DEPARTMENT OF ECONOMICS OxCarre Oxford Centre for the Analysis of Resource Rich Economies

Manor Road Building, Manor Road, Oxford OX1 3UQ Tel: +44(0)1865 281281 Fax: +44(0)1865 271094

oxcarre@economics.ox.ac.uk www.oxcarre.ox.ac.uk

Direct tel: +44(0) 1865 281281 E-mail: oxcarre@economics.ox.ac.uk

_

OxCarre Policy Paper 34

Stranded Nations? The Climate Policy Implications for Fossil Fuel-Rich

Developing Countries

David Manley

National Resource Governance Institute

James Cust*

The World Bank

&

Giorgia Cecchinato

LSE

STRANDED NATIONS? THE CLIMATE POLICY IMPLICATIONS FOR FOSSIL FUEL-RICH DEVELOPING COUNTRIES

David Manley, James Cust and Giorgia Cecchinato1

This version: February 2017

First version: October 2016

ABSTRACT

Developing countries rich in fossil-fuels face a unique challenge posed by climate change. They seek

to extract fossil fuels at a time when the global community must reduce carbon emissions. Effective

global climate policies and low carbon technologies will likely reduce the demand for fossil fuels,

creating the risk of ‘stranded nations’—where resources under the ground become commercially

unattractive to extract and a substantial share of a nation’s wealth may permanently lose its value.

This constitutes a parallel to the stranded assets challenge faced by the private sector.

We identify three key challenges faced by fossil-fuel rich developing countries. First, we show that

these countries are highly exposed to a decline in fossil fuel demand, with their median ratio of oil

and gas reserves to GDP is 3.66, compared with a median for non-FFRCs of 0.58. Second, they are

less able to diversify away from this risk than fossil fuel companies or investors - oil companies on

average hold only around 13 years of reserves on their balance sheets, whereas FFRDCs hold a

median of 45 years of known reserves at current production rates. Third, these countries often find

themselves under pressures to implement policies that may expose them to further risk. For

example, supporting fossil fuel linked infrastructure and skills that relies on long time horizons for

payoffs to the country, subsidising fossil fuel consumption that extends carbon-intensity of

production, or by investing state capital in fossil fuel linked assets such as national oil companies.

In response, we identify four policy implications arising from this carbon market risk that fossil-fuel

rich developing countries should consider. While several of the policy recommendations align with

general good practice, we show that the prioritisation, sequencing and in some cases direction of

these policies require modification when considering the risks posed by a global shift away from

fossil fuels.

Keywords: stranded assets, fossil fuel producers, natural resource management, climate policy,

technological innovation, rapacious depletion

JEL codes: H23, O13, O34, Q32, Q38, Q54

1 The authors can be contacted at: dmanley@resourcegovernance.org, jimcust@gmail.com, gi.cecchinato@gmail.com. We would like to acknowledge the Natural Resource Governance Institute who supported the preparation of this paper. Additional author affiliations: Cust- OxCarre and World Bank. Cecchinato: LSE. The views presented in this paper are those of the authors, and not necessarily those of their associated organizations. The authors would like to thank the following for helpful comments and discussion: Saleem Ali, Nicola Barnfather, Jelson Garcia, Patrick Heller, Daniel Kaufmann, Katarina Kuai, Thomas Lassourd, Carlos Monge, Keith Myers, Alex Tilley, Rick van der Ploeg and Samuel Wills. We also with to thank the anonymous reviewers of our related article in Finance and Development, March 2017: Unburnable Wealth of Nations.

2

INTRODUCTION

Developing countries rich in carbon-emitting fossil fuel resources such as oil, gas, and coal face risks

from climate change—both in terms of the direct impact of a changing climate, but also the effects

of climate action and technological innovations that seek to limit greenhouse gas emissions, and

therefore may give rises to a decline in fossil fuel demand. This is because a significant portion of

their national wealth and a major sector of their economy is tied to fossil fuels at a time when the

world needs to transition away from such carbon-intensive energy.

These countries face a dilemma. They seek to drive economic development and reduce poverty

using revenues and investment associated with extracting these resources, but doing so drives the

consumption of oil, gas and coal contributing to climate change. In turn, such climate change is likely

to severely impact these very same countries (IPCC 2001, 2014). On the other hand, if their fossil fuel

resources remain in the ground, they may miss out on a significant opportunity to develop their

economies and alleviate poverty.2 This challenge is combined with currently low commodity prices, a

failure by many of these countries to build up net savings during the boom and global pressure to

curb new fossil fuel consumption. As such these countries find themselves under serious and urgent

pressure to chart the right way forwards for their economies (van der Ploeg 2016).

CARBON MARKET RISK TO STRANDED NATIONS

Building on the emerging “stranded asset” debate (Leaton 2013, Rozenberg et al. 2014) that has

focused on the repercussions for fossil fuel companies, in this paper we consider the impact on fossil

fuel-rich developing countries. The carbon market risk we examine occurs if, over the long term,

fossil fuel demand permanently declines, and with it prices for fossil fuels.

The prospects for a global shift away from fossil fuel consumption are growing. Governments at the

Paris COP21 talks agreed to limit the global temperature increase to 2 degrees, with an aspiration of

keeping temperatures below 1.5 degrees warming. This 2 degrees’ target implies that to ensure that

temperatures stay within this limit, more than two thirds of current fossil fuels must remain in the

ground (IEA 2012). To achieve this, governments may impose climate change policies that reduce

demand for, or even supply of fossil fuels. Further, the advancement of alternative energy sources to

fossil fuels may begin to take substantial market share from fossil fuel producers. This risk of a

permanent fall in fossil fuel prices for producers we term here as the ’carbon market risk’.

If this carbon market risk is already priced into commodity markets, governments of fossil fuel-rich

developing countries can use current market prices signals to guide long term decision making.

However, market prices may not be a perfect signal of these long term risks. First, the significant

year-on-year fluctuations in prices make it difficult to discern the long-run trend in prices and

complicate long-term planning based on these prices for governments. Second, market prices of

commodities may not hold significant informational content of expectations past more than a few

years at most. As an example, the Chicago Mercantile Exchange (one of the world’s largest markets

for commodity futures) offers trading on oil futures contracts up to nine years forward, but figure 1

shows that trading volumes of contracts with maturities past 12 months are currently minimal (CME

2016). Given the cost of storing commodities such as crude oil, and therefore the difficulty in

2 Venables (2016) shows on average for the top 15 oil and gas rich countries in the world, 54 percent of government revenues were derived from oil and gas industries from 2006 to 2010. Although he argues that transforming this revenue into economic development has proven difficult for most resource-rich countries.

3

speculators making long-term bets on prices, the value of using commodity prices to understand

long-term risks such as the carbon market risk described here is severely limited.

Figure 1. Proportion of total daily trades of Crude Oil Futures contracts

Source: Chicago Mercantile Exchange. “Crude Oil Futures Quotes”. Accessed August 2, 2016. http://www.cmegroup.com/trading/energy/crude-oil/light-sweet-crude.html Notes: Contracts are for Light Sweet Crude, CME Globex product code “CL”.

Given the lack of market signals on the one hand, and the significant impact if carbon market risks

are realised on the other, it may be important to consider the implications of an uncertain future for

fossil fuel demand. Under a climate-optimistic scenario, a dramatic shift away from fossil fuels would

be expected. Meanwhile a climate-pessimistic scenario would also be costly for many developing

nations.

