Reform of environmentally harmful subsidies: distributional issues

Category	Articles
Sub-Categories	n/a
Year	2014
Author(s)	Annegrete Bruvoll Haakon Vennemo
	http://vista-analyse.no/site/assets/files/5683/2014_reform_of_environmentally_harmful_subsidies_distributional_issues.pdf
Download	file_download (206.2 kB)
Read in browser	PDF

263	
12. \beform of environmentally harmful 
subsidies: distributional issues	
Annegrete Bruvoll and Haakon Vennemo	
12.1 INT\bODUCTION
According to the International Monetary Fund (IMF, 2013) global sub-
sidies on energy amounted to US$480 billion in 2011. The International 
Energy Agency (IEA, 2012) estimates subsidies on fossil fuels, a part 
of energy, to be US$523 billion in 2011. These are subsidies on energy 
and fossil fuels consumed by households and enterprises. Further sub-
sidies, such as government in- kind support, are available for enterprises. 
According to IEA et al. (2010a), producer subsidies add at least another 
US$100 billion. In many developing countries, subsidies on fossil fuels exceed 5 per cent 
of GDP. Top of this particular league is Uzbekistan, where in 2011 fossil 
fuel subsidies amounted to 28 per cent of GDP, followed by Turkmenistan 
(23 per cent), Iraq (19 per cent), and Iran (17 per cent).	
1 Subsidies to energy 
are discussed in Chapter 6 of this book. Several countries have in recent years attempted to reduce subsidies to 
energy, but attempts are often met with resistance. In fact, the International 
Monetary Fund (2013) recently evaluated 28 reform attempts carried 
out over a period of 20 years in 22 developing countries. In at least five 
instances reforms were met with widespread public protests. The resistance to subsidy reform is striking since removal of environ-
mentally harmful subsidies tends to increase GDP and economic welfare. 
There will be winners and there will be losers, but the gains to winners 
will be higher than the loss to losers. Ellis (2010) reviews six major multi- 
country, multi- fuel studies undertaken since the early 1990s. All six studies 
found that a subsidy reform would increase GDP in the countries studied 
by 0.1–0.7 per cent per year to 2050. In addition, pollution would be lower, 
as would greenhouse gas emissions. The strain on public budgets would be 
lower too. In Uzbekistan, the top subsidizer, fossil fuel subsidies are equal 
to 60 per cent of public revenue, according to the International Monetary 
Fund (2013).	
OOSTERHUIS 9781782\P545309 PRINT (M3374)\P (\b).indd 263	07/03/2014 09:57\P

264 Paying the polluter
A leading explanation for the resistance across the world is that the 
losers from reform will in fact not receive compensation. Using a social 
science term there is little or no ‘transition management’ including com-
pensation and other policies to help people adjust. This insight leads to 
the theme of this chapter: how can policy- makers design compensation to 
accompany a reform of environmentally harmful subsidies and improve 
its support? Since it is not obvious that the wealthy deserve compensation, 
this design question includes an assessment of who should get compensa-
tion, in which form, and by how much. In the sections that follow we first discuss historical attempts at lowering 
environmentally harmful subsidies on coal production in Europe and fossil 
fuel subsidies on consumption mainly outside Europe. Based on the histori -
cal attempts at lowering subsidies we outline three steps of a successful com -
pensation scheme to support reform of environmentally harmful subsidies.
12.2 CASE STUDIES
12.2.1 Subsidies to \boal P\troduction	2	
Coal production has been supported in many countries and over decades. 
Subsidies were defended as a vehicle to maintain jobs and domestic energy 
supply. In the EU, coal is still heavily subsidized in Germany’s \buhrgebiet 
region, Northwest Spain, and \bomania’s Jiu Valley (World Nuclear 
Association, 2013).
Over time the economic burdens created by the subsidies push through 
reform, and several countries have gradually reduced or removed coal 
subsidies. The major drivers of reforms have been budgetary considera-
tions and a reduced perceived need to secure energy independence via coal 
mining. Climate policy and environmental pressure groups have become 
increasingly important drivers in recent decades. State aid for coal production was nominally prohibited in Europe since 
the 1952 European Coal and Steel Community (ECSC), but multiple 
exceptions were allowed. After 2002, direct state support to coal produc-
tion was prohibited in the EU (IEEP et al., 2007). In 2007, eight of the 
EU Member States produced coal: Poland, Germany, Hungary, the UK, 
Spain, Czech \bepublic, Slovakia and Greece. France closed its last mine in 
2004. In 2010, some €3.2 billion in coal subsidies were handed out in six 
EU countries: Germany, Hungary, Poland, \bomania, Slovakia and Spain 
(World Nuclear Association, 2013). In 2010, the European Commission 
decided that member governments must stop granting subsidies to loss- 
making coal mines by the end of 2014.	
OOSTERHUIS 9781782\P545309 PRINT (M3374)\P (\b).indd 264	07/03/2014 09:57\P

