Where will the money come from?: How to finance a Green New Deal

By William White (ex-BIS, OECD), panel comments for the 5th Workshop of the New Paradigm Initiative (now Forum New Economy) held in Paris.




1. NOVEMBER 2019


15 MIN.

1. What is the Green New Deal (GND)?

Before costing any proposal we have to know exactly what that  proposal entails. The phrase “Green New Deal” is commonly associated with efforts to mitigate climate change, but it has in fact been used by a  variety of groups and people whose objectives and specific proposals often differ. A UK group was perhaps the first to use the phrase in 2008, and the UN also called for a GND in 2009. More recently, various European groups, including the Democracy in Europe Movement and the European Commission, have also made a variety of policy proposals either directed to or including measures to mitigate climate change.

Perhaps the recent set of proposals that has attracted most attention has been the GND proposed for the US by Alexandia Ocasio-Cortez. However, these proposals (the New Deal part) go well beyond environmental concerns (the Green part) and include plans for guaranteed jobs (affecting 15 million workers), universal health care, stopping “endless wars” and much more. The American Action Forum (2019) says these proposals could cost as much as $ 96 trillion over ten years; an annual charge of 50 percent of current US GDP. In contrast, a paper by Nersisyan and L Randall Wray (2019) treats many of the costs identified by the American Action Forum as benefits; for example, reforming the US health system would free up a lot of resources that could be used for other purposes. Nersisyan and Wray add it all up and conclude the AOC proposals would cost (net) only 1 ½ percent of GDP.  It is rather alarming that these cost estimates should differ so widely.

2. How much will it cost to mitigate climate change?

A more tractable question might be, what would be the cost of mitigating only climate change? Unfortunately, answering this question is also very hard since it demands more precision and still more supplementary questions. Note as well, that climate change is only one of the many ways that human beings are negatively affecting the environment.

A first question is, how much climate mitigation? The more the mitigation targeted, the higher the costs. Moreover, costs might rise non-linearly as the target becomes more ambitious.

William Nordhaus seems reasonably comfortable with accepting a 3 increase in temperature from preindustrial norms[1]. Although he just received the Nobel prize for his work, the Integrated Assessment Models which he pioneered has in fact been subject to some severe criticism. Keen (2019) contends that Nordhaus underestimates the costs of climate change, not least the possibility of catastrophic tipping points. Others say he overestimates the costs of mitigation by ignoring economies of scale and the possibility of significant technological progress once a serious process of mitigation begins. Still others feel that he discounts the future in an “ethically unacceptable” way.

This last consideration raises a fundamental question. Is a cost -benefit approach appropriate when a potential benefit is avoiding human annihilation. To rephrase this, can we not just assume that the rate of return on investments in climate change mitigation will be very, very, high?

At the time of the Paris agreements of 2015, there seemed to be general agreement that the target should be holding the global temperature increase (above the pre-industrial level) to 2 degrees Celsius. However, the most recent report of the Intergovernmental Panel on Climate Change (2018) lays out 1 ½ degrees Celsius as a minimum acceptable goal, which implies going to zero net carbon emissions by 2050. It is worrisome that successive reviews, going back decades, seem to indicate an ever more urgent need to act. This reflects not only the passage of time but also a growing scientific understanding of the dynamic processes at work.

To achieve zero net carbon by 2050, urgent action is required. This is all the more the case since net emissions are continuing to increase year by year, not decrease as necessary. This last point raises another issue. How rapidly should the mitigation process proceed, since shortening the adjustment time will increase costs as well as benefits. Moreover, costs will also be materially affected if a significant number of countries do not participate[2], particularly countries that have the potential for major mitigation at relatively low cost (in particular, China and India). Should countries willing to act begin now, or wait until potential “free riders” have agreed to participate?

A third question has to do with mitigation strategy. Should it include nuclear power or not? Should natural gas be allowed for a time, or is it imperative to go directly to non renewables? It is likely that many other questions of a similar nature could be added.

Finally, in addition to all these questions pertaining to how best to estimate the costs of climate mitigation, Flyvberg (2009) reminds us of an unpleasant fact. In the past, big infrastructure projects have almost always come in massively over budget and far behind schedule. This fact applies especially to “tech heavy projects” and to projects in emerging markets.

