Swedish environmental policy often emphasizes the importance of “taking the lead”. For example, Sweden has chosen a more ambitious climate policy target than required by the European Union (EU), namely a reduction of Swedish emissions of greenhouse gases by 40 percent by 2020 compared to the 1990 level. Government Bill 2008/09:162 emphasizes Sweden’s role as a good example in making an effort to re-duce climate change by showing that an offensive climate policy can indeed be com-bined with high economic growth. This view of environmental policy is, however, the subject of constant debate. <p>A common argument is that environmental requirements induce private costs by forc-ing firms to make investments that crowd out other more productive investments, which hampers productivity growth and therefore competitiveness. Professor Mi-chael E. Porter of Harvard questioned this argument, and his view has become known as the Porter hypothesis (Porter, 1991). This hypothesis implies that levying stringent environmental regulations on firms enhances their productivity compared to competi-tors not subject to, or subject to lax, environmental regulations. A central message is that the connection between environmental regulation and competitiveness should be scrutinized within a dynamic framework (Porter and van der Linde, 1995). <p>The main objective of this paper is to test the Porter hypothesis by assessing static and dynamic effects of environmental policy on productivity within the Swedish manufac-turing industry, specifically on the component total efficiency. The paper adds mainly to previous literature by using unique data on environmental protection investments, divided into investments in pollution control and pollution prevention, as a proxy for envi-ronmental regulation. The distinction between these types of investments is crucial to the understanding of the outcomes anticipated by the Porter hypothesis. <p>The international literature studying the Porter hypothesis is extensive. A comprehen-sive review reveals that neither theoretical nor empirical literature gives general sup-port for the hypothesis (Brännlund and Lundgren, 2009). We argue that, to some ex-tent, the Porter hypothesis has not yet been given a fair chance in the empirical litera-ture, as dynamic effects are often neglected in empirical tests. Two exceptions are Managi et al. (2005) and Lanoie et al. (2008), who first estimate Total Factor Produc-tivity (TFP) scores that then are used as dependent variables in regression analyses where explanatory lagged environmental stringency measures model dynamic effects. A disadvantage with these studies is, however, that environmental stringency is ap-proximated by the cost of complying with environmental command- and-control regulations, such regulations are not emphasized by the Porter hypothesis. <p>The empirical test of the Porter hypothesis is performed as a two-step procedure, where total efficiency scores are first estimated by adopting a stochastic production frontier function approach. In the second step, the efficiency scores are used as the dependent variable in random effects regression analyses, where the independent vari-ables are, e.g., investment in pollution control and pollution prevention. In order to assess whether these investments have dynamic effects on total efficiency these vari-ables are also lagged. If positive effects are established we cannot reject the claim that environmental leadership will benefit the Swedish industry. The estimations are based on firm level data from five Swedish industries for the period 1999-2004, and carried out for the pooled data as well as for the industries separately.
