DEMOCRACY_INDEX

THE WORLD IN 2OO7

Democracy index

The Economist Intelligence Unit’s

index of democracy

By Laza Kekic, director, country forecasting services, Economist Intelligence Unit

Defining and measuring democracy

There is no consensus on how to measure democracy,

deﬁ nitions of democracy are contested and there is an

ongoing lively debate on the subject. The issue is not

only of academic interest. For example, although de-

mocracy-promotion is high on the list of American

foreign-policy priorities, there is no consensus within

the American government on what constitutes a de-

mocracy. As one observer recently put it, “the world’s

only superpower is rhetorically and militarily promot-

ing a political system that remains undeﬁ ned—and it

is staking its credibility and treasure on that pursuit”

(Horowitz, 2006, p 114).

Although the terms “freedom” and “democracy”

are often used interchangeably, the two are not syn-

onymous. Democracy can be seen as a set of practices

and principles that institutionalise and thus ultimately

protect freedom. Even if a consensus on precise deﬁ ni-

tions has proved elusive, most observers today would

agree that, at a minimum, the fundamental features of a

democracy include government based on majority rule

and the consent of the governed, the existence of free

and fair elections, the protection of minorities and re-

spect for basic human rights. Democracy presupposes

equality before the law, due process and political plural-

ism. Is reference to these basic features sufﬁ cient for a

satisfactory concept of democracy? As discussed below,

there is a question of how far the deﬁ nition may need

to be widened.

Some insist that democracy is necessarily a dichoto-

mous concept—a state is either democratic or not. But

most measures now appear to adhere to a continuous

concept, with the possibility of varying degrees of de-

mocracy. At present, the best-known measure is pro-

duced by the US-based Freedom House organisation.

The average of its indexes, on a 1 to 7 scale, of political

freedom (based on 10 indicators) and of civil liberties

(based on 15 indicators) is often taken to be a measure

of democracy.

The index is available for all countries, and stretches

back to the early 1970s. It has been used heavily in em-

pirical investigations of the relationship between de-

mocracy and various economic and social variables. The

so-called Polity Project provides, for a smaller number

of countries, measures of democracy and regime types,

based on rather minimalist deﬁ nitions, stretching back

to the 19th century.

Freedom House also measures a narrower concept,

that of “electoral democracy”. Democracies in this mini-

mal sense share at least one common, essential charac-

teristic. Positions of political power are ﬁ lled through

regular, free, and fair elections between competing par-

ties, and it is possible for an incumbent government

to be turned out of ofﬁ ce through elections. Freedom

House criteria for an electoral democracy include:

1. A competitive, multiparty political system.

2. Universal adult suffrage.

3. Regularly contested elections conducted on the basis

of secret ballots, reasonable ballot security and the

absence of massive voter fraud.

4. Signiﬁ cant public access of major political parties to

the electorate through the media and through gener-

ally open campaigning.

The Freedom House deﬁ nition of political freedom is

somewhat (though not much) more demanding than

its criteria for electoral democracy—ie, it classiﬁ es more

countries as electoral democracies than as “free” (some

“partly free” countries are also categorised as electoral

democracies). At the end of 2005, 122 states were clas-

siﬁ ed as electoral democracies; of these, 89 states were

classiﬁ ed as free. The Freedom House political-freedom

measure covers the electoral process and political plu-

ralism and, to a lesser extent, the functioning of govern-

ment and a few aspects of participation.

A key difference in the various measures of democ-

racy is between “thin” or minimalist ones and “thick”

or wider concepts (Coppedge, 2005). The thin concepts

correspond closely to an immensely inﬂ uential aca-

demic deﬁ nition of democracy, that of Robert Dahl’s

concept of polyarchy (Dahl, 1970). Polyarchy has eight

components, or institutional requirements: almost all

adult citizens have the right to vote; almost all adult citi-

zens are eligible for public ofﬁ ce; political leaders have

the right to compete for votes; elections are free and fair;

all citizens are free to form and join political parties and

other organisations; all citizens are free to express them-

selves on all political issues; diverse sources of informa-

tion about politics exist and are protected by law; and

government policies depend on votes and other expres-

sions of preference.

The Freedom House electoral democracy measure

is a thin concept. Its measure of democracy based on

political rights and civil liberties is thicker than the

measure of electoral democracy. Other deﬁ nitions of

democracy have broadened to include aspects of society

and political culture in democratic societies.

Democracy index

THE WORLD IN 2OO7

The Economist Intelligence Unit’s measure of

democracy

The Economist Intelligence Unit’s index is based on the

view that measures of democracy that reﬂ ect the state

of political freedoms and civil liberties are not thick

enough. They do not encompass sufﬁ ciently or at all

some features that determine how substantive democ-

racy is or its quality. Freedom is an essential component

of democracy, but not sufﬁ cient. In existing measures,

the elements of political participation and functioning

of government are taken into account only in a marginal

way.

The Economist Intelligence Unit’s democracy index

is based on ﬁ ve categories: electoral process and plu-

ralism; civil liberties; the functioning of government;

political participation; and political culture. The ﬁ ve

categories are interrelated and form a coherent con-

ceptual whole. The condition of having free and fair

competitive elections, and satisfying related aspects of

political freedom, is clearly the basic requirement of all

deﬁ nitions.

All modern deﬁ nitions, except the most minimalist,

also consider civil liberties to be a vital component of

what is often called “liberal democracy”. The principle of

the protection of basic human rights is widely accepted.

It is embodied in constitutions throughout the world as

well as in the UN Charter and international agreements

such as the Helsinki Final Act. Basic human rights in-

clude freedom of speech, expression and the press; free-

dom of religion; freedom of assembly and association;

and the right to due judicial process. All democracies are

systems in which citizens freely make political decisions

by majority rule. But rule by the majority is not neces-

sarily democratic. In a democracy majority rule must be

combined with guarantees of individual human rights

and the rights of minorities.

Most measures also include aspects of the minimum

quality of functioning of government. If democratically

based decisions cannot or are not implemented then the

concept of democracy is not very meaningful or it be-

comes an empty shell.