While extensive research examines the expectation of threats posed by climate change, the

probability and impact of the climate-optimistic scenario. Although the Paris COP21 talks provided

some indication of future policies, there remains significant uncertainty about how this will actually

happen. The lack of certainty creates a mismatch between the opportunities to develop and invest in

resources (Mitchell et al. 2015). This paper argues that even in the face of considerable uncertainty,

countries should consider these three challenges since the impact, if it occurs, will be severe.

THE SCALE OF THE CARBON MARKET RISK FOR FOSSIL FUEL PRODUCERS

THE WORLD’S CARBON BUDGET AND STRANDED RESERVES

Fossil fuels may be plentiful from a geological point of view, but if the world burns all available fossil

fuels the impact on the climate will be severe. Emerging research suggests that the world should aim

to restrict the accumulated carbon and other greenhouse gases emitted between 2011 and 2050 to

1,100 Gigatonnes of carbon dioxide equivalent (McGlade and Ekins, 2014; Allen et al. 2009; Allen et

al. 2015, Meinhausen et al. 2009). This limit is referred to as the world’s ‘carbon budget’. Breaking

this budget significantly risks deteriorating global environmental and social conditions and ‘runaway’

climate change—in which natural positive feedback loops contribute to ever greater climate change

effects.

4

In comparison, the estimated combustible carbon dioxide content of the world’s reserves3 of oil, gas,

and coal in 2014 was 2,900 gigatonnes (McGlade and Ekins 2014). Therefore, to avoid using up the

world’s carbon budget and causing extreme climate change, two thirds of existing fossil fuel reserves

must remain undeveloped.

Keeping two thirds of reserves in the ground necessitates a significant reduction in the world’s use of

fossil fuels. This, in turn, necessitates a long-term decline in the demand for fossil fuels, which could

occur if governments impose carbon taxes or similar policies4; if there is a broad transition to

alternative energy supplies, such as nuclear or renewables; or if economic activity becomes

significantly more efficient in using energy. Reserves may also remain undeveloped if governments

impose policies to limit the market supply of fossil fuel resources.5

A fall in fossil fuel demand, other things equal, will lead to a fall in fossil fuel prices. This might lead

to existing reserves remaining undeveloped or ’stranded‘. Meanwhile projects that remain

commercially viable will be less valuable. This in turn may diminish the rents and therefore tax

revenues fossil fuel extraction can generate for producing countries.

To understand this, consider an example from the global oil market. The price received for a barrel

of oil is roughly the same for all projects after accounting for relatively small differences in the

quality of oil (such as viscosity) and transport costs. However, costs will differ considerably across

location and geology. Figure 2 shows a representative global supply curve that ranks all oil projects

from least to most expensive in terms of unit costs of production. For any given market price, lower

cost projects, on the left part of the supply curve, will have a higher asset value per barrel than high

cost projects. If the producer price falls, say in response to a climate policy, some projects with costs

higher than the prevailing price will become stranded, while some value of all remaining operating

projects will be foregone.

3 Reserves are deposits of fossil fuels that are discovered and commercially viable. This includes P10 to P90 reserves, so includes some deposits with quite some uncertainty associated with them. Resources are deposits that are discovered but are not commercially viable under current economic conditions. See Chapter 14. Downey M. (2009) Oil 101. Wooden Table Press. 4 ‘Cap and trade’ of permits, carbon production taxes and carbon consumption taxes are all policies that are being used in some form. See Tickell, O. Kyoto2: How to Manage the Global Greenhouse. 2008. Zed Books. 5 The concept of ‘carbon leakage’ suggests that government need to have both demand- and supply-side

policies. If merely demand-side the outcome is for global prices to fall stimulating greater consumption. Collier and Venables (2014) suggest one such supply-side policy.

5

Figure 2. How a fall in demand reduces value and strands fossil fuel reserves

Asset stranding and significant falls in values in the fossil fuel industry are not new, the recent

commodity price slump has stranded many projects already and fossil fuel companies have seen the

value of their assets reduced by many millions of dollars. However, the scale of such an outcome

under, for example, an effective global climate policy would be entirely new. It would also imply a

permanence not normally considered during commodity price slumps. For example, figure 3 shows

estimates of the proportion of reserves already discovered that could become permanently stranded

if the world keeps to its carbon budget.

Figure 3. Regional distribution of reserves unburnable before 2050 for the two degree Celsius

scenario

Oil Gas Coal

Country or region Billions of barrels

% of total reserves Trillions of cubic ft

% of total reserves Gt

% of total reserves

Africa 28 26% 4.4 34% 30 90%

Canada 40 75% 0.3 24% 5.4 82%

China and India 9 25% 2.5 53% 207 77% Former soviet union 28 19% 36 59% 209 97% Central & South Asia 63 42% 5 56% 11 73%

Europe 5.3 21% 0.3 6% 74 89%

Middle East 264 38% 47 61% 3.4 99%

OECD Pacific 2.7 46% 2 51% 85 95% Other Developing Asia 2.8 12% 2.1 22% 17 60% United States of America 4.6 9% 0.5 6% 245 95%

Global 449 35% 100 52% 887 88%

Source: McGlade, Christopher and Paul Ekins. The geological distribution of fossil fuels unused when limiting

global warming to 2 degrees. Nature, Vol 517. 2015

Notes: The regional distribution is determined using regional reserves and estimates of differences in regional

extraction costs. Percentages are the estimated unburnable reserves as a proportion of total reserves in the

region. See appendix for the definition of reserves.

6

THE LIKELIHOOD OF CARBON MARKET RISK

Whether governments can implement effective climate policies to reduce global fossil fuel

consumption or whether alternative energy sources gain sufficient market share is highly uncertain.

This paper does not directly address this uncertainty but assumes that the probability is greater than

zero. Further, even if the probability is low, the impact on fossil-fuel rich countries may be very large

for the reasons discussed. This has some corollary to the so-called ‘fat-tailed risk’ or the

‘precautionary principle’ of climate damages—where the probability of an impact is low, but the

impact itself is high, necessitating precautionary actions (Weitzman 2011). Given these conditions

fossil-fuel rich countries should carefully consider how their policy choices might be modified by

carbon market risk and what type of, and how urgent, such precautionary actions should be. We will

address these policies in the final part of this paper.

The likelihood of a permanent fall in fossil fuel demand, or ‘carbon market risk’ rest on four factors.

First, governments of carbon emitting countries could impose policies that keep emissions within the

global carbon budget. This could take the form of demand-side policies, such as consumer taxes, or

supply-side measures such as producer taxes or quantity restrictions, or some mix of these policies.

The Paris COP21 made some advance towards countries imposing climate policies but it is unclear

how the result of these negotiations will influence policy.

Second, the market share of renewables, nuclear and other alternatives to fossil fuels could increase

substantially. Solar and wind power capacity has increased exponentially during the last decade,

although starting from a small base in comparison to fossil fuel-derived energy. Currently 13 percent

of global primary energy supply is from renewables. Projected shares of renewable energy to total

energy in 2030 in this share range from 14 percent—by Exxon—to 43 percent—by Geotechnical and

Environment Associates (Meister Consultants Group 2015). The International Energy Agency, an

often quoted authority on the subject, forecasts that renewables will still account for only 15

percent of global energy supply by 2040 (IEA 2015). However, the IEA has consistently

underestimated renewables growth in the past, and so can be considered relatively pessimistic

(Metayer et al. 2015).