Reform of EHS: dist\lributional issues 
265
Examples from individual countries demonstrate the resilience of sub -
sidies. The reform in France took 40 years to complete. Poland started 
reforming its coal subsidies in 1990. \bestructuring provided for a strong 
emphasis on environmental protection, restructuring programmes, liq -
uidation of consistently unprofitable mines and privatization of those 
that turned a profit. Poland still has a very high extraction rate of coal 
relying on a complex system of export and restructuring subsidies. In 
Germany, subsidies were reduced by 50 per cent from 1998 to 2008, and 
the aim is a complete phase- out by 2018. However, the average mining 
job was subsidized by €75 000 in 2007. The continuing subsidies in 
Germany are upheld by strong and organized stakeholders with connec -
tions to electoral power. In Spain the perceived importance of the energy 
supply industry, and contracts the industry had made with governments, 
enabled its expansion when the rest of the European coal industry was 
in decline. In 2004, the average cost per tonne of Spanish coal was up 
to nine times higher than the cost of imported coal. The energy market 
was liberalized in 1998, with the introduction of a long- term restructur-
ing plan including increased competition. In 2012, thousands of Spanish 
coal miners marched to protest a cut in government subsidies by 63 
per cent that they say will eliminate their industry. Overall, the subsidy 
reforms have led to significant fall in coal extraction and employment 
within the coal sector.
\bompensatory measures
The costs associated with the reforms are the main causes for continued 
coal subsidies around the world: social consequences, employment costs, 
fuel costs to consumers and security of supply. In France, extensive meas-
ures were used to promote alternative economic activity in the affected 
regions. Former miners were guaranteed employment until the age of 45, 
when those with at least 25 years of service became eligible for a leave, 
during which they would receive 80 per cent of their final working salary 
until retirement (IEA et al., 2010b). In Poland, relatively generous sever -
ance packages were provided to miners leaving work, based on individual 
arrangements (IEEP et al., 2007). In the UK, compensatory subsidies were 
given to stimulate competitiveness of the industry with a viable future and 
aid the remaining industry to compete in the reformed electricity market. 
Support was given to create or safeguard jobs within some socially and 
economically disadvantaged areas. The aid given to those entering unem-
ployment was focused on creating alternative economic activities or sup-
porting skill development to facilitate a transition to another progression. 
This partly helped reduce some of the opposition to the reform and aimed 
to support long- term economic viability in otherwise disadvantaged areas. 	
OOSTERHUIS 9781782\P545309 PRINT (M3374)\P (\b).indd 265	07/03/2014 09:57\P

266 Paying the polluter
The gradual removal of subsidies has also helped to minimize dislocation 
of environmentally damaging activities (ibid.).
Evaluation
Despite the potential economic and environmental benefits, coal subsidy 
reforms in Europe have faced severe resistance and the subsidies proved 
hard to get rid off. The restructuring of the UK coal sector, while driven 
by political concerns and economic objectives, could be seen as offering 
positive environmental benefits in terms of lower CO	
2 emissions and 
other pollution to air and water. It also enabled the country to main -
tain a reasonably competitive coal industry, albeit a much smaller one. 
However, it was at the cost of extensive mine closures, significant social 
opposition, with some shift in environmental problems where domestic 
coal was replaced by imported coal. Coal- mining jobs in the coalfields of 
England and Wales accounted for about a quarter of all male jobs located 
in these areas. A lesson from coal reforms is that compensation in terms of 
stimulation of new job opportunities can increase support to the reform 
and foster economic growth in the longer run. Strong political will was a 
fundamentally key factor. Still, it shows how the economic, environmental 
and indeed party political benefits can be paid by increased social costs 
for a few. In the UK coal mining experience, the compensation of the 
losing parties was clearly not sufficient to avoid a rise in unemployment 
and social problems.
12.2.2 Subsidies to Fossil Energy \bonsump\ttion in Some Low- income 
\bountries
Currently, subsidies to consumption of energy are mostly found in low- 
income countries. We gave some aggregate figures in the introduction – see 
also Chapter 6. As another indication, it was found in a recent assessment 
of 101 countries in Africa, Asia, the Middle East and Latin America 
that an increase in international fuel prices did not increase domestic fuel 
prices to the same extent. Between 40 and 90 per cent of the increase was 
passed on in a median country (Arze del Granado et al., 2012). Evidently, 
someone picks up pa\frt of the bill. The stated aim of subsidizing consumption of energy is often to support 
low- income households. However, as a matter of fact, only a small share 
of the expenditure on energy subsidies reaches its aim. In a study of 20 
countries it was found that for each dollar spent on energy subsidies, 
only 7 cents reached the bottom fifth of households, the 20 per cent of 
households with the lowest income. Forty- three per cent went to the top 
fifth (ibid.). Similar distributional impacts are reported from individual 	
OOSTERHUIS 9781782\P545309 PRINT (M3374)\P (\b).indd 266	07/03/2014 09:57\P

Reform of EHS: dist\lributional issues 
267
countries. In Indonesia, for example, only 2 cents of every dollar spent 
on gasoline subsidies reach the bottom fifth of households (Dartanto, 
2013). The top fifth take away 60 per cent of the subsidy. There are differ -
ences between fuels, but even in the case of kerosene, which is used more 
intensively by the poor, the bottom fifth in Indonesia obtain only 9 cents 
of every dollar spent. In many poor countries consumer subsidies to energy amount to signifi-
cant shares of GDP and even higher shares of public revenue. Uzbekistan, 
Turkmenistan, Iraq and Iran have already been mentioned. These are 
exporters of oil and gas that may believe they can afford domestic subsi-
dies. But consider Bangladesh. This poor country spends 43 per cent of its 
public revenue on energy subsidies. Or consider Egypt, which despite its 
location in the Middle East is not a big petroleum nation. Egypt spends 
48 per cent of its public revenue on energy subsidies. Yemen spends 24 per 
cent. Cameroon spends 20 per cent, and so on. The numbers are all from 
the International Monetary Fund (20\f13). For each dollar spent on energy subsidies there is one less to spend on 
social programmes, poverty reduction, and the like. By making a prior -
ity of energy subsidies, countries forego opportunities to help their poor. 
Some countries have realized this, or at least realized that energy subsidies 
do cost a lot of money, and initiated reforms. However, according to the 
International Monetary Fund (2013) the reforms have had mixed success. 
Among the 28 reforms studied by the Fund, 12 were successful. Eleven 
more were deemed a partial success, and five were unsuccessful. A partially 
successful reform is one that was later reversed or was not completely 
implemented.
Reform stories
The details some of these reforms are revealing. We mention six reform 
stories. The first reform story is about Iran (Arze del Granado et al., 
2012). When international gasoline prices peaked above US$2 per litre in 
2008, they stood at US$0.10 in Iran. In late 2010 Iran announced plans 
to increase domestic prices to 95 per cent of international prices. This 
was an announcement that could be met with resistance since poor and 
rich alike would have to pay significantly more for energy. Subsidies were 
to be cut US$60 billion. But prior to the announcement the government 
had deposited US$30 billion in special bank accounts created for nearly 
80 per cent of the Iranian population, 61 million accounts in total. The 
ATM network was extended to every corner of the country to enable the 
population to confirm with their bank accounts that the promised com-
pensation did materialize. In addition, US$15 billion was set aside for 
enterprises to finance investments in energy savings. Prior to the reform the 	
OOSTERHUIS 9781782\P545309 PRINT (M3374)\P (\b).indd 267	07/03/2014 09:57\P

