How much would the Kyoto Protocol cost to consumers?
Erik Dietzenbacher* and Mònica Serrano**
*
Faculty of Economics and Business, University of Groningen, The Netherlands
Department of Economic Theory, University of Barcelona, Spain
**
PRELIMINARY VERSION (April 2012)
Please, do not quote.
Abstract:
In recent decades there has been a growing interest in the consequences of
globalization for environmental problems, especially for those generated by
greenhouse gas emissions. This concern has been intensified in the last years since
the five-year accounting period (2008-2012) of the Kyoto protocol is near to be
accomplished. The numerous published papers related to this topic reveal that little
attention has been paid to the macro-economic effects of this protocol. This paper
starts from the assumption that industries incur extra costs per ton of emitted
pollutants (e.g. CO2 abatement costs, emission allowances in the EU Emissions
Trading Scheme, environmental taxes, opportunity costs), which are necessary to
reduce the present amount of pollution. It is further assumed that all producers fully
pass these costs on to the buyers of their products. As a consequence, this will affect
the value of consumption. This ‘extra’ cost to consumers can be viewed as a change
in the consumption price index. The aim of this paper is to give some insights in the
extra costs for consumers if Annex B countries had fulfilled the targets of the Kyoto
protocol. Using the World Input-Output Tables of the WIOD project, this paper
analyses different scenarios. The main results show that, in general, consumers might
well be able to bear the economic costs of the Kyoto protocol.
Keywords: Price input-output model, World Input-Output Tables, cost of pollution
1
1. Introduction
In recent decades there has been a growing interest in the consequences of
globalization for environmental problems, especially for those generated by
greenhouse gas (GHG) emissions. This concern has been intensified in the last years
since the five-year accounting period (2008-2012) of the Kyoto protocol is near to be
accomplished.
Studies concerned with the analysis of the broader economic repercussions of
adopting the Kyoto protocol are usually based on modelling the interaction between
the environment, energy sectors, and the rest of the economy. Basically there are two
broad approaches: the so-called bottom-up and top-down models. The bottom-up
models accentuate the analysis on a detailed, technologically-based treatment of the
energy system; whereas the top-down models stress a theoretically consistent
description of the general economy. Regardless their differences, both mainly
coincide with respect the emphasis on the energy system as the sector where all
policy instruments or technology changes are applied. Weyant (1999) offers an
interesting and comprehensive report on a comparative set of modelling analyses of
the economic and energy sector impacts of the Kyoto protocol. And in-depth studies
for some countries can be found in ICCF (2005).
However, the numerous published papers related to environmental emissions
monitoring by the Kyoto protocol using input-output approach reveal that little
attention has been paid to the macro-economic effects of this protocol within this
analytical framework. In this paper we would like to have some insights about what
would have been the cost for final consumers (i.e. households) if Annex B countries
had fulfilled their Kyoto emission targets. We assume that all industries, not only the
energy sectors, incur extra cost per ton of emitted pollutants in order to reduce the
present amount of pollution employing the current technology. These costs can be
materialized differently; for instance, in form of CO2 abatement costs, emission
allowances in the European Union (EU) Emissions Trading Scheme (ETS),
environmental taxes, or economic opportunity costs. We further assume that the costs
of emissions will pass along to consumers in the form of higher prices for all goods
2
and services. The consumers’ purchasing power will reduced by the higher cost of
the consumption. Ultimately, this “extra” cost to consumers can be viewed as a
change in the value of the same purchased bundle. In this paper, various scenarios
are analysed using an input-output price model and the World Input-Output Database
(WIOD).
The remainder of the paper is as follows. Section 2 provides the methodology
applied in this study. In Section 3 we describe the database used and all the
assumptions we have made in order to accommodate the available data to our model.
In Section 4 we analyse the results for different scenarios. They show that in the
most likely situation, i.e. if all Annex B countries would have ratified the protocol
and fulfilled their respective binding emission ceilings, consumers might well be able
to bear the economic costs of the Kyoto protocol. However, the important role that
China could play would offer different outcomes. Finally, conclusions are presented
in Section 5.
2. Methodology
The analytical framework used in this study is the price input-output model based on
monetary input-output table as presented in Chapter 2 of Miller and Blair (2009).