Democracy is more than the sum of its institu-

tions. A democratic political culture is also crucial for

the legitimacy, smooth functioning and ultimately the

sustainability of democracy. A culture of passivity and

apathy, an obedient and docile citizenry, are not consist-

ent with democracy. The electoral process periodically

divides the population into winners and losers. A suc-

cessful democratic political culture implies that the los-

ing parties and their supporters accept the judgment of

the voters, and allow for the peaceful transfer of power.

Participation is also a necessary component, as

apathy and abstention are inimical to democracy. Even

measures that focus predominantly on the processes of

representative, liberal democracy include (although in-

adequately or insufﬁ ciently) some aspects of participa-

tion. In a democracy, government is only one element in

a social fabric of many and varied institutions, political

organisations and associations. Citizens cannot be re-

quired to take part in the political process, and they are

free to express their dissatisfaction by not participating.

However, a healthy democracy requires the active, freely

chosen participation of citizens in public life. Democ-

racies ﬂ ourish when citizens are willing to take part in

public debate, elect representatives and join political

parties. Without this broad, sustaining participation,

democracy begins to wither and become the preserve of

small, select groups.

At the same time, even our thicker, more inclusive

and wider measure of democracy does not include

other aspects—which some authors argue are also cru-

cial components of democracy—such as levels of eco-

nomic and social wellbeing. Thus our index respects the

dominant tradition that holds that a variety of social

and economic outcomes can be consistent with political

democracy.

The Economist Intelligence Unit’s index provides a

snapshot of the current state of democracy worldwide

for 165 independent states and two territories. This cov-

ers almost the entire population of the world and the

vast majority of the world’s 192 independent states (27

micro-states are excluded).

Several things stand out. Although almost half of the

world’s countries can be considered to be democracies,

the number of “full democracies” is relatively low (only

28). Almost twice as many (54) are rated as “ﬂ awed de-

mocracies”. Of the remaining 85 states, 55 are authori-

tarian and 30 are considered to be “hybrid regimes”. As

could be expected, the developed OECD countries (with

the notable exception of Italy) dominate among full de-

mocracies, although there are two Latin American, two

central European and one African country, which means

that the level of development is not a binding constraint.

Only one Asian country, Japan, makes the grade.

More than half of the world’s population lives in a

democracy of some sort, although only some 13% reside

in full democracies. Despite the advances in democracy

in recent decades, almost 40% of the world’s population

still lives under authoritarian rule (with a large share of

these being, of course, in China). Given the most recent

trends, that are tantamount to a retreat from democ-

racy as discussed in our article in The World in 2007, it is

unlikely that this proportion will decrease signiﬁ cantly

soon. On our ten-country watchlist for likely signiﬁ cant

changes in 2007 (see box below) only one country is on

positive watch and nine are on negative watch.

The relationship between the level of development

(income per head) and democracy is not clear-cut.

There is an apparent association, although even in the

full democracy category there are a few that are not rich

OECD countries. The simple correlation between our

democracy index and GDP per head ($ at PPP) in 2006

is 0.6. This may look surprisingly low—it implies that in

a simple two-variable regression of the democracy index

on income per head, less than 40% of the inter-country

variation in democracy is explained by income levels. If

we also control for oil wealth (with a so-called dummy

variable that takes a value of 1 for major oil exporting

countries and 0 otherwise), the explanatory power of the

regression rises sharply to almost two-thirds of the inter-

country variation in the democracy index. Although this

still leaves more than one-third of the variation unex-

plained, it illustrates the often-observed strong negative

impact on democratic development of a reliance on oil.

