Runar Brännlund, Umeå University

Objective with the project

The overarching objective with the project is to increase the body of knowledge about the long run relationship between economic growth and the environment, especially emissions. The project builds on complementary research areas, mainly economic history and economics. More specifically the objectives with the project can be summarized by the following points:
1. To empirically test for the occurrence of Environmental Kuznets Curves (EKC), and possibly explain the development in terms of technological change and changes in preferences.
2. To estimate technological change, change in total factor productivity (TFP) using an extended definition of production that includes emissions (bad outputs).
3. To estimate demand elasticities for different goods, and combining income data with data on emissions for the purpose of investigating the role of income for emissions.

Results

The project has contributed with several important results within both economic-history and economics. Taken together the results show that the development of technological progress and preferences, and their interaction, in Sweden work in favour of an EKC for airborne emissions such as sulphur and carbon dioxide.
Within economic history a substantial effort has been made to collect and refine a detailed dataset for the manufacturing industry 1913 to the present. Besides annual output data, constant price volumes and labour input from 1913, the dataset includes fuel and electricity consumption (approximately 10 categories of fuels) from 1936. A common classification system was elaborated, approximately corresponding to the presently used ISIC-4 level. The data set is useful for various economic historical investigations, for instance analysis of energy consumption and structural change (see for instance Balk et al 2005, Lindmark 2007a and 2007b). The dataset is described in detail by Andersson (2005).
We have furthermore collected and elaborated the household budget surveys from 1913 up to 2000. This data was for instance used in the articles by Brännlund & Ghalwash (2008), and Ghalwash (2007b).
Additional economic-historical research includes Lindmark (2004) which deals with transition patterns of GDP and carbon dioxide intensities among rich and poor countries in a long time perspective. The main results are that the carbon dioxide intensities have develop according to some specific patterns, which are depending on when a country was industrialised. This shows the importance of time specific technology. Today, the differences according to carbon dioxide intensity between countries are less pronounced as compared to historical periods, such as the interwar period. Lindmark & Kander (2005) have investigated the implicit carbon dioxide and energy exports through Swedish foreign trade. The study shows that Sweden has been a net energy and carbon exporter with respect to the mix of traded goods. In another study, Lindmark & Kander (2004) gives a very long term perspective on Swedish environmental and energy efficiency performance during the last 200 years. Worth to notice is that the article has triggered further international efforts in the field. As a sequel to this investigation, Lindmark has started an investigation of Norwegian energy consumption and carbon dioxide emissions since 1830. Lindmark & Vikström (2003) further investigated carbon dioxide productivity in a panel over several developed and developing countries. The conclusion is that the total efficiency is not obviously a function of income levels, if carbon dioxide is included as a bad output. The manuscript will be submitted to an international journal for consideration. Vikström (2004) did study capital and energy substitution in the manufacturing industry 1954 to 1964 (based on the previously mentioned data set). The main goal was the estimation of substitution elasticities. The theoretical foundations for studying the re-bound effect in a dynamic CGE setting was elaborated in Vikström (2004b). Both manuscript will be published in the Economic History Department working paper series.
Among follow-up studies from the project we can mention Lindmark & Bergquist (2007), which provides a long-term overview and analysis of Swedish energy use and carbon dioxide emissions from a historical land-use perspective. Also Lindmark & Bergquist (2008) focuses the environmental adaptation of the Rönnskär metal smelter, using the theoretical framework of this project, which has been presented in Kriström & Lundgren (2005) and Johansson & Kriström (2007).
The pure economic history part of the project can be viewed as the base on which the empirical economic research within the project rests. The basic theoretical framework for the so called Environmental Kuznets Curve (EKC) is provided in Johansson & Kriström (2007). They effectively show that the EKC can be viewed as a particular equilibrium relationship, where technology and preference parameters determine the shape of the curve. The main conclusion drawn from the theoretical analysis is essentially that we can neither reject nor accept the hypothesis of an EKC, on theoretical grounds. The theory in Johansson & Kriström is confronted with Swedish long run data on sulfur emissions. The empirical results show that we can't reject an EKC for sulfur. But it is also shown that the relationship is structurally instable.
Balk et.al. (2006) use similar historical data for the Swedish manufacturing industry to estimate what they denote an "environmental performance index". The index is calculated using data on emissions of sulfur and carbon dioxide. The results from the calculations show that there is a significant change in the index over time in the sense that environmental performance improves over time. The calculated index is used in a second step as the dependent variable in a statistical analysis of the relationship between GDP and the environmental performance index. The results show from the second step show a clear EKC pattern. These results are supported by the results in Kriström & Lundgren (2007).
The studies referred to above can be classified as macro studies. However, within the project there is a number of micro studies, especially concerning household preferences. Ghalwash (2007) studies demand for consumption goods that are complementary to the "consumption" of environmental services, such as recreation. More specifically the objective of the paper is to estimate the income elasticitity, and its change, for goods that are complements to environmental goods and services. The study make use of household specific survey data that cover a major part of the 2000'th century, including the 1913 household survey mentioned above. The results show that a given percentage increase in income leads to higher percentage increase in consumption of goods that are complements to the environment, i.e., as income grows we tend spend a larger share of the income increase on goods related to environmental services. Brännlund & Ghalwash (2008) is using essentially the same dataset, appended with data on good specific emission data, to study also the role of income distribution for the EKC. The results support the EKC hypothesis, but they also show that the relationship depends on the income distribution in the society. An equalization of income, at a given income level, implies that the turning point on the EKC shifts to the right as a result of a more even income distribution. In both of the above cited studies it was assumed that an individual, or household, base its consumption decision on its own income level. In Broberg (2007) this assumption is relaxed, and instead it is assumed that the consumption choice may also depend on the individual's income relative to others income. The study deviates from the previous ones also with respect to methodology. Instead of using actual observed data, Broberg (2007) use a stated preference approach in the form of a willingness to pay survey. The results show that the willingness to pay for a virgin forest area is positively correlated with the income level, but also that the willingness to pay depends on the individual's relative income.

