, which bear asymmetric turnover responses to firm-level demand shocks, can rationalize this finding. These findings provide a more clear picture on the distressed firms’ strategic layoff behavior given existing public institutional incentives. At the same time, this opens up avenues for essential further research. For instance, the authors are currently engaged in an extension of this project, pursuing a structural estimation of firm-level adjustment costs, given changes in business and institutional conditions.

The second pillar of the project targets the nexus between various internal and external employment adjustment costs. Without a good understanding of the nature of adjustment costs, e.g. whether they are mainly due to search and mobility frictions of workers or to hire/fire fees on employers, premature policy interventions might worsen existent equilibria. In the paper “Performance Pay, Wage Dispersion, and Job Separation”, Dr. Sebastian Buhai together with Dr. Miguel Portela of Minho University use a specific firm remuneration policy, incentive compensation, investigating how this affects individual wage growth, within- and between-firm earnings inequality, as well as worker inflow and outflow patterns from firms. The project uses uniquely suited Portuguese linked employer-employee with information on the pay-for-performance component (hereinafter PP) of each employee’s wage. PP is found to have large effects on the workers selection across firms and on their pecuniary welfare. These findings can be rationalized by means of an extended monitoring costs/incentive pay theory model. Within the same project theme, in a paper entitled “Returns to tenure or seniority?” Dr. Sebastian Buhai, together with co-authors Dr. Miguel Portela (Minho Univ.), Prof. Coen Teulings (Univ. of Cambridge and Amsterdam) and Dr. Aico van Vuuren (VU Univ. Amsterdam), investigate worker-firm wage dynamics and firm hiring/ firing pattern in large firms with specific investment (e.g. training) costs. Using exhaustive linked employer employee data from both Denmark and Portugal, the authors show that firms will adjust their workforce based on a Last-in-First-Out rule, where junior employees are fired before their more senior colleagues. Essential, the firm’s wage profile is also function of the relative seniority ranking: workers hired earlier are able to capture more of the firm’s profits, relative to their more recent colleagues. This research brings therefore together the career concerns of workers and the profitability concerns of firms, when costs of adjustment materialize as specific investment. On top of their direct implications, this paper also stimulates important follow-up research. In an ongoing generalization of the first paper within this theme, Sebastian Buhai has been extending a dynamic equilibrium model of labor demand and supply to incorporate both mobility costs of workers across firms as well as costs of their inter-sectoral and inter-regional mobility, while on the firm’s side both labour and capital are production inputs. Whereas this model cannot be tested with the available Portuguese data, it can again be verified with a subset of the linked-employer-employee Danish data mentioned earlier, augmented with information on the firms’ balance sheets.

Finally, the third pillar of the project investigates dynamic sorting patterns between workers and firms. Within this theme we target sorting patterns of worker and firm features based both on pecuniary and non-pecuniary factors, potentially caused by and/or causing labor adjustment, with full information on individual work (dis)amenities, wage dynamics, as well as employment duration histories. In the paper “Job Hazard Premia and Worker Risk Preferences”, in collaboration with Dr. Elena Cottini of Cattolica University Milano, Dr. Sebastian Buhai is revisiting an old question in the economic literature, namely the (in)existence of “compensating wage differentials” offered by firms for various disamenities. Given access to a rich data source on self-reported work environment conditions and worker risk attitudes that can be linked to the Danish matched employer-employee data, the authors are able to control both for unobserved time-fixed worker and firm components, as well for the worker’s preferences for work conditions, in particular health and safety risks. They find negative selection of risk-lovers into shift-jobs, with sizable hourly wage premia, while positive selection, and no compensation, in other types of hazard work. The empirical results are rationalized by means of a parsimonious Roy-model of job hazard premia with worker risk profiles. These findings indicate that firms offer indeed pecuniary compensation for certain type of contractual disamenities, such as inflexible time schedules or shift work, on top of the workers’ risk profiles. This research is by not means a close end, but has fruitful further implications. For instance, a next step is to investigate more general sorting patterns of workers and firms using an augmented production function, that includes inter alia a skill distribution indices for cross-section of the matched-employer-employee data, extending related earlier work by Sebastian Buhai. In addition, related to the same third pillar of the project, Sebastian Buhai in joint work with Prof. Coen Teulings, entitled “Tenure Profiles and Efficient Separation in a Stochastic Productivity Model” study sorting of workers in firms as function of their match quality, focusing on the selectivity of wage-tenure profiles on both the workers’ inside- and outside (alternative) job productivities. The authors use US panel data on wage dynamics and employment durations and show, among other results, that about 80% of the estimated wage returns to tenure is due to negative selection in realized outside productivities. In other words, workers sorting in long-tenure jobs are in fact “locked-in”, without feasible outside alternatives