(Mis)alignment of productivity and wages: firm-level evidence

Jozef Konings (Department of Economics, KU Leuven, Leuven, Belgium)
François Rycx (Centre Emile Bernheim and DULBEA, Université libre de Bruxelles, Brussels, Belgium; IRES, Université catholique de Louvain, Louvain-la-Neuve, Belgium and IZA, Bonn, Germany,)
Vincent Vandenberghe (Institut de recherches économiques et sociales, Université catholique de Louvain, Louvain-la-Neuve, Belgium)

International Journal of Manpower

ISSN: 0143-7720

Article publication date: 3 May 2016



Konings, J., Rycx, F. and Vandenberghe, V. (2016), "(Mis)alignment of productivity and wages: firm-level evidence", International Journal of Manpower, Vol. 37 No. 2. https://doi.org/10.1108/IJM-12-2015-0222



Emerald Group Publishing Limited

(Mis)alignment of productivity and wages: firm-level evidence

Article Type: Guest editorial From: International Journal of Manpower, Volume 37, Issue 2.


This special issue of the International Journal of Manpower contains eight empirical papers which were presented at the Firm-level Analysis of Labour Issues conference, held in Louvain-la-Neuve (UCL-Belgium), 28 May 2014. They are all based to a large extent on a few key considerations.

First, focusing on firms is key. This is more than just the consequence of the growing availability of good-quality firm-level data on productivity and other outcomes. Productivity is, in essence, a firm-level phenomenon and should be primarily assessed at that level. In modern economies, where most people work in firms, labour-related productivity gains cannot possibly exist at the individual level (as highlighted in Mincer-type analyses) if they do not show up at the firm level. Productivity is probably intrinsically determined by the (heterogeneous) ability of firms to successfully aggregate individual productivities, in conjunction with other factors of production (e.g. capital). A similar reasoning applies to countries: the benefits of labour should show clearly in the performance of firms, if they are to emerge at a more aggregate level. We thus argue that a study of the relationship between workers’ characteristics (e.g. age, education and gender), productivity and remuneration requires analysing data at the level of the firm. Individual workers’ productivity is hardly ever observed[1]. By contrast, many datasets now contain good-quality information about what firms are able to produce (e.g. firm value added). Similarly, the alignment of productivity and pay at the individual level is hard to assess. But it can be evaluated with firm-level aggregates, conditional on adoption of an adequate analytical framework. Workers’ characteristics (e.g. their educational attainment) can be aggregated at the firm level and introduced into firm-level equations in order to explore how they influence productivity and pay/remuneration. Productivity can be examined at macro level but aggregation bias exist, and many unobserved macro-level factors can drive country-level productivity (e.g. quality of institutions, importance of public/non-profit sector, effect of taxation on incentives).

Second, many of the papers collected in this special issue examine firm productivity in conjunction with one or several other key economic issues, that have historically been treated separately by the economic literature: for example, the internal functioning of the firm (e.g. labour force structure, payment schemes, employment contracts, workforce absenteeism), its organisational and financial structure (e.g. being part of a multinational company or not) or the role the firm’s environment (e.g. its product or labour market regime). For instance, the growing availability of good-quality firm-level data, providing information on the characteristics of both employees and their workplaces, has created a promising research avenue, bridging industrial organisation and labour economics. Economists active in this area investigate the relationship between the structure of the workforce, or whether or not the latter is remunerated based on performance, and the productivity of firms. The ubiquity of studies exploiting firm-level evidence enables researchers to consider a large set of topical issues: the impact of labour characteristics (e.g. education, age, gender, part- vs full-time assignment) on productivity and employability, the magnitude of wage discrimination based on gender or national origin, the importance and consequences of job turnover, the link between company-based training, productivity and pay, or the impact of multinationals on the disconnection of productivity and pay caused by rent sharing. The purpose of this special issue is to assemble the best and most recent contributions of researchers active in this area.