However, the direction of causality between de-

mocracy and income is debatable. The standard mod-

ernisation hypothesis that economic development leads

to—and is a necessary pre-condition for—democracy,

THE WORLD IN 2OO7

Democracy index

Table 1

Economist Intelligence Unit democracy index 2006

Category scores

Overall I Electoral process II Functioning III Political IV Political V Civil

Rank score and pluralism of government participation culture liberties

Full democracies

Sweden 1 9.88 10.00 10.00 10.00 9.38 10.00

Iceland 2 9.71 10.00 9.64 8.89 10.00 10.00

Netherlands 3 9.66 9.58 9.29 9.44 10.00 10.00

Norway 4 9.55 10.00 9.64 10.00 8.13 10.00

Denmark 5 9.52 10.00 9.64 8.89 9.38 9.71

Finland 6 9.25 10.00 10.00 7.78 8.75 9.71

Luxembourg 7 9.10 10.00 9.29 7.78 8.75 9.71

Australia 8 9.09 10.00 8.93 7.78 8.75 10.00

Canada 9 9.07 9.17 9.64 7.78 8.75 10.00

Switzerland 10 9.02 9.58 9.29 7.78 8.75 9.71

Ireland 11= 9.01 9.58 8.93 7.78 8.75 10.00

New Zealand 11= 9.01 10.00 8.57 8.33 8.13 10.00

Germany 13 8.82 9.58 8.57 7.78 8.75 9.41

Austria 14 8.69 9.58 8.21 7.78 8.75 9.12

Malta 15 8.39 9.17 8.21 6.11 8.75 9.71

Spain 16 8.34 9.58 7.86 6.11 8.75 9.41

US 17 8.22 8.75 7.86 7.22 8.75 8.53

Czech Republic 18 8.17 9.58 6.79 7.22 8.13 9.12

Portugal 19 8.16 9.58 8.21 6.11 7.50 9.41

Belgium 20= 8.15 9.58 8.21 6.67 6.88 9.41

Japan 20= 8.15 9.17 7.86 5.56 8.75 9.41

Greece 22 8.13 9.58 7.50 6.67 7.50 9.41

UK 23 8.08 9.58 8.57 5.00 8.13 9.12

France 24 8.07 9.58 7.50 6.67 7.50 9.12

Mauritius 25= 8.04 9.17 8.21 5.00 8.13 9.71

Costa Rica 25= 8.04 9.58 8.21 6.11 6.88 9.41

Slovenia 27= 7.96 9.58 7.86 6.67 6.88 8.82

Uruguay 27= 7.96 10.00 8.21 5.00 6.88 9.71

Flawed democracies

South Africa 29 7.91 8.75 7.86 7.22 6.88 8.82

Chile 30 7.89 9.58 8.93 5.00 6.25 9.71

South Korea 31 7.88 9.58 7.14 7.22 7.50 7.94

Taiwan 32 7.82 9.58 7.50 6.67 5.63 9.71

Estonia 33 7.74 9.58 7.50 5.00 7.50 9.12

Italy 34 7.73 9.17 6.43 6.11 8.13 8.82

India 35 7.68 9.58 8.21 5.56 5.63 9.41

Botswana 36= 7.60 9.17 7.86 5.00 6.88 9.12

Cyprus 36= 7.60 9.17 6.79 6.67 6.25 9.12

Hungary 38 7.53 9.58 6.79 5.00 6.88 9.41

Cape Verde 39= 7.43 9.17 7.86 5.00 6.88 8.24

Lithuania 39= 7.43 9.58 6.43 6.67 5.63 8.82

Slovakia 41 7.40 9.58 7.50 6.11 5.00 8.82

Brazil 42 7.38 9.58 7.86 4.44 5.63 9.41

Latvia 43 7.37 9.58 6.43 6.11 5.63 9.12

Panama 44 7.35 9.58 7.14 5.56 5.63 8.82

Jamaica 45 7.34 9.17 7.14 5.00 6.25 9.12

Poland 46 7.30 9.58 6.07 6.11 5.63 9.12

Israel 47 7.28 9.17 6.64 7.78 7.50 5.29

Trinidad and Tobago 48 7.18 9.17 6.79 6.11 5.63 8.24

Bulgaria 49 7.10 9.58 5.71 6.67 5.00 8.53

Romania 50 7.06 9.58 6.07 6.11 5.00 8.53

Croatia 51 7.04 9.17 6.07 6.11 5.63 8.24

Ukraine 52 6.94 9.58 5.71 5.56 5.63 8.24

Mexico 53 6.67 8.75 6.07 5.00 5.00 8.53

Argentina 54 6.63 8.75 5.00 5.56 5.63 8.24

Serbia 55 6.62 9.17 5.36 5.00 5.63 7.94

Democracy index

THE WORLD IN 2OO7

Mongolia 56 6.60 9.17 6.07 3.89 5.63 8.24

Sri Lanka 57 6.58 6.92 5.00 5.56 7.50 7.94

Montenegro 58 6.57 9.17 5.71 5.00 5.63 7.35

Namibia 59= 6.54 4.75 4.00 6.67 8.75 8.53

Papua New Guinea 59= 6.54 7.33 6.43 4.44 6.25 8.24

Suriname 61 6.52 9.17 6.07 4.44 5.00 7.94

Moldova 62 6.50 9.17 4.29 6.11 5.00 7.94

Lesotho 63= 6.48 7.92 6.43 4.44 6.25 7.35

Philippines 63= 6.48 9.17 5.36 5.00 3.75 9.12

Indonesia 65= 6.41 6.92 7.14 5.00 6.25 6.76

Timor Leste 65= 6.41 7.00 5.57 5.00 6.25 8.24

Colombia 67 6.40 9.17 4.36 5.00 4.38 9.12

Macedonia 68 6.33 8.25 4.50 7.22 3.75 7.94

Honduras 69 6.25 8.33 6.43 4.44 5.00 7.06

El Salvador 70 6.22 9.17 5.43 3.89 4.38 8.24

Paraguay 71= 6.16 7.92 5.00 5.00 4.38 8.53

Benin 71= 6.16 6.83 6.43 3.89 6.88 6.76

Guyana 73 6.15 8.33 5.36 4.44 4.38 8.24

Dom Rep 74 6.13 9.17 4.29 3.33 5.63 8.24

Bangladesh 75= 6.11 7.42 5.07 4.44 6.25 7.35

Peru 75= 6.11 8.75 3.29 5.56 5.00 7.94

Guatemala 77 6.07 8.75 6.79 2.78 4.38 7.65

Hong Kong 78 6.03 3.50 5.71 5.00 6.25 9.71

Palestine 79 6.01 8.25 2.71 7.78 6.88 4.41

Mali 80 5.99 8.25 5.71 3.89 5.63 6.47

Malaysia 81= 5.98 6.08 5.71 4.44 7.50 6.18

Bolivia 81= 5.98 8.33 5.71 4.44 3.75 7.65

Hybrid regimes

Albania 83 5.91 7.33 5.07 4.44 5.63 7.06

Singapore 84 5.89 4.33 7.50 2.78 7.50 7.35

Madagascar 85= 5.82 5.67 5.71 5.56 6.88 5.29

Lebanon 85= 5.82 7.92 2.36 6.11 6.25 6.47

Bosnia and Hercegovina 87 5.78 8.25 3.29 4.44 5.00 7.94

Turkey 88 5.70 7.92 6.79 4.44 3.75 5.59

