6. How Big Are Green Spending Multipliers?1
© 2022 Chapter Authors, CC BY-NC 4.0 https://doi.org/10.11647/OBP.0328.06
Introduction
Fixing the twin climate and biodiversity crises is still possible, but it requires stewarding the global economy within limits set by nature (Rockström et al. 2017; Attenborough 2020; Georgieva 2020; Stiglitz 2020; Carney 2021). Although some have argued that cutting emissions and protecting wildlife clashes with job creation and growth (see, for example, Walley and Whitehead 1994; NERA 2017; and Christian 2021), analysis based on a global survey of experts found that green projects are widely perceived as capable of creating more jobs and delivering higher short-term returns per dollar spent by comparison with traditional fiscal stimuli (Hepburn et al. 2020).
In a recent paper (Batini et al. 2022), we contribute to this debate. To our knowledge, it is the first study directly estimating the effect on GDP of money spent to foster the transition to a zero-carbon, nature-friendly world for a variety of green expenditure typologies. Although ‘green’ expenditure has historically tended to be defined as spending that helps reduce greenhouse gas emissions, we expand the definition to include examples of nature-based negative emissions technologies (“nature-based solutions” or NBSs) in the form of expenditure on biodiversity conservation and rewilding. These are increasingly regarded by science as solutions that support the Earth’s natural capabilities to sequester carbon and mitigate climate change. Moreover, these measures have been shown to be a vital complement of planetary climate and global temperature stabilisation strategies (IPCC 2019; IPBES 2019; Foley et al. 2020; Dasgupta et al. 2021).
Employing a new international dataset, part of which was especially put together for the analysis, Batini et al. (2022) find that every dollar (private and public) spent on green activities—from zero-emission power plants to the protection of wildlife and ecosystems—can generate more than a dollar’s worth of economic activity: the total increase in GDP is greater than the original increase in green spending. These economic effects appear significantly larger and more long-lasting than ‘non-green’ spending in alternative energy technologies or land/sea uses, a result connected to the fact that green spending tends to be both more labour intensive and richer in domestic content than non-green spending, as we discuss later. Although green and non-ecofriendly expenditures are not always strictly comparable due to data limitations, the estimated multipliers associated with green spending are found to be generally bigger than those associated with non-green expenditure. In the case of renewable versus fossil fuel energy investments, where country and time samples are homogeneous and allow for a formal statistical comparison, green spending multipliers are about twice as large as their non-green counterparts. The point estimates of the multipliers are 1.1–1.7 for renewable energy investment and 0.4–0.7 for fossil fuel energy investment, depending on horizon and specification.
These findings suggest that, in crafting a post-COVID-19 recovery, investments in energy and land/sea use transitions may be economically superior to those offered by supporting economic activities involving unsustainable ways to produce energy and food.
The remainder of this chapter is organised as follows. Section 2 summarises the empirical results by Batini et al. (2022). Section 3 draws policy implications and concludes.
6.1 Results
While for the details on the construction of the dataset and on the empirical methodology, we refer to Batini et al. (2022), here we summarise the main results. We discuss results by sector, starting with energy then moving to land use, and comparing output effects of green and non-green spending. We use cumulated spending multipliers, defined as the cumulative change in GDP divided by the cumulative change in spending on energy or land use, at various time horizons, following the approach proposed by Gordon and Krenn (2010) and Ramey and Zubairy (2018).
Multiplier values should be interpreted in the standard way. For example, a value of the cumulated spending multiplier equal to, say, 1.5 in the third year would indicate that, after three years from the occurrence of the spending shock, the cumulative increase in output, in dollar terms, is one and a half times the size of the cumulative increase in green (or non-green) expenditure. In this case, then, a change of, for example, US$100 in public or private investment in clean energy infrastructure or power generation will have an effect of more than US$100 (and precisely US$150) on the level of real GDP.
6.2 Green Energy versus Non-Green Energy Spending Multipliers
In this subsection we report cumulated multipliers of spending on clean energy (renewable and non-renewable) versus spending on non-green energy (fossil fuel energy generation). It is worth noting upfront that multipliers related to fossil fuel and renewable energy generation are fully comparable because their underlying data cover the same country and time sample. The data on nuclear energy spending cover a smaller set of countries and a larger number of years, therefore they are not strictly comparable.
For both short and longer horizons the green renewable energy spending multiplier is systematically higher than the non-green energy multiplier (Table 6.1). Specifically, the impact multiplier for green renewable energy is 1.19. For non-green energy, the impact multiplier is 0.65, suggesting that these kinds of expenditures tend to crowd out private investment or consumer spending that would have otherwise taken place to a larger extent.