Third, despite the availability of good-quality firm-level (and longitudinal) data, establishing rigorously how internal or external traits of firms affect their productivity remains a challenge; not only because productivity is firm specific, but also because (short term) productivity changes can be spuriously correlated with the explanatory variable under investigation, e.g. the workforce age structure of firms. Most contributors to this special issue assume that the identification of causal links requires dealing with: time-invariant, firm-specific characteristics that are unobservable, but that simultaneously drive the firm’s average productivity and the main explanatory variable; and the short-term simultaneity or endogeneity bias (e.g. the aforementioned spurious correlation between productivity shocks and the workforce age structure). The panel structure of their data allows them to use fixed-effects methods (mean-centring or first differences) to cope with non-randomly distributed time-invariant unobservables. To account for the presence of an endogeneity bias, many papers estimate the relevant parameters using IV-GMM methods. This is a strategy regularly used in the production function literature with labour heterogeneity (Aubert and Crépon, 2003; Garnero et al., 2014a; Kampelmann and Rycx, 2014). The key idea is to instrument potentially endogenous first-differenced explanatory variables (e.g. age shares) by their lagged values. An alternative to IV-GMM, notably used by Dostie (2011), Cataldi et al. (2012), Kampelmann and Rycx (2012), Vandenberghe (2013a, b), Garnero et al. (2014b), Konings and Vanormelingen (2015), Parrotta et al. (2014), Mahy et al. (2015) and Rycx et al. (2015), is to adopt the more structural approach initiated by Olley and Pakes (1996) and further developed by Levinsohn and Petrin (2003) or Ackerberg et al. (2006). The essence of the Levinsohn and Petrin strategy is to use some function of a firm’s demand for intermediate inputs (i.e. raw materials, electricity) in order to infer a value for the short-term productivity shocks causing the endogeneity bias. Firms can swiftly (and also at a relatively low cost) respond to productivity developments by adapting the volume of the intermediate inputs they buy on the market.

Workers’ and jobs’ characteristics. The three first contributions in this special issue examine the consequences of workers’ and jobs’ characteristics (i.e. age, education, gender and employment contacts) on productivity and/or productivity-wage gaps. Commentators often tout workforce diversity (i.e. matching of unlikes) as performance (productivity) enhancing. Is this really true? To answer that question, Vincent Vandenberghe (Université catholique de Louvain) departs from the approach used in most recent empirical papers exploiting firm-level evidence, where output is regressed on traditional inputs plus an index of diversity (Parrotta et al., 2014; Garnero et al., 2014b). He addresses the question by adopting a more structural framework. The idea is to root the empirical strategy applied to firm-level data in the theoretical literature on population heterogeneity/stratification and growth (Bénabou, 1994). Essentially, what that literature suggests is that diversity is optimal when the technology displays concavity in the share of workers considered (e.g. decreasing marginal contribution of rising shares of more productive/skilled workers). What is also shown in the paper is that a production function à-la-Hellerstein-Neumark (1995), where workforce diversity is captured via an index of labour shares, is suitable for estimating the concavity of the technology, and thus for assessing the case for/against workforce diversity. Finally, the paper contains an application of this Bénabou-Hellerstein-Neumark framework to two panels of Belgian firms covering the 1998-2012 period. The main result is that of an absence of strong evidence that age, gender or educational diversity is good or bad for efficiency.

Next, Philipp Grunau (Institute for Employment Research, Nuremberg) analyses the impact of overeducated and undereducated workers among an establishment’s workforce on its productivity, providing first representative evidence for Germany. Many contributions to the educational mismatch literature address the productivity effects of both excess and deficit educational attainments for workers on the individual level. Due to the limited transferability of their results to firm-level performance, especially when allowing for the possibility of spillover effects from mismatched workers to their well-matched colleagues, from an employer’s vantage point, it is highly important to know the net effect of educationally mismatched employees on productivity at the establishment level. Using linked employer-employee data from Germany, Philipp Grunau estimates dynamic panel production functions using a system GMM estimator. He finds that undereducated workers among an establishment’s workforce impair its (establishment level) productivity, implying that an establishment’s HR management should avoid the recruitment of undereducated workers, at least if they follow a short-term personnel policy. The effect for overeducated employees is also negative, albeit small and insignificant.

The third contribution in this special issue investigates the role of temporary jobs in shaping wage and productivity differentials. While mass production tends to favour the planning of production and therefore the smoothing of changes in the labour force (Holt et al., 1960), the spread of elements of the Japanese "lean production" model (e.g. just-in-time production) in advanced economies since the 1990s increased the need for flexible labour management. Enterprises nowadays have more flexibility to adjust the size of their labour force to changes in the business cycle. Temporary contracts may notably be used as a buffer to product demand fluctuations (Jahn et al., 2012). In turn, this may lead to a division of the labour force into a core component, which is relatively well protected from demand fluctuations, and a peripheral component, which is at risk to demand fluctuations (Boeri, 2011). In light of this evolution, an accurate understanding of the different repercussions of temporary employment contracts has emerged as an increasingly salient problem in labour and industrial economics. The paper of Andrea Garnero (OECD), Romina Giuliano (University of Mons), Benoît Mahy (University of Mons) and François Rycx (Université libre de Bruxelles), in this special issue, is one of the first to examine the impact of fixed-term contracts (FTCs) on labour productivity, wages (i.e. labour cost) and productivity-wage gaps (i.e. profits) while addressing three important methodological issues related to the state dependency of the three explained variables, to firm time-invariant heterogeneity, and to the endogeneity of FTCs. Their dynamic panel data estimates, based on detailed Belgian linked employer-employee data, indicate that FTCs exert stronger positive effects on productivity than on wages and (accordingly) that the use of FTCs increases firms’ profitability.