Nicaragua 89 5.68 8.25 5.71 3.33 3.75 7.35

Thailand 90 5.67 4.83 6.43 5.00 5.63 6.47

Fiji 91 5.66 6.50 5.21 3.33 5.00 8.24

Ecuador 92 5.64 7.83 4.29 5.00 3.13 7.94

Venezuela 93 5.42 7.00 3.64 5.56 5.00 5.88

Senegal 94 5.37 7.00 5.00 3.33 5.63 5.88

Ghana 95 5.35 7.42 4.64 4.44 4.38 5.88

Mozambique 96 5.28 5.25 5.71 4.44 6.88 4.12

Zambia 97 5.25 5.25 4.64 3.33 6.25 6.76

Liberia 98 5.22 7.75 2.14 5.00 5.63 5.59

Tanzania 99 5.18 6.00 3.93 5.06 5.63 5.29

Uganda 100 5.14 4.33 3.93 4.44 6.25 6.76

Kenya 101 5.08 4.33 4.29 5.56 6.25 5.00

Russia 102 5.02 7.00 3.21 5.56 3.75 5.59

Malawi 103 4.97 6.00 5.00 3.89 4.38 5.59

Georgia 104 4.90 7.92 1.79 3.33 5.00 6.47

Cambodia 105 4.77 5.58 6.07 2.78 5.00 4.41

Ethiopia 106 4.72 4.00 3.93 5.00 6.25 4.41

Burundi 107 4.51 4.42 3.29 3.89 6.25 4.71

Gambia 108 4.39 4.00 4.64 4.44 5.63 3.24

Haiti 109 4.19 5.58 3.64 2.78 2.50 6.47

Armenia 110 4.15 4.33 3.21 3.89 3.13 6.18

Kyrgyzstan 111 4.08 5.75 1.86 2.78 5.00 5.00

Iraq 112 4.01 4.75 0.00 5.56 5.63 4.12

Category scores

Overall I Electoral process II Functioning III Political IV Political V Civil

Rank score and pluralism of government participation culture liberties

THE WORLD IN 2OO7

Democracy index

Authoritarian regimes

Pakistan 113= 3.92 4.33 5.36 0.56 4.38 5.00

Jordan 113= 3.92 3.08 3.79 3.89 5.00 3.82

Comoros 115= 3.90 3.00 3.21 4.44 5.63 3.24

Morocco 115= 3.90 3.50 3.79 2.78 5.63 3.82

Egypt 115= 3.90 2.67 3.64 2.78 6.88 3.53

Rwanda 118 3.82 3.00 3.57 2.22 5.00 5.29

Burkina Faso 119 3.72 4.00 1.79 2.78 5.63 4.41

Kazakhstan 120 3.62 2.67 2.14 3.33 4.38 5.59

Sierra Leone 121 3.57 5.25 2.21 2.22 3.75 4.41

Niger 122 3.54 5.25 1.14 1.67 3.75 5.88

Bahrain 123 3.53 3.50 2.57 2.78 5.00 3.82

Cuba 124= 3.52 1.75 4.64 3.89 4.38 2.94

Nigeria 124= 3.52 3.08 1.86 4.44 4.38 3.82

Nepal 126 3.42 0.08 3.57 2.22 5.63 5.59

Côte d’Ivoire 127 3.38 1.25 2.86 3.33 5.63 3.82

Belarus 128 3.34 2.58 2.86 3.33 4.38 3.53

Azerbaijan 129 3.31 3.08 0.79 3.33 3.75 5.59

Cameroon 130 3.27 0.92 3.21 2.78 5.63 3.82

Congo Brazzaville 131 3.19 1.42 2.86 2.22 5.63 3.82

Algeria 132 3.17 2.25 2.21 2.22 5.63 3.53

Mauritania 133 3.12 1.83 4.29 2.22 3.13 4.12

Kuwait 134 3.09 1.33 4.14 1.11 5.63 3.24

Afghanistan 135= 3.06 6.17 0.00 2.22 2.50 4.41

Tunisia 135= 3.06 0.00 2.36 2.22 6.88 3.82

Yemen 137 2.98 2.67 2.71 2.78 4.38 2.35

China 138 2.97 0.00 4.64 2.78 6.25 1.18

Swaziland 139= 2.93 1.75 2.86 2.22 3.13 4.71

Iran 139= 2.93 0.08 3.57 3.89 5.63 1.47

Sudan 141 2.90 2.25 2.36 1.67 5.00 3.24

Qatar 142 2.78 0.00 3.43 1.67 5.00 3.82

Oman 143 2.77 0.00 3.07 1.67 5.00 4.12

Democratic Republic of Congo 144 2.76 4.58 0.36 2.78 3.75 2.35

Vietnam 145 2.75 0.83 4.29 2.78 4.38 1.47

Gabon 146 2.72 0.50 3.21 2.22 5.63 2.06

Bhutan 147= 2.62 0.08 4.64 1.11 3.75 3.53

Zimbabwe 147= 2.62 0.17 0.79 3.89 5.63 2.65

Tajikistan 149 2.45 1.83 0.79 2.22 6.25 1.18

UAE 150 2.42 0.00 3.07 1.11 5.00 2.94

Angola 151 2.41 0.50 2.14 1.11 5.63 2.65

Djibouti 152 2.37 2.50 1.43 0.56 5.00 2.35

Syria 153 2.36 0.00 1.79 1.67 6.88 1.47

Eritrea 154 2.31 0.00 2.14 1.11 6.25 2.06

Laos 155 2.10 0.00 3.21 1.11 5.00 1.18

Equatorial Guinea 156 2.09 0.00 2.86 1.11 5.00 1.47

Guinea 157 2.02 1.00 0.79 2.22 3.75 2.35

Guinea-Bissau 158 2.00 2.08 0.07 3.33 1.88 2.65

Saudi Arabia 159 1.92 0.00 2.36 1.11 4.38 1.76

Uzbekistan 160 1.85 0.08 0.79 2.78 5.00 0.59

Libya 161 1.84 0.00 1.64 1.11 5.00 1.47

Turkmenistan 162 1.83 0.00 0.79 2.78 5.00 0.59

Myanmar 163 1.77 0.00 1.79 0.56 5.63 0.88

Togo 164 1.75 0.00 0.79 0.56 5.63 1.76

Chad 165 1.65 0.00 0.00 0.00 5.00 3.24

Central Africa 166 1.61 0.42 1.43 1.67 1.88 2.65

North Korea 167 1.03 0.83 2.50 0.56 1.25 0.00

Category scores

Overall I Electoral process II Functioning III Political IV Political V Civil

Rank score and pluralism of government participation culture liberties

Democracy index

THE WORLD IN 2OO7

is no longer universally accepted. Instead it has been ar-

gued that the primary direction of causation runs from

democracy to income (Rigobon and Rodrik, 2005; Ace-

moglu et al, 2005).

One advantage of our index compared with others is

that it provides for considerable differentiation of scores,

including among developed countries. The “near-per-

fect” democracy is Sweden, the country with the highest

score. The other Nordic countries also have high ranks.