The standard input-output approach assumes that each industry produces a single
good by means of combination of intermediate inputs and primary factors in fixed
proportions. It also assumes that all production processes operate under constant
returns to scale and, in consequence, the economic benefits of each sector are equal
to zero. Hence, the value of output of any sector j must be equal to the value of its
inputs. That is, for sector j the value of its production (or sales) x j should equal the
cost of production given by the cost of intermediate inputs zij plus the cost of
primary inputs μ g . So, let p j be the price per unit of sector’s j output, and π g be
the exogenous unit price of primary input g (for instance, wage per person-year,
rents of land per hectare and per year, or return on capital per year), we can write the
3
following accounting equation, which states that for each sector its receipt equals its
outlays:1
m
p j x j = p1 z1 j + p2 z2 j + " + pn znj + ∑ π g μ gj
(1)
g =1
Dividing this expression by x j gives:
m
p j = p1a1 j + p2 a2 j + " + pn anj + ∑ π g lgj
(2)
g =1
Where aij and lgj are the intermediate and primary input coefficients, respectively.
This expression shows the price of one unit of output j as the unit cost of production,
that is, it expresses the price of one physical unit of product j equals the cost
involved in producing that unit. In matrix notation, the set of equations yields:
p' = p'A + v'
(3)
Where v = πL is a vector of primary cost per unit of output (the elements of this
vector represent the total monetary value of all primary inputs required for unit of
sector j ’s output).
Expression (3) leads to:
p ' = v '(I − A )-1 = v ' L
(4)
Where L = (I − A)-1 is the Leontief matrix. This expression indicates that product
prices are proportional to the cost of primary inputs. If the data are obtained from a
standard input-output table, the price of each product will be equal to 1. That is, the
base year index price will be represented by the summation vector consisting of 1’s
i ' . This fact reflects that there is a unique unit of measurement, i.e. the amount that
1
Matrices are indicated by bold, upright capital letters; vectors by bold, upright lower case letters; and
scalars by italicized lower case letters. Vectors are columns by definition, so that row vectors are
obtained by transposition, indicated by a prime. A diagonal matrix with the elements of any vector on
its main diagonal and all other entries equal to zero is indicated by a circumflex.
4
can be purchased for 1 monetary unit (let say American dollars $, Euros €, Japanese
yens ¥, pound sterling £, etc.).
The price model presented above is generally used to measure the impact on prices
throughout the economy of a change in the cost of primary inputs in one or more
sectors.2 However, this price model can also be applied to analyse the impact of new
costs. Imagine the situation that industries must incur extra costs per ton of emitted
pollutants (e.g. CO2 abatement costs, emission allowances in Emissions Trading
Scheme, environmental taxes, opportunity cost, etc.), which are necessary to reduce
the present amount of pollution. In such a case, the new price of one unit of output j
will be:
p ' = p ' A + v '+ d '
(5)
Where d is a vector of extra cost per unit of output. The impact on prices of this
extra cost will be determine by the difference between initial prices p and new
prices p , such that Δp ' = p '− p ' = (Δp) ' A + d ' . Solving this expression the price
increase will be:
Δp ' = d ' L
(6)
In this study, the extra cost has been computed as d = τ e , where , τ is the
environmental cost and e is the vector of extra emission coefficients. In that case,
however, the emissions considered to calculate coefficients in e are those emissions
that had exceed the target fixed by the Kyoto protocol ( w ) to each country, such
−1
that: e = w ' x .
The input-output price model defined above are “long run” or “supply” prices
(Bulmer-Thomas, 1982). That is, the prices of products are defined without
considering the interaction of final demand from households or final consumers.
Moreover, in this model it is further assumed that all producers fully pass their costs
on to the buyers of their products. Hence, in the cost-push input-output price model
2
This price model is called cost-push input-output price model. See Oosterhaven (1996) and
Dietzenbacher (1997).
5
is the consumer (the final buyer) who fully bears the extra cost as a last resort.
Consumers will face higher prices for all goods and services and their consumers’
purchasing power would be reduced. This “extra” cost to consumers can be measured
as a change in the value of consumption defining, for instance, a Laspayres type
price index such as:
n
PI L =
∑ ip c
i =1
n
i i
∑pc
i =1
(7)
i i
Where ci is quantity of good i purchased by consumers in the base year, pi is the
initial price of good i in the base year (i.e. the price without the extra cost), and ip i is
the new price of good i including the extra cost. This Laspayres price index
compares the value of the bundle of base year using new prices with the value of the
same bundle using the base year or initial prices. So, a Laspayres index of 1 would
state that the consumer can afford to buy the same bundle of goods as he consumed
before including the extra cost, given that income has not changed.