Variable payment schemes. The two following contributions examine the effects of variable payment schemes on productivity/absenteeism and compensation. Mirella Damiani (University of Perugia), Fabrizio Pompei (University of Perugia) and Andrea Ricci (ISFOL, Rome) first analyse the role of performance-related pay (PRP) in a sample of Italian manufacturing and service firms and present quantile estimates to investigate heterogeneity in pay-performance impacts on labour productivity and wages. In a second stage, the endogeneity of PRP is taken into account by using instrumental variable quantile regression techniques. They find considerable heterogeneity across the distribution of labour productivity and wages, with the highest role of PRP obtained at the lowest and highest quantiles. However, for all quantiles, the comparison of productivity and wage estimates suggests that PRP might not only be rent-sharing devices, but also incentive schemes that substantially lead to efficiency enhancements. These findings are confirmed for firms under union governance and suggest that well designed policies, that circumvent the limited implementation of PRP practices, would guarantee productivity improvement.

Despite its potential to raise productivity, PRP is not widespread in market-oriented economies. Furthermore, despite secular changes conducive to its take-up, there is mixed evidence as to whether it has become more prominent over time. The next contribution, by John Forth, Alex Bryson and Lucy Stokes (NIESR), is the first to present firm-level evidence for Britain on both the incidence and size of bonus payments in the 2000s. The authors decompose the share of the total wage bill accounted for by bonuses into the shares of employment in the PRP and non-PRP sectors, the ratio of base pay between the two sectors, and the gearing of bonus payments to base pay within the PRP sector. They show that there was some growth in the share of total pay accounted for by bonuses in Britain in the mid-2000s. However this rise – and subsequent fluctuations since the onset of recession in 2008 – can be almost entirely explained by changes in the gearing of bonus to base pay within the PRP sector. There has been no substantial change in the percentage of employment accounted for by PRP firms; if anything it has fallen over the past decade. Furthermore, the movements in the gearing of bonuses to base pay in the economy at large are heavily influenced by changes in the finance industry: a sector which accounts for a large proportion of all bonus payments in the British economy.

A typical outcome of the introduction of PRP systems is an increase in wage dispersion. This is due to the fact that there is a greater underlying variation in the individual endowments that determine worker performance (e.g. cognitive or physical ability, risk propensity) than in those that determine input (e.g. ability to put in eight hours per day). Starting from these premises, Benoît Mahy (University of Mons), François Rycx (Université libre de Bruxelles) and Mélanie Volral (University of Mons) examine the impact of intra-firm wage dispersion on sickness absenteeism, i.e. an important determinant of firm productivity (Eurofound, 2010). Using detailed matched panel data enabling to compute a conditional wage dispersion indicator, following the Winter-Ebmer and Zweimüller (1999) methodology, and to control for time-invariant workplace characteristics, they find a positive and hump-shaped relationship between intra-firm wage dispersion and sickness absenteeism, though the turning point of this relation is extremely high. Their results could therefore suggest that wage dispersion (i.e. a stronger linkage between pay and performance), decreases worker satisfaction and the workplace climate in general. Only a specific type of workers, who are less sensitive to equity and cohesion considerations, would be less absent as pay-for-performance mechanisms are more intensively used.

Imperfect product/labour markets. The next three papers estimate the extent of product/labour market imperfections and their consequences on earnings notably through rent sharing. Sabien Dobbelaere (VU Amsterdam), Rodolfo Lauterach (Maastrict University) and Jacques Mairesse (CREST) start from the belief that product and labour markets are intrinsically characterised by imperfections and by the fact that variable input factors’ estimated marginal products are not equal to their measured payments. There are a number of reasons why the latter discrepancy could exist. Paramount among these are economic factors like imperfections in product and/or factor markets, variable factor utilisation, factor adjustment costs and measurement issues. Focusing on the former, the authors investigate, on the basis of unbalanced firm-level panel data, how different manufacturing industries in Chile and France are in their factor shares, in their marginal products and in their imperfections in the product and labour markets in which they operate. Consistent with differences in institutions and in the industrial relations system in the two countries, they find regime differences across the two countries. In addition, they observe cross-country differences in the levels of product and labour market imperfections within regimes.