By contrast, the United States and Britain are near the

bottom of the full democracy category, but for some-

what different reasons. America falls down on some

aspects of governance and civil liberties. Despite low

election turnouts, political participation in the United

States is comparatively high. In Britain low political par-

ticipation (the lowest in the developed world) is a major

problem, and to a lesser extent, for now, so are eroding

civil liberties. The rating for France is also comparatively

low as a result of modest scores for the functioning of

government, political participation and political cul-

ture. Italy performs even worse, and falls in the ﬂ awed

democracies category—as a result of problems in func-

tioning of government and the electoral process, as well

as weaknesses in the political culture.

These results seem to highlight the interesting hy-

pothesis that large countries, other things being equal,

tend to be less democratic. But this appears to be the case

only among the developed countries. It does not hold

across the whole sample—there is no signiﬁ cant relation-

ship between the value of the democracy index and the

size of population for the entire 167-country sample.

Looking at the regional distribution of regime types,

ﬂ awed democracies are concentrated in Latin America

and eastern Europe, and to a lesser extent in Asia. De-

spite progress in Latin American democratisation in

recent decades, many countries in the region remain

fragile democracies. Levels of political participation are

generally very low and democratic cultures are weak

(with the caudillismo phenomenon still widespread ac-

cording to opinion surveys). There has also been sig-

niﬁ cant backsliding in recent years in some areas such

as media freedoms.

Much of eastern Europe illustrates the difference be-

tween formal and substantive democracy. The new EU

members from the region have pretty much equal levels

of political freedoms and civil liberties as the old devel-

Table 2

Democracy index 2006 by regime type

Countries % of countries % of world population

Full democracies 28 16.8 13.0

Flawed democracies 54 32.3 38.3

Hybrid regimes 30 18.0 10.5

Authoritarian regimes 55 32.9 38.2

“World” population refers to total population of the 167 countries that are covered. Since this excludes only micro states this is nearly equal to

the entire actual estimated world population in 2006.

Source: Economist Intelligence Unit; CIA World Factbook

2007 watchlist

Positive watch

Hong Kong: further improvements in civil

liberties and democratic political practices

after Donald Tsang’s election as chief execu-

tive in March 2007.

Negative watch

Taiwan: risk of a no-conﬁ dence vote in the

government that could trigger a constitu-

tional crisis; increased pressures in the run-

up to the 2007 parliamentary elections.

Bangladesh: caretaker government will

oversee general elections in early 2007. An

unclear or disputed election result could

trigger political crisis and rollback of de-

mocracy.

Armenia: parliamentary election in May 2007

could be highly ﬂ awed, tipping the country

into an outright authoritarian regime.

Russia: at present a hybrid regime, with a

trend towards curtailment of media and

other civil liberties. A potentially highly

ﬂ awed parliamentary election at the end of

2007 would reﬂ ect a further intensiﬁ cation

of the country’s apparent slide in an au-

thoritarian direction.

Nigeria: a disputed April 2007 election to

be followed by political turbulence and the

possible installation of a military-backed

interim government.

Burundi: president and government inten-

sify crackdown on opponents. The country

could slide from a hybrid regime to authori-

tarianism.

Guinea and Guinea-Bissau are already rated

as authoritarian, but things could get even

worse in 2007 as there is a high risk of mili-

tary coups in both. In Guinea there is a risk

of a military takeover in 2007 if ailing Presi-

dent Lansana Conte dies. In Guinea-Bissau

rising discontent in the army increases the

risk of a coup.

Mauritania: the country is undergoing a

democratic transition following a military

coup in August 2005. But there is a high risk

of a backlash, especially as the move from

military to civilian rule has potentially

destabilising inter-ethnic implications.

Hopes of democratisation are unlikely to

be fulﬁ lled.

THE WORLD IN 2OO7

Democracy index

oped EU, but lag signiﬁ cantly in political participation

and political culture—a reﬂ ection of widespread anomie

and weaknesses of democratic development. Only two

countries from the region—the Czech Republic and

Slovenia (just)—are in the full democracy category. Hy-

brid and authoritarian regimes dominate heavily in the

countries of the former Soviet Union, as the momentum

towards “colour revolutions” has appeared to peter out.

Most of the world’s authoritarian regimes are to be

found in the Middle East and Africa, although there is

also a fair number in Asia. The dearth of democratic

regimes in the Middle East and North Africa is a well-

known phenomenon, with much debate about the

causes. In the statistical relationship between democ-

racy and income discussed above, a dummy variable

for Middle East and North Africa is negative and highly

signiﬁ cant statistically even when oil wealth is control-

led for in our 167-country sample—that is, Middle East

and North Africa has much lower levels of democratisa-

tion than could be inferred on the basis of income levels.

A similar variable for Asia is also negative, although at

lower levels of statistical signiﬁ cance. And there is some

evidence that western Europe’s average democracy lev-

els are higher than even its high income levels would

suggest. For other regions—Sub-Saharan Africa, eastern

Europe and Latin America—average level of democratic

development correspond to what would be expected on

the basis of average income levels.

Table 3

Democracy across the regions

Democracy Number of Full Flawed Hybrid Authoritarian

index average countries democracies democracies regimes regimes

North America 8.64 2 2 0 0 0

West Europe 8.60 21 18 2 1 0

Eastern Europe 5.76 28 2 14 6 6

Latin America & the Caribbean 6.37 24 2 17 4 1

Asia & Australasia 5.44 28 3 12 4 9

Middle East & North Africa 3.53 20 0 2 2 16

Sub-Saharan Africa 4.24 44 1 7 13 23

Total 5.52 167 28 54 30 55

Democracy index

THE WORLD IN 2OO7

The Economist Intelligence

Unit’s index of democracy, on a

0 to 10 scale, is based on the rat-

ings for 60 indicators grouped

in ﬁ ve categories: electoral proc-

ess and pluralism; civil liberties;

the functioning of government;

political participation; and po-

litical culture. Each category has

a rating on a 0 to 10 scale, and

the overall index of democracy

is the simple average of the ﬁ ve

category indexes.