268 Paying the polluter
government ran a campaign to inform the public why reform was neces-
sary. For instance, the public was informed about the waste of energy from 
low prices, the low benefit to the poor compared to expenditures, and the 
costs of smuggling. As noted above, data from 2011 show that subsidies 
in Iran are still large and the reform was put on hold in 2012 amid fears 
of inflation, according to the International Energy Agency (2013). The 
International Monetary Fund (2013) calls the reform in Iran a partial 
success.
A second country that has tried reform is Ghana. In 2004 Ghana spent 
roughly 2.2 per cent of its GDP subsidizing fuel and 1 per cent on support 
to its national refinery company (e.g., Bruvoll et al., 2011). These are much 
lower figures than Iran, but Ghana is an oil importer and a low- income 
country. A government commission found that Ghana’s rich received the 
greatest benefits from subsidies and quantified how and to what extent 
the poor would be affected by future deregulation (Bacon and Kojima, 
2006). The research of the commission was an important foundation 
for communicating the necessity of reform and for designing policies to 
reduce impacts of higher fuel prices on the poor. A formula was put in 
place to link domestic prices of oil to international prices. The formula was 
intended to reduce political interference in fuel prices. The reform also included a communications campaign, and policies to 
assist the poor. The government (1) eliminated fees for state- run primary 
and secondary schools, (2) increased the number of public transport buses, 
(3) put a price ceiling on public transport fares, (4) channelled extra funds 
into a healthcare scheme for poor areas, (5) raised the daily minimum wage 
by 22 per cent and (6) started programmes to help spread electrification to 
rural areas and purchase essential equipment for workers. It also (7) contin-
ued its previous policy of cross- subsidizing kerosene and LPG. Expenses 
were partly paid for by an explicit earmarked social mitigation tax incor -
porated into the pricin\fg formula. In early 2008 the pricing formula was set aside and a new government 
reduced fuel taxes in 2009. This, however, led to a dramatic increase in 
fuel consumption and smuggling out of the country. In 2011 the govern-
ment increased petroleum prices by 30 per cent, to full cost recovery. This 
generated widespread protests. The International Energy Agency (2013) 
reported a similar attempt in early 2013. The International Monetary 
Fund (2013) calls Ghana a partial success. A third example of reform is presented by Indonesia. In 2005, the gov-
ernment started to gradually liberalize the fuel market with the aim of 
completely eliminating fuel subsidies. That year prices more than doubled, 
with kerosene prices rising 185 per cent, for example (Arze del Granado 
et al., 2012). A public information campaign that included newspapers, 	
OOSTERHUIS 9781782\P545309 PRINT (M3374)\P (\b).indd 268	07/03/2014 09:57\P

Reform of EHS: dist\lributional issues 
269
TV talk shows, village notice boards, and the distribution of pamphlets 
and brochures preceded the price increases. In addition, the government 
initiated temporary unconditional cash transfers to one- third of the 
Indonesian population, 19 million households. Each household received 
about US$120 (ibid.). The government budget savings from the reduced 
cost of fuel subsidies were estimated to be about US$10.1 billion in 
2005–06 (Bruvoll et al., 2011). Expenses on the cash transfers were about 
US$2.3 billion. The government also financed programmes in education, 
rural development and health. While the IEA et al. (2010a) consider the 
compensation programme to be largely successful, Dartanto (2013, p.  2) 
writes that ‘the drastic reduction of fuel subsidies in 2005 resulted in 
misery for the poor’. A new phase of the Indonesian reform programme 
was initiated in 2008, with the cash compensation programme briefly 
reinstated. According to the International Energy Agency further reforms 
are taking place in 2012–13. However, as we have seen there are still large 
subsidies in place in the country. The International Monetary Fund char -
acterizes the reforms of 2005 and 2008 as p\fartial successes. A fourth reform example is Malaysia. Malaysia has had a cap on the 
price of electricity and petroleum products in place for some years. The 
difference between world market prices and the caps has been subsidized 
by the government. \bising oil prices in 2007 and 2008 substantially 
increased subsidies as the gap between world market prices and the price 
caps on electricity and petroleum products widened, putting pressure on 
the budget and prompting the Malaysian government to review its subsidy 
policies. Subsidies were reported to have cost the Malaysian government 
US$14 billion in 20\f08, or about 4 per cent of GDP (IEA et al., 2010b). In 2008, the Malaysian government introduced a broad package of 
reforms to its energy subsidies. In 2008 the price of natural gas for power 
generation was raised by 124 per cent, and the average electricity tariff for 
all sectors of the economy was increased by 24 per cent, gasoline prices by 
41 per cent and diesel prices by 63 per cent (Hamid, 2008). To offset the 
increased prices, the Malaysian government offered compensation in the 
form of (1) lower annual road taxes, (2) cash rebates, (3) windfall taxation 
on certain sectors and (4) an expansion of the social safety net. Despite 
these measures there were widespread protests against the subsidy reform 
(IEA et al., 2010b). In 2011 energy subsidies in Malaysia amounted to 
about 2 per cent of GDP according to the International Monetary Fund 
(2013). A fifth example is the Mexican cash transfer programme (the 
Oportunidades programme) used to compensate very low- income house-
holds for reduced energy subsidies. Subsidized energy prices in Mexico 
have represented a serious economic strain on the government budget, 	
OOSTERHUIS 9781782\P545309 PRINT (M3374)\P (\b).indd 269	07/03/2014 09:57\P