A chief source of firms’ ability to pay wages above the competitive rate arises from firms’ position of dominance in their product markets. Such rents may be "shared" in a number of ways. If such firms are unionised, then employees may bargain for a share of these rents. But they not need to be unionised to benefit from rent sharing. Managers in such firms may also concede higher pay in order to keep their employees happy and get a "quiet life". They may also use part of these rents in order to develop policies to motivate their employees, or to build loyalty. Jozef Konings (KU Leuven), Luca Marcolin (OECD) and Ilke Van Beveren (KU Leuven) provide empirical evidence of international rent sharing in multinational enterprises. To this end, the authors use a unique firm-level longitudinal data set of M&As in Belgium between 1998 and 2010. They can therefore analyse rent sharing before and after the deal, and highlight the role of target and acquirer profitability in the wage setting process for the target firm. They thus identify changes in the bargaining power of workers due to the takeover. What is more, their empirical setup permits to net their estimates from selection effects in the choice of target firm. They find evidence that the deal does not significantly affect the degree of domestic rent sharing, but it enables international rent sharing. They qualify their results in terms of the acquirer’s controlling stake, location and industry link with the target. Further robustness specifications include a placebo test and the comparison with a sample of domestic acquisitions.

Wage inequality has substantially increased in Germany over the last two decades. Part of the literature on this development has sought to clarify the extent to which this rise can be explained by market forces or institutions (Dustmann et al., 2009). At the same time, there is the striking finding of increasing heterogeneity across firms, which draws attention to the process of rent generation and rent sharing at the firm level (Card et al., 2013). In the final paper of this special issue, Clemens Ohlert (Hamburg University) investigates the sources of rising firm heterogeneity and its impact on overall wage inequality in Germany between 2000 and 2010. Using linked employer-employee data and regression-based variance decomposition techniques, he shows that the rise in wage inequality in Germany is to a great extent associated to rising wage variance across establishments, implying that establishment-specific wage premiums have grown. By further decomposing across establishments’ components of wage inequality, it is found that changes in collective bargaining, worker co-determination and internal labour markets together account for only 3 per cent of the rise in total inequality. In contrast, changes in establishments’ skill and occupational composition account for about 11 per cent and establishment size for about 18 per cent of the rise in total wage inequality.

Jozef Konings -Department of Economics, KU Leuven, Leuven, Belgium

François Rycx - Centre Emile Bernheim and DULBEA, Université libre de Bruxelles, Brussels, Belgium;IRES, Université catholique de Louvain, Louvain-la-Neuve, Belgium and IZA, Bonn, Germany

Vincent Vandenberghe - Institut de recherches économiques et sociales, Université catholique de Louvain, Louvain-la-Neuve, Belgium


1. Productivity, as most economists conceive it, should not be amalgamated with individual capabilities, either physical or (more related to what is discussed here) cognitive ones. There is evidence, stemming from the International Adult Literacy Surveys (IALS), for example, that individuals who have completed more years of education and possess college or university degree achieve better in terms of literacy or other aspect of cognitive performance.


The authors are grateful to the Editor, Professor Adrian Ziderman, for asking them to act as Guest Editors of this special issue. The collection of papers in this special issue was originally presented at the Firm-level Analysis of Labour Issues conference, held in Louvain-la-Neuve (UCL-Belgium), 28 May 2014. The authors would like to express their gratitude to the referees who did an excellent job in discussing the papers included in this issue. Financial support was provided by the Belgian Federal Government: SPP Politique scientifique, programme "Société et Avenir", research contract TA/00/046/EDIPO.