The category indexes are

based on the sum of the indicator

scores in the category, converted

to a scale of 0 to 10. Adjustments

to the category scores are made

if countries do not score a 1 in

the following critical areas for

democracy:

1. Whether national elections are

free and fair;

2. The security of voters;

3. The inﬂ uence of foreign pow-

ers on government;

4. The capability of the civil serv-

ice to implement policies.

If the scores for the ﬁ rst three

questions are 0 (or 0.5), one

point (0.5 point) is deducted

from the index in the relevant

category (either the electoral

process and pluralism or the

functioning of government). If

the score for 4 is 0, one point is

deducted from the functioning-

of-government category index.

The index values are used

to place countries within one of

four types of regimes:

1. Full democracies—scores of

8-10.

2. Flawed democracies—scores

of 6 to 7.9.

3. Hybrid regimes—scores of 4

to 5.9.

4. Authoritarian regimes—scores

below 4.

Threshold points for regime

types depend on overall scores

that are rounded to one decimal

point.

The scoring system

We use a combination of a di-

chotomous and a three-point

scoring system for the 60 indica-

tors. A dichotomous 1-0 scoring

system (1 for a yes and 0 for a no

answer) is not without problems,

but it has several distinct advan-

tages over more reﬁ ned scoring

scales (such as the often-used 1-

5 or 1-7). For many indicators,

the possibility of a 0.5 score is in-

troduced, to capture “grey areas”

where a simple yes (1) or no (0)

is problematic, with guidelines

as to when that should be used.

Thus for many indicators there

is a three-point scoring system,

which represents a compromise

between simple dichotomous

scoring and the use of finer

scales.

The problems of 1-5 or 1-7

scoring scales are numerous.

For most indicators under such

a system, it is extremely difﬁ cult

to deﬁ ne meaningful and com-

parable criteria or guidelines for

each score. This can lead to ar-

bitrary, spurious and non-com-

parable scorings. For example, a

score of 2 for one country may

be scored a 3 in another and so

on. Or one expert might score an

indicator for a particular coun-

try in a different way to another

expert. This contravenes a basic

principle of measurement, that

of so-called reliability—the de-

gree to which a measurement

procedure produces the same

measurements every time, re-

gardless of who is performing

it. Two- and three-point systems

do not guarantee reliability, but

make it more likely.

Second, comparability be-

tween indicator scores and aggre-

gation into a multi-dimensional

index appears more valid with

a two- or three-point scale for

each indicator (the dimensions

being aggregated are similar

across indicators). By contrast,

with a 1-5 system, the scores are

more likely to mean different

things across the indicators (for

example a 2 for one indicator

may be more comparable to a 3

or 4 for another indicator, rather

than a 2 for that indicator). The

problems of a 1-5 or 1-7 system

are magniﬁ ed when attempting

to extend the index to many re-

gions and countries.

Some features of the Economist

Intelligence Unit’s index

Public opinion surveys

A crucial, differentiating aspect

of our measure is that in addition

to experts’ assessments we use,

where available, public opinion

surveys—mainly the World Val-

ues Survey. Indicators based on

the surveys predominate heavily

in the political participation and

political culture categories, and a

few are used in the civil liberties

and functioning of government

categories.

In addition to the World Val-

ues Survey, other sources that

can be leveraged include the

Eurobarometer surveys, Gallup

polls, Latin American Barom-

eter, and national surveys. In

the case of countries for which

survey results are missing, survey

results for similar countries and

expert assessment are used to ﬁ ll

in gaps.

Participation and voter turnout

After increasing for many dec-

ades, there has been a trend of

decreasing voter turnout in most

established democracies since the

1960s. Low turnout may be due

to disenchantment, but it can

also be a sign of contentment.

Many, however, see low turnout

as undesirable, and there is much

debate over the factors that affect

turnout and how to increase it.

A high turnout is generally

seen as evidence of the legitimacy

of the current system. Contrary

to widespread belief, there is in

fact a close correlation between

turnout and overall measures of

democracy—ie, developed, con-

solidated democracies have, with

very few exceptions, higher turn-

out (generally above 70%) than

less established democracies.

The legislative and executive

branches

The appropriate balance between

these is much-disputed in politi-

cal theory. In our model the clear

predominance of the legislature

is rated positively as there is a

very strong correlation between

legislative dominance and meas-

ures of overall democracy.

Methodology

THE WORLD IN 2OO7

Democracy index

The model

I Electoral process and pluralism

1. Are elections for the national

legislature and head of govern-

ment free?

Consider whether elections

are competitive in that electors

are free to vote and are offered a

range of choices.

1: Essentially unrestricted con-

ditions for the presentation

of candidates (for example,

no bans on major parties)

0.5: There are some restrictions

on the electoral process

0: A single-party system or

major impediments exist

(for example, bans on a

major party or candidate)

2. Are elections for the national

legislature and head of govern-

ment fair?

1: No major irregularities in

the voting process

0.5: Significant irregularities

occur (intimidation, fraud),

but do not affect signifi-

cantly the overall outcome

0: Major irregularities occur

and affect the outcome

Score 0 if score for question 1 is

3. Are municipal elections both

free and fair?

1: Are free and fair

0.5: Are free but not fair

0: Are neither free nor fair

4. Is there universal suffrage for

all adults?

Bar generally accepted exclusions

(for example, non-nationals;

criminals; members of armed

forces in some countries).

1: Yes

0: No

5. Can citizens cast their vote

free of signiﬁ cant threats to their

security from state or non-state

bodies?

1: Yes

0: No

6. Do laws provide for broadly

equal campaigning opportuni-

ties?

1: Yes

0.5: Yes formally, but in practice

opportunities are limited

for some candidates

0: No

7. Is the process of financing

political parties transparent and

generally accepted?

1: Yes

0.5: Not fully transparent

0: No

8. Following elections, are the

constitutional mechanisms for

the orderly transfer of power

from one government to another

clear, established and accepted?