Third, energy efficiency measures could improve to reduce future energy demand. There are signs

that this will happen. The world economy is gradually ‘decarbonising’—greenhouse gas emissions

per dollar of GDP are falling. From 2000 to 2014, the carbon intensity of economic activity has fallen

by 1.3 percent each year on average, although total greenhouse gas emissions are still rising as the

global economy grew 3.7 percent a year on average over the same period (PWC 2015). This

decarbonisation trend is set to continue. The carbon intensity of the Chinese economy, already the

largest greenhouse gas emitter in the world, fell by two percent in this period, and 6.7 percent in

2013 to 2014. In addition, Green and Stern (2015) forecast that Chinese carbon emissions will peak

by 2025.

A fourth factor is the success or failure of carbon capture and storage (CCS) methods—including

both initiatives to expand ‘carbon sinks’ such as forests, and technologies that directly prevent

carbon emission from entering the atmosphere. If these develop, there is less need to restrict fossil

fuel production for climate change purposes: the carbon and other greenhouse gases can be

removed from the atmosphere and stored. However, CCS faces three problems: first, aside from

forests, man-made CCS methods are not currently commercially viable; second, retrofitting CCS

technologies to transport, power stations and other carbon emitters appears prohibitively

expensive; and third, there is currently few viable methods to safely store carbon without

7

sufficiently reducing the risks of carbon leaking into the atmosphere after being ‘captured (Helm

2015).

FROM STRANDED ASSETS TO STRANDED NATIONS: THREE CHALLENGES

The risk of stranded assets for fossil fuel investors is receiving increasing attention and concern from

researchers and analysts, in particular the work by Carbon Tracker (see Leaton 2013). If stranded

assets are a concern for investors, we argue they should be an even bigger concern for many fossil

fuel-rich developing countries.

We focus on fossil fuel-rich developing countries (FFRDCs), which we define as those countries: 1)

whose value of known fossil fuel reserves is at least 25 percent of their total wealth (Produced,

Intangible, Foreign and Natural assets)6 OR the value of fossil fuel production is at least 10 percent of

GDP; AND 2) whose GNI per capita is less than USD 12,736 (Middle, Low middle or Low income

country definition according to World Bank classification). Figure 4 maps these FFRDCs; the appendix

describes how we calculated fossil fuel values.

Figure 4. Fossil fuel rich developing countries

Source: BP Statistical Review 2015 and authors’ calculations.

Notes: Figure 5 provides the full list and summary statistics of FFRDCs.

For these FFRDCs, the realisation of carbon market risk—the widespread adoption of carbon

policies, rise in alternative energy use or the decarbonisation of the world economy—resulting in a

permanent fall in the producer price of fossil fuels presents three challenges.

CHALLENGE 1: FFRDCS ARE HIGHLY EXPOSED TO CARBON MARKET RISK

According to one estimate, of the USD 25 trillion of fossil fuel value at risk, fossil fuel-rich

governments face 80 percent of the risk (Nelson 2014). A fall in fossil fuel prices for producers

significantly reduces the rents available from fossil fuel extraction on existing investments, and

makes further development of reserves less profitable, potentially stranding much of their fossil fuel

reserves and related assets. This reduces government revenues collected from fossil fuel extraction

and non-fiscal benefits to the domestic economy. A decline in government revenues in particular

restricts the ability of governments to support economic development—although strong public

6 Calculated from World Bank Changing Wealth of Nations. World Bank. The Changing Wealth of Nations:

Measuring Sustainable Development in the New Millennium. 2011.

8

financial management practices can help governments, in the short term, to shield their budgets to

abrupt changes to government revenues.

Fossil fuel-rich developing countries (FFRDCs) currently hold a significantly proportion of their

national wealth in the form of fossil fuel reserves and related assets (figure 5) and their fossil fuel

wealth is more at risk of a permanent decline in prices than non-FFRDCs (figure 7). The possibility of

a permanent fall in fossil fuel use exposes these countries to the risk of losing this portion of national

wealth.7

Figure 5 shows that FFRDCs ranked according to the value of fossil fuel reserves to GDP. Their

median ratio of fossil fuel reserves to GDP is 3.66, compared with a median for non-FFRCs of 0.58.

Note that prospective countries that are potentially rich in fossil fuels but with few developed

projects and currently low production rates—Tanzania, Uganda, Guatemala, among others—do not

fall within our definition of ‘fossil fuel-rich developing countries’. However, the analysis and policy

implications in this paper are highly relevant for these countries as they create policies that will

govern their sectors over the next few decades.

Figure 5. Fossil fuel rich developing countries

Country Region Fossil fuel reserves to GDP

Fossil fuel production value to GDP

Fossil fuel reserves to total assets

GNI per capita

Turkmenistan Central Asia 1330% 63% - 5,808

Iraq MENA 687% 28% - 5,742

South Sudan Sub-Saharan Africa

637% 28% - 946

Kazakhstan Central Asia 507% 23% - 9,764

Iran MENA 490% 20% 55% 6,720

Azerbaijan Central Asia 482% 29% 72% 6,426

Uzbekistan Central Asia 431% 32% 61% 1,714

Yemen MENA 421% 22% - 1,198

Equatorial Guinea Sub-Saharan Africa

378% 31% - 10,430

Angola Sub-Saharan Africa

376% 27% 66% 4,194

Libya MENA 372% 15% - 9,222

Gabon Sub-Saharan Africa

368% 24% 45% 9,290

Russian Federation Europe 364% 19% 41% 12,112

Venezuela Latin America & Caribbean

353% 14% 42% 11,940

Indonesia Asia Pacific 280% 14% 37% 3,298

Chad Sub-Saharan Africa

254% 12% 42% 944

Ecuador Latin America & Caribbean

232% 12% 26% 5,324

Algeria MENA 229% 10% 32% 5,052

Nigeria Sub-Saharan Africa

222% 11% 40% 2,264

7 FFRDCs’ value at risk lies predominantly in their fossil fuel reserves. However, mature producers in particular

hold some of their national wealth in the form of produced capital—such as domestically owned pipelines and

refineries—and human capital, such as engineers and workers in the petroleum supply chain.

9

Malaysia Asia Pacific 167% 14% 24% 9,906

Vietnam Asia Pacific 163% 12% 25% 1,570

Myanmar Asia Pacific 129% 11% - 1,270

Bolivia Latin America & Caribbean

121% 12% 46% 2,318

Timor-Leste Asia Pacific 0% 177% - 3,340

Source: BP Statistical Review 2015 and authors’ calculations.

Notes: See appendix for assumptions and calculations.

Important factors in valuing FFRDCs’ reserves are the assumptions on fossil fuel prices, costs and the social discount rate. We have chosen to use prices realised in 2015 which are low by historical standards, and so produce a relatively optimistic -for FFRDCs- estimate of the value at risk. We analyse the value of production from reserves, rather than rent, so ignore costs of exploration, development and production. In addition, we have chosen a social discount rate of four percent, as used by the World Bank to estimate wealth values in World Bank (2011).

Figure 7 shows the effect of a simulated decline in prices for the whole set of fossil fuel-rich developing countries, comparing the value of reserves given constant prices and the value of reserves given a steady decline in prices of two percent year-on-year. This price decline is assumed purely to illustrate the effect of a declining price path. In practice, the price decline may be higher or lower than this, and is unlikely to be a smooth decline.8 The assumption of a smooth decline in prices abstracts away from the likely path of prices. The combination of climate change policies and the rising market share of alternative energy sources may result in an abrupt and permanent fall in prices at some point in the future. For further discussion of our modelling of a price decline, please refer to the Appendix.

