Elsevier

Economic Modelling

Volume 61, February 2017, Pages 439-447
Economic Modelling

Green versus brown: Comparing the employment impacts of energy efficiency, renewable energy, and fossil fuels using an input-output model

https://doi.org/10.1016/j.econmod.2016.11.012Get rights and content

Highlights

  • We estimate 9 clean energy and 2 fossil fuel industry employment multipliers.

  • We find $1 million spending creates 7.49 FTE jobs in renewables, 7.72 in EE.

  • Fossil fuel industries support 2.65 FTE jobs per $1 million spending.

  • We show how to create a “synthetic” industry such as EE or RE using I-O tables.

  • The synthetic industry approach is transparent and easily replicable.

Abstract

Global carbon emissions have reached unsustainable levels, and transforming the energy sector by increasing efficiency and use of renewables is one of the primary strategies to reduce emissions. Policy makers need to understand both the environmental and economic impacts of fiscal and regulatory policies regarding the energy sector. Transitioning to lower-carbon energy will entail a contraction of the fossil fuel sector, along with a loss of jobs. An important question is whether clean energy will create more jobs than will be lost in fossil fuels. This article presents a method of using Input-Output (I-O) tables to create “synthetic” industries – namely clean energy industries that do not currently exist in I-O tables. This approach allows researchers to evaluate public and private spending in clean energy and compare it to the effects of spending on fossil fuels. Here we focus on employment impacts in the short-to-medium term, and leave aside the long-term comparison of operations and maintenance employment. We find that on average, 2.65 full-time-equivalent (FTE) jobs are created from $1 million spending in fossil fuels, while that same amount of spending would create 7.49 or 7.72 FTE jobs in renewables or energy efficiency. Thus each $1 million shifted from brown to green energy will create a net increase of 5 jobs.

Introduction

Employment in the clean energy sector, or so-called “Green Jobs,” has become an important issue politically as an avenue to reduce unemployment while growing the economy on a sustainable path. Carbon emissions have reached an unsustainable level, and the global energy system must be transformed in order to limit global climate change to a 2 degree Celcius rise above pre-industrial levels by 2100 (Bruckner et al., 2014). Many studies have evaluated the relationship between economic growth, energy use, and emissions (Bloch et al., 2015, Chen et al., 2016; Narayan et al., 2016). There is debate both among economists and environmentalists as to whether economic growth is necessary and what level is sustainable. “Green Growth” is sometimes seen as a way for standards of living to rise while carbon emissions fall (e.g. Pollin et al., 2014).

National governments have long used both fiscal policy and regulation in the energy sector. Fiscal policy has included tax preferences, direct public spending, loan guarantees and other financing mechanisms, investments in research and development, and many other forms of financial support and incentives. In simultaneously addressing questions of job creation, emissions reduction, and energy use, it is important to understand the economic effects of energy policy and public spending within the energy sector. Particularly when the policy goal is to reduce unemployment, it is useful to compare the employment effects of clean energy as opposed to fossil fuels. For a given level of public spending, how many jobs would be supported by renewable energy (RE) and energy efficiency (EE) industries compared to fossil fuel (FF) industries? What would be the net effect on employment if spending shifts from fossil fuels to clean energy?

There is a small but growing body of literature examining the economic benefits of a clean energy transition, including those of job creation (Pollin et al., 2014, Lehr et al., 2012). While the peer-reviewed literature is still limited and the impacts on employment are sometimes noted as being disputed (Lambert and Silva, 2012), a number of researchers using input-output modeling have found positive employment impacts resulting from the growth of renewable energy industries (Malik et al., 2014; Markaki et al., 2013; Tourkolias and Mirasgedis, 2011) Input-output models are often used for these studies because they have the advantage of being transparent, having few assumptions built in, are easily replicable, and are built from current or recent data from national accounts.

One current drawback of using I-O models to study clean energy impacts is that renewable energy industries and energy efficiency industries are not explicitly identified as industrial categories in national accounts. While industries such as oil extraction, natural gas distribution, and petroleum refining exist in most national accounting systems, industries such as wind, solar, or home weatherization do not. Thus we cannot readily compare the employment impacts of demand for fossil fuel to demand for clean energy, making policy assessment difficult. Some researchers have overcome this obstacle by disaggregating and re-aggregating the industries in the I-O tables in order to separate green from traditional activities, or by adding a new set of industries to the existing tables (Malik et al., 2014, Garrett-Peltier, 2011).

A re-aggregation approach is useful for analyzing mature industries and has been used to study “satellite accounts” such as the Health Care Satellite Account and the Travel and Tourism Satellite Accounts of the U.S. Bureau of Economic Analysis (U.S. BEA). The main drawback to such an approach is the collection or availability of data sufficient for detailed disaggregation, a problem that is particularly concerning with nascent industries such as clean energy. This re-aggregation approach is generally time-consuming, requiring detailed survey data as well as compilation and re-balancing of the input-output tables.