1: All three criteria are ful-

ﬁ lled

0.5: Two of the three criteria are

fulﬁ lled

0: Only one or none of the cri-

teria is satisﬁ ed

9. Are citizens free to form polit-

ical parties that are independent

of the government?

1. Yes

0.5: There are some restrictions

0: No

10. Do opposition parties have

a realistic prospect of achieving

government?

1: Yes

0.5: There is a dominant two-

party system in which other

political forces never have

any effective chance of tak-

ing part in national govern-

ment

0: No

11. Is potential access to public

ofﬁ ce open to all citizens?

1: Yes

0.5: Formally unrestricted, but

in practice restricted for

some groups, or for citi-

zens from some parts of the

country

0: No

12. Are citizens free to form po-

litical and civic organisations,

free of state interference and

surveillance?

1: Yes

0.5: Ofﬁ cially free, but subject

to some restrictions or in-

terference

0: No

II Functioning of government

13. Do freely elected representa-

tives determine government

policy?

1: Yes

0.5: Exercise some meaningful

inﬂ uence

0: No

14. Is the legislature the supreme

political body, with a clear su-

premacy over other branches of

government?

1: Yes

0: No

15. Is there an effective system of

checks and balances on the exer-

cise of government authority?

1: Yes

0.5: Yes, but there are some seri-

ous ﬂ aws

0: No

16. Government is free of undue

inﬂ uence by the military or the

security services.

1: Yes

0.5: Inﬂ uence is low, but the de-

fence minister is not a civil-

ian. If the current risk of a

military coup is extremely

low, but the country has a

recent history of military

rule or coups

0: No

17. Foreign powers do not de-

termine important government

functions or policies.

1: Yes

0.5: Some features of a protec-

torate

0: No (signiﬁ cant presence of

foreign troops; important

decisions taken by foreign

power; country is a protec-

torate)

18. Special economic, religious or

other powerful domestic groups

do not exercise signiﬁ cant politi-

cal power, parallel to democratic

institutions?

1: Yes

0.5: Exercise some meaningful

inﬂ uence

0: No

19. Are sufficient mechanisms

and institutions in place for as-

suring government accountabil-

ity to the electorate in between

elections?

1: Yes

0.5. Yes, but serious ﬂ aws exist

0: No

20. Does the government’s au-

thority extend over the full ter-

ritory of the country?

1: Yes

0: No

21. Is the functioning of govern-

ment open and transparent, with

sufﬁ cient public access to infor-

mation?

1: Yes

0.5: Yes, but serious ﬂ aws exist

0: No

22. How pervasive is corrup-

tion?

1: Corruption is not a major

problem

0.5: Corruption is a signiﬁ cant

issue

0: Pervasive corruption exists

23. Is the civil service willing and

capable of implementing gov-

ernment policy?

1: Yes

0.5. Yes, but serious ﬂ aws exist

0: No

24. Popular perceptions of the

extent to which they have free

choice and control over their

lives

1: High

0.5: Moderate

0: Low

If available, from World Values

Survey

% of people who think that they

have a great deal of choice/con-

trol

1 if more than 70%

0.5 if 50-70%

0 if less than 50%

25. Public conﬁ dence in govern-

ment.

1: High

0.5: Moderate

0: Low

If available, from World Values

Survey

% of people who have a “great

deal” or “quite a lot” of conﬁ -

dence in government

1 if more than 40%

0.5 if 25-40%

0 if less than 25%

Democracy index

THE WORLD IN 2OO7

26. Public conﬁ dence in political

parties.

1: High

0.5: Moderate

0: Low

If available, from World Values

Survey

% of people who have a “great

deal” or “quite a lot” of conﬁ -

dence

1 if more than 40%

0.5 if 25-40%

0 if less than 25%

III Political participation

27. Voter participation/turnout

for national elections.

(average turnout in parliamen-

tary and/or presidential elec-

tions since 2000. Turnout as

proportion of population of

voting age).

1 if consistently above 70%

0.5 if between 50% and 70%

0 if below 50%

If voting is obligatory, score 0.

Score 0 if scores for questions 1

or 2 is 0.

28. Do ethnic, religious and

other minorities have a reason-

able degree of autonomy and

voice in the political process?

1: Yes

0.5: Yes, but serious ﬂ aws exist

0: No

29. Women in parliament.

% of members of parliament

who are women

1 if more than 20% of seats

0.5 if 10-20%

0 if less than 10%

30. Extent of political participa-

tion. Membership of political

parties and political non-gov-

ernmental organisations.

1 if over 7% of population for

either

0.5 if 4% to 7%

0 if under 4%.

If participation is forced, score

31. Citizens’ engagement with

politics.

1: High

0.5: Moderate

0: Low

If available, from World Values

Survey

% of people who are very or

somewhat interested in politics

1 if over 60%

0.5 if 40% to 60%

0 if less than 40%

32. The preparedness of popula-

tion to take part in lawful dem-

onstrations.

1: High

0.5: Moderate

0: Low

If available, from World Values

Survey

% of people who have taken part

in or would consider attending

lawful demonstrations

1 if over 40%

0.5 if 30% to 40%

0 if less than 30%

33. Adult literacy.

1 if over 90%

0.5 if 70% to 90%

0 if less than 70%

34. Extent to which adult popu-

lation shows an interest in and

follows politics in the news.

1: High

0.5: Moderate

0: Low

If available, from World Values

Survey

% of population that follows

politics in the news media (print,

TV or radio) every day

1 if over 50%

0.5 if 30% to 50%

0 if less than 30%

35. The authorities make a seri-

ous effort to promote political

participation.

1: Yes

0.5: Some attempts

0: No

Consider the role of the educa-

tion system, and other promo-

tional efforts. Consider measures

to facilitate voting by members

of the diaspora.

If participation is forced, score

IV Democratic political culture

36. Is there a sufﬁ cient degree of

societal consensus and cohesion

to underpin a stable, functioning

democracy?

1: Yes

0.5: Yes, but some serious doubts

and risks

0: No

37. Perceptions of leadership;

proportion of the population

that desires a strong leader who

bypasses parliament and elec-

tions.

