COVID-19 vaccinations, cases, excess mortality, and much more
Explore our COVID-19 data

Notice: This is only a preliminary collection of relevant material

The data and research currently presented here is a preliminary collection or relevant material. We will further develop our work on this topic in the future (to cover it in the same detail as for example our entry on World Population Growth).

If you have expertise in this area and would like to contribute, apply here to join us as a researcher.

All our charts on Democracy

A majority of countries are democracies

About half of the world’s countries are now democracies. The charts here use two different classifications to depict the slow rise of the number of democracies over the last two centuries.

The end of World War I led to the birth of many democracies. However, during the 1930s, many of these young democracies then reverted to being autocratic.

After World War II, the number of democracies began growing again. But it was the end of the Cold War that led to a more dramatic increase in the number of democracies.

200 years ago, everyone lacked democratic rights. Now, billions of people have them

When French revolutionaries stormed the Bastille prison in 1789 in pursuit of liberty, equality, and fraternity (and weapons), they could not have imagined how far democratic political rights would have spread a mere 200 years later. In the 19th century, there were few countries one could call democracies. Today, the majority are.

It is an astonishing achievement that many countries are now governed democratically. But the mere number of countries does not tell us how many people enjoy democratic rights. When Tunisia became democratic in 2012, its population of 11 million gained the political rights that came with it. When India democratized in the 1950s, this same transition affected almost 400 million people.

If we adopt the common and famous understanding of democracy as rule by the people, we should also look at how many people get to have a say in their government. How many people have democratic political rights around the world? And how has their number changed over the last two hundred years?

To answer these questions, we need to combine long-term data on countries’ populations1 with information on their political systems. This tells us how the political rights of the world’s population have changed over the past two hundred years.

How do researchers identify which countries are democracies?

Identifying which countries are democracies comes with many challenges. People disagree  about what characterizes a democracy, and whether actual political systems can even come close to such an ideal. If they agree on what democracy is and that countries can come meaningfully close to it, its characteristics — such as whether an election is free and fair — still are difficult to assess. If knowledgeable researchers can be found, their assessments are still to some degree subjective, and they may disagree with others. Even if researchers align in their assessment of specific characteristics of a political system, they may disagree about how to reduce the complexity of these many characteristics into a single variable: a binary measure that says whether a country is a ‘democracy’ or not.

Because of these difficulties, classifying political systems is unavoidably controversial.

In our work we therefore rely on sources that work hard to address these many challenges, and are transparent, so that they can be interrogated and criticized by those who disagree. 

Here, we identify the political systems of countries with the Regimes of the World (RoW) classification by political scientists Anna Lührmann, Marcus Tannenberg, and Staffan Lindberg.2 The classification uses data from the Varieties of Democracy (V-Dem) project3 and distinguishes between four types of political systems: closed autocracies, electoral autocracies, electoral democracies, and liberal democracies.

Which political systems does the ‘Regimes of the World’ classification distinguish?

  • In closed autocracies, citizens do not have the right to choose either the chief executive of the government or the legislature through multi-party elections.
  • In electoral autocracies, citizens have the right to choose the chief executive and the legislature through multi-party elections; but they lack some freedoms, such as the freedoms of association or expression, that make the elections meaningful, free, and fair.
  • In electoral democracies, citizens have the right to participate in meaningful, free and fair, and multi-party elections.
  • In liberal democracies, citizens have further individual and minority rights, are equal before the law, and the actions of the executive are constrained by the legislative and the courts.

While we use RoW’s classification and V-Dem’s data, we expand the years and countries covered and refine the coding rules. This post details how the political systems are measured, which changes we made, and what shortcomings and strengths the measure has. It is important to know that this measure describes when many people in a country had certain political rights, not that everyone had them.4 It is not a perfect classification, but still allows us to approximate how many people have had democratic rights.

Using the RoW classification, the interactive map shows how each country is classified at the end of each year, going back in time as far as 1789. To explore changes over time, you can drag the time-slider below the map.

Almost everyone lacked democratic political rights in the 19th century, but many have gained them since.