Figure 7. Decline in fossil fuel values if fossil fuel prices fall year-on-year by one percent

Country Region

Fossil fuel reserves value, constant prices (USD billion)

Fossil fuel reserves value, one percent price decline year on year (USD billion)

Value at risk

Venezuela Latin America & Carib. 1,345 1,076 20%

Libya MENA 305 244 20%

Iran MENA 2,734 2,196 20%

Iraq MENA 1,497 1,203 20%

Turkmenistan Central Asia 468 378 19%

South Sudan Sub-Saharan Africa 66 55 16%

Kazakhstan Central Asia 1,033 863 16%

Algeria MENA 467 395 16%

Chad Sub-Saharan Africa 31 27 15%

Indonesia Asia Pacific 2,566 2,183 15%

Nigeria Sub-Saharan Africa 1,024 874 15%

8 To show the strength of this price assumption, we also calculated the impact of price declines of one percent and two percent. This results in a difference in the value-at-risk of 10 percentage points on average. For instance, the value-at-risk with a price decline of one percent year-on-year for Nigeria is 15 percent compared with 26 percent for a price decline of two percent.

10

Russian Federation Europe 7,347 6,334 14%

Ecuador Latin America & Carib. 203 176 13%

Vietnam Asia Pacific 254 222 13%

Yemen MENA 135 118 13%

Azerbaijan Central Asia 331 293 11%

China Asia Pacific 8,465 7,589 10%

Gabon Sub-Saharan Africa 65 59 9%

Uzbekistan Central Asia 221 202 8%

Angola Sub-Saharan Africa 434 398 8%

Equatorial Guinea Sub-Saharan Africa 62 58 7%

Myanmar Asia Pacific 96 89 7%

Malaysia Asia Pacific 510 476 7%

Bolivia Latin America & Carib. 33 31 6%

Timor-Leste Asia Pacific - - -

Source: BP Statistical Review 2015 and authors’ calculations.

Notes: See appendix for assumptions made. Mean value at risk = 17 percent, median = 18 percent.

The value-at-risk column in figure 7 shows the difference in the value of reserves assuming current prices are maintained, and the value of reserves assuming a year-on-year decline of two percent. The difference is largely determined by the rate at which countries can deplete their reserves, which we assume to be equal to the current production rate in each country. Countries with lower reserve-to-production ratios are likely able to extract their resources faster than those with high reserve-to-production ratios and avoid more of the effect of a fall in future values. Venezuela, for example, has large reserves and low production rates, so could see a fifth of the value of its reserves cut under the assumed fall in prices.

CHALLENGE 2: FFRDCS COULD BE LESS ABLE TO DIVERSIFY THEIR CARBON MARKET RISK

Not only are FFRDCs exposed to a possible permanent fall in fossil fuel use, but—compared with

fossil fuel companies—FFRDCs are less able to diversify their exposure (Mitchell et al. 2015). This

suggests that the arguments made of the dangers of stranded assets for fossil fuel companies are

even more relevant for fossil fuel-rich developing countries.

The ability to diversify or reduce their exposure to carbon market risk depends on two factors. First,

the time and cost of converting their fossil fuel related assets into other non-related assets, such as

cash—known in the financial sense as the asset’s liquidity. Second, the ability to hold a diversified

portfolio of assets.

In terms of market liquidity, companies can relatively easily divest from fossil fuel related assets into

assets less exposed to carbon market risk. While it is likely that as more companies seek to divest

liquidity will fall, but with sufficient time, an orderly transition can occur. Companies own the

extraction rights to relatively few booked reserves, with relatively high production rates. For

example, in 2013, the reserve-to-production ratios for all oil and gas companies were 12.8 years and

13.9 years respectively (EY 2013). Therefore, companies can, in principle, stop replacing their

reserves and run down their existing reserves over the next 13 to 14 years. Although, given that

most oil companies currently seek to maintain or increase their reserves, this is unlikely to happen

immediately.

11

In comparison to companies, FFRDCs hold fossil fuel assets that are less financially liquid. For

FFRDCs, fossil fuel assets can typically only be converted at the rate based on the time taken to

develop and produce from an extraction project. Using past reserve-to-production ratios as an

indicator of the speed at which countries can convert their subsoil assets into cash, figure 8 shows

that most countries must wait many decades (a median of 45 years) to liquidate their fossil fuel

wealth, unless they can find ways to increase their rates of production.

For those countries that enter development agreements with private sector companies, the

depletion of reserves will be equal between countries and companies—the same reserve is being

depleted. What accounts for difference in reserve-to-production ratios is the national ownership and

production of reserves led by state-owned enterprises in many FFRDCs. For instance, 18 of the 23

countries in figure 8 have national oil companies.9

Figure 8. Fossil fuel-rich developing countries’ weighted reserves-to-production ratios (years)

Source: BP Statistical Review 2015, authors’ calculations

Notes: Figure 8 uses a measure of country’s reserve-to-production ratios for all three fossil fuels (oil, gas and

coal) that is weighted by the proportional value of production of each commodity (see Appendix for details).

Reserve-to-production ratios are a function of geological and technological constraints on one hand, and a

policy choice on the other. Some countries may be in a position to increase production rates over time while

others may not, depending on cost, maturity of the sector and governance considerations.

In addition to being able to liquidate their fossil fuel assets quicker than FFRDCs, companies also own

or manage a more diversified portfolio of fossil fuel assets, for example across a variety of different

countries and with different cost profiles; in contrast to countries whose fossil fuel reserves are

geographically bound.

Further, few resource rich countries have successfully diversified their economies and holding

foreign assets has been limited by the rate of depletion and ability to hold revenues as savings

9 Additionally, for a given value of fossil fuel reserves, the expected state share is typically higher than 50% due to the rents associated with extraction. This is particularly true for hydrocarbons where state shares can rise as high as 80-90% of the total extraction value of oil.

12

(Venables 2016). Sovereign wealth funds (SWFs), in which funds are invested in foreign assets, are

one way for a government to hold a wider range of non-fossil fuel linked assets. However, making

funds available for this purpose can be difficult given government expenditure needs within the

country, and can increase the risk of inappropriate use of these funds (Bauer 2014). Further, the

assets of sovereign wealth funds owned by FFRDC governments represent only three percent of

their fossil fuel reserves on average (see figure 9).

Figure 9. FFRDCs hold few foreign assets relative to their fossil fuel reserves

Country Fossil fuel reserve value (USD billion)

SWF funds (USD billions)

Proportion of SWF funds to reserves

Azerbaijan 331 72 21.7%

Kazakhstan 1,033 71 6.9%

Algeria 467 25 5.4%

Russian Federation 7,347 288 3.9%

Uzbekistan 221 6 2.7%

Gabon 65 <1 0.2%

Nigeria 1,024 1 0.1%

Source: Sovereign Wealth Fund Institute, BP Statistical Review 2015, authors’ calculations

CHALLENGE 3: FFRDCS MAY PURSUE POLICIES THAT INCREASE EXPOSURE TO CARBON MARKET RISK

In addition to being exposed and limited in their ability to diversify from carbon market risk, some of

the economic policies common to fossil fuel-rich countries may increase FFRDCs’ exposure.

First, National Oil Companies (NOCs), common in oil-rich countries, involve the investment of state

capital into fossil fuel assets. If the expected life of these assets is long enough that declining oil, gas

or coal prices will impact returns, or a government cannot liquidate these assets at a reasonable

value, then governments, by investing in an NOC, may be increasing the exposure of national wealth

and public assets to carbon market risk. This exposure increases for NOCs that operate reserves

abroad in the same manner as other oil companies are exposed to carbon market risk (see above).

