Lessons Learned on Green Stimulus

Case Studies from the Global Financial Crisis

Global Overview

Green Stimulus Spending

Globally, about US$520 billion out of the $3.25 trillion fiscal stimulus went to green measures, when using a broad definition of the term (Robins et al. 2009; Robins et al. 2010; ILO 2011). Almost all of the global green stimulus was spent by the Group of 20 (G20), which comprises 19 countries and the EU. China allocated $218 billion, about a third of its stimulus package, to green measures. Almost all of this was spent on rail transportation, grid, and water management. The United States announced $118 billion in green stimulus, about 12 percent of its total. This included $33 billion on renewable power, three times more than the rest of the world combined. South Korea announced 69 percent of its stimulus for green measures, the highest share of any country, though it did not provide much detail on the breakdown. Japan and the EU round out the top five. Globally, the biggest category of green spending was on rail transportation (26 percent of the total), followed by grid modernization (18 percent), building energy efficiency (17 percent), water/waste management (15 percent), renewable power (8 percent), and low-carbon vehicles (4 percent). These data are presented in Figure 1 in total terms and in Figure 2 as a share of each country’s gross domestic product (GDP).

Figure 1 | Announced Green Stimulus by Country (September 2008–December 2009)

 

Notes: Water/Waste management includes sewage treatment, dams and flood defenses, canals and waterways, and environmental restoration. Rail transportation includes railways and public transit. Vehicle efficiency includes low-carbon vehicles, vehicle efficiency, and electric vehicles. Other includes projects defined by governments as green, but without details, plus carbon capture and storage and nuclear power. In most cases these estimates rely on government definitions of categories. These estimates are based on stimulus as announced, not as implemented.

Sources: Robins et al. 2009; Robins et al. 2010; ILO 2011.

Figure 2 | Announced Green Stimulus as Percentage of Total Stimulus (September 2008–December 2009)

 

Notes: Data from Robins et al. (2010)—cited in ILO (2011) and others—put the percentage of South Korean green stimulus to total stimulus at around 80 percent, but those data do not include a $10.9 billion non-green stimulus that South Korea announced in November 2008 (Wassener 2008; Tienhaara 2018).

Sources: Robins et al. 2009; Robins et al. 2010; ILO 2011; Barbier 2011; Tienhaara 2018.

Looking only at clean energy stimulus (including renewable energy, efficiency, smart grid, and low-carbon vehicles), $177 billion was announced by the 13 largest spenders in 2008 and 2009. The United States was the top spender, with $67 billion; China was second with $47 billion; and South Korea was third with $16 billion (Liebreich et al. 2010).

These estimates must be treated cautiously because they are based on the green stimulus measures as announced, not as disbursed. For some countries that have closely tracked their green stimulus spending, it is possible to see how much was disbursed: for example, in the United States 89 percent of the funds that were initially allocated for green stimulus were spent, in Canada 77 percent of the initial allocation was spent, while in Australia only 34 percent was spent (Tienhaara 2018). In some countries the disbursement process was opaque or the amount of money that was announced for green measures was later reduced (Horn-Phathonothai 2009; Liebreich et al. 2010; Tienhaara 2018).

Economic Impact

The goal of a stimulus is to help the economy recover, including stimulating demand, raising output, creating jobs, and advancing new industries. The most effective fiscal stimulus has impact on the ground as fast as possible and has a high economic return for every dollar spent in both the short and long term (Hepburn et al. 2020). Some investments are better for stimulating the economy quickly, like unemployment compensation, while others set an economy up for future growth, like investment in new technologies; an ideal stimulus package has elements of both. There are also choices to be made about what types of financial and policy instruments to use to make green investments. A review of the literature found that direct government investment delivers higher economic multipliers than tax reductions (Mafouz et al. 2002; Hepburn et al. 2020).

The economic effects of green investments in stimulus packages are difficult to measure. Wherever possible this paper uses after-the-fact, ex post evaluations, but our survey of the literature and other recent surveys (Agrawala et al. 2020; Varro et al. 2020) uncover only a few such assessments. There are other complicating factors that make it difficult to understand the economic effects, including the lack of a counterfactual no stimulus scenario for comparison purposes; the difficulty in disentangling the effects of green components compared to the rest of the stimulus; and the fact that many studies only report gross jobs added rather than the net effects. When reporting jobs data, this paper strives to specify net jobs effects rather than gross effects and identify the time span of the jobs created by using units of job-years (full-time jobs over one year) or job-hours.