Ackerberg, D., Caves, K. and Frazer, G. (2006), "Structural identification of production functions", Working Paper No. 195, Department of Economics, UCLA, Los Angeles, CA

Aubert, P. and Crépon, B. (2003), "La productivité des salariés âgés: Une tentative d’estimation", Economie et statistique, Vol. 368 No. 1, pp. 95-119

Bénabou, R. (1994), "Education, income distribution and growth: the local connection", Working Paper No. 4798, NBER, Cambridge, MA

Boeri, T. (2011), "Institutional reforms and dualism in European labor markets", in Ashenfelter, O. and Card, D. (Eds) Handbook of Labor Economics, Elsevier, North Holland, pp. 1173-1236

Card, D., Heining, J. and Kline, P. (2013), "Workplace heterogeneity and the rise in West German wage inequality", Quarterly Journal of Economics, Vol. 128 No. 3, pp. 967-1015

Cataldi, A., Kampelmann, S. and Rycx, F. (2012), "Does it pay to be productive? The case of age groups", International Journal of Manpower, Vol. 33 No. 3, pp. 264-283

Dostie, B. (2011), "Wages, productivity and aging", De Economist, Vol. 159 No. 2, pp. 139-158

Dustmann, C., Ludsteck, J. and Schönberg, U. (2009), "Revisiting the German wage structure", Quarterly Journal of Economics, Vol. 124 No. 2, pp. 843-881

Eurofound (2010), "Absence from work report", Dublin, OH

Garnero, A., Kampelmann, S. and Rycx, F. (2014a), "Part-time work, wages and productivity: evidence from Belgian matched panel data", Industrial and Labor Relations Review, Vol. 67 No. 3, pp. 926-954

Garnero, A., Kampelmann, S. and Rycx, F. (2014b), "The heterogeneous effects of workplace diversity on productivity, wages and profits", Industrial Relations: A Journal of Economy and Society, Vol. 53 No. 3, pp. 430-477

Hellerstein, J. and Neumark, D. (1995), "Are earnings profiles steeper than productivity profiles: evidence from Israeli firm-level data", Journal of Human Resources, Vol. 30 No. 1, pp. 89-112

Holt, C., Modigliani, F., Muth, J. and Simon, H. (1960), Planning Production, Inventories and Work Force, Prentice Hall, Upper Saddle River, NJ

Jahn, E., Riphahn, R. and Schnabel, C. (2012), "Flexible forms of employment: boon and bane", Economic Journal, Vol. 122 No. 562, pp. 115-124

Kampelmann, S. and Rycx, F. (2012), "Are occupations paid what they are worth? An econometric study of occupational wage inequality and productivity", De Economist, Vol. 160 No. 3, pp. 257-287

Kampelmann, S. and Rycx, F. (2014), "The impact of educational mismatch on firm productivity: evidence from linked panel data", Economics of Education Review, Vol. 31 No. 6, pp. 918-931

Konings, J. and Vanormelingen, S. (2015), "The impact of training on productivity and wages: firm-level evidence", Review of Economics and Statistics, Vol. 97 No. 2, pp. 485-497

Levinsohn, J. and Petrin, A. (2003), "Estimating production functions using inputs to control for unobservables", Review of Economic Studies, Vol. 70 No. 2, pp. 317-341

Mahy, B., Rycx, F. and Vermeylen, G. (2015), "Educational mismatch and firm productivity: do skills, technology and uncertainty matter?", De Economist, Vol. 163 No. 2, pp. 233-262

Olley, G. and Pakes, A. (1996), "The dynamics of productivity in the telecommunications equipment industry", Econometrica, Vol. 64 No. 6, pp. 1263-1297

Parrotta, P., Pozzoli, D. and Pytlikova, M. (2014), "Does labor diversity affect firm productivity?", European Economic Review, Vol. 66, February, pp. 144-179

Rycx, F., Saks, Y. and Tojerow, I. (2015), "Does education raise productivity and wages equally? The moderating roles of age, gender and industry", Discussion Paper No. 9043, IZA, Bonn

Vandenberghe, V. (2013a), "Are firms willing to employ a greying and feminizing workforce?", Labour Economics, Vol. 22, June, pp. 30-42

Vandenberghe, V., Rigo, M. and Waltenberg, F. (2013b), "Ageing and employability. Evidence from Belgian firm-level data", Journal of Productivity Analysis, Vol. 40 No. 1, pp. 111-136

Winter-Ebmer, R. and Zweimüller, J. (1999), "Intra-firm wage dispersion and firm performance", Kyklos, Vol. 52 No. 4, pp. 555-572

Further reading

Becker, G. (1973), "A theory of marriage: part I", Journal of Political Economy, Vol. 81 No. 4, pp. 813-846

Hanka, G. (1998), "Debt and the terms of employment", Journal of Financial Economics, Vol. 48 No. 3, pp. 245-282

Related articles