270 Paying the polluter
equivalent to more than 0.5 per cent of GDP over the period 2005–09. The 
programme targets human capital, that is, education, health and nutrition 
of children. The distribution mechanism is cash transfer and health clinic 
visits. Another programme in Mexico stimulates employment in rural and 
small communities. When labour demand is low and opportunities are few, 
work is provided in projects typically related to infrastructure and environ-
mental preservation (G- 20, 2010; IEA et al., 2010a).
Finally, consider an example that has so far not turned out a success, 
the case of subsidies to liquefied petroleum gas (LPG, in practice propane 
and/or butane) in Senegal. In the 1970s, Senegal had a goal of reducing 
deforestation by substituting part of charcoal consumption with LPG. 
LPG was subsidized first through cooking equipment and from 1988 the 
government started to subsidize LPG fuel itself. The LPG subsidy pro-
gramme reduced household pollution and it reduced pressure on forests. 
But the policy also led to an unsustainable fiscal burden, disproportionate 
benefits for the relatively rich, and fuel smuggling (Laan et a\fl., 2010). A phased reduction of the subsidy in annual increments of 20 per cent 
was started in 1998. The plan was put on hold in 2002 due to increasing 
global LPG prices, exchange rate variations and inflation, resulting in 
continuing high subsidies. In 2006, the cost of LPG subsidies amounted 
to 1.4 per cent of GDP. In 2008, the IMF found that the 40 poorest per 
cent of the population gained only 19 per cent of the total improvement in 
welfare from the LPG subsidy, while the richest 40 per cent gained 61 per 
cent (ibid.). The Senegal experience does not bear the mark of a successful policy. The 
government has attempted to reform the subsidy for over a decade and has 
made little headway. However, removal of the subsidy risks an increase in the 
use of charcoal and firewood by some households. Senegal could learn from 
the packages of measures used by, for example, Ghana, that is, research to 
identify those most likely to be negatively impacted by de- subsidization, 
information campaigns about the benefits of reform, cash transfers to the 
poor, and greater independence and transparency of fuel prices (ibid.). 
Disincentives for the use of charcoal and wood, rather than fuel subsidies, 
would be a more direct way to address the goal of reducing deforestation.
Evaluation
The advice to Senegal summarizes some of the points that emerge from 
the reform stories. Information campaigns have been part of the reforms 
of Iran, Ghana and Indonesia. It is also an issue highlighted by the 
International Monetary Fund (20\f13, p. 28) in its evaluation of 28 reforms. 
The Fund wrote that ‘a far reaching communications campaign can help 
generate broad political and public support and should be undertaken 	
OOSTERHUIS 9781782\P545309 PRINT (M3374)\P (\b).indd 270	07/03/2014 09:57\P

Reform of EHS: dist\lributional issues 
271
throughout the reform process’, and cited further examples to this effect 
from Namibia, the Phi\flippines and Uganda.\f
Cash transfers to the poor was another recommendation we gave for 
Senegal, and more broadly, compensation in the form of cash or cash- 
like spending such as lower school fees are included in the reforms of 
Iran, Ghana, Indonesia and Malaysia. Again this is an issue that the 
International Monetary Fund highlighted, writing that ‘well- targeted 
measures to mitigate the impact of energy price increases on the poor are 
critical for building public support for subsidy reforms’ and ‘targeted cash 
transfers or near- cash transfers (vouchers) are the preferred approach to 
 compensation . . . when cash transfers are not feasible, other programmes 
can be expanded while administrative capacity is developed’ (IMF, 2013, 
p. 30). A third piece of advice to Senegal was to conduct research on who 
are the most likely to be negatively affected by a lifting of subsidies. This 
knowledge is important to work out well- targeted compensation measures, 
and also to support the information campaign. The Ghana experience sets 
an example to follow. More generally the research on impacts is part of the 
preparation and planning that a thoughtful government will do. The prep-
aration of the Iranian government in terms of setting up accounts, extend-
ing the ATM network and so on, is another example. The International 
Monetary Fund (2013) highlighted research and other preparatory work, 
as well as a timetable for reform, as elements of a comprehensive reform 
plan. A successful r\feform needs a comprehensive reform plan. A final piece of our advice to Senegal was greater independence and 
transparency of fuel prices. The International Monetary Fund (2013) high-
lighted the automatic pricing formula of the kind that was tried in Ghana. 
Although the formula was set aside in Ghana, the Fund argued that by 
itself such a formula makes reform more likely to succeed. ‘Depoliticize 
energy pricing’ was the message from the Fund (IMF, 2013, Executive 
Summary). A further recommendation of the Fund, based on the 28 reform epi-
sodes, is to phase and sequence price increases in order to give people 
time to adjust, and to accommodate inflationary pressures as in the case 
of Iran. The Fund also called for an improvement in the efficiency of 
the state- owned enterprises that commonly supply energy in low- income 
countries. A suggested list of key elements of a successful reform is then:	
●
●a comprehensive reform plan;
●
●a communications strategy;
●
●appropriately phased and sequenc\fed price increases;	
OOSTERHUIS 9781782\P545309 PRINT (M3374)\P (\b).indd 271	07/03/2014 09:57\P