Given the available evidence from 2008 to 2009, green stimulus spending did help economies recover and did create jobs. Table 2 provides some key ex post evaluations of the economic impacts of green spending in the United States, South Korea, China, and the European Union. In these economies green stimulus spending increased GDP, created jobs, and helped build up new industries, generally with success but sometimes with difficulty or more slowly than expected. Our case studies go more in-depth. In recent months a few high-level assessments have also concluded that green interventions after the global financial crisis provided effective economic stimulus (Hepburn et al. 2020; Agrawala et al. 2020).

Table 2 | Key Ex Post Economic Evaluations of Green Stimulus in Response to the Great Recession

Country

Evaluations

United States

• The American Recovery and Reinvestment Act (ARRA) supported 900,000 job-years (full-time jobs over one year) in clean energy fields from 2009 to 2015 (Council of Economic Advisors 2016).

• Each $1 million of green ARRA investments created 15 new jobs, which arose mostly from 2013 to 2017 (Popp et al. 2020).

• The ARRA was successful in stimulating job creation in renewable and energy efficiency sectors (Lim et al. 2020).

South Korea

• South Korea’s unemployment rate in 2009 was 3.6%, compared to projections of 4.3% if it hadn’t been for the stimulus. Short-term public employment increased by a net 165,000 jobs in 2009 (OECD 2010).

• South Korea’s Green Growth Plan directly created 156,000 new green jobs from 2009 to 2011 (Korean Development Institute, cited in Jung 2015).

• The South Korean economy recovered from the economic crisis faster than expected, with green stimulus measures a key contributor (Mundaca and Damen 2015).

China

• The rapid investment in rail and grid networks and other green initiatives in China led to a large and immediate boost to GDP, around 4.2% above the baseline in 2009 and 3.6% above the baseline in 2010 (Pollitt 2011).

EU

• The green elements of the stimulus packages had a small positive impact on European countries’ economies in the short run. Each $1 in green investment boosted GDP by $0.60 to $1.10 at the national level and up to $1.50 at the European level. Most green investment policies also led to higher employment levels (Pollitt 2011).

Notes: EU = European Union; GDP = Gross domestic product.

Source: WRI.

 

Box 1 | Green Investment and Job Creation in the Economic Literature

The economic literature suggests that many types of green investments create more jobs than fossil fuel investments do. According to the International Energy Agency (IEA) (2020), investing $1 million in building efficiency, clean urban transport, or solar photovoltaics (PV) would create more than twice as many gross jobs as investing $1 million in coal or gas power (see Figure B1). Restoration and sustainable forest management, electric vehicle (EV) charging infrastructure, walking and cycling infrastructure, biofuels, and recycling have also been identified as having high employment multipliers (Edwards et al. 2013; IEA 2019; Garrett-Peltier 2011; Blythe et al. 2014). Other studies and systematic reviews arrive at different employment multipliers for individual technologies (e.g., wind investments have a lower multiplier in the IEA report than elsewhere), but overall the takeaway is the same: clean energy often creates more jobs than other infrastructure investments in the short term (Blythe et al. 2014; Garrett-Peltier 2017; IRENA 2020; Edwards et al. 2013). This is because green projects like installing solar panels or planting trees are more labor-intensive than highly automated, capital-intensive fossil fuel projects. They may also require less-skilled labor or be part of the gig economy, which means that wages and benefits would be lower. Employment multipliers vary by region—they are highest in Latin America and Southeast Asia and lowest in Europe, North America, and Oceania (IRENA 2020). More research is needed on employment multipliers in low-income countries.

Figure B1 | Jobs Created per $1 Million

 

Notes: PV = Photovoltaics; IEA = International Energy Agency.

Source: IEA 2020.

Emissions Impact

Global CO2 emissions from fossil fuels and cement fell 1.4 percent in 2009 because of the global recession but increased 4.3 percent in 2010, the biggest annual increase in decades, and continued to climb upward for years (Global Carbon Project 2020). Even among the five countries that spent the most on green stimulus, in the years after the recovery the amount of CO2emissions per unit of GDP did not improve or remained on about the same trajectory as before (Figure 3). Some early projections when the stimulus packages were being announced estimated that they would decrease emissions compared to the baseline (US EIA 2009; WWF 2010; Pollitt 2011); however, it is impossible to form conclusions about green stimulus overall because there have been no ex post counterfactual analyses. Nonetheless, we do have research that calculates the emissions impact of some individual stimulus interventions, as presented in the case studies.