Focusing on the impact multiplier, however, may be misleading because investments in energy can only be implemented over time and the economy may only respond gradually. The cumulative multiplier for green renewable energy spending falls only marginally over the years and plateaus to a five-year value of 1.11, very close to the first-year effect. This may reflect the fact that renewables are built sequentially and the persistence of the multiplier as well as the fact that the composition of their investment vector typically includes different types of activities (construction itself, networks for transmission and distribution, smart meters, etc.). For non-green energy spending, however, the multiplier becomes even smaller at year five (0.52). In other words, when an additional dollar of public or private money is spent to build more fossil fuel energy infrastructure and power generation plants, this expenditure crowds out some other component(s) of GDP (investment, consumption, and/or net exports) by 48 cents in the medium run. When the same dollar is spent on solar, wind or geothermal, 11 cents are instead crowded in. In addition, while the green multiplier is statistically significant up until four years after the shock occurrence, the non-green multiplier loses its significance after three years.2
Horizon |
Green (Renewable) Energy Investments Multiplier |
Non-Green Energy Investments Multiplier |
Impact |
1.19* |
0.65* |
1 Year |
1.20* |
0.64* |
2 Years |
1.19* |
0.62* |
3 Years |
1.17* |
0.59* |
4 Years |
1.14* |
0.55 |
5 Years |
1.11 |
0.52 |
These results are intuitive on two grounds. First, clean energy is more labour intensive than carbon-based fuels spending. In relation to spending within fossil fuel industries, spending on clean energy—including the direct spending on specific projects plus the indirect spending of purchasing supplies—uses far more of its overall investment budget on hiring people, and relatively less on acquiring land (either on- or offshore), machines, and supplies and energy itself (Wiser et al. 2017; IRENA 2016; Garrett-Peltier 2017). In addition to the jobs directly created in the renewable energy industry, growth in clean energy can create positive economic “ripple” effects. For example, both industries in the renewable energy supply chain and unrelated local businesses benefit from increased household and business incomes (EPA 2020; IEA 2020). Moreover, clean-energy investments produce far more jobs at all pay levels—higher as well as lower-paying jobs—than the fossil fuel industry (E2-ACORE-CELI 2020). For the United States, Muro et al. (2019) find that workers in clean energy earn mean hourly wages that are between 10% and 20% above the national average; and their wages are more equitable, with workers at lower ends of the income spectrum earning up to US$10 more per hour than other jobs. At the same time, clean-energy investments also produce more jobs for a given dollar of expenditure due to the larger number of entry-level jobs relative to the fossil fuel industry. Second, clean energy implies a higher domestic content than fossil fuel energy, which explains the crowding out of demand from spending on the latter, as money spent on fossil fuel plants or generation tends to “leak” abroad.3 Considering direct plus indirect spending, clean energy spending relies much more on economic activities taking place within the domestic economy—such as retrofitting homes or upgrading the electrical grid system locally—than spending within conventional fossil fuel sectors (IRENA 2016; EPA 2020). These considerations help rationalise the much stronger multiplier effect of clean spending than that of non-green spending on the larger economy.
Table 6.2 reports cumulated spending multipliers of non-renewable clean energy (nuclear energy), indicating that spending on nuclear energy has a large output effect, about six times larger than the output effect associated with spending on fossil fuel energy. However, nuclear spending multipliers lose statistical significance after two years from the occurrence of the shocks.
Horizon |
Nuclear Energy Investments Multiplier |
Impact |
4.11* |
1 Year |
3.97* |
2 Years |
3.88 |
3 Years |
3.83 |
4 Years |
3.80 |
5 Years |
3.78 |
Although nuclear spending multipliers are not strictly comparable to the other two sets of multipliers, its initially larger values may be linked to their nature. Relative to other forms of clean energy (e.g., solar and wind) investments in nuclear energy may lead to larger employment of both high- and lower-skilled resources for the construction of nuclear reactors relative to lighter energy-producing infrastructure. In addition, while building and operating nuclear reactors tends to take time (5.1 years on average for large reactors of recent construction) spending is not sequential like in the case of renewables and tends to be more frontloaded, which could explain the stronger near-term impact and subsequent loss of statistical significance. Findings in studies comparing a steady-state employment estimate for the generation of electricity using nuclear versus wind power indicate that investment in nuclear power produces about 25% more employment per unit of electricity than wind power (WNA 2020). Moreover, research comparing pay across nuclear, wind and solar direct workforces in the United States in 2017 indicates that pay of nuclear workers is one-third higher than that in the wind and solar sectors, and that they were paid more than twice the mean for power sector workers (Oxford Economics 2019). In the medium term, the point estimate of the nuclear energy spending multiplier is still larger than the renewable energy counterpart, but not being statistically significant, does not allow us to draw definite conclusions.