In the core chart of this post we see how many people lived under each of the four political systems since 1800. To see what share of the world’s population lived in each regime, you can tick the ‘Relative’ box.

Very few people had democratic political rights in the 19th century. In 1800, almost everyone lived in regimes that are classified as closed autocracies by RoW. No country was a democracy, and only 22 million people lived in the two countries classified as electoral autocracies: the United Kingdom and the United States.

Most people continued to live in closed autocracies over the course of the 19th century, with 3 out of 4 people still having few political rights by 1900, while those who did not mostly lived in electoral autocracies in the Americas and Western Europe. Only the 14 million people in Australia, Belgium, and Switzerland enjoyed a wide range of electoral and liberal political rights. A further 41 million in France and New Zealand enjoyed many democratic (but not liberal) political rights.

The first half of the 20th century made clear that a spread of democratic rights was both possible and uncertain. Democratic progress in the first decades of the century was set back when countries such as Germany reverted to autocratic rule in the 1930s and 1940s. 

Over the course of the second half of the 20th century, large numbers of people then gained democratic political rights. In 1950, more than 200 million people — mostly in Western Europe — lived in liberal democracies, and another 240 million lived in electoral democracies in Western Europe and the Americas. This number increased in the next decades, and by the late 1990s the majority of the world’s population — around 3 billion people — lived in electoral and liberal democracies.

The spread of democratic political rights continued during the early 21st century. By 2018, more than 2.7 billion people lived in electoral democracies in all regions of the world: most coming from the populous countries of India,5 Indonesia, Brazil and Nigeria. Another billion people lived in liberal democracies, such as those living in South Korea and Ghana. Almost all of the 1.9 billion people still living in a closed autocracy now reside in just one country: China.

Democratic political rights are still far from universal — and far from inevitable

While democratic rights have spread far, they are also still far from universal, and there have been recent setbacks. Even though many people now have them, the total number of people not having democratic rights is higher than ever. This is because the world’s population grew faster than democracy spread. And some people have recently lost political rights; most prominently the 1.4 billion people living in India, which became an electoral autocracy in 2019.6 This means that now more than two thirds of the world’s population live in closed and electoral autocracies.

These setbacks should serve as a reminder that continued political progress is not inevitable. Nonetheless, we see that a staggering number of people have gained democratic rights in a relatively short period of time. Many people still lack them, but the pace of this progress is a sign that this can change quickly. The French revolutionaries’ — and our — pursuit of liberty, equality, and fraternity is far from over. But we have come a long way already.

In most countries, democracy is a recent achievement. Dictatorship is far from a distant memory

To young people living in democracies, authoritarianism may seem like a long-forgotten part of their country’s history. For as long as they can remember, their fellow citizens have had the right to voice their opinion and to organize freely, political parties have competed in meaningful elections, and the legislature and courts have checked their governments’ actions.

But these experiences are far from universal. Many countries are not democracies; and almost all countries that are democratic are younger than a lifetime. This means that for most people, life under authoritarianism is either their current experience, or they remember a time when it was.

How can researchers measure the age of democracies?

Identifying which countries are democracies comes with many challenges. People disagree  about what characterizes a democracy, and whether actual political systems can even come close to such an ideal. If they agree on what democracy is and that countries can come meaningfully close to it, its characteristics — such as whether an election is free and fair — still are difficult to assess. If knowledgeable researchers can be found, their assessments are still to some degree subjective, and they may disagree with others. Even if researchers align in their assessment of specific characteristics of a political system, they may disagree about how to reduce the complexity of these many characteristics into a single variable: a binary measure that says whether a country is a ‘democracy’ or not.

Because of these difficulties, classifying political systems is unavoidably controversial.

In our work we therefore rely on sources that work hard to address these many challenges, and are transparent, so that they can be interrogated and criticized by those who disagree. We also use multiple sources to see how the assessments of different researchers compare.

The first source we use here is the Regimes of the World (RoW) classification by political scientists Anna Lührmann, Marcus Tannenberg and Staffan Lindberg.7 The classification uses data from the Varieties of Democracy (V-Dem) project8 and distinguishes between two types of democracies: electoral democracies and liberal democracies.