Figure 10 shows the significant values of state ownership in NOCs in FFRDCs (for which there is data).

Figure 10. State owned oil companies

Country State-owned company State share

(%)

Total assets

(USD bn),

State share

(USD bn)

China China National Petroleum Corporation (incl. Petrochina) 100% 576 576

China Sinopec Group 100% 321 321

Russian

Federation Gazprom 100% 319 319

Russian

Federation Rosneft 100% 228 228

Venezuela Petróleos de Venezuela 100% 227 227

13

Iran National Iranian Oil 100% 200 200

China China National Offshore Oil Corporation 100% 167 167

Malaysia Petronas 100% 165 165

Bolivia Yacimientos Petrolíferos Fiscales Bolivianos 82% 104 85

Angola Sociedade Nacional de Combustíveis de Angola Unidade

Empresarial Estatal 100% 54 54

Indonesia Pertamina 100% 51 51

Kazakhstan Kazmunaigaz 66% 49 33

Azerbaijan State Oil Company of the Azerbaijan Republic 100% 31 31

Ecuador Petroecuador 100% 9 9

Timor-Leste TIMOR GAP 100% 0.004 0.004

Source: Most recent annual reports of companies (2014, 2015); Natural Resource Governance Institute and

author’s calculations. Note: The table does not include a number of smaller national oil companies in fossil-fuel-rich

developing countries for which data are not available: Sontrach, Algeria; Société des Hydrocarbures

du Tchad, Chad; Petroamazonas, Ecuador; Sociedad Nacional de Gas, Equatorial Guinea; Gabon

Oil Company, Gabon; Myanmar Oil and Gas Enterprise, Myanmar; Nigeria National Petroleum

Corporation, Nigeria; Turkmengaz, Turkmenistan; and Uzbekneftegaz, Uzbekistan.

Second, some governments seek to capture value by encouraging the domestic private sector to

participate in oil, gas, and coal supply chains, often known as promoting ‘local content’. This aims to

increase the share of the proceeds from extraction retained by the domestic economy. In addition,

some local content policies aim to develop skills of local workers and advance business practices that

spread to other sectors in the economy not directly related to extraction. However, whatever the

objectives, such policies to promote local content may increase a country’s exposure to the carbon

market risk by increasing the total share of a country’s assets (either foregone public revenues, or a

share of the nation’s human or physical capital) that would be exposed to a fall in fossil fuel demand.

If the public, physical and human capital investment in local content delivers high returns—both in

the financial and social sense—in the earlier rather than later years, or if the skills and business

developed to supply fossil industries can be applied to other sectors, the carbon market risk is

reduced. However, if local content takes many years to develop, foregoes significant tax revenues, or

has limited value outside the fossil fuel sector, these policies may increase countries’ exposure.

Third, fossil fuel rich countries have tended to develop economies that are relatively carbon-

intensive. Figure 11 shows that petroleum and coal producers (highlighted in red and blue

respectively) emit a larger amount of carbon per dollar of GDP than non-fossil fuel producers). A

chief policy that has led to this carbon intensive development is the tendency for fossil fuel

producers to subsidise fuel consumption (Friedrichs and Inderwildi 2013).

14

Figure 11. The carbon intensity of economies in terms of the ratio of carbon emitted per dollar of

GDP.

Source: Friedrichs, J., & Inderwildi, O. R. (2013)

Note: Red highlights denote predominantly petroleum producers, Blue highlights denote predominantly coal

producers.

This is a concern for countries wishing to reduce global carbon emissions, but also a concern for

fossil fuel-rich countries seeking to reduce exposure to the carbon market risk. These countries could

suffer in two ways. First, if their trading partners (such as the European Union) impose a carbon

consumption tax: a policy in which the consumers of products are taxed according to the carbon

content of the product, rather than a tax imposed on carbon emitters (Helm 2015). Second, they

would suffer if climate finance initiatives reward those countries that do reduce fossil fuel

consumption or emissions.

POLICY PROSPECTS

These three related challenges are highly significant for FFRDCs. Countries’ fossil fuel wealth

represents an opportunity to fund a significant expansion in their productive capital base and to

promote economic development. Not extracting their fossil fuel wealth could eliminate an important

path to economic development and write off a significant portion of national wealth.

In the face of these challenges, we discuss four policy options FFRDC governments should consider.

1) Diversify the economy to reduce exposure to carbon market risk

1a. Diversify the economy away from fossil fuel industries. Resource rich countries cannot liquidate

their fossil fuel assets easily—unlike investors who can sell the financial assets in a market—but

fossil fuel revenues can kick start industrial growth if used well (see, for example, precept 10 of the

Natural Resource Charter). Diversification means increasing the relative size of sectors not positively

related to the carbon-adjusted value of fossil fuels. This also means carefully considering the risk of

promoting fossil fuel intensive industries, including downstream sectors benefitting from fossil fuel

inputs.

1b. Reduce state investment in the fossil fuel industry. The third challenge we have described shows

that it is not only reserves that can become stranded. State capital invested in state owned

companies, infrastructure and the supply chains may also fail to provide a sufficient return to the

state if the world reduces it use of fossil fuels. If the expected life and payoff of these assets is long

enough that declining oil, gas or coal prices will impact returns, or a government cannot liquidate

15

these assets at a reasonable value, then governments may wish to limit investments in these assets.

This may be the case for instance if a state company is expected to provide a sufficient positive

return only in the long-term, or if its international operations will expose additional state capital to

carbon market risk. Building up a profitable state owned company could potentially take longer than

the time taken for carbon-adjusted fossil fuels values to fall. Consistent with this notion, Marcel

(2016) advises that governments should base decisions about how much to invest in NOCs (and how

ambitious to craft their NOC agendas) in part on how long the country expects to have viable

reserves to produce. Climate market risk presents a potential further reduction in the expected

period of economic returns from the sector.

1c. Reconsider investing state and human capital in fossil fuel related supply chains. Some value of

local businesses and a workforce specialised in fossil fuel extraction may also become stranded.

Where local suppliers and labour can participate in supply chains relatively quickly without

protection or subsidies, then the country may be able to benefit from investment in human capital.

However, if education or training workers takes decades, and if these skills are not transferrable to

other industries, not only will state capital invested in workers be left stranded, but so may the

human capital. The workers trained exclusively for working in fossil fuel industry could have

improved their lives by developing a different set of skills suitable for sectors not at risk. Similar to

countries with high reserve-to-production ratios, a policy of investing in human capital is at risk if

returns are expected only in the long term. Marcel (2016) suggests that governments build local

content policies that maximise the associated benefits for other sectors of the economy. Further,

such strategies should carefully consider their value in a potential post-fossil fuel global economy.

1d. Reconsider additional exposure to assets whose value is correlated with the climate policy

adjusted price of energy. State assets whose value positively correlates with the oil price in a world

in which a widespread climate policy is adopted will add to the risk countries face from their fossil

fuel assets being stranded. As such, certain assets in sovereign wealth funds and state equity in

companies who themselves are not well diversified from carbon market risk are vulnerable. Instead,

governments could hedge against the carbon market risk by minimising capital held in such assets,

or otherwise investing in assets that are negatively correlated with or at least not correlated to the

carbon adjusted prices of fossil fuels in a world in which a climate policy is widely adopted.10

2) Promote the competitiveness of the fossil fuel sector to reduce the potential extent of

stranding

2a. Consider whether to promote continued or increased exploration through licensing

While the world may already have more reserves than can be safely burned, it does not follow that

the best course of action for a country is to stop exploration. Development and extraction are costly,

so it is better for the country and investors to explore for reserves that are less costly to extract, if

the cost of exploration does not outweigh this difference in development and carbon-adjusted

extraction costs. Referring to the supply and demand graph in figure 2, some new discoveries may

be situated further on the left of the global supply curve than the country’s existing projects and

thus less likely to become stranded as prices fall. In other words, new, cheaper discoveries that can

be brought to production quickly can displace more expensive fields—the pace of which is subject to

10 See van den Bremer, T., Ploeg, F. Van Der, & Wills, S. (2013). The Elephant in the Ground : Managing Oil and

Sovereign Wealth (OxCarre Research Paper 129) (Vol. 44) for a similar analysis on hedging reserve value risk

for sovereign wealth fund management.

