Wage rigidity – for example, due to collective contracts – can generate employment fluctuations, particularly during economic downturns. This column studies the impact of downward nominal wage rigidity on wage and employment dynamics during the 1993 and 2009 recessions in Spain. It shows that collective contracts renegotiated just before the recession led to higher wage growth for the first year of the downturn compared to those signed afterwards. Furthermore, it finds that wage inflexibility increases non-employment for workers close to the minimum wage floors due to lengthy collective contracts (particularly pronounced in 2009) signed before the onset of the downturn.
The role of wage rigidity in generating employment fluctuations has been extensively debated in both academic and policy circles, particularly during economic downturns (e.g. Faia and Pezone 2019, Lennard 2022). Several macroeconomic studies highlight that binding collective contracts, signed at different stages of the business cycle, contribute to wage inflexibility and act as a propagation mechanism during recessions (e.g. Olivei and Tenreyro 2010, Gertler and Trigari 2009). Specifically, collective contracts set nominal wages for a fixed period, so during a fall in aggregate demand, high labour costs may induce employers to reduce employment. In policy terms, these adjustments have led to proposals for temporary deviations from collective contracts to prevent job destruction (see Jimeno and Thomas 2013 for a theoretical justification).
Several factors can attenuate the employment consequences of downward wage rigidity. For example, wage rigidity is less restrictive if collective contracts are frequently renegotiated or if firms respond by reducing hours of work instead. Moreover, some wage components may be more flexible than others. For example, collective contracts in Spain typically set minimum wages by skill, province, and year, but do not prevent wage cuts above those minimum levels.
In a recent study (Adamopoulou et al. 2024), we combine comprehensive data on sectoral collective contracts with longitudinal information on workers’ employment histories to estimate the impact of downward nominal wage rigidity on employment during major recessions in Spain. These recessions differ in their duration, inflation levels, and/or the availability of other adjustment mechanisms, offering a valuable context to understand how downward wage rigidity shapes employment adjustments.
The thought experiment
In Spain, unions and employer federations set pre-specified minimum wages by province and occupation level, which are compulsory for all covered workers and firms for a predetermined period. Since negotiations mostly take place at the province-industry level, at any given time, social partners in some provinces and industries are engaged in ongoing negotiations over minimum wage floors, while others have already agreed on these minima. That is, collective contracts are staggered.
Consider a large, unanticipated shock, such as the 1993 recession or the 2009 recession. After such a shock, some collective contracts are under negotiation, while others are not. Contracts being negotiated can incorporate new information from the aggregate shock and adjust wages downwards. In contrast, contracts settled before the recession began reflect the labour market conditions of the prior expansion and are subject to ‘downward wage rigidity’. Our empirical strategy exploits the exogenous exposure of different sectors and provinces to this rigidity.
Panels A and B in Figure 1 illustrate the adjustment of negotiated wage growth as stipulated in collective contracts in the 1993 and 2009 recessions. Collective contracts negotiated just before the recession (indicated with a vertical line) agreed on higher wage growth for the first year of the recession compared to those signed afterward. In both recessions, the adjustment is sluggish, with wage growth in collective contracts signed after the onset of the recession settling about 1.5-2.0 percentage points below those signed pre-recession.
Figure 1 The wage and employment effects of staggered collective contracts during the 1993 and 2009 recessions
Source: Adamopoulou et al. (2024), using matched data for Spain of the Register of Collective Contracts and the Continuous Sample of Working Histories.
The results: The response of wages
We first examine how the differences in negotiated wage growth in collective contracts propagate over the entire distribution of wages earned by workers. The behaviour in both recessions is remarkably similar: higher negotiated wage growth in contracts signed pre-recession leads to higher wage growth among workers whose earnings are closest to the minimum wage floors (up to 20% above the minimum for their skill and province), but less so among other workers. Overall, a one percentage point increase in minimum wage growth set in collective contracts results in an increase of between 0.40 percentage point (in 2009) and 0.45 percentage point (in 1993) increase in workers’ wage growth.
The response of employment
A natural question that arises is whether this differential wage rigidity also led to differential employment dynamics. To answer this, we estimate event studies at the worker level to examine the short- and medium-run effects of wage rigidity on the probability of being non-employed (unemployed or out of the labour force). The results shown in Figure 1, panel D, show the average monthly probability of non-employment each year for workers covered by collective agreements signed before the 2009 recession, compared to those whose collective contract were signed afterward. One year after the start of the 2009 recession, workers covered by pre-recession agreements were 0.5 percentage point more likely to be non-employed. Four years later, this group had a one percentage point higher likelihood of being non-employed. During this recession, the employment response to downward nominal wage rigidity was substantial –we estimate firm-specific labour demand elasticities to negotiated wages at about 0.52. Consistent with the wage dynamics described above, this response is entirely driven by employees whose pre-recession earnings were at most 20% higher than the minimum wages set in collective contracts.
Figure 1, panel C presents a remarkably different result for the 1993 recession. In this case, the employment trajectories of workers covered by collective contracts signed before the recession and those signed during the recession are very similar, with the differences not being statistically significant.
Using heterogeneity across collective contracts to unpack mechanisms
A key source of wage rigidity due to staggered collective contracts is that current wages reflect the macroeconomic conditions at the time of the negotiation, which may differ from present conditions. Consequently, a longer period until renegotiation extends the impact of these differences in labour costs and their real effects. To explore this mechanism, we compare the employment outcomes of contracts signed prior to the recession with lengthy renegotiation periods (two years or more) to those with shorter renegotiation periods (one year). Focusing on the 2009 recession, we find that workers under contracts signed during the expansion and renegotiated shortly after the recession began experienced employment dynamics similar to those with contracts renegotiated during the recession. Instead, the dis-employment effects were concentrated among workers covered by longer contracts lasting two or three years.
This finding is relevant because, possibly due to the volatility and uncertainty during the inflationary period, most collective contracts during the 1993 recession were signed for short durations, averaging about one year. In other words, if the average contract duration of collective contracts in 2009 had been similar to those in 1993, employment losses would have been much more muted. Thus, a possible explanation for the lack of employment responses to downward wage rigidity in 1993 was the short renegotiation period. Therefore, assessing the employment impacts of collective contracts crucially depends on contract duration, which may explain the wide range of estimated elasticities reported in the literature.
Finally, the richness of our data allow us to rule out other features of collective bargaining, such as inflation indexation clauses or layoff costs, as determinants of employment dynamics.
Overall, our results support the hypothesis that nominal wage rigidity, caused by the automatic extension of sector-province agreements and contract staggering, can amplify employment destruction during recessionary periods with long collective contracts.
Source : VOXeu