1: Low

0.5: Moderate

0: High

If available, from World Values

Survey

% of people who think it would

be good or fairly good to have

a strong leader who does not

bother with parliament and elec-

tions

1 if less than 30%

0.5 if 30% to 50%

0 if more than 50%

38. Perceptions of military rule;

proportion of the population

that would prefer military.

1: Low

0.5: Moderate

0: High

If available, from World Values

Survey

% of people who think it would

be very or fairly good to have

army rule

1 if less than 10%

0.5 if 10% to 30%

0 if more than 30%

39. Perceptions of rule by experts

or technocratic government;

proportion of the population

that would prefer rule by experts

or technocrats.

1: Low

0.5: Moderate

0: High

If available, from World Values

Survey

% of people who think it would

be very or fairly good to have

experts, not government, make

decisions for the country

1 if less than 50%

0.5 if 50% to 70%

0 if more than 70%

THE WORLD IN 2OO7

Democracy index

40. Perception of democracy and

public order; proportion of the

population that believes that de-

mocracies are not good at main-

taining public order.

1: Low

0.5: Moderate

0: High

If available, from World Values

Survey

% of people who disagree with

the view that democracies are

not good at maintaining order

1 if more than 70%

0.5 if 50% to 70%

0 if less than 50%

41. Perception of democracy

and the economic system; pro-

portion of the population that

believes that democracy beneﬁ ts

economic performance.

If available, from World Values

Survey

% of people who disagree with

the view that the economic sys-

tem runs badly in democracies

1 if more than 80%

0.5 if 60% to 80%

0 if less than 60%

42. Degree of popular support

for democracy.

1: High

0.5: Moderate

0: Low

If available, from World Values

Survey

% of people who agree or

strongly agree that democracy

is better than any other form of

government

1 if more than 90%

0.5 if 75% to 90%

0 if less than 75%

43. There is a strong tradition

of the separation of church and

state.

1: Yes

0.5: Some residual inﬂ uence of

church on state

0: No

V Civil liberties

44. Is there a free electronic

media?

1: Yes

0.5: Pluralistic, but state-con-

trolled media are heavily

favoured. One or two pri-

vate owners dominate the

media

0: No

45. Is there a free print media?

1: Yes

0.5: Pluralistic, but state-con-

trolled media are heavily

favoured. There is high

degree of concentration of

private ownership of na-

tional newspapers

0: No

46. Is there freedom of expres-

sion and protest (bar only gener-

ally accepted restrictions such as

banning advocacy of violence)?

1: Yes

0.5: Minority viewpoints are

subject to some ofﬁ cial har-

assment. Libel laws restrict

heavily scope for free ex-

pression

0: No

47. Is media coverage robust? Is

there open and free discussion

of public issues, with a reason-

able diversity of opinions?

1: Yes

0.5: There is formal freedom,

but high degree of con-

formity of opinion, includ-

ing through self-censorship,

or discouragement of mi-

nority or marginal views

0: No

48. Are there political restric-

tions on access to the internet?

1: No

0.5: Some moderate restric-

tions

0: Yes

49. Are citizens free to form pro-

fessional organisations and trade

unions?

1: Yes

0.5: Ofﬁ cially free, but subject to

some restrictions

0: No

50. Do institutions provide citi-

zens with the opportunity to

successfully petition government

to redress grievances?

1: Yes

0.5: Some opportunities

0: No

51. The use of torture by the

state

1: Torture is not used

0: Torture is used

52. The degree to which the judi-

ciary is independent of govern-

ment inﬂ uence.

Consider the views of interna-

tional legal and judicial watch-

dogs. Have the courts ever issued

an important judgment against

the government, or a senior gov-

ernment ofﬁ cial?

1: High

0.5: Moderate

0: Low

53. The degree of religious tol-

erance and freedom of religious

expression.

Are all religions permitted to

operate freely, or are some re-

stricted? Is the right to worship

permitted both publicly and pri-

vately? Do some religious groups

feel intimidated by others, even

if the law requires equality and

protection?

1: High

0.5: Moderate

0: Low

54. The degree to which citizens

are treated equally under the

law.

Consider whether favoured

members of groups are spared

prosecution under the law.

1: High

0.5: Moderate

0: Low

55. Do citizens enjoy basic secu-

rity?

1: Yes

0.5: Crime is so pervasive as to

endanger security for large

segments

0: No

56. Extent to which private prop-

erty rights protected and private

business is free from undue gov-

ernment inﬂ uence.

1: High

0.5: Moderate

0: Low

57. Extent to which citizens

enjoy personal freedoms.

Consider gender equality, right

to travel, choice of work and

study.

1: High

0.5: Moderate

0: Low

58. Popular perceptions on

human rights protection; pro-

portion of the population that

think that basic human rights

are well-protected.

1: High

0.5: Moderate

0: Low

If available, from World Values

Survey

% of people who think that

human rights are respected in

their country

1 if more than 70%

0.5 if 50% to 70%

0 if less than 50%

59. There is no signiﬁ cant dis-

crimination on the basis of peo-

ple’s race, colour or creed.

1: Yes

0.5: Yes, but some signiﬁ cant ex-

ceptions

0: No

60. Extent to which the gov-

ernment invokes new risks and

threats as an excuse for curbing

civil liberties.

1: Low

0.5: Moderate

0: High

References

Acemoglu, Daron, Simon Johnson,

James A. Robinson, and Pierre Yared

(2005), “Income and democracy”,

NBER Working Paper No. 11205,

March.

Coppedge, Michael (2005), “Deﬁ n-

ing and measuring democracy”,

Working paper, International Politi-

cal Science Association, April.

Dahl, Robert A(1970), “Polyarchy”,

New Haven, Yale University Press.

Freedom House, various, www.

freedomhouse.org.

Horowitz, Irving Louis (2006) “The

struggle for democracy”, National

Interest, spring.

Rigobon, Roberto and Dani Rodrik

(2005), “Rule of law, democracy,

openness, and income: estimating

the interrelationships”, Economics of

Transition, Volume 13 (3).