Estimates for the costs of achieving a 2 degree Celsius target, above pre-industrial temperature levels, range from 1-2 % of global GDP (annually) to 6-7 %. Obviously, as the numbers rise, critics of mitigation say it is not “doable”, but in fact this is not obvious. First, the costs of mitigation might be large, but the costs of not mitigating might be much larger still, as suggested above. Second. we are not talking about the need for totally new (green) resources but, in large part, ways to reallocate resources from existing (brown) uses. Third, if one accepts that much infrastructure investment will happen any way, then the marginal costs of making those investments in a climate sustainable way will be much less[3]. Finally, the benefits of moving away from fossil fuels are by no means limited to mitigating climate change and this must also be factored in. Local air pollution kills millions of people a year, in both advanced and emerging markets.  Surely avoiding this has some value.

As long as these questions remain unanswered, it is simply not possible to say how much climate mitigation might cost. The rest of this paper therefore focuses on a simpler set of questions. How might fiscal policy, financial (regulatory) policy and monetary policy contribute to climate change mitigation?

The literature seems to imply there is a natural ordering here[4]. Fiscal policies should be relied upon first and foremost; in particular, raising carbon taxes and reducing energy subsidies. This approach would seem the most effective in “internalizing the externality” of global warming associated with the burning of fossil fuels and seems, on the surface, to satisfy the “polluter pays” principle[5]. However, if “political economy” arguments imply that fiscal measures will only be used suboptimally,  then second best solutions might have to be turned to. In fact, that seems to be what is happening already. Financial regulators and central banks are already doing a great deal in this regard. Unfortunately, this is as much bad news as good news. Governments remain unwilling to do what only they can do: in particular, cut subsidies for fossil fuels and raise the price of emitting carbon.

3. How fiscal, financial (regulatory) and monetary policy might contribute to climate change mitigation?

 Fiscal Policies

There are a large number of ways in which Governments might wish to increase their expenditures to contribute to the mitigation of climate change. These would include fundamental research and development, tax credits for pre-commercial technologies, public infrastructure, transportation alternatives to cars and planes, financial incentives (subsidies and public guarantees) to support private sector investments, taking on contingent risks, public-private partnerships and funding for national development banks and multilateral development banks to support climate change mitigation.

While these expenditures would increase the government’s financing needs (the deficit), other climate related fiscal policies should work in the opposite direction. Virtually all economists would call for a large and rising carbon tax to shift the balance of energy supplies away from fossil fuels to renewables. Further, most economists would call for the reduction or elimination of current government subsidies (estimated by the International Energy Authority to be around $500 billion globally) to promote less energy use and specifically less use of the energy derived from fossil fuels.

The balance of these expenditures and taxes would likely leave  governments with a substantial surplus rather than a deficit. However, there is also a growing recognition that a public backlash might then endanger many of these desirable fiscal policies, not least the need for a significant increase in carbon taxes[6]. This has led in turn to various proposals for remitting all or part of the resulting fiscal surplus back to taxpayers in a relatively progressive way that might help avoid such a  backlash[7]. Such an approach would then support two government objectives; climate change mitigation and more equal income distribution. Cutting energy subsidies would also help reduce income inequality since richer people tend to use more energy and therefore benefit the most from such subsidies.

It could then still be the case that more government expenditures might increase the government’s deficit and stock of debt. While this might provide a welcome increase in aggregate demand and employment, such an expansion raises other questions in turn.

Could capacity constraints and rising inflation become a problem caused by larger government deficits arising from climate mitigation? Modern Monetary Theory (MMT)[8] is correct in suggesting that this is a more relevant question than “how is the government to be financed?”[9]I In this regard, MMT takes the same approach as Keynes (1940) did in his famous tract “How to pay for the war?” This approach first asks whether the economy has adequate unused production resources and then, assuming the answer is no, how the needed resources might be freed up to avoid inflation becoming a problem.

On the one hand, inflation could become a problem, in spite of our recent experience of very low inflation. These disinflationary trends (and low interest rates) have been largely driven by positive demographic shocks (baby-boomers and the return of China and other countries to the global trading system). However, these demographic trends are now going into reverse. We will soon face an ongoing negative supply shock from this source. The inflationary potential will be made worse (further negative supply shocks) by costs associated with global warming already in progress, and by temporary supply constraints associated with the need to reallocate resources from “brown” to “green” purposes. Should this happen, tightening will be necessary, either of other fiscal instruments or of monetary policy.

