The Effect of Environmental Protection Expenditures on Industrial Employment in Sweden

Environmental and Resource Economics https://doi.org/10.1007/s10640-025-00961-7 The Effect of Environmental Protection Expenditures on Industrial Employment in Sweden Golnaz Amjadi1,5 · Moriah Bostian1,2 · Hanna Lindström1,3 Mattias Vesterberg1,3 · Tommy Lundgren1,4 · Accepted: 9 January 2025 © The Author(s) 2025 Abstract In this paper, we empirically investigate how environmental protection expenditures affect sector-level employment within manufacturing industries, using detailed firm-level data for Sweden for the years 2002–2021. We use a structural model that allows for a decomposition of the total employment effect of environmental protection expenditures within a sector into a cost effect, a factor shift effect, and a demand effect. We add to previous literature by using instrumental variables in our empirical framework, to account for endogenous environmental spending stemming from, e.g., corporate social responsibility and self-regulation. Our results reveal that increased environmental protection expenditures generally have no statistically significant effect on employment among the sectors studied, with the paper and pulp sector being the exception, showing non-negligible negative effects on employment. Keywords Environmental protection · Labor demand · Environmental regulation Hanna Lindström 1 Centre for Environmental and Resource Economics (CERE), Umeå University and Swedish University of Agricultural Sciences, Umeå, Sweden 2 Department of Economics, Lewis and Clark College, Portland, OR, USA 3 Department of Economics, Umeå School of Business, Economics and Statistics, Umeå University, Umeå, Sweden 4 Department of Forest Economics, Swedish University of Agricultural Sciences, Umeå, Sweden 5 STATEC Research , (National Institute of Statistics and Economic Studies), Luxemburg, Luxemburg 13 G. Amjadi et al. 1 Introduction In this paper, we investigate the effect of environmental regulation on sector-level employment levels for the Swedish manufacturing industry, using self-reported environmental protection expenditures to measure regulatory burden. This paper contributes to previous literature by providing estimates of the employment effects of increased environmental protection expenditures1 in a European context, using a rich data set that spans several sectors and years. We also add to previous literature by taking into account that environmental protection expenditures may be partly endogenous if firms choose to self-regulate, e.g., engage in corporate social responsibility (CSR), thereby incurring costs beyond what environmental regulation stipulates. This endogeneity in environmental protection expenditures has been repeatedly highlighted in previous literature as a potentially important feature (see, for example, the discussion in Belova et al. 2015), but to the best of our knowledge, no previous study has addressed this concern. The motivation for studying how environmental protection expenditures affect employment originates from recent policy debates about environmental regulation, where concerns are raised about a potential “jobs versus the environment” conflict. Labor unions and trade groups argue that environmental regulation imposes extra costs on producers, which reduce production level and demand for labor. An alternative to this pessimistic perspective on environmental regulation is brought forward by the concept of “green jobs” which has become part of the policy discussion in recent years. This perspective argues that “green jobs” (i.e., jobs created directly or indirectly as a consequence of environmental regulation), can solve challenges related both to climate change and to high rates of unemployment in industrialized countries.2   In many cases, this debate concerns the effect of environmental regulation on the total employment level in the economy, even though such effects will only be present if the economy deviates from its normal rate of full employment (Belova et al. 2015). However, partial-equilibrium and sector-level effects are also important to policy makers, and these latter effects are what the current paper studies. For example, the size of sector-level effects may be important from the perspectives of labor unions and regional development, and can inform policy design regarding support for structural change following the green transition. However, previous empirical studies on the sector-level employment effects of environmental regulation point to mixed results and apply different metrics and approaches when estimating these effects. For example, Berman and Bui (2001) and Morgenstern et al. (2002) do not find any substantial effects of environmental regulation on employment when looking at industry averages, although Morgenstern et al. (2002) find small positive and signifi1 In previous literature, these expenditures are sometimes referred to as regulatory costs. However, both Eurostat and Statistics Sweden refer to these costs as environmental protection expenditures; see https://e c.europa.eu/eurostat/statistics-explained/index.php?title=Environmental_protection_expenditure_accounts and https://www.scb.se/en/finding-statistics/statistics-by-subject-area/environment/environmental-account s-and-sustainable-development/environmental-protection-expenditure/. We follow this notation. 2 This policy debate can thus to some extent be tied to the Porter Hypothesis (PH) which states that wellcrafted and well-enforced environmental regulation can benefit both the environment and firms’ productivity by promoting innovation to overcome compliance costs (Porter 1991; Porter and Van der Linde 1995). However, while studies aiming to prove or disprove the PH analyze outcomes in terms of innovation (e.g., R&D investments, patent applications) and productivity (e.g., total factor productivity, export dynamics) (see, e.g., Rubashkina et al. 2015 and references therein), the current study focuses on the relation between environmental regulation and employment. 13 The Effect of Environmental Protection Expenditures on Industrial… cant total employment effects for the individual sectors of plastic material and petroleum, respectively. Although using data and empirical approach that are similar to Morgenstern et al. (2002), Belova et al. (2013a) find substantially larger (positive) employment effects. As pointed out in Belova et al. (2015), the variation in both metrics (gross vs. net effects) and estimated impacts (both in magnitude and direction) raises questions on how feasible it is to extrapolate from one study’s context to other contexts. Since a majority of the empirical studies on this topic is based on data from the U.S., there is a current lack of studies from countries where the structure and composition of the labor market and industrial sector differ compared to the U.S. context. For example, differences in labor market structures between Europe and the U.S. have repeatedly been highlighted in the literat (...truncated)