This article presents an alternative approach for analyzing the employment impacts of renewable energy and energy efficiency industries. I use an input-output model, but rather than re-aggregating the tables to separate green from non-green activities, I treat clean energy spending as a demand shock. That is, I treat additional spending as investments in the industry, and simulate the effects of expanding the clean energy sector by creating vectors of demand that include the manufacturing, construction, and service industries that comprise the clean energy sector. As shorthand, I refer to this as the “synthetic industry” approach.

In essence, what the “synthetic industry” approach does is use existing data in the national accounts, in existing I-O tables. While “wind” or “solar” are not identified as industries as such, the activities and materials making up the wind and solar industries are already implicitly captured in existing I-O tables. In the synthetic industry approach, we use existing data from national accounts and create a proxy, a vector of demand for the package of goods and services making up each synthetic industry.

The advantages of the “synthetic industry” approach are twofold: First, the data requirements are fewer than those necessary for re-aggregation; and second, this method could be used as a complementary modeling technique to re-aggregation, where this “synthetic industry” approach is used to assess the initial impacts of investments in creating or expanding an industry, and then the re-aggregation technique is used to model ongoing operations and maintenance of an established industry.

In this article I perform synthetic industry analysis for renewable energy industries such as wind, solar, and bioenergy, as well as for energy efficiency industries including building weatherization, mass transit and freight rail, and electrical grid upgrades. I build from the methods first presented in Garrett-Peltier (2011) but update with the most recent data available, which are the 71-industry “Summary” tables for 2013 from the U.S. BEA. In addition, I generate new cost vectors for various renewable and efficiency industries and compare the results with those using previously-established cost structures. I then present a method of analyzing the sensitivity of the results to the choice of specification.

I present the methodology for simulating demand for RE and EE industries, then use existing literature and survey data to form vectors of demand. I estimate employment multipliers for 9 clean energy industries and 2 fossil fuel industries. I then present a simple policy example estimating the overall impact of shifting $1 billion in fossil fuel subsidies into investments in renewable energy or energy efficiency. I provide the weights used in composing all of these 11 energy industries to allow replication by other researchers.

I find that on average, renewable energy creates 7.49 full-time-equivalent (FTE) jobs per $1 million spending, energy efficiency creates 7.72 FTE jobs per $1 million spending, and fossil fuels create 2.65 FTE jobs per $1 million spending. These job numbers include both direct and indirect (supply-chain) jobs, and the multipliers are the same regardless of whether the source of demand is public or private spending. It is important to note that these employment effects are relevant for the short-to-medium term, in which an expansion of clean energy involves significant increases in manufacturing and installation of renewable and efficiency technologies. Here we leave aside the long-run comparison of operations and maintenance employment in the energy sector. Overall, I find that the expansion of clean energy creates three jobs for each job lost in the fossil fuel sector, and for each $1 million shift from fossil fuels to clean energy, an average of five additional jobs are created.

Section snippets

Brief background on input-output analysis

Input-output (I-O) analysis is useful in estimating the impact of changes in demand for the output of an industry or group of industries. I-O tables provide a “snapshot” of the economy. In any given year, they show the inputs used by each industry, the outputs produced by each industry, and the relationship between industry output and final demand among various users.

I-O tables are constructed from surveys of businesses as well as from administrative records and are generally available at

Employment multipliers

The employment multipliers for RE (wind, solar, bioenergy, hydro, and geothermal), EE (building weatherization, freight rail & transit, industrial EE, and Smart Grid), as well as for FF (oil & gas and coal) are presented in Table 5. It is important to note that the multipliers for RE and EE relate specifically to the expansion of these industries, and thus are composed primarily of manufacturing and construction industries as well as related goods and services. The multipliers do not represent

Conclusion

In this article I have presented a method to estimate employment multipliers for industries that are not explicitly identified in input-output tables, termed “synthetic industries.” Specifically, I have estimated employment multipliers for clean energy industries including wind, solar, bioenergy, geothermal, hydropower, building weatherization, mass transit & freight rail, industrial EE, and Smart Grid. These clean energy industries are not identified as such in national accounts or in

References (26)

  • Bloomberg New Energy Finance (BNEF)

    Sustainable Energy in America: 2013 Factbook

    (2013)
  • T. Bruckner et al.

    Energy systems

  • Congressional Budget Office (CBO), 2015. Federal Support for the Development, Production, and Use of Fuels and Energy...
  • Cited by (215)

    • Measuring and characterising green jobs: A literature review

      2024, Energy Research and Social Science
    View all citing articles on Scopus
    View full text