3. Database and assumptions
In this study we basically used the information offered by the World Input-Output
Database (WIOD).3 This is database includes a worldwide time series of national
input-output tables for 40 countries covering at least 80% of world GDP and 35
sectors. These national tables are fully linked through bilateral trade data, generating
a full multi-country input-output table. The time series covers the period 1995-2009
and contains tables in current and constant international prices. In addition, the
WIOD also includes socio-economic and environmental satellite accounts.
Specifically, we used the so-called World Input-Output Table (WIOT) analytical at
current prices and data on the three main greenhouse gases (GHG) monitoring by the
Kyoto protocol (i.e. CO2, N20 and CH4) by sector expressed in kilotons (kton) of CO2
3
The current version of this paper refers to the last release 22 December 2011
6
equivalent.4 These three GHG have been aggregated in accordance with the global
warming potential (GWP100) of each gas as established by the IPCC (1997). These
conversion factors are: 1 for CO2, 21 for CH4, and 310 for N2O.
Since the aim of this study is to know what would have happened if ratifier countries
of the Kyoto protocol had paid a cost for the emissions that had exceed the own
target,5 we need to quantify the amount of these extra emissions for each country and
to determine the corresponding cost. On one hand, due to the WIOD availability we
take 1995 emissions as an approximation of the Kyoto base year 1990 for all
countries. Also due to WIOD characteristics we take 2006 as the hypothetical year of
actuation. Hence, if the country already fulfilled their Kyoto target in 2006 or the
country is a non-Annex B country, the cost will be 0. Otherwise, we will apply a cost
on the extra emission coefficients. These coefficients are calculated as the difference
between emissions in 2006 and emission target in 2012 divided by 2006 output.
On the other hand, we determine the hypothetical cost that countries would have paid
for its extra emissions. In that case, we analyse three different scenarios. In scenario
1, we establish is a kind of minimum cost of 0.02634 million US$ per kton of CO2.
This amount is the result of considering the price of one European Union Allowance
(EUA) in the European Union Emissions Trading Scheme (EU-ETS) as an
approximation to the “cost of the CO2 emissions”. In that case we take the mean of
monthly prices for 2006 from the December 2008 series according to the study
reported by Ellerman and Joskow (2008). This cost agrees with the other studies that
also consider GHG abatement cost (IEA, 2009; McKinsey & Company, 2009; Russ
et al., 2009). Scenario 2 considers a kind of maximum cost of 0.1 million US$ per
kton of CO2. This information provides from the IEA (2009) report, which states that
almost all the emissions reductions come at cost below $100 per tonne of CO2 in the
period to 2030. Finally, a third scenario has been computed. In that case, we took
into account the emission reduction actions that are possible with technologies that
either are available today or offer a high degree of certainly about their potential in a
4
Only emissions from sectors are considered; so, household emissions have been discarded.
The different targets are presented in the Appendix of this paper. For more information about the
Kyoto protocol and each country target see United Nations (1997), and also the website from the
Framework Convention on Climate Change http://unfccc.int/kyoto_protocol/items/2830.php.
5
7
2030 time horizon. The cost of reducing 1 kton of CO2 would be 0.07527 million
US$.
For each scenario we have considered various assumptions. The first assumption (I)
reflects the present context, that is, non-Annex B countries have not any target. The
second assumption (II) tries to reproduce a hypothetical situation in which China had
the same target as USA, i.e. it should have reduced their emissions by 93 percent of
the base year emissions. Finally, a last assumption (III) has also been considered in
which all non-Annex B countries had the same target as USA (i.e. they should have
reduced their emissions by 93 percent of the base year emissions). In all scenarios we
assumed USA ratified the Kyoto protocol on February 2005.
4. Results
In this section we present some preliminary results obtained from the last three
scenarios taking into account the different assumptions (see Figure 1 for a summary).
It is important to bear in mind that the information is not the final version and
therefore some odd results have to be carefully checked.
