It Does Matter Which Side of the Labour Market is Taxed

Abstract

In this paper we discuss the influence of tax shifting on wages and employment. The paper is related to earlier research in this field, both for the Netherlands and for other European welfare states. Our approach differs since we pay explicit attention to the well-known theoretical result that it does not matter which side of the market is taxed (Dalton's Law). We will analyse the mechanisms behind tax shifting. Further we want to analyse whether a shift from employers' to employees' burden has an influence on wages and employment.The paper discusses the influence of taxes on wages and employment in various bargaining settings: the perfect competition model, right-to-manage models (including that of bilateral monopoly) and efficiency wage models are analysed. We conclude that the results depend on the framework that is used in the description of wage setting behaviour. The theorem that it is irrelevant which side of the market is taxed, does not hold for right-to-manage and efficiency wage models.In estimations for the Netherlands, the elasticity of wage costs with respect to employers' taxes is usually found to lie around 0.9, whereas the elasticity with respect to employees' taxes usually is found to lie around 0.4. This apparent violation of Dalton's Law has never been explained before. However, it can be explained from our analysis. Moreover, we show the importance of this result for the impact of a recent tax reform in the Netherlands on wages. 3rd June 1996We would like to thank Lex Borghans for his stimulating comments on an earlier version of this * paper. Further we are grateful to Egbert Jongen for his computational assistance. We thank the Netherlands Association of Science (NWO) for their financial support to this research.

The Wedge

“…Muysken and van Veen (1996) and Picard and Toulemonde (1999) also demonstrate that the legal incidence has an impact on the economic incidence in a right-to-manage union model. arises: what determines the alternative income?…”

The tax system incidence on unemployment: A country-specific analysis for the OECD economies

“…As Rasmussen (1997), Muysken and van Veen (1996), Koskela and Schöb (1999) and Picard and Toulemonde (2001) have shown, this frame of analysis seems to be the most suitable one when dealing with non-competitive markets. 9 …”

“…Our empirical analysis considers a linear approximation of (18) and attempts to quantify the influence of these determinants (lagged unemployment, prices, productivity and taxes) by estimating equation (19).…”

“…In particular, it has become standard to take 5-years averages and increase the number of degrees of freedom by enlarging the amount of cross-section units. 19 In this context, the use of panel data estimation techniques is highly restrictive in terms of individual (country) results. Our methodology differs crucially from this strategy: it uses long time-series to conduct country-specific regressions and only restricts the pool of explanatory variables.…”

“…∆π it−j +ε t , (19) where α i are the constants, i = 1, ...21 are the 21 economies considered, and j = 1, ...n is the lagged structure of the rest of the explanatory variables. Because all the variables considered are either rates or growth rates, we treat them as I (0) and we do not apply cointegration techniques; furthermore, since we do not estimate multi-equation models we use OLS and not instrumental variables methods such as 2SLS or 3SLS.…”