16

a country’s reserve-to-production ratios, as discussed above. Further, state capital is not usually

committed to the exploration process making it lower risk than other policy trade-offs.

An important caveat is whether there is a high likelihood that companies will discover cheaper

resources. If a country is at the point where further discoveries are likely to be smaller fields with

higher costs than what companies have previously discovered, further exploration is less likely to

reduce countries’ exposure to carbon market risk.

As discussed by Collier and Venables (2014) and Caney (2016), there may be a moral case made for

the lowest income countries being permitted to extract fossil fuels even after other countries have

halted their extraction. Under such circumstances companies and countries may still find it attractive

to explore for oil in the poorest societies- who may stand to benefit the most from the dwindling

demand for fossil fuels.

This policy fits with established good practice for countries. Companies’ interests in acquiring a license will rest on prospectivity. For frontier regions, investing in geological and geophysical surveys before licensing reduces geological uncertainty for prospective license holders—whether it indicates an abundance or otherwise of resources. Nor is this policy necessarily in conflict with the recommendation to diversify the economy. Discoveries of reserves with cheaper extraction costs does not necessarily imply a relative expansion in the extractive sector.

2b. Reduce ‘above ground risks’ and costs to further reduce the risk of stranding

A second option to mitigate stranding risk is to improve governance and reduce ‘above ground’ risks

and therefore increase risk-adjusted returns expected by investors. Cust and Harding (2015) suggest

that countries with strong political institutions have an advantage over countries with weak

institutions in attracting exploration investment given the likely lower risk-adjusted returns imposed

by weak governance. Arezki and van der Ploeg (2016) argue that economic openness can improve

investor attractiveness. Consistent with these findings, governments can undertake a host of

reforms that make their country more attractive to investment and, in a similar manner to the first

proposal above, projects will be situated further to the left of the global supply curve if above

ground risks fall.

Countries can also improve their attractiveness to investors by improving infrastructure such as

power, roads and pipelines, and skilled labour. However, a government must consider how much

public capital to invest in these improvements to attract company investment given the potentially

shortened period for making returns. A government may wish to choose improvements that

minimise public capital commitments relative to attracting company investment (some public

investment in facilitating exploration might be one policy) and choose improvements that have large

spillover benefits to other areas of the economy. For example, energy or transportation

infrastructure that also benefits other sectors of the economy.

3) Avoid subsidising fuel consumption and promote policies consistent with higher future

carbon prices

The IMF, World Bank, OECD and G20, as well as guides such as the Natural Resource Charter already

advise countries against using fuel subsidies. They can put a large strain on the national budget,

leverage the risk from oil price volatility as subsidy payments must increase when the oil price

increases, and are often socially regressive—the primary beneficiaries are citizens with access to

17

automobiles and central heating, not necessarily the poorest in a country. For instance, Del Granado

et al. (2012) estimates that the wealthiest 20 percent in developing countries get a disproportionate

43 percent of the benefit from fossil fuel subsidies, while the poorest 20 percent get only 7 percent

(Arze et al. 2012).

For those countries that subsidize fossil fuel consumption already (Ross et al. 2017), reducing

subsidies during periods of low prices reduces the immediate impact on consumers (as the effective

subsidy is low). Further, governments need to consider how long industries that rely on fossil fuels,

such as transport, can take to switch to alternative energy sources, and calibrate that with a plan to

eliminate subsidies.

4) Consider the time-profile trade-offs regarding the depletion rate of the resource base

A key result from this paper is that countries with high reserve-to-production ratios are most at risk

from seeing some reserves stranded. One implication of this is that producer countries may choose

to accelerate the pace of licensing (the main lever by which governments can control the rate of

resource depletion): converting their fossil fuel wealth into cash before falling prices reduce their

value.

However, there are two factors that may deter accelerated production. Stevens et al. (2015) argue

that for low income countries a slower pace of licensing may give time for government to improve

institutions and with it, potentially earn more income in the future. In addition, accelerated recovery

from individual oil wells can reduce well pressure and reduce the total amount recovered. This

poses a dilemma for countries who may need to time to develop strong institutions governing their

fossil fuel sectors, but with the risk that fossil fuel prices fall before they can do so.

Further, even if faster development is an optimum strategy for countries on an individual basis, if all

producers do the same, supply rises and reduces prices for a given demand; a result known as the

“green paradox” (Sinn 2008; Ploeg and Withagen 2013). Previous authors have expressed the

paradox in terms of an attempt to reduce carbon emissions in the form of a carbon tax ends up

increasing emissions as producers increase production in the near term. For FFRDCs, the same

response to reduce exposure to the carbon market risk brings forward the decline in fossil fuel prices

and weakens the logic of accelerating production.

The optimal time profile of depletion under demand uncertainty will also likely depend on how

much sunk state capital has already occurred. In other words, the relatively maturity of the sector as

well as extent of state involvement will play a role. Less mature and prospective sectors may be

better off deferring or cancelling extraction as the prospects for declining demand for fossil fuels

increases.

APPROPRIATE POLICY ACTION UNDER TWO SCENARIOS FACED BY FFRDCS

The above discussion lays out a range of policies that FFRDC governments may find appropriate if

fossil prices did permanently decline. However, in the short term, some of these policies may impose

a cost to countries. Without clear policy action on climate change these scenarios remain highly

uncertain for fossil fuel rich countries, so deciding whether to impose these policies is difficult. To

help consider policy makers’ choices it is useful to consider two scenarios for FFRDCs. These

represent two contrasting versions of the future; reality is likely to be somewhere between these

two extremes.

18

In the first scenario there is no significant fall in demand in fossil fuels (the carbon market risk is not

realised), carbon emissions continue past the carbon budget, and the resulting change in the climate

continues to impact livelihoods. In this scenario, FFRDCs are able to draw benefits from their fossil

fuel resources but find that climate change may adversely impact other areas of the economy, such

as farming, water and low-lying coastal settlements.

In the second scenario, there is a significant fall in fossil fuel demand (carbon market risk is realised),

FFRDCs are unable to fully benefit from their fossil fuel resources, but the worst of climate change

may be avoided. In this scenario, FFRDCs may not face significant climatic costs, but the opportunity

to fund economic development via fossil fuel extraction is significantly diminished.

Each scenario might require a different policy response. Should policy makers continue to develop

their fossil fuel reserves and invest public capital in doing so, or diversify as quickly as possible

before fossil fuel demand falls? Figure 12 summarises the policy options available to FFRDC

governments and whether they are more or less likely to be appropriate under our two scenarios:

one in which carbon market risk or a permanent fall in fossil demand does not occur, and one in

which it does occur. With reality likely being somewhere between these two extreme scenarios.

While we suggest a level of appropriateness bearing in mind that the appropriateness of most

policies are always subject to local factors in the country at the time.