On the other hand, a case could be made that we are already on the verge of another, large global downturn. Rising global debt levels, along with other financial and real imbalances all point in this direction. The Trump tariff wars might then act as the specific trigger. If this downturn does materialize, then near-term inflation will not be a problem. Many well-respected economists are now suggesting that fiscal stimulus would be the appropriate response[10]. Their logic seems to be that the stimulative effects on aggregate demand of easy money have declined over time, largely due to the “headwinds” of debt.  If that is the case, focussing on fiscal expenditures directed to mitigating climate change would kill two birds with one stone.

Both supporters of MMT, and the mainstream economists calling for  fiscal easing in the next downturn, would seem to prefer to keep monetary conditions very expansionary. Indeed, MMT theory seems to imply this could go on forever, provided inflation did not return. Here, however, two objections can be raised.

First, there is the point made by Keynes in his negative reaction to Abba Lerner’s very similar proposals for “functional finance”. Keynes allegedly said[11], at a meeting in Washington in 1943, “No that’s humbug. The national debt can’t keep on growing.” Most governments today are already technically insolvent[12] and have access to market finance only because the markets believe that the governments will deal with this problem over time in a non- inflationary way. However, persistent and growing recourse to the central bank (fiscal dominance) for financing can cause this assumption to be questioned. In turn, this can result in a sudden panic, flight from the currency and even hyperinflation. Sargent and Wallace (1981) have laid out the theory, while Bernholz (2006), Edwards (2019) and other students of Latin American history have documented how it works in practice.

Second, it is simply wrong to suggest that possible inflation is the only unwelcome side effect of maintaining ultra easy monetary policy[13]. As is being increasingly recognized, such policies contribute to financial instability (by lowering the profits of financial institutions and encouraging risk taking) and to financial and real distortions which reduce productivity growth and living standards over time. This implies that, as the economic recover proceeds under the influence of fiscal stimulus, monetary policy should be carefully “renormalized”.  Companies made vulnerable, either through climate mitigation policies or too high debt levels, should be restructured or made bankrupt in as orderly a way as possible. Procedures for doing this are still inadequate in many countries[14].

Financial (regulatory) policies

Government agencies have already begun to explore ways in which financial regulation might be changed to support a carbon free future. One reason is a tacit recognition that the first best option of a significant carbon tax might not happen. Allied to this is the recognition that finance is a crucial component of virtually all spending decisions. A financial “nudge” could materially improve the reallocation of resources to climate change mitigation. A second reason is the recognition that market failures can occur within the financial system itself; free rider problems, incomplete markets, short sightedness and a failure to incorporate climate change risks into lending conditions.

Proposed changes to regulatory policies can be divided into those changes that are consistent with “core mandates” (stability of individual financial institutions and of the financial system as a whole) and policy changes that go beyond this and are “promotional” in nature. The former is a form of adaptation to climate change, while the latter is a vehicle for mitigation. It is important to note that, if we want mitigation, “promotional” policies will be required and this will immediately raise questions about the mandates of both regulators and central banks.

The most important regulatory policy suggestion, consistent with a “core mandate” for preserving the stability of individual lenders, is that the risks associated with climate change should be estimated and priced into each loan. These risks include losses from climate adaptation (floods, hurricanes etc.), losses from climate mitigation (stranded assets) and losses from lawsuits against those that have contributed to the losses of others. The Financial Stability Board (supported by the G 20) set up the Task Force on Climate Change Financial Disclosure (TCFD) in 2015 to pursue these objectives[15].

To date, the biggest obstacle to success has been the unwillingness of borrowers to reveal their exposures and face tighter credit conditions. However, other problems can also be identified. The magnitude of these expected losses is very hard to estimate, since both the probability of loss events, and the losses that might be associated with such events are not at all clear. Moreover, there is a negative correlation, not well understood, between losses associated with climate change mitigation (losses due to sunk assets) and physical losses associated with rising temperatures.