6.3 Green Land Use versus Non-Green Land Use Multipliers
Lastly, we consider spending on ecosystem conservation (green land use spending) versus a shock of the same size to spending on subsidies to conventional agriculture (non-green land use spending).
Interpreting differences in multipliers from spending in these two land use categories requires caution for two reasons. First the multipliers have been estimated over different country and time samples, and in two separate econometric specifications, because of data coverage and availability constraints explained in Batini et al. (2022). This is also the reason why a statistical test on their difference cannot be constructed. In addition, spending in conservation reflects a mix of public spending in wages, education, training and recreational programming (which are thus part of public consumption) and some public investment,4 whereas spending on conventional agriculture here reflects primarily public transfers and subsidies to crop and animal producers in industrial farm systems. However, even coarse comparisons of average output effects of spending on sustainable versus unsustainable land uses can be informative, as a consensus is emerging that subsidies to unsustainable land use and conventional agriculture should be quickly redirected toward sustainable uses (see for example UNEP-UNDP-FAO 2021). Getting a sense of the potential economic gains (or losses) of redressing land use subsidies to sustainable and land regenerative goals is key for policymaking and budgetary decisions.
Table 6.3 reporting cumulated spending multipliers on green versus non-green land use shows that, while green land use spending multipliers are not significantly different from zero on impact and over the first year’s horizon, cumulated multipliers at horizons greater than one year are large and grow over time. This suggests that spending to sustain natural ecosystems exerts powerful positive ripple effects on the economies that practice it: for every dollar spent in conservation, almost seven more are generated in the larger economy in the medium term, a result in line with findings in bottom-up analyses of local and regional impacts (see Batini et al. 2022).
Horizon |
Green Land Use Multiplier |
Non-Green Land Use Multiplier |
Impact |
-5.36 |
0.55* |
1 Year |
-1.60 |
0.85* |
2 Years |
1.45* |
0.95* |
3 Years |
3.75* |
0.96* |
4 Years |
5.45* |
0.95 |
5 Years |
6.67* |
0.94 |
By contrast, the multipliers of spending to support industrial agricultural production are below one at every horizon. This reflects the high mechanisation of industrial agriculture, the typically low value added associated with high costs of machinery, fossil fuel energy, and imported chemical inputs and foreign-patented GMO seeds, all of which tend to have low domestic content, given the high global market concentration of suppliers of all these inputs (FOLU 2019; UNEP 2020; UNEP-UNDP-FAO 2021).
The high multipliers associated with green land use are expected and can be ascribed to two main determinants. First, as documented by Waldron et al. (2020) the conservation activity has a strong labour intensity. Much of the economic impact of conservation is in driving a visitor economy, with associated creation of opportunity and income in sectors such as hospitality and tourism in rural and coastal communities which, in developing countries, tend to have below average income, a higher marginal utility of income, and thus are more likely to have higher propensities to spend. Second, by limiting land available for agricultural expansion, conservation spending lifts the prices paid to rural producers (Waldron et al. 2020). More generally, protecting biodiversity helps underpin the ecosystem services upon which economic activity and lives depend, such as food production, fresh water, natural resources and the protection from extreme weather events. These activities all create jobs and inspire innovation through biomimicry (Kennedy and Marting 2016; OECD 2020). While keeping in mind the caution on comparability made above, this finding is a potential indication that repurposing spending from unsustainable land uses toward more labour intensive and high-domestic-content sustainable land uses may promise important economic gains and may hold the keys to a successful green recovery.
6.4 Conclusions
Drawing on the work of Batini et al. (2022), in this chapter we discussed empirical evidence about output multipliers of spending in green and non-green energy and land use. Spending on the green economy is both efficient—returning more than the initial investment in all cases—and superior to spending on non-green activities. In the case of renewable versus fossil fuel energy investments, where country and time samples are homogeneous and allow for a formal statistical comparison, multipliers on green spending, at 1.1–1.7, are about twice as large as their non-green counterparts, at 0.4–0.7, depending on the estimation horizon and specification used.5
These findings can be rationalised by noting that, compared with fossil fuel technologies, which are typically mechanised and capital intensive, the renewable energy industry is more labour intensive, and investments have a higher domestic content. This feature is highlighted in sector studies, showing that, on average, more jobs are created for each unit of electricity generated from renewable sources than from fossil fuels. Similar results emerge for spending on nuclear energy, for which there is an even greater multiplier than renewable energy—albeit obtained on a different dataset and thus not formally comparable.