Figure 1: Summary of the scenarios and assumptions considered
SCENARIO 1: Cost 0,02634 SCENARIO 2: Cost 0,1 SCENARIO 3: Cost 0,07527 million US$ per kton of CO2 million US$ per kton of CO2 million US$ per kton of CO2 Ass. I: Kyoto Targets only for Annex B countries Ass. II: China has the same target as USA (i.e. 93) Ass. III: All non‐Annex B countries have the same target as USA (i.e. 93) Table 1 shows the consumption index for each country under each scenario and
assumption. The results of this tables indicates that regardless the cost considered
under the assumption I (i.e. the current situation) the consumers might well be able to
bear the economic costs of the Kyoto protocol. The average consumer index would
be 1.0013 under Scenario 1.I, 1.0050 under Scenario 2.I, and 1.0038 under Scenario
3.I. In all cases the maximum consumer index would correspond to Bulgaria (1.0047,
1.0178, and 1.0134) and the minimum to India (1.0001, 1.0004, and 1.0003,
respectively).
8
If we compare Scenario 1.I with Scenario 1:II we see that when the cost is almost 4
times higher, the maximum “inflation” also but the maximum “inflation” also is
approximately 4 times as high (in that case correspond to Spain). But when the cost
is multiplied by 4 the average “inflation” only doubles (2.45056). The minimum
consumption index under scenario 1.II (which corresponds to Brazil) only increases
about 1.6 percent. These results are quite reasonable; however, comparing
assumption I, II and III under scenario 2 we find some odd results. Specifically, some
of the consumption indexes would increase dramatically (especially under 2.II). This
is the case of Spain (13.02502), Estonia (13.01886) and Latvia (13.01886). If China
would agree to have the same target as USA has the country who would bear the
lowest “inflation” would be Brazil, but if this target would be accepted by all nonAnnex B countries the surprising result is that this position would be occupied by
USA.
All these results have to be taken with the maximum caution since the calculus have
not been doing with the latest version of the WIOD. However, they are a sample of
the potentially of this database.
Table 1: Consumption Indexes
COUNTRY AUS AUT BEL BGR BRA CAN CHN CYP CZE DEU DNK ESP EST FIN FRA GBR GRC SCENARIO 1 Ass. I Ass. II Ass. III 1,00197 1,00221 1,00271 1,00468 1,00010 1,00188 1,00018 1,00028 1,00128 1,00218 1,00260 1,00280 1,00089 1,00147 1,00058 1,00153 1,00465 1,25682 1,15722 1,15331 1,09067 1,01591 1,02448 1,06635 1,07676 1,16256 1,08588 1,64464 4,16788 4,16626 1,13526 1,09989 1,35208 1,08236 1,25727
1,15756
1,15393
1,09155
1,03355
1,02481
1,06706
1,07712
1,16298
1,08627
1,64505
4,16825
4,16670
1,13561
1,10018
1,35242
1,08268
SCENARIO 2 Ass. I Ass. II Ass. III SCENARIO 3 Ass. I Ass. II Ass. III 1,00748 1,97485 1,97656 1,00563 1,73376 1,73505
1,00838 1,59680 1,59810 1,00631 1,44921 1,45018
1,01030 1,58194 1,58432 1,00776 1,43802 1,43981
1,01776 1,34418 1,34751 1,01337 1,25906 1,26157
1,00039 1,06041 1,12735 1,00029 1,04547 1,09585
1,00713 1,09294 1,09418 1,00537 1,06995 1,07089
1,00068 1,25187 1,25455 1,00051 1,18958 1,19160
1,00107 1,29136 1,29272 1,00080 1,21931 1,22033
1,00484 1,61708 1,61864 1,00365 1,46447 1,46565
1,00829 1,32599 1,32747 1,00624 1,24537 1,24648
1,00986 3,44701 3,44857 1,00742 2,84184 2,84301
1,01064 13,02502 13,02644 1,00801 10,05109 10,05215
1,00336 13,01886 13,02055 1,00253 10,04645 10,04772
1,00559 1,51343 1,51477 1,00421 1,38645 1,38746
1,00220 1,37919 1,38026 1,00166 1,28541 1,28622
1,00580 2,33648 2,33777 1,00436 2,00595 2,00693
1,01765 1,31265 1,31384 1,01328 1,23533 1,23622
9
HUN IDN IND IRL ITA JPN KOR LTU LUX LVA MEX MLT NLD POL PRT ROM RUS SVK SVN SWE TUR TWN USA RoW 1,00183 1,00023 1,00009 1,00046 1,00109 1,00068 1,00021 1,00272 1,00125 1,00073 1,00031 1,00057 1,00084 1,00073 1,00213 1,00039 1,00238 1,00060 1,00232 1,00063 1,00023 1,00030 1,00136 1,00046 1,12985 1,02722 1,01650 1,10707 1,08954 1,03039 3,14605 1,34718 1,08161 4,12680 1,02194 1,14549 1,42795 1,19727 1,39309 1,08332 2,98788 2,31642 1,11006 2,10055 1,05655 1,03832 1,01640 1,07978 1,13023