Central banks have also recently set up the Network for Greening the Financial System (NGFS). This network has two objectives, one of which is estimating risks posed by climate change to the stability of the financial system as a whole. The principal problem here is that the analytical underpinnings for systemic risk assessment and policy response (so called “macro prudential policies”) are not well developed, even in the face of traditional threats to financial stability[16]. As for climate change, since this is effectively unprecedented in human history, there is no past data to guide policy reactions. Another complication is that climate change might not affect the financial system directly, but only indirectly, as worsening economic conditions slowly grind down the capacity of borrowers to service debts.

Turning now to “promotional” polices, the NGFS is actively involved in these as well. One set of efforts has to do with identifying market failures that penalize “green” loans by financial institutions and encourage “brown” ones. These can then be either fixed at source or offset through other policy actions. For example, it could be contended that regulatory liquidity requirements impede insurance companies from making the longer-term loans that match their liabilities. A second suggestion is to alter regulatory capital requirements to reflect explicitly “green-brown” considerations. Still more radical, it could be suggested that all lenders must hold a minimum proportion of their   portfolio in “green” assets.

Evidently, “promotional policies” raise risks of diluting “core mandates”, and they also raise governance issues pertaining to “who does what” and ultimate accountability. Some commentators conclude that, at the least, governments should give explicit mandates to the appropriate regulatory bodies, to prevent accusations of overreach. Finally, worry has been expressed that the shift into green assets could turn into a bubble (like sub prime mortgages) that might burst with unfortunate effects on both the financial system and climate change mitigation itself. However, there is no evidence that we are anywhere near such an inflection point.

Since the financial crisis of 2008, a notable trend has been corporate borrowing through markets rather than from financial institutions. Bond issues have risen sharply, as has the size of asset management companies. The issue of “Green Bonds”, where the money raised is used for pursuing Environmental, Social and Governance (ESG) goals, has also risen significantly. This has been supported by industry groups such as Climate Action 100+. Nevertheless, the total stock of such issues still amounts to a very small percentage of all bonds outstanding, and there are also impediments to further growth in the future.

The central problem is that the ESG criteria are not well specified. As an illustration, the Global Sustainability Alliance has identified $31 trillion of ESG assets, while J P Morgan has identified only $3 trillion using stricter criteria. The principal danger this poses is that the whole categorisation process could fall into disrepute. What is needed is a trusted body to set criteria and evaluate proposals using those criteria[17]. A related concern is that, absent a trusted body to set criteria, Green Bonds will be judged on the basis of the credibility or trustworthiness (in terms of pursuing ESG objectives) of the national authority standing behind them. Since richer countries tend to be better endowed in this respect, poorer countries might well find themselves disadvantaged[18]

Monetary Policy

A number of suggestions have been made for incorporating environmental concerns into the monetary policy functions of central banks. Perhaps the most important effect of this would be inspirational; a high-profile government agency would be seen to be taking the transitional issue seriously.

Arguments can be put forward to reconcile such efforts with the “core” central bank function of pursuing price stability. Climate change is effectively a negative supply shock, lowering potential (level and perhaps growth) and raising inflation. This would then require central banks to face up to some difficult trade-offs, as occurred with the oil price shocks of the 1970’s. Central banks should therefore take steps today to avoid such developments later. Some also argue that central bank mandates are often fuzzy and could easily be reinterpreted to support such action[19].

While superficially attractive, this argument has a number of shortcomings. First, while central banks generally act to forestall forecast increases in inflation, the forecast horizon is much shorter than the horizon for climate change. Second, all sorts of economic malfunctions (eg badly functioning labour markets or trade wars) make the pursuit of price stability more difficult, but it is rarely argued that the central bank should take responsibility for fixing the malfunction itself. Finally, what would such an extended mandate mean operationally? Would it imply tighter monetary policy to restrict investment in capital intensive industries that are highly polluting (like steel and cement) or easier policy to support capital intensive industries like renewable energy?

Whatever the implication for mandates, some very specific suggestions have been made for central bank action in pursuit of climate change mitigation. The first has to do with central bank collateral for the loans it makes. It could provide easier terms for “green” collateral or could provide easier terms for those financial institutions judged to be pursuing a green agenda. A second suggestion is that central bank purchase Green Bonds more aggressively when doing Quantitative Easing. Both suggestions imply distributional effects that central banks have traditionally sought to avoid. Further, both policy suggestions refer to central bank actions in response to temporary problems. It is hard to see how either might have a permanent effect on climate change mitigation.

