ANALYSISIs it wise to combine environmental and labour market policies? An analysis of a Swedish subsidy programme
Introduction
In 1998 the Swedish government launched the largest investment in ecological sustainability to date, the Local Investment Programme (LIP). The LIP, in effect between 1998 and 2002, was an investment subsidy programme with dual objectives: to speed up Sweden's transformation into an ecologically sustainable society and to reduce unemployment. During the programme period, SEK 6.2 billion1 were allocated to more than 1800 different projects in 161 municipalities.2 By having dual objectives, the LIP policy did not conform to policy guidelines advocated by economists.3
The purpose of this paper is to evaluate how effectively a subset of the LIP projects achieved the separate objectives. Effectiveness in the (un)employment objective is relatively easy to evaluate since the LIP projects were required to state the number of annual full-time employment opportunities generated in the projects.4 Effectiveness in the ecological sustainability objective is, however, more difficult to evaluate. Ecological sustainability is a heterogeneous objective, involving many different ecological systems, emissions and environmental problems. To evaluate the LIP projects effectiveness in achieving the ecological sustainability objective we need a yardstick to which ecological sustainability can be compared. For this purpose, we concentrate on the LIP projects for remediation of contaminated sites, i.e. landfills or areas of soil, groundwater or sediment contaminated by anthropogenic activities.
The remediation projects are interesting from several points of views. Even though only twenty-four of the projects (slightly more than one percent of the projects) were aimed at remediation of contaminated sites they received a disproportionate amount, six percent (SEK 400 million), of the total LIP subsidies. The degree of subsidization, i.e. the ratio between the subsidy and the investment cost, was consequently higher in the remediation projects than in all other LIP projects. Remediation of contaminated sites is an interim target under the Swedish parliament's environmental quality objective “A Non-Toxic Environment” (Environmental Objectives Portal, 2005), stating that highest priority should be given to contaminated sites posing the highest risks to human health and the environment. Thus, an explicit target for the Swedish parliament is to adhere to the “worst things first” principle in the choice of sites to remediate. The “worst things first” principle conveniently offers us a yardstick for ecological sustainability to which the allocation of LIP subsidies to remediation can be evaluated.
Our paper contributes to the existing literature in at least four ways. First, our data is unique for evaluating the outcomes of a policy mixing environmental and labour market objectives. Second, our results tentatively indicate that the Swedish government may have used the LIP subsidies for remediation tactically, i.e. to “purchase” votes. Third, because remediation of contaminated sites is an expensive venture that only just has begun in Europe,5 our analysis gives important policy implications for future remediation decisions. Fourth, since most analyses of contaminated sites have concerned the U.S., our analysis contributes with a European perspective. Although there are similar initiatives in the Netherlands, Denmark, Germany and the United Kingdom, the LIP is unique among the EU states in terms of the amount of funding and its diverse scope (Hanberger et al., 2002, Swedish EPA, 2005c).
To evaluate the remediation projects' effectiveness in achieving the LIP's separate objectives we test four hypotheses. If the remediation projects were effective in achieving the LIP's sustainability objective: i) we expect, according to the “worst things first” principle, that the most hazardous sites were most likely to receive LIP subsidies for remediation; ii) we also expect that the most hazardous sites received higher subsidies— basically because these sites are more expensive and cumbersome to remediate. If the remediation projects were effective in achieving the LIP's employment objective: iii) we expect that the projects with the most employment opportunities were most likely to receive subsidies; iv) we also expect that projects with more employment opportunities received higher subsidies. Furthermore society's marginal cost for employment under the LIP should not exceed society's marginal cost for employment under regular labour market policy. By using data on both accepted and rejected projects, we test hypotheses i) and iii) controlling for variables that have previously proved important in remediation decisions (e.g. municipal political preferences, wealth and unemployment). By using data on accepted projects, we test hypotheses ii) and iv). Through estimation of the marginal subsidy per employment opportunity, i.e. the public's marginal cost for an employment opportunity in the remediation projects, we can compare society's marginal cost for employment under the LIP with society's marginal cost for employment in traditional labour market policies.
Somewhat surprisingly, we find that the probability of receiving a subsidy for remediation increases the lower the health and environmental risk posed by the site. In another, more conservative analysis, we find, contrary to the “worst things first” principle, that the site's risk to environment and health had no effect on the probability of receiving a subsidy for remediation at all. Thus, depending on analysis, the “worst things first” principle is either contradicted or ignored. However, if granted a subsidy, the most hazardous sites were granted more subsidies than the less hazardous sites. Altogether, about 75% of the SEK 400 million were granted to the most hazardous sites, implying that approximately SEK 100 million could have been spent more wisely.
