Misalignment of productivity and wages across regions: evidence from Belgium

ABSTRACT

This paper is one of the first to estimate how regions affect the productivity, wage cost and cost competitiveness (i.e., the productivity–wage gap) of firms. Detailed linked employer–employee panel data for Belgium and the Hellerstein–Neumark framework are used to estimate dynamic models at the establishment level. The findings show that interregional differences in productivity and wages are significant, but to a large extent due to drivers at the individual and/or firm level. The research provides evidence that the specificity of the Brussels region can be linked to its higher density compared with the rest of Belgium. Robustness tests suggest that the relatively better ceteris paribus performance of firms in Brussels is limited to the service sector.

KEYWORDS:

• regions
• agglomeration effects
• density
• productivity
• labour costs
• linked panel data

JEL:

• C33
• J24
• J31
• R30

DISCLOSURE STATEMENT

No potential conflict of interest was reported by the authors.

Notes

1. Article 19 of the law, dated 5 December 1968, specifies that a collective agreement is automatically binding upon the signatory organizations, employers who are members of those organizations or who have personally concluded the agreement, employers joining those organizations after the date of the conclusion of the agreement, and, finally, all workers, whether or not unionized, who are employed by an employer so bound.

2. For more detailed information on the data sets used in this paper, see Appendix A in the supplemental data online.

3. The share of workers with at most lower secondary education serves as a reference category.

4. All independent variables are measured in terms of shares in total work hours. For instance, the fraction of part-time workers is computed on the basis of the proportion of hours worked by employees working fewer than 30 hours per week over the total amount of hours worked within the establishment.

5. Given that the cost of capital is not taken into account in the computation of the dependent variable, the latter does not reflect overall competiveness. If wage costs increase faster than productivity, overall competitiveness will be deteriorated only if other costs are not adjusted in compensation (OECD, 2008).

6. Technically, this figure is obtained by taking the antilog (to base e) of the estimated coefficient from which 1 is subtracted (× 100). As demonstrated by Halvorsen and Palmquist (1980), this transformation is required when interpreting a dummy variable in a log-linear model.

7. These firm-level agreements are therefore negotiated in addition to national and sectoral agreements which they cannot undercut.

8. No significant wage cost differential is observed between Flanders and Wallonia.

9. Apart from regions, industries, the level of collective wage bargaining and time dummies, all explanatory variables have been treated as endogenous.

10. In the regressions the densities of the resident, economically active and employed population are measured in thousands.

11. Industry sectors refer to NACE (Statistical Classification of Economic Activities in the European Community) codes C (Mining and quarrying), D (Manufacturing), E (Electricity, gas and water supply) and F (Construction). Services sectors include NACE codes G (Wholesale and retail trade; repair of motor vehicles, motorcycles and household goods), H (Hotels and restaurants), I (Transport, storage and communication), J (financial intermediation), and K (Real estate, renting and business activities).

Additional information

Funding

This work was funded by the National Bank of Belgium.