272 Paying the polluter	
●
●improving the efficiency of state- owned enterprises in the energy 
sector;
●
●targeted compensation measures;
●
●depoliticizing of energy prices.
Taken together such a list suggests a programme for transition manage-
ment before, under and after a reform of energy prices. Compensation is 
at the heart of such a programme. The next section describes how to carry 
out the compensation aspect.	
12.3 A TH\bEE- STEP P\bOCEDU\bE FO\b USING 
COMPENSATION TO GOOD EFFEC\fT
The examples we have just given are fairly representative of reforms to 
environmentally harmful subsidies in the sense that they cover the pro-
ducer–consumer dimension; the Europe–non- Europe dimension and the 
dimension of exemptions from optimal taxation versus outright subsidies. 
More examples are found in reports and books devoted to the issue (e.g., 
Bruvoll et al., 2011; International Energy Agency et al., 2011; International 
Monetary Fund, 2013)\f. Economic theory teaches that a reform of an environmentally harmful 
subsidy is a reform that increases GDP and welfare. Empirical studies 
confirm this (Ellis, 2010). This means that everybody in theory can be 
compensated for the loss of subsidies, and in addition there will be a net 
gain to share. If the increase is distributed well, everybody could gain. On 
the other hand the purpose of the reform is to increase the price of the 
environmentally harmful goods, which means that people will have to pay 
more to obtain the sam\fe quantities of these goods. By this, they lose. One of the keys to a successful reform is to make sure that the potential 
gain is turned into an actual gain for the majority of those involved, and 
low- income groups in particular. We have seen above that low- income 
groups benefit surprisingly little from energy subsidies. Still the perceived 
benefit may be larger, and whatever the size of the benefit, compensation 
for their loss should be offered to low- income groups. More concretely our 
review offers the following three lessons, which we call steps, in a successful 
compensation policy of environmentally harmful subsidies:	
●
●Step 1: Identify who gains from subsidies. These are the ones who 
stand to lose when subsidies are lifted, at least in the shor\ft term.
●
●Step 2: Identify who should be compensated when lifting subsidies. 
It is neither possible nor desirable to compensate everyone, but 	
OOSTERHUIS 9781782\P545309 PRINT (M3374)\P (\b).indd 272	07/03/2014 09:57\P

Reform of EHS: dist\lributional issues 
273	
those who qualify on certain cri\fteria (it may not just be low income) 
should be given compensation.
●
●Step 3: Identify how to compensate and what level of compensation 
is appropriate over what timescale. Compensation may potentially be 
offered in different ways with consequence\fs that differ.
12.3.1 Step 1: Identify Who\t Gains
Step 1 in designing a successful compensation policy is estimating who 
gains from the subsidies and thus, who stands to lose from a reform and 
hence likely oppose attempts at reform. How to do this? Consider a sug-
gestion to reduce subsidies on energy consumption. A reasonable starting 
point for estimating who gains from current subsidies is the household 
expenditure survey. A recent household expenditure survey is available in 
most countries. It surveys household expenditures on a number of goods 
including heating, food, transport and so on, and also socioeconomic 
characteristics of the household such as income, education, size of house-
hold and so forth. When working out the gains from subsidies on energy 
we are particularly interested in how much each group spends on energy. 
For instance, let us assume that a particular household category spends 
US$100 and that the subsidized price is US$1. If the true price without 
subsidy is US$2 a quick estimate would suggest that the household gains 
US$100 from the subsidy. The analysis could also be worked out in relative terms. If the household 
in question spends 10 per cent of its expenditure on energy at subsidized 
prices of US$1 and the true price is US$2 we may assume that the house-
hold gain equals 10 per cent of its expenditure. The two approaches will in fact give different answers because they 
assume different degrees of price responsiveness in the household. The first 
approach, which concludes that the household gains US$100, assumes that 
the change in purchase when the price changes is zero. The second approach, 
which concludes that the household gains 10 per cent of expenditure, 
assumes that the percentage change in quantity consumed is equal (with a 
minus sign) to the percentage change in price. Both assumptions are valid 
but the first one (zero response) seems to be more commonly used (Arze del 
Granado et al., 2012). In a detailed analysis, suitable price elasticities that fit 
local conditions should be worked out and so- called income effects should 
be excluded in order to work out ‘consumer surplus’. Interested readers 
should consult textbooks in microeconomics or Google the term. An analysis based on consumer expenditure on energy helps to estimate 
the direct effect of energy subsidies and their removal. The indirect effects 
are also important. The indirect effects concern the fact that enterprises 	
OOSTERHUIS 9781782\P545309 PRINT (M3374)\P (\b).indd 273	07/03/2014 09:57\P