The project's number of employment opportunities did not affect the probability of receiving a subsidy. Nevertheless, the subsidy's magnitude was positively affected by the project's employment opportunities, i.e. the more employment, the greater the subsidy. Society's marginal cost for an employment opportunity however appears to be much higher in the remediation projects than in a traditional labour market policy.
Viewed separately, neither the ecological sustainability objective nor the employment objective were achieved satisfactory in the remediation projects. Although other subsets of the LIP projects may have been more successful, a combination of environmental and labour market policies appears – based on these findings – to lead to an inefficient use of public resources, possibly because focus is lost when there are several objectives in a single policy. Our message to policy-makers is therefore to be cautious of mixing disparate objectives in a single policy. A mixing policy is most likely only useful when the objectives are complementary, i.e. outcomes of the same process, but even under such circumstances there is evidence (Budh, 2005) that the sub-optimization costs from using separate policies are non-substantial.
The rest of the paper is organized as follows. In the next section, the environmental problem of contaminated sites in Sweden is described, as is the LIP. Section 3 specifies the model, and Section 4 describes the data. Estimation results are presented in Section 5. The findings are discussed in the concluding section.
Section snippets
Assessing risk
In 1990, the Swedish EPA was assigned the task of planning for remediation of contaminated sites. Between 1992 and 1994 a nation-wide inventory of industries was carried out to identify the sites in greatest need of remediation (Swedish EPA, 1995). The sites were classified by category of risk, ranging from “very high risk” (risk class 1) to “low risk” (risk class 4), according to their historical and on-going industrial activities.6
Model specification
Because the guidelines for the LIP (Regeringskansliet, 2000) state that applications for the LIP subsidy should be assessed according to the anticipated outcome of the project, we formulate a subsidy function (Eq. (1)) where the magnitude of the LIP subsidy to a remediation project depends on the project's anticipated beneficial environmental and employment effects, i.e.
Thus, we assume that the LIP subsidy to remediation project i, in municipality j, at time t,
Data
Data on the accepted projects were available from the LIP database at the Swedish EPA. Data on the rejected projects could only be found in the LIP applications, filed at the MoE or, for 2002, at the Swedish EPA.16
Results
Table 4 gives the estimation results from the discrete margin of the allocation decision (Eq. (2)).22
Discussion
In this paper we evaluate how successful a major Swedish subsidization scheme, the Local Investment Programme (LIP), was in fulfilling its dual objectives; to improve ecological sustainability and to reduce unemployment. In order to get a yardstick to which the ecological sustainability objective can be compared, we focus on a subset of the LIP projects, i.e. the projects aimed at remediation of contaminated sites, and employ the Swedish parliament's interim target for remediation of such sites.
Acknowledgements
The authors would like to thank Henrik Hammar, Per Johansson, Pelle Marklund, Magnus Sjöström, Göran Östblom, Anna Öster, participants at the NIER seminar and two anonymous referees for valuable comments, and Therese Lager and Anna Torkelstam for collecting portions of the data. Financial support from The Swedish Research Council for Environment, Agricultural Sciences and Spatial Planning (FORMAS) is gratefully acknowledged.
References (45)
Intergovernment grants as a tactical instrument: empirical evidence from Swedish municipalities
Journal of Public Economics
(2003)Arbetsmarknadspolitiska åtgärder
- et al.
Evaluation of a national investment program for local sustainable development with a focus on environmental work (in Swedish)
(2003) Can Integrated Control Strategies for Multiple Emissions Enhance Cost-efficiency in Environmental Policy? Evidence from Sweden
(2005)- et al.
Microeconometrics. Methods and Applications
(2005) - et al.
The determinants of pesticide regulation: a statistical analysis of EPA decision making
The Journal of Political Economy
(1992) - et al.
Direkta undanträngningseffekter av arbetsmarknadspolitiska åtgärde (in Swedish)
Ekonomisk Debatt
(2000) - et al.
On the vote-purchasing behaviour of incumbent governments
American Political Science Review
(2002) - et al.
The determinants of success of special interests in redistributive politics
Journal of Politics
(1996) - Environmental Objectives Portal, 2005. http://www.miljomal.nu/english/objectives.php. Accessed...