New research issues emerging from the project

The research within the project has of course resulted in new questions, or at least issues that should be penetrated further. One such issue is the role of income distribution. This is an important question for several reasons, not the least concerning the problem with climate change. When discussing solutions (or non-solutions) to the climate change problem it becomes very clear that the question of income distribution, both within and between countries, becomes of great importance. Another issue that should be penetrated further is the source of technical change and the role of policy in technical change. The studies in this project, referred to above, have treated technical change as exogenous. Future research in this area should take a step further and more systematically consider endogenous technical change as well as policy induced technical change.

The two most important publications

An important publication within the project is Andersson (2005), although not published in a peer review journal. The publication can be viewed as the base for most of the empirical studies that have been published within the project. Concerning the peer reviewed studies from the project it is hard to rank them according to importance. Most of them are complement to each other in different ways. In the "results" section above we refer to a subset of all the publications in the project, which means that we view them as very central.

Outreach

The project has resulted in several international publications, and the project participants have attended actively in several international workshops and conferences. Worth to mention is the European Historical Energy Project, EGP, and the annual conference in environmental and resource economics arranged by the European Association of Environmental and Resource Economics.
Internal work
The internal work within the project has worked well. In spite of different subject areas the participants have become a "group" which has contributed significantly to the research environment in environmental and resource economics at the Umeå campus. There has been regular workshops and seminars, as well as a larger workshop in Ammarnäs which included invited researchers from USA, Spain and Scotland.

Human capital build up

The project has been valuable and important for the human capital build up in Umeå within the subject are of environmental and resource economics. The project has resulted in two doctoral dissertations, and a third dissertation that is partially financed by the project (Broberg, 2007).