Likewise, findings on ecosystem conservation spending show that it is associated with large economic gains. In contrast, spending to support unsustainable land uses—highly mechanised and imported-input-dependent industrial crop and animal agriculture—returns less than the initial expenditure. While these estimates originate in different datasets and preclude a formal discussion on their statistical difference, they are indicative of a potential economic advantage of a sustainable use of land relative to the widespread conventional farming practices.
All in all, these findings lend support to those post-COVID-19 stimulus programmes that prioritise green investments. For instance, the European Union’s Next Generation EU plan, approved to help member states repair the economic damages caused by the COVID-19 pandemic, features the green transition as one of its core elements.
References
Attenborough, D. (2020) A Life on Our Planet: My Witness Statement and a Vision for the Future, New York: Grand Central Publishing.
Batini, N., M. Di Serio, M. Fragetta, G. Melina and A. Waldron (2022) “Building back Better: How Big are Green Spending Multipliers?”, Ecological Economics 193: 107305, https://doi.org/10.1016/j.ecolecon.2021.107305.
Carney, Mark (2021) Value(s): Building a Better World for All, New York: PublicAffairs Books.
Christian, W. (2021) “Opinion: An Assault from All Fronts on Energy Independence”, Speech posted on 5 Jan 2021, Texas Railroad Commission, https://www.worldoil.com/news/2021/1/5/opinion-an-assault-from-all-fronts-on-energy-independence.
Dasgupta, P. et al. (2021) Report of the Independent Review on the Economics of Biodiversity led by Professor Sir Partha Dasgupta. London: HM Treasury, https://www.gov.uk/government/publications/final-report-the-economics-of-biodiversity-the-dasgupta-review.
Environment Protection Agency (EPA) (2020) Quantifying the Multiple Benefits of Energy Efficiency and Renewable Energy: A Guide for State and Local Governments. Washington D.C.: U.S. Environment Protection Agency, https://www.epa.gov/statelocalenergy/quantifying-multiple-benefits-energy-efficiency-and-renewable-energy-guide-state.
Foley, J., K. Wilkinson, C. Frischmann, R. Allard, J. Gouveia, K. Bayuk, M. Mehra, E. Toensmeier, C. Forest, T. Daya, et al. (2020) The Drawdown Review (2020)—Climate Solutions for a New Decade. Project Drawdown, https://drawdown.org/sites/default/files/pdfs/TheDrawdownReview%E2%80%932020%E2%80%93Download.pdf.
FOLU (2019) Growing Better: Ten Critical Transitions toTransform Food and Land Use, https://www.foodandlandusecoalition.org/wp-content/uploads/2019/09/FOLU-GrowingBetter-GlobalReport.pdf.
Garrett-Peltier, H. (2017) “Green versus Brown: Comparing the Employment Impacts of Energy Efficiency, Renewable Energy, and Fossil Fuels Using an Input-output Model”, Economic Modelling 61: 439–47.
Georgieva, K. (2020) “The Long Ascent: Overcoming the Crisis and Building a More Resilient Economy.” Speech delivered for the 125th Anniversary of the London School of Economics on 6 October 2020. Washington D.C.: International Monetary Fund.
Gordon, R.J. and R. Krenn (2010) “The End of the Great Depression 1939–41: Policy Contributions and Fiscal Multipliers”, NBER Working Paper, National Bureau of Economic Research.
Hepburn, C., B. O’Callaghan, N. Stern, J. Stiglitz, and D. Zenghelis (2020) “Will COVID-19 Fiscal Recovery Packages Accelerate or Retard Progress on Climate Change?”, Smith School Working Paper 20–02.
Intergovernmental Panel on Climate Change (2019) Climate Change and Land: An IPCC Special Report on Climate Change, Desertification, Land Degradation, Sustainable Land Management, Food Security, and Greenhouse Gas Fluxes in Terrestrial Ecosystems. Geneva: IPCC.
Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services (IPBES) (2019) Global Assessment Report on Biodiversity and Ecosystem Services. Bonn: IPBES.
International Energy Agency (IEA) (2020) Renewable Energy Market Update: Outlook for 2020 and 2021. Paris: International Energy Agency, https://www.iea.org/reports/renewable-energy-market-update.
International Renewable Energy Agency (IRENA) (2016) Renewable Energy: Measuring the Economics, https://www.irena.org/-/media/Files/IRENA/Agency/Publication/2016/IRENA_Measuring-the-Economics_2016.pdf.
Kennedy, E. and T. Marting (2016) “Biomimicry: Streamlining the Front End of Innovation for Environmentally Sustainable Products”, Research-Technology Management, 59(4): 40–48, https://doi.org/10.1080/08956308.2016.1185342.