1,05104
1,03654
1,10745
1,08985
1,03069
3,15825
1,34747
1,08201
4,12716
1,02761
1,14606
1,42861
1,19763
1,39360
1,08385
2,98848
2,31693
1,11095
2,10089
1,06164
1,04468
1,01668
1,08061
1,00695 1,49292 1,49433 1,00523 1,00088 1,10332 1,19375 1,00066 1,00036 1,06264 1,13870 1,00027 1,00174 1,40643 1,40789 1,00131 1,00416 1,33988 1,34108 1,00313 1,00258 1,11537 1,11650 1,00194 1,00080 9,14622 9,19255 1,00060 1,01033 2,31788 2,31895 1,00778 1,00474 1,30979 1,31129 1,00357 1,00277 12,86907 12,87044 1,00209 1,00116 1,08328 1,10479 1,00088 1,00215 1,55226 1,55444 1,00162 1,00318 2,62447 2,62697 1,00240 1,00277 1,74882 1,75017 1,00209 1,00810 2,49214 2,49407 1,00610 1,00149 1,31628 1,31827 1,00112 1,00902 8,54584 8,54809 1,00679 1,00227 5,99704 5,99894 1,00171 1,00881 1,41780 1,42116 1,00663 1,00240 5,17758 5,17890 1,00181 1,00088 1,21465 1,23399 1,00066 1,00115 1,14545 1,16959 1,00086 1,00515 1,06227 1,06332 1,00388 1,00175 1,30283 1,30599 1,00131 1,37101 1,07776 1,04715 1,30591 1,25582 1,08684 7,13156 1,99195 1,23318 9,93371 1,06269 1,41568 2,22272 1,56363 2,12312 1,23806 6,67967 4,76121 1,31447 4,14442 1,16156 1,10948 1,04687 1,22793 1,37207
1,14583
1,10440
1,30701
1,25672
1,08769
7,16643
1,99276
1,23430
9,93474
1,07888
1,41732
2,22460
1,56465
2,12457
1,23956
6,68136
4,76264
1,31700
4,14540
1,17612
1,12765
1,04766
1,23032
Source: Own calculations. Note: For the meaning of the countries acronyms see Table A of the Appendix section. 5. Conclusions
This paper starts from the assumption that industries incur extra costs per ton of
emitted pollutants (e.g. CO2 abatement costs, emission allowances in the EU
Emissions Trading Scheme, environmental taxes, opportunity costs), which are
necessary to reduce the present amount of pollution. It is further assumed that all
producers fully pass these costs on to the buyers of their products. As a consequence,
this will affect the value of consumption. This ‘extra’ cost to consumers can be
viewed as a change in the consumption price index. The aim of this paper is to give
some insights in the extra costs for consumers if Annex B countries had fulfilled the
targets of the Kyoto protocol. Using the World Input-Output Tables of the WIOD
project, this paper analyses different scenarios. The main results show that, in
10
general, consumers might well be able to bear the economic costs of the Kyoto
protocol.
In this paper we analyse what would have been the cost for final consumers if Annex
B countries had fulfilled their Kyoto emission targets. The preliminary results show
that in the most likely situation, i.e. if all Annex B countries would have ratified the
protocol and fulfilled their respective binding emission ceilings, consumers might
well be able to bear the economic costs of the Kyoto protocol. However, different
results have been obtained from other hypothetical situations exhibiting the
important role played by China as it has been reflected in the strained negotiations.
Notwithstanding, these outcomes have to be taken with extremely caution and
always as an approximation of what would have been the real situation. On one hand,
the calculus has been carried out using a preliminary version of the WIOD. And on
the other hand, due to the availability and characteristics of this database, neither the
base year (1995) nor the year of action (2006) coincide with the reality (1990 and
2012, respectively). This fact would be relevant for the latter since the economic
crisis started in 2009 have had important influenced on the emission levels.
References
Bulmer-Thomas, V. (1982) Input-Output Analysis in Developing countries: sources,
methods and applications, Chichester [etc.]: John Wiley & Sons.