274 Paying the polluter
that use energy intensively, tend to charge less for their goods and serv-
ices when energy is cheap, and conversely, if the price of energy increases 
because subsidies are removed, they will have to charge more. The indirect 
effects affect households differently depending on their overall expenditure 
and pattern of expenditure.
To work out the indirect effect it is necessary to estimate how much 
prices increase. It is reasonable to start by considering the energy costs of 
enterprises. Unless it can respond by using less energy, an enterprise that 
spends 20 per cent of its costs on energy will experience that a doubling of 
the price of energy means a 20 per cent increase in overall costs. Hence, a 
20 per cent increase in price, and similar for other enterprises and indus-
tries depending on how intensively the use energy (and how easy it is to 
switch from energy to other inputs)\f. In addition, enterprises sometimes buy from other enterprises in order 
to produce goods. Hence there is for most enterprises a second- round 
impact on costs – which leads to a third, and a fourth and so on. To keep 
track of these rounds and work out the accumulated impacts on costs, 
economists make use of the ‘input–output table’. The input–output 
table records economic transactions between sectors in the economy, for 
example, from the energy sector to manufacturing, and from manufactur -
ing to the energy sector and so on, in a consistent loop. In low- income 
countries input–output tables are sometimes not up to date, which creates 
an element of uncertainty in the analysis (Arze del Granado et al., 2012). 
Another uncertainty is created by methodological choices. The approach 
we have just described for instance assumes that the rate of return to 
owners is unaffectd by the energy price changes. Wages and the exchange 
rate are also assumed to be un\faffected. Indirect impacts – those showing up as a consequence of cost changes 
of firms – tend to be quite important compared to direct impacts. For 
instance, in the Arze del Granado et al.’s (2012) survey the direct impact of 
energy subsidy removal on the poorest fifth of households was only 40 per 
cent of the total impact. In the case of producer subsidies including prefer -
ential treatment, free access to land, infrastructure and so on, the indirect 
impact is 100 per cent. This is the case since producer subsidies target costs 
of firms, that is, the indirect impact. The assessment of who gains from the 
subsidy will in this case rely solely on the estimate of the indirect impact, 
guided by the input–output\f table.
12.3.2 Step 2: Identify Who\t Should be \bompensa\tted
Step 2 in designing a successful compensation policy is to determine who 
should be compensated. So far in this chapter we have focused the discus-	
OOSTERHUIS 9781782\P545309 PRINT (M3374)\P (\b).indd 274	07/03/2014 09:57\P

Reform of EHS: dist\lributional issues 
275
sion on compensation based on income – low- income groups should be 
compensated. But compensation based on income is not the only possibil-
ity. Compensation, for example, to laid- off workers is quite often seen as 
fair, independently of the relative income of workers. Sometimes, com-
pensation has a regional dimension. In addition, compensation to vocal 
stakeholders may be necessary to help buy support for the reform in the 
political process.
\bompensation based \ton income and wealth
Low income is arguably the most common and most important criterion 
for compensation (e.g., International Monetary Fund, 2013). In particular 
this is the case when considering consumption subsidies on energy, mostly 
in developing countries. The primary reason why low income is impor -
tant is that compensation is seen in the context of income distribution. A 
person with low income is entitled to compensation because he or she has 
low income to begin with. A person with high income is not, as a rule, con-
sidered for compensation, even if his or her loss from the subsidy reform in 
isolation may be higher than t\fhe loss of a poor person. The primary reason for focusing on low income is confirmed and rein-
forced by the fact that the benefit to low- income groups quite often is 
used as a political motivation for continued subsidies. Politicians may fear 
removing subsidies because they to some extent benefit people with a very 
low budget. It may not matter very much that other groups also benefit, 
and in fact benefit more. The low- income people concerned often feel the 
same way and as reported in the introduction to this chapter proposals to 
remove subsidies in poor \fcountries have sometimes met sign\fificant protest 
from the person in th\fe street. From an analytical point of view there is no consensus of what consti-
tutes a low- income group. Sometimes the poorest fifth or tenth of house-
holds are in focus (Arze del Granado et al., 2012). Sometimes a measure 
of poverty is used, for example, ‘less than 60 per cent of median income’ 
(Dartanto, 2013). Household size and possibly other factors such as region 
and climate could be adjusted for. The final choice of who constitutes a 
low- income person depen\fds on national circumstances.
\bompensation based \ton isolated losses
Sometimes compensation based on the degree of loss generated by the 
reform itself is seen as fair. Consider a reform that inflicts heavy losses on 
a group that in a general sense is neither particularly wealthy nor poor. 
It is, in short, a fairly average group. If a reform leads to heavy losses for 
this group, and few or no losses for other groups, some compensation 
could be seen as fair. A prominent example is compensation to people laid 	
OOSTERHUIS 9781782\P545309 PRINT (M3374)\P (\b).indd 275	07/03/2014 09:57\P

276 Paying the polluter
off when, for example, subsidies to coal mining are ended. The issue of 
compensation to laid off miners has been heavily debated in Europe, with 
some arguing that compensation offered is not nearly enough, with others 
arguing that those laid off should seek support from the channels available 
in a welfare state to anybody that is laid off. Those arguing for compensa-
tion have not framed their argument in terms of the overall low income of 
the miners, but rather in terms of the heavy loss inflicted from the reform 
in isolation. The need for social stability and concerns for add- on effects 
on other businesses in cities and regions have also been important. A lesson 
from the UK coal subsidy reform is that measures to stimulate economic 
development, and therefore creating new job opportunities in areas where 
industrial activities are to be scaled down or closed, can increase support 
to the reform. Another example is that the provisions offered to former 
miners helped to make the closure of coal mines possible in France. There is no analytical consensus on when and in what circumstances 
the loss from the reform in isolation could be used as a valid argument for 
compensation. However, it is in our view important not to overuse this 
argument. Any reform will imply that some people will have to change 
behaviour, whether in the labour market or in their purchases. Indeed, such 
changes are part of the point.
\boncerns about policy processes
Compensation may be relevant for political purposes, for example, if and 
when some group has disproportional influence. Politically motivated com-
pensation is hardly an end in itself, but is rather a side- constraint on the 
reform of environmentally harmful subsidies. Such a side- constraint arises 
from the understandable fact that nobody likes to suffer from a reform, 
especially one that increases GDP and overall welfare. In this case com-
pensation is a way of spreading the gains, hopefully without compromising 
too much of the gain itself. Put differently, compensation can be justified 
to mitigate transitional costs and to secure political support for improved 
market efficiency in the lon\fger term.
12.3.3 Step 3: Identify How to \bompensate
Step 3 in designing a successful compensation policy is deciding how com-
pensation should be carried out, which is sometimes called a compensation 
scheme, or compensation mechanism. There is ample scope for compensat-
ing in a more direct way than through environmentally harmful subsidies. 
We stated above that in average across 16 energy subsidy reforms the 
poorest fifth of the population obtained 7 cents out of each dollar spent. 
Put another way it takes US$14 on the public budget to transfer US$1 to 	
OOSTERHUIS 9781782\P545309 PRINT (M3374)\P (\b).indd 276	07/03/2014 09:57\P