Muro, M., A. Tomer, R. Shivaram, and J. Kane (2019) Advancing Inclusion Through Clean Energy Jobs, Metropolitan Policy Program at Brookings. Washington DC: The Brookings Institution, https://www.brookings.edu/research/advancing-inclusion-through-clean-energy-jobs/.
NERA (2017) Impacts of Greenhouse Gas Regulations on the Industrial Sector, Report prepared for the American Council for Capital Formation Center for Policy Research, https://www.globalenergyinstitute.org/sites/default/files/NERA%20Final%20Report%202.pdf.
OECD (2020) Biodiversity and the Economic Response to COVID-19: Ensuring a Green and Resilient Recovery, https://www.oecd.org/coronavirus/policy-responses/biodiversity-and-the-economic-response-to-covid-19-ensuring-a-green-and-resilient-recovery-d98b5a09/.
Oxford Economics (2019) “Nuclear Power Pays—Assessing the Trends in Electric Power Generation Employment and Wages”, Oxford Economics, https://www.oxfordeconomics.com/resource/nuclear-power-pays-assessing-the-trends-in-electric-power-generation-employment-and-wages/.
Ramey, V.A. and S. Zubairy (2018) “Government Spending Multipliers in Good Times and in Bad: Evidence from U.S. Historical Data”, Journal of Political Economy 126 (2): 850–901, https://doi.org/10.1086/696277.
Rockström, R. J., O. Gaffeny et al. (2017) “A Roadmap for Rapid Decarbonization”, Science 355(6331): 1269–71, https://doi.org/10.1126/science.aah3443.
Stiglitz, J. (2020) Green Dealing—The Green Recovery Event, Project Syndicate, 15 October 2020, https://www.project-syndicate.org/videos/green-dealing-the-green-recovery-event?referral=d582d5.
United Nations Environment Programme (UNEP) (2020) 10 Things You Should Know About Industrial Farming. Nairobi: UNEP, https://www.unep.org/news-and-stories/story/10-things-you-should-know-about-industrial-farming.
United Nations Environment Programme-United Nations Development Programme-Food and Agriculture Organization (UNEP-UNDP-FAO) (2021) A Multi-Billion-Dollar Opportunity: Repurposing agricultural support to transform food systems. Nairobi: UNEP, https://www.unep.org/resources/repurposing-agricultural-support-transform-food-systems.
Waldron, A., V. Adams, J. Allan, A. Arnell, G. Asner, S. Atkinson, A. Baccini, J. E. M. Baillie, A. Balmford, and J. Austin Beau (2020) Protecting 30% of the Planet for Nature: Costs, Benefits and Economic Implications. Cambridge: Cambridge University, https://www.conservation.cam.ac.uk/files/waldron_report_30_by_30_publish.pdf.
Walley, N. and B. Whitehead (1994) “It’s Not Easy Being Green”, Harvard Business Review, May-June 1994, https://hbr.org/1994/05/its-not-easy-being-green.
Wiser, R. and M. Bolinger (2017). 2016 Wind Technologies Market Report. Washington D.C.: U.S. Department of Energy, https://www.energy.gov/eere/wind/downloads/2016-wind- technologies-market-report.
World Nuclear Association (WNA) (2020) Employment in the Nuclear and Wind Electricity Generating Sectors. WNA Report No. 2020/006. London: World Nuclear Association, https://www.world-nuclear.org/getmedia/690859bf-ebe6-43a2-bedd-57ddf47ee3ac/Employment-in-Nuclear-Report-Final.pdf.aspx.
1 This chapter is based on the findings of Batini et al. (2022). The views expressed in this paper are those of the authors and do not necessarily represent those of the IMF, its Executive Board, the Independent Evaluation Office, IMF management, or the UK’s FCDO.
2 For the sake of simplicity, we prefer to use the terminology of statistical significance, in analogy to the frequentist approach to inference. However, the Bayesian approach used in the analysis formally leads to credible intervals around the estimates. We consider “significant” those multipliers with credible intervals, delimited by the 16th and the 84th percentiles, that exclude zero.
3 In addition, network effects may be important: oil fields and gas wells tend to be economic and geographic enclaves which may lead to smaller multipliers.
4 For example this includes the construction and the maintenance of infrastructure such as fences, boardwalks, observation platforms, and other durable machinery such as communication equipment and optical devices for distant viewing, vehicles or satellite monitoring and GPS tracking devices necessary to perform conservation services.
5 Please see the detailed results and the robustness analysis in Batini et al. (2022).