Figure 12. Appropriate policies under two scenarios: no carbon market risk vs carbon market risk

Policy May be appropriate if no decline in prices

May be appropriate if decline in prices

Improve general governance environment

Yes Yes

Accelerate development of resources

Depends on other factors

Potentially, but depends on other factors (and beware “green paradox”)

Facilitate exploration Yes Yes

Heavy state investment in companies and fossil-fuel sector assets

Depends on other factors

Depends on other factors. Not appropriate if discounted payoff and payoff horizon unfavourable against carbon market risk

Avoid fuel subsidies Yes Yes

Invest in foreign assets negatively correlated with fossil fuel prices

Depends on other factors

Yes

Encourage local content and investment in fossil-fuel sector skills

Depends on other factors

Depends on other factors. Not appropriate if discounted payoff and payoff horizon unfavourable against carbon market risk

Diversify economy away from fossil fuel dependence

Yes Yes

Source: NRGI. Natural Resource Charter, 2014 and author’s views

Several of the policies are appropriate under both scenarios. In these cases, countries are more likely

to benefit from the policy no matter which scenario is realised and so the risk of governments

choosing an inappropriate policy depending on the climate policy scenario is relatively low. Fewer

19

policies are likely to be appropriate under scenario of declining demand for fossil fuels but depend

on other factors.

CONCLUSION

While still highly uncertain, there is a growing likelihood that fossil fuel use will decline in the future.

This is not only from the positive outcome of the Paris COP21 talks, but the emerging evidence of

the decarbonisation of global economic activity and the rise in renewable energy sources. In the

long-run, global demand for carbon-intensive energy relative might fall, reducing the market price

for fossil fuels. This creates the risk of ‘stranded nations’ – where resources under the ground

become commercially unviable to extract and a substantial share of a nation’s wealth may

permanently lose its value.

This paper identifies three challenges for fossil fuel resource rich developing countries. First, we

show that these countries are highly exposed to a decline in fossil fuel demand, with their median

ratio of oil and gas reserves to GDP is 3.66, compared with a median for non-FFRCs of 0.58. Second,

they are less able to diversify away from this risk than fossil fuel companies or investors- oil

companies on average hold only around 13 years of reserves on their balance sheets, whereas

FFRDCs hold an average of 45 years of known reserves at current production rates. Third, these

countries often find themselves under political pressure to implement policies that may expose

them to further risk. For example, supporting fossil fuel linked infrastructure and skills that relies on

long time horizons for payoffs to the country, subsidising fossil fuel consumption that extends

carbon-intensity of production, or by creating large, internationally operating national oil companies

whose operations may exposes national wealth to additional carbon market risk.

In response, we examine four broad policy implications arising from this carbon market risk that

fossil-fuel rich developing countries should consider. First, developing countries rich in fossil fuel

resources should seek to reduce their exposure to a decline in fossil fuel demand by diversifying

away from their dependence on fossil fuels and avoiding locking up additional state capital in fossil

fuel linked investments and expenditures. Second, they should seek to promote the competitiveness

of their fossil-fuel sectors by reducing costs and investor risks. Third, they should avoid subsidising or

otherwise failing to internalise the carbon costs linked to fossil fuel production. Fourth, they should

consider the time profile of their resource base and the implications of climate policy for optimal

depletion rates and savings rates.

The uncertainties regarding the timing and likelihood of the imposition of climate policies make

these decisions difficult. This suggests governments of fossil fuel-rich developing countries would

benefit from detailed analysis of future scenarios (perhaps along the lines of what we present in this

paper). Yet in many cases, our policy recommendations align with generally held good practice for

such countries. While their prioritisation, sequencing and in some cases direction may require

modification when the expectations of future global climate policies are considered, the ideas

expressed here suggest good governance of extractives and the fight to alleviate poverty need not

conflict with global efforts to prevent climate change. Many of the principles of extractives

governance apply no matter what the future holds.

20

REFERENCES

Allen, M., Frame, D., Huntingford, C., Jones, C., Lowe, J., Meinshausen, M., and Meinshausen, N. “Warming Caused by Cumulative Carbon Emissions Towards the Trillionth Tonne”, Nature 458, (2009): 1163-1166 Allen, M., Millar, R., Rogeij, J., and Friedlingstein, P. “The Cumulative Carbon Budget and its Implications”, Oxford Review of Economic Policy (2015).

Arezki, Rabah, Toscani, Frederik., and van der Ploeg, Frederick, 2016, “Shifting Frontiers in Global Resource Wealth” CEPR Discussion Paper DP11553, Centre for Economic Policy, London. Arze del Granado, Francisco Javier, David Coady, and Robert Gillingham. “The Unequal Benefits of Fuel Subsidies: A Review of Evidence for Developing Countries.” World Development 40 (2012): 2234–48. Loh, Tim. “JPMorgan Won’t Back New Coal Mines to Combat Climate Change”. Bloomberg, March 7, 2016. Accessed: 9 May, 2016. http://www.bloomberg.com/news/articles/2016-03-07/jpmorgan-won-t-finance-new-coal-mines-that-worsen-climate-change BP. BP Statistical Review of World Energy. 2015. Bremer, Ton van den., Frederick. Van Der Ploeg and Samuel Wills. The Elephant in the Ground : Managing Oil and Sovereign Wealth. OxCarre Research Paper no. 129, Vol. 44. (2013). Caney, Simon, 2016, “Climate change, equity and stranded assets”, Oxfam America Research Backgrounder series, https://www.oxfamamerica.org/static/media/files/climate_change_equity_and_stranded_assets_backgrounder.pdf Ceres. “Shareholder resolutions”. Accessed May 9, 2016. http://www.ceres.org/investor-network/resolutions Chicago Mercantile Exchange. “Crude Oil Futures Quotes”. Accessed August 2, 2016. http://www.cmegroup.com/trading/energy/crude-oil/light-sweet-crude.html Collier, Paul and Anthony Venables. “Closing Coal: Economic and Moral Incentives”. Oxford Review of Economic Policy. Vol 30, no . (2013): 492 - 512. Covert, Thomas, Michael Greenstone, and Christopher R. Knittel. “Will We Ever Stop Using Fossil Fuels?” Journal of Economic Perspectives Vol 30, no. 1. (2016): 117–138. Cust, James and Torfinn Harding. 2015. Institutions and the location of oil exploration. OxCarre Working Paper 127. Downey M. Oil 101. Wooden Table Press, 2009. EY. Global oil and gas reserves study. 2013. Financial Times. “Mark Carney warns investors face ‘huge’ climate change losses”. 29 September 2015. Pilita Clark in London.