Reform of EHS: dist\lributional issues 
277
the poorest fifth. As alternative, Arze del Granado et al. (2012) offer the 
following example of a cash transfer: assume that 15 cents of each dollar 
is spent on administration, and that 80 per cent of the rest reaches the 
bottom fifth (20 per cent is assumed wasted on people posing as belonging 
to this group, and people who transitorily belong to the group). With this 
scheme it will take around US$1.5 to transfer US$1 to the poorest fifth. 
The example is hypothetical, but not unrepresentative of real cash transfer 
programmes.
There is waste in the cash transfer scheme, even visible waste in the form 
of administration, but clearly it is much less than the waste inherent in the 
energy subsidy. The key difference between the schemes is that the cash 
transfer scheme does no\ft transfer huge amounts \fto the wealthy. Compensation in cash is fairly easy to administer in developed econo-
mies that have compulsory social security registration and full income tax 
coverage. In developing countries it may be more difficult to administer. 
In Iran and Indonesia authorities decided to overcome the difficulties 
and according to the International Monetary Fund (2013) targeted cash 
transfers were used in nine out of 28 reform episodes. In other countries 
and situations, targeted compensation that resembles cash has been used 
instead (18 of 28 episodes according to the Fund). Summarizing from the 
case studies of developing countries above and further examples listed by 
the International Monetary Fund (20\f13), these include:	
●
●social spending on health, education, rural development and the 
social safety net (Mexico, Brazil, Ghana, Gabon, Niger, Nigeria, 
Mozambique, Indonesia, Malaysia, Philippines);
●
●low electricity tariffs on minimum consumption (Armenia, Brazil, 
Kenya, Uganda, Philippin\fes);
●
●support for electrification (Ghana, Kenya);
●
●support for public transport (Ghana, Malaysia, Philippines).
These examples indicate that developing countries can transform environ-
mentally harmful subsidies to comprehensive social protection packages 
efficiently targeting the low- income groups. Other examples are the sever -
ance packages and direct aid provided to Polish, UK and French miners 
leaving work in the restructuring of the coal sector mentioned \fabove. High- income countries offer other types of examples of direct income 
support. (1) ‘Diversified income taxation’ is a textbook example of a direct 
measure to prioritize low- income groups. For example, the Norwegian 
income taxation is progressive with a marginal tax of per cent for low- 
income groups and increasing to maximum 50 per cent. Differentiation 
of the pay- roll tax and energy taxes is used to avoid negative incentives to 	
OOSTERHUIS 9781782\P545309 PRINT (M3374)\P (\b).indd 277	07/03/2014 09:57\P

278 Paying the polluter
regional development from tax changes. The Dutch tax increase on natural 
gas was followed with reduced personal income taxes. Consequently, net 
wage levels rose, and the large majority of households saw an overall 
increase of net disposable income. (2) ‘Unemployment benefits’ – the 
labour market has significant elements of wage coordination securing 
minimum incomes. Severance packages can potentially be used to alleviate 
transitional cost for workers when adjusting to ne\fw jobs.
\bompensation to regions
Countries often have political targets for local and regional economies. In 
this situation lump- sum support can be distributed to the local municipali-
ties for their own distribution according to their preferences. The General 
Purpose Grant Scheme for Norwegian municipalities serves as an example 
(Ministry of Local Government and \begional Development, 2013). This 
system is meant to even out the differences between different localities. 
Municipalities with low incomes, high population growth and high level 
of decentralization are particularly emphasized. Based on own incomes 
and governmental support, local politicians can stimulate regional devel-
opment by prioritizing local infrastructure to attract new industries over 
municipal budgets. The government can also stimulate regional development over the gov-
ernmental budgets by investing directly in infrastructure, culture, decen-
tralization of governmental institutions, supporting health care and social 
support programmes.
12.4 CONCLUSIONS
A well- formulated environmentally harmful subsidy reform increases 
GDP. Potentially everybody could gain from the reform, although it is not 
always desirable to compensate all parties. The subsidies are formulated to 
support low- income groups, and compensation to these groups is usually 
necessary to facilitate the implementation of reform through the political 
process. In order to preserve the overall gains from reform the compensa-
tion itself should be as targeted and non- distorting as possible.
With these principles in mind we have outlined three steps of a suc-
cessful compensation scheme that can be embedded in a transition 
 management process. Step 1 is to analyse who gains from the pre- reform 
subsidies. In the case of fuel subsidies, for instance, there are low- income 
groups that benefit from cheap fuel. However, the largest gains are reserved 
for those who consume large quantities of fuel, namely the well- off. In the 
case of coal subsidies to production there were segments of the working 	
OOSTERHUIS 9781782\P545309 PRINT (M3374)\P (\b).indd 278	07/03/2014 09:57\P