Dietzenbacher, E. (1997) “In Vindication of the Ghosh Model: A Reinterpretation as
a Price Model”, Journal of Regional Science, 37, 629-651.
Ellerman, A.D., P.L Joskow (2008) “The European Union’s Emissions Trading
System in perspective”, document prepared for the Pew Center on Glogal
Climate Change, Cambridge, US: Massachusetts Institute of Technology,
available at http://www.pewclimate.org/docUploads/EU-ETS-In-PerspectiveReport.pdf
11
ICCF, The International Council for Capital Formation (2005) Reports on the
Economic Cost of the Kyoto Protocol to Italy, Spain and the UK, Brussels:
ICCF, available at http://www.iccfglobal.org/research/climate/index.html
IEA, International Energy Agency (2009) “World Energy Outlook 2009”, Paris:
OCDE, available at http://www.worldenergyoutlook.org/
IPCC (1997) Revised 1996 IPCC Guidelines for national greenhouse gas inventories,
3 volumes, London: Intergovernmental Panel on Climate Change, available at
http://www.ipcc-nggip.iges.or.jp/
McKinsey & Company (2009) “Pathways to a Low-Carbon Economy. Version 2 of
the Global Greenhouse Gas Abatement Cost Curve”, McKinsey Solutions,
available at https://solutions.mckinsey.com/ClimateDesk/default.aspx
Miller, R.E., P.D. Blair (2009) Input-Output Analysis. Foundations and Extensions.
2nd edition. Cambridge [etc]: Cambridge University Press.
Oosterhaven, J. (1996) “Leontief versus Ghoshian Price and Quantity Model”,
Southern Economic Journal, 62, 750-759.
Russ, P., J.C. Ciscar, B. Saveyn, A. Soria, L. Szábó, T. van Ierland, D. van
Regemorter, R. Virdis (2009) “Economic Assessment of Post-2012 Global
Climate Policies. Analysis of Greenhouse Gas Emission Reduction Scenarios
with the POLES and GEM-E3 models”, EUR 23768 EN-2099, Insitute for
Prospective
Technological
Studies,
Joint
Research
Centre,
European
Commission, Luxembourg: Office for Official Publications of the European
Communities.
United Nations (1997) Kyoto Protocol to the United Nations Framework Convention
on Climate Change, 10 December 1997, (FCCC/CP/L.7/Add.1).
Weyant, J. (Eds) (1999) “The Costs of the Kyoto Protocol: A Multi-Model
Evaluation”, The Energy Journal, Special Issue, May 1999, 390 pp.
12
Appendix
Table A: List of countries and corresponding Kyoto targets
AUS AUT BEL BGR BRA CAN CHN CYP CZE DEU DNK ESP EST FIN FRA GBR GRC HUN IDN IND IRL ITA JPN KOR LTU LUX LVA MEX MLT NLD POL PRT ROM RUS SVK SVN SWE TUR TWN COUNTRY Australia Austria Belgium Bulgaria Brazil Canada China Cyprus Czech Republic Germany Denmark Spain Estonia Finland France United Kingdom Greece Hungary Indonesia India Ireland Italy Japan South Korea Lithuania Luxembourg Latvia Mexico Malta Netherlands Poland Portugal Romania Russia Slovak Republic Slovenia Sweden Turkey Taiwan TARGET 108 87 92.5 92 ‐‐ 94 ‐‐ ‐‐ 92 79 79 115 92 97.4 98.1 87.5 125 94 ‐‐ ‐‐ 113 93.5 94 ‐‐ 92 72 92 ‐‐ ‐‐ 94 94 127 92 100 92 92 104 ‐‐ ‐‐ NOTES (1) (1) (2) (3) (3) (2) (2) (1) (1) (1) (2) (1) (1) (1) (1) (2) (3) (3) (1) (1) (3) (2) (1) (2) (3) (2) (1) (2) (1) (2) (2) (2) (1) (3) 13
USA RoW United States 93 Rest of the World ‐‐ Not ratified in 2005 (3) Source: United Nations, Framework Convention on Climate Change. Information available at http://unfccc.int/kyoto_protocol/items/2830.php Notes: (1) European Union 15: they have a global target of 92. (2) European Union: they enter in the EU later than 2004 and 2007; they have their own Kyoto targets. (3) Non‐Annex B countries. 14