21

Friedrichs, J., & Inderwildi, O. R. “The carbon curse: Are fuel rich countries doomed to high CO2 intensities?” Energy Policy, 62 (2013): 1356–1365. Green, Fergus and Stern, Nicolas. “China’s “new normal”: structural change, better growth and peak emissions”. Policy Brief Grantham Research Institute on Climate Change and the Environment. 2015. Helm, Dieter. The Carbon Crunch, revised and updated. New Haven and London: Yale University Press, 2015. International Energy Agency. World Energy Outlook 2012. International Energy Agency. World Energy Outlook 2015. IPCC. Climate change 2001: Impacts, Adaptation and Vulnerability. Chapter 19.4.2 Distribution of Impacts by Sector. IPCC. 5th Assessment Report 2014: Impacts, Adaptation and Vulnerability. Chapter 21.5.1.2, p.1177 Regional context Lahn, Glada and Sian Bradley. Left Stranded? Extractives-Led Growth in a Carbon Constrained World. Chatham House, 2016. Leaton, James. Unburnable Carbon – Are the world’s financial markets carrying a carbon bubble? Carbon Tracker Initiative, 2013. Marcel, Valerie (editor) 2016. Guidelines for Good Governance in Emerging Oil and Gas Producers. Chatham House. McGlade, Christophe and Paul Ekins. The geological distribution of fossil fuels unused when limiting global warming to 2 degrees. Nature, Vol 517 (2015): 187–190. Meinshausen, M., Meinshausen, N., Hare, W., Raper, S.C.B., Frieler, K., Knutti, R., Frame, D.J., and Allen, M.R. “Greenhouse-Gas Emission Targets for Limiting Global Warming to 2 oC” Nature, 458, (2009): 1158-1162 Meister Consultants Group. “Renewable Energy Revolution” March 16, 2015. Metayer, Matthieu; Breyer, Christian; and Fell, Hans-Josef. “The projections for the future and quality in the past of the World Energy Outlook for solar PV and other renewable energy technologies”. Published in the proceedings of the 31st European Photovoltaic Solar Energy Conference, September 14 - 18, 2015, Hamburg, Germany. Mitchell, John., Valerie Marcel and Beth Mitchell. Oil and Gas Mismatches : Finance, Investment and Climate Policy. Chatham House, 2015. Natural Resource Governance Institute. Natural Resource Charter. Second edition, 2014. Natural Resource Governance Institute. Resource Governance Index 2013. Nelson, David et al. Moving to a Low-Carbon Economy : The Impact of Policy Pathways on Fossil Fuel Asset Values. Climate Policy Initiative, 2014.

22

Proctor, Steve. Re-risking the Geology Natural Resources Magazine (Feb 25 2010). PWC. Conscious uncoupling? Low Carbon Economy Index 2015. Ross, Michael., Hazlett, Chad., & Mahdavi, Paasha, 2017, Global progress and backsliding on gasoline taxes and subsidies. Nature Energy 2, Article number: 16201 (2017) doi:10.1038/nenergy.2016.201 Rozenberg, Julie, Adrien Vogt-Schilb and Stephane Hallegatte. 2014. “Transition to Clean Capital, Irreversible Investment and Stranded Assets.” Policy Research Working Paper 6859. Washington DC: World Bank. http://www-wds.worldbank. org/external/default/WDSContentServer/WDSP/IB/2014/05/06/000158349_2014 0506114857/Rendered/PDF/WPS6859.pdf Sinn, H.W. The Green Paradox. Cambridge, MA, MIT Press, 2008 Spencer Dale, “New Economics of Oil”, 13 October 2015, Society of Business Economists Annual Conference, London. Stevens, Paul; Glada Lahn and Jaakko Kooroshy. The Resource Curse Revisited. Chatham House Research Paper, 2015. Tickell, O. Kyoto2: How to Manage the Global Greenhouse. Zed Books, 2008. Toews, Gerhard and Naumov, Alexander (2015). The Relationship Between Oil Price and Costs in the Oil and Gas Industry. OxCarre Research Paper 152 van der Ploeg, F. 2016, Fossil fuel producers under threat. Oxcarre Policy Paper 28. van der Ploeg, F., & Withagen, C. On the Relevance of Green Paradoxes. OxCarre Policy paper 18. 2013. Venables, Anthony. “Using Natural Resources for Development: Why has it proven so difficult?” Journal of Economic Perspectives. Vol 30. No. 1. (2016): 161 – 184. Weitzman, M. 2011. Fat-Tailed Uncertainty in the Economics of Catastrophic Climate Change. Review of Environmental Economics and Policy, volume 5, issue 2, summer 2011, pp. 275–292 World Bank. The Changing Wealth of Nations: Measuring Sustainable Development in the New Millennium. 2011. World Bank. Towards a sustainable energy future for all: directions for the World Bank Groups energy sector. 2013.

23

APPENDIX

The set of Fossil Fuel Rich Developing Countries listed in figure 5 is defined as those countries with:

GNI per capita 2010 – 2014 average of an USD 12,736 or less

Proportion of value of oil, gas and coal production (2014) to GDP (2014) of 10 percent or more

Proportion of oil, gas and coal reserves values to total national assets of 25 percent or more.

Production values are calculation by multiplying the unit amount by the following respective prices:

Oil price taken as Brent Crude spot price USD 36 per barrel

Natural Gas for European, Central Asia and Middle East and North Africa regions USD 5.81 per mmBtu

Natural Gas for Asia Pacific and East Sub Saharan Africa regions USD 9.50 per mmBtu

Natural Gas for North America, Latin America & Caribbean and West Sub Saharan Africa USD 1.92 per mmBtu

Coal is taken as USD 92.7 for European producers (equivalent to the Northwest Europe marker coal price), USD 74.3 for North American and LAC producers (equivalent to the US Central Appalachian coal spot price), USD 101 for Central Asian producers (equivalent to Asian marker coal price), and USD 166.9 for Asia-Pacific producers (equivalent to the Japan coking coal import cif price).

Reserve values are calculated by assuming a constant production rate for the number of years of available reserves (equivalent to the reserve to production ratio) with future values discounted using a rate of four percent, as used in World Bank (2011). A country’s weighted reserve to production ratio is the reserve to production ratios of oil, gas and coal weighted by the proportion value of production of each commodity:

Weighted reserve-to-production (R-to-P) ratio = [(Value of Oil production/Value of total production)*R-to-P of oil] *[(Value of gas production/Value of total production)*R-to-P of gas] *[(Value of coal production/Value of total production)*R-to-P of coal]

Modelled price decline of fossil fuels

To illustrate the effect of a long-run decline in prices, we assumed a smooth decline in prices (two percent each year from the base case) for both oil, gas and coal. Figure 6. Illustration of the effect of a decline in fossil fuel prices

Source: authors’ calculations. Notes: See appendix for assumptions and method.

24

We calculate the value of reserves as the amount of production each year until countries deplete their current reserve amounts multiplied by the assumed prices of production, with future values discounted by the social discount rate of four percent. Crucially, this does not account for extraction costs. The income derived from extraction that is typically shared between companies and governments would therefore be less than revenue estimates presented here, and the proportional value-at-risk larger.11 The value-at-risk column in figure 7 shows the difference in the value of reserves assuming current prices are maintained, and the value of reserves assuming a year-on-year decline of two percent. The difference is largely determined by the rate at which countries can deplete their reserves, which we assume to be equal to past production rate in each country. Countries with lower reserve-to-production ratios are likely able to extract their resources faster than those with high reserve-to-production ratios and avoid more of the effect of a fall in future values. Venezuela, for example, has large reserves and low production rates, so sees a fifth of the value of its reserves cut given the assumed fall in prices. Fixed production rates is a strong assumption. As we discuss below, countries may seek to increase production to monetise their fossil fuel wealth before a significant decline in prices. However, there will be limits to how much countries can do this based on the infrastructure they have and the ability to attract sufficient capital to enhance existing infrastructure. Furthermore, investing more capital to increase production rates simultaneously increases the capital stock that is exposes to the carbon market risk. So countries face a dilemma in this regard that does not necessarily mean greater production capacity is optimal.

11 This calculation also makes the simplifying assumption that extraction costs do not change. However, in practice, costs are shown to be endogenous and positively related to the market price of the commodity extracted (Toews & Naumov 2015), so our simplifying assumption is still useful as the proportional rent does not change as much as if costs were constant.