Figure 3 | Carbon Intensity of GDP before and after the Global Financial Crisis (2000–2018)

 

Notes: tCO2e = Tons carbon dioxide equivalent; GDP = Gross domestic product.

Source: Global Carbon Project 2020.

There are several possible explanations for the bounce back in CO2 emissions. First, green investments may take time to have an impact on emissions. For example, stimulus efforts to support clean energy manufacturing and research will not immediately reduce emissions, but can be foundational in building up the supply chains and infrastructure to make clean energy widespread in the long term.

Second, the green stimulus measures were relatively small compared to the rest of the fiscal stimulus; in some cases the green investments were directly counterbalanced by additional investments in high-emitting sectors (Robins et al. 2010; ILO 2011). At least eight countries provided support to the automobile industry, and heavy infrastructure projects like railways and grid expansion led to higher production of iron, steel, and cement (Schweinfurth 2009; WWF 2010).

Third, the green stimulus measures were not accompanied by broader structural reforms that could have incentivized a low-carbon transition (Agrawala et al. 2020). For example, while fossil fuel subsidies fell during the recession, they rebounded strongly in 2010 and 2011, despite a G20 pledge to phase them out (IEA 2016, 2019; OECD 2018b). Fossil fuel subsidies totaling $2.8 trillion were spent from 2009 to 2013, 15 times higher than the $177 billion in clean energy stimulus spending disbursed in the same time frame and also higher than the amount spent on renewable subsidies outside of the stimulus (see Figure 4). These numbers pertain only to the energy sector; there are still other distortions such as agricultural subsidies that incentivize polluting activities like excessive use of fertilizer (Searchinger 2020). Moreover, less than 5 percent of the world’s GHG emissions were covered by some sort of carbon price during the financial crisis (World Bank 2020).

Figure 4 | Fossil Fuel Subsidies Outweighed Clean Energy Stimulus

 

Notes: Green stimulus disbursement includes the 12 major economies that announced $1 billion or more in green stimulus. Renewable subsidies are global. Fossil fuel subsidies include consumption subsidies from 41 developing economies and production and consumption subsidies from Organisation for Economic Co-operation and Development (OECD) countries.

Sources: Liebreich et al. 2010; IEA 2016, 2019; OECD 2018b; Authors.

The Role of Central Banks

Many of the world’s central banks responded to the Great Recession with aggressive monetary policy, including quantitative easing (QE), the large-scale purchase of assets such as sovereign or supranational bonds, asset-backed securities, covered bonds, corporate bonds, or equities (Williamson 2017). In the United States, the Federal Reserve (the Fed) implemented three phases of QE starting in 2008: Purchases of $175 billion in agency securities and $1.25 trillion in mortgage-backed securities (QE1); purchases of $600 billion in long-maturity Treasury securities; and purchases of at least $40 billion per month for mortgage-backed securities and $45 billion per month for long-maturity Treasury securities (Williamson 2017). The Bank of England started QE in 2009, and by 2012, QE totaled $603 billion.1 Likewise, the European Central Bank (ECB)’s QE program included the purchase of government bonds totaling $306 billion and two rounds of loans totaling $1.39 trillion (Anderson 2015).2 The Bank of Japan, which was the first to use QE, began a modest program in 2010 to purchase $60 billion in assets, but after 2013 significantly scaled up QE (“Abenomics”) (Agostini et al. 2016).

However, these QE programs did not target green sectors. The Fed deliberately targeted mortgage-backed securities in its first round of QE to “provide support to mortgage and housing markets.” The ECB and the Bank of England aimed for “neutrality” in the sense of avoiding market distortions, and allocated purchases according to the makeup of the market. This has been criticized as maintaining the high-carbon status quo (Matikainen et al. 2017). Indeed, an analysis of the ECB’s and Bank of England’s corporate bond purchase program this decade indicates a skew toward carbon-intensive areas such as manufacturing and electricity and gas production, disproportionate to their contribution to the economy (Matikainen et al. 2017).