4. A Concluding Warning

Focussing only on the Green part of the GND still leaves many uncertainties about the final resource costs of climate change mitigation. Yet, if the costs of not mitigating climate change are likely to be existentially high, then almost any price is a price worth paying. There are also grounds for belief that the effect on government deficits might be quite manageable if governments raise carbon taxes and cut subsidies as suggested above.

The greatest danger is that governments will not act as aggressively as the situation requires, for whatever reason. Within countries, “losers” from the transition will lobby hard (and have adequate financing) to ensure it does not happen. Voters might be insufficiently aware of the dangers of climate change, or simply too short sighted, to support radical action. Across countries, the free rider problem always has to be contended with. The greatest danger is that a few free riding countries provide cover for other countries to also reduce their efforts to support the mitigation of climate change.

What must be emphasized in concluding is that it is an illusion to suppose that second best solutions, focussed on financial issues, can be relied upon to solve the very difficult problem of climate change. The difficult is not primarily the need to change the mandate of central banks and regulators, as inconvenient as this might seem to incumbents. Such changes have in fact occurred many times in history[20]. The real problem is that that these second-best measures will  prove inadequate to deal with the serious problems associated with climate change.

The recurrent tendency of government to turn to inadequate solutions is attested to by the policies followed in the post crisis period. Governments preferred to rely almost totally on ineffective central bank monetary stimulus, to avoid having to face up to the need for painful structural reforms and extensive debt restructuring. This strategy threatens to end in another deep recession. Should this happen, it might then lead to an eventual (albeit belated and more costly) implementation of the fundamental policy changes that were required earlier. In the case of climate change, however, there will be no second chance to get the policies right.


[1] He looks for the temperature point at which the costs of damage due to climate change equal the costs of mitigation. See Nordhaus W (2013)

[2] See Nordhaus (2013)

[3] For example, the OECD (2017) estimates that the level of global infrastructure spending required on a “business as usual” basis, at $6.3 trillion annually. This rises only to $7.0 trillion if done in a climate sustainable way.

[4] Two recent papers deserve careful reading. See Krogstrup and  Oman (2019) and Buiter and Nabarro (2019) .

[5] Actually, this is not obvious. The problem of global warming arises from a stock of GHG’s, to which the advanced economies have made the largest contribution historically. That said, emerging market countries (especially China) are now contributing materially to new emissions.

[6] This desirable policy outcome would also be endangered by resistance from companies producing carbon intensive fuels, and by companies using such fuels. The latter would suggest they should be exempt from such a tax on grounds of “competitiveness”. This issue then raises the difficult question of how to calculate the carbon content of imports, so that a cross border adjustment could be made to ensure that a higher carbon tax did not reduce “competitiveness”.

[7] Over three thousand American economists, including 27 Nobel Laureates, have recently signed a petition to the US government to this effect. See the Wall Street Journal (2019)

[8] For an overview and a critical assessment see Buiter and Mann (2019)

[9] The central bank automatically meets the government’s financial requirements if a country has its own currency and a floating exchange rate.

[10] Blanchard and Summers (2019) and Fleming and Giles (2019).


[11] Aspromourgos T (2014)

[12] Recent estimates by J Miron (2015), based on the earlier work on intergenerational accounting by Larry Kotlikoff, indicate that satisfying the government’s inter-temporal budget constraint would require taxes to rise by 30 % in Europe and 50 % in the United States.

[13] White (2016)

[14] Working Party 1 at the OECD has carried out extensive research on this question in recent years. Also see warnings given by the Group of Thirty (2018).

[15] The Task Force is made up of private sector representatives under the Chairmanship of Michael Bloomberg. In 2017 over 100 large companies signed a statement of support for TCFD recommendations.

[16] White (2019a)

[17] It is important to note that a binary distinction between Brown and Green bonds might be insufficient and even dangerous. Just as rating agencies provide various grades of “credit worthiness”, bonds might have varying grades of “greenness”. What, for example, is one to say about an investment that produces clean water but demands the use of large amounts of electricity?

[18] Emerging markets already pay an interest premium, and to the extent their debt load is heavy (as in many African countries) that premium rises. Since the same countries might also be hardest hit by climate change, this comes down to saying that those who most need Green Finance are the least likely to receive it.

[19] Dikau and Volz (2019)

[20] See W White (2019b)