Reform of EHS: dist\lributional issues 
279
population who had little alternative employment in the vicinity of their 
neighbourhoods. These groups would lose from a subsidy removal both in 
the labour market and in the housing market if the value of owned prop-
erty fell.
Step 2 in a successful compensation scheme is to determine who should 
be compensated. Low- income groups, and their supporters in politics, 
usually require firm confirmation that they will indeed receive sufficient 
compensation. By focusing compensation on low- income groups the 
reform of environmentally harmful subsidies may contribute to improved 
income distribution. By contrast, the rich and well- off are seldom a prior-
ity for compensation for reasons of distribution. Political realities often 
dictate support to organized interests who are not candidates for compen-
sation on pure equity grounds to help obtain buy- in to the reform. Any 
such support should be explicitly directed at the purpose of getting the 
reform through, and time limited to avoid new inefficient subsidies. A third 
group of recipients of compensation are laid- off workers. Inhabitants of 
affected regions may also require compensation, if a region stands to lose 
and it is deemed necessary to compensate the whole region. A general 
point of step 2 is that compensation should be carried out for an explicit 
reason – either distributional reasons, unfair losses or sometimes the politi-
cal realities of having well- organized interests.
Step 3 in a successful compensation scheme is to determine how com-
pensation should be carried out. Economic efficiency suggests that com-
pensation should be cash- based, similarly to programmes in Indonesia 
and Iran. Compensation in cash was also used in coal- producing Western 
countries, along with support to re- education and so on. Countries that 
have well- developed taxation system (and high taxes) may use the progres-
sivity of the income tax sch\fedule for redistribution.
NOTES
1. International Energy Association interactive database on fossil fuel subsidies, www.iea.
org/subsidy/index.html; accessed 20 \fJanuary 2014.
2. Main sources: IEEP et al. (2007); IEA et \fal. (2010b); Laan e\ft al. (2010).
\bEFE\bENCES
Arze del Granado, J., D. Coady and \b. Gillingham (2012), ‘The unequal ben-
efits of fuel subsidies: a review of evidence for developing countries’, World 
Development, 40(11), 2234–48.	
OOSTERHUIS 9781782\P545309 PRINT (M3374)\P (\b).indd 279	07/03/2014 09:57\P

280 Paying the polluter
Bacon, \b. and M. Kojima (2006), Coping with Higher Oil Prices, \beport No. 323/06, 
ESMAP, World Bank.
Bruvoll, A., J.M. Skjelvik and H. Vennemo (2011), Reforming Environmentally 
Harmful Subsidies, Nordic Council of Ministers TemaNord 551 report, 
accessed 21 January 2014 at http://www.norden.org/en/publications/publika 
tioner/2011–551.
Dartanto, T. (2013), ‘\beducing subsidies and the implication on fiscal balance and poverty in Indonesia: \fa simulation analysis’, Energy Policy, 58, 117–34.
Ellis, J. (2010), Untold Billions: Fossil- fuel Subsidies, their Impacts and the Path 
to Reform: The Effects of Fossil- fuel Subsidy Reform: \b Review of Modelling 
and Empirical Studies, Geneva: The Global Subsidies Initiative, International 
Institute for Sustainable Development.
G20 (2010), \bNNEX: G20 Initiative on Rationalizing and Phasing Out Inefficient 
Fossil Fuel Subsidies: Implementation Strategies & Timetables, accessed 21 
January 2014 at http://www.eenews.net/assets/2010/06/\f28/document_cw_03.p\fdf.
Hamid, J. (2008), ‘Malaysia PM faces bigger protest, dissent over fuel’, 12 June, \beuters.
IEA (International Energy Agency) (2012), World Energy Outlook 2012, Paris: 
International Energy Agency.
IEA (International Energy Agency) (2013), ‘\becent developments in energy subsi- dies’, World Energy Outlook \besources web page.
IEA, OECD and World Bank (2010a), The Scope of Fossil Fuel Subsidies in 2009 
and a Roadmap for Phasing Out Fossil- fuel Subsidies, joint report prepared for 
the G- 20 Summit, Seoul, November 2010.
IEA, OPEC, OECD and World Bank (2010b), \bnalysis of the Scope of Energy 
Subsidies and Suggestion for the G- 20 Initiative, joint report prepared for submis-
sion to the G- 20 Summit meeting T\foronto, June 2010.
IEA, OPEC, OECD and World Bank (2011), Joint \beport by IEA, OPEC, OECD and World Bank on Fossil- Fuel and Other Energy Subsidies: An Update of the 
G20 Pittsburgh and Toronto Commitments, accessed 21 January 2014 at http://
www.oecd.org/env/49090716.pdf.
IEEP, Ecologic, FEEM and IVM (2007), Reforming Environmentally Harmful 
Subsidies, final report to the European Commission’s DG Environment, March 
2007.
IMF (2013), Energy Subsidy Reform: Lessons and Implications, 28 January, 
Washington, DC: Inter\fnational Monetary Fund.
Laan, T., C. Beaton and B. Presta (2010), Untold Billions: Fossil- Fuel Subsidies, 
their Impacts and the Path to Reform: Strategies for Reforming Fossil- Fuel 
Subsidies: Practical Lessons from Ghana, France and Senegal, The Global 
Subsidies Initiative, International Institute for Sustainable Development.
Ministry of Local Government and \begional Development (2013), ‘The General Purpose Grant scheme’, accessed 21 January 2014 at http://www.regjeringen.
no/en/dep/kmd/subjec\fts/municipal- economy/the- general- purpose- grant- scheme.
html?id5540083.
World Nuclear Association (2013), ‘Energy subsidies and external costs’, accessed 21 January 2014 at http://world- nuclear.org/info/Economic- Aspects/Energy- 
Subsidies- and- External- Costs/#.Ua96V0CpXSg.\f	
OOSTERHUIS 9781782\P545309 PRINT (M3374)\P (\b).indd 280	07/03/2014 09:57\P

Parse pdf-doc by clicking the button below. Parsed pdf-documents will be searchable.

Metadata Found
CreationDate
2014-03-07T09:58:18+00:00
ModDate
2014-06-12T11:56:09+02:00
Producer
Adobe PDF Library 9.0
Pages
18