The Response of Development Finance Institutions

International finance institutions increased their portfolios by $101 billion in 2009 (Figure 5). The multilateral development banks (MDBs) all responded similarly, ramping up crisis lending and addressing secondary crisis effects, such as credit contraction and the drying up of trade finance, largely relying on preexisting instruments (Independent Evaluation Group 2012) Among MDBs the World Bank and African Development Bank (AfDB) increased lending by the largest amounts (Independent Evaluation Group 2012). The International Monetary Fund (IMF) increased its lending by a staggering 2,131 percent between 2005–07 and 2008–10 ($166 billion in 2010) (Independent Evaluation Group 2012).

Figure 5 | Multilateral Development Bank Lending during the Economic Crisis

 

Notes: Lending includes new commitments across all financial instruments in that year, while portfolio refers to all gross outstanding loans, equity investments, guarantees. EIB = European Investment Bank; EBRD = European Bank for Reconstruction and Development; ADB = Asian Development Bank; IFC = International Finance Corporation; IDB = Inter-American Development Bank; AfDB = African Development Bank.

Source: Independent Evaluation Group 2012.

Some development finance institution (DFI) green finance did increase markedly in response to the economic crisis, most notably for renewables. Overall development assistance for renewables in developing countries went up from $2 billion in 2008 to more than $5 billion in 2009. World Bank Group finance for renewables increased fivefold in 2009 to $1.38 billion. The Inter-American Development Bank (IDB) committed more than $1 billion in loans for renewable energy, while the Asian Development Bank (ADB) invested approximately $93 million in renewables (including large hydropower). Among the national development banks, Germany’s KfW Banking Group increased its lending to renewable energy from €5.4 billion ($7.9 billion) in 2008 to €6.3 billion ($8.8 billion) in 2009,3 while renewables investment by the Brazilian National Bank of Economic and Social Development (BNDES) went from $2.4 billion in 2007 to $7.0 billion in 2008 and $6.4 billion in 2009 (REN21 2010). Moreover, the China Development Bank reported that its “environmental protection” lending (e.g., urban and river basin environmental protection, industrial pollution control and recycling, clean energy, and ecological projects) doubled to $33.8 billion between 2009 and 2010 (Matisoff 2012).4

While overall green DFI finance did increase and some DFIs scaled up renewable energy finance, there is no comprehensive assessment of the proportion of the overall finance that was green versus polluting. Official development assistance from multilateral institutions5 for agriculture increased slightly between 2007 and 2010 (4.9 percent to 5.8 percent of total), while for energy it decreased marginally (7.1 percent to 6.2 percent), but Organisation for Economic Co-operation and Development (OECD) Development Assistance Committee data from that period are not disaggregated further into green categories (OECD DAC 2020). Likewise, there are no comprehensive data on whether there was an overall increase in the green lending at national development banks.

At the MDBs, at least, green finance overall did not grow disproportionately during the crisis, except at the European Investment Bank (EIB). The EIB made climate change one of its three foci in response to the economic crisis. It increased lending to initiatives to act against climate change from $14.4 billion in 2008 to $23.6 billion in 2009, a significant proportion of EIB’s $69.5 billion crisis response between 2008 and 2010 (EIB 2010). In contrast, one of the main responses at the ADB during the crisis was the expansion of its Trade Finance Program to $1 billion;6 in 2010, 95 percent of this finance supported the import of raw materials, and 39 percent supported oil, gas, and energy-related products (ADB, Independent Evaluation Department 2011). The European Bank for Reconstruction and Development (EBRD) reported that the sectoral distribution of projects did not show any major change over 2008 (EBRD, Evaluation Department 2010), and the percentage of lending and guarantees in the environment sector at the IDB did not change significantly between 2008 and 2009 (IDB 2008, 2009). At the AfDB, infrastructure sector approvals increased more than 177 percent between 2008 and 2009 and environment sector approvals declined by 37 percent (AfDB 2009). Likewise, at the World Bank, while all sectors increased their lending, the biggest increases during the crisis period—exceeding 150 percent—were for poverty reduction and economic management, financial and private sector development, and human development.7 In fact, the percentage of disbursements for the environment sector declined slightly from 2005–07 (13.2 percent) to 2009–10 (12.0 percent) (Independent Evaluation Group 2012).

Start reading