In electoral democracies, citizens have the right to participate in meaningful, free and fair, multi-party elections. Liberal democracies go further: citizens have individual and minority rights, equality before the law, and the actions of the executive are constrained by the legislature and the courts.9

While we use RoW’s classification and V-Dem’s data, we expand the years and countries covered and refine the coding rules. This post details how RoW measures democracies, which changes we made, and what shortcomings and strengths the measure has.

Although RoW allows us to look at two different understandings of democracy, Lührmann et al. (2018) acknowledge that it is a demanding measure of which countries are democracies.10 RoW has high standards for how free, fair, and meaningful elections have to be to classify as a democracy. It therefore classifies some political systems with limited flaws, such as some restrictions on the freedoms of expression and association, as autocratic. Other measures are more accepting of such flawed elections and consider these political systems as sufficiently democratic.

One alternative source we use here is the BMR classification by political scientists Carles Boix, Michael Miller, and Sebastian Rosato.11 The classification distinguishes between democracies, in which political leaders are elected under broad suffrage in free and fair elections, and non-democracies, which do not meet these criteria.12 They apply the classification to 222 countries, some of which they cover as far back as 1800.13

To calculate the age of democracy in a country, we count the number of years since a country is considered democratic, looking in turn at electoral democracies and liberal democracies based on RoW, and democracies as identified by BMR. If the classification categorizes the country as non-democratic at any point, the age of its democracy is set back to zero.14

Just over half of countries are democracies today

The interactive map shows how old electoral democracy was in each country in 2020 when using the RoW classification.15 The bar chart sums up the number of countries by age.

Based on this classification and the assessments of V-Dem’s experts, we see that only half of all countries were democracies in 2021: of the 179 countries for which data is available, 89 held meaningful, free and fair, and multi-party elections.

This means that for the people living in the other 90 countries, authoritarian government is not a memory at all, but their current experience.

To explore how the ages of democracies and their frequencies have changed over time, you can drag the time-slider below the charts as far back as 1789, when no country was democratic.16 

Most democracies are young

The data shows that democracy is young in most of the countries that are democratic today.

Many democracies are less than a generation old. Eighteen of them are younger than 18, not older than a child. These include the democracies in Ukraine, Sri Lanka, Nepal, or Liberia. Others are only as old as the country’s young adults, such as the democracies in Indonesia, Mexico, and South Africa.

This means that in these countries even most young people have experienced authoritarian rule, and that older people have lacked democratic political rights for a large part of their lives.

A smaller group of countries have been electoral democracies for two or three generations. This includes the democracies in Botswana, Costa Rica, Italy, Portugal and Spain.

In these countries, children and young adults have only known life in a democracy. But their parents and grandparents, and thus large parts of the population, have still experienced non-democratic rule.

Only a few countries have been electoral democracies for a long time. Eight countries — Canada, Finland, Iceland, Ireland, New Zealand, Sweden, the United Kingdom, and the United States — have been electoral democracies for a century or more. And according to this data, just two countries have been electoral democracies even longer: Australia and Switzerland have been democracies since the mid-19th century.

Democracy in these countries is therefore older than all or close to all of their citizens.This does not mean, however, that everyone in these countries has enjoyed democratic political rights since then. For example, the Australian and Swiss governments forbade women to vote and stand in elections until 1902 and 1971, respectively.

Liberal democracy is an even more recent achievement

If we look at liberal democracies, defined in the RoW-classification as political systems in which citizens have further individual and minority rights in addition to the meaningful, free and fair, multi-party elections that characterize electoral democracies, we see that they are an even rarer and more recent achievement.17

In the interactive map and bar chart we see the age of liberal democracy in any given country and their frequencies by year.18 In 2021, only 34 of the 179 countries covered were considered to be liberal democracies.

These liberal democracies split evenly into four age groups: democracies the age of children or young adults, such as Chile, Estonia, and South Korea; middle-aged democracies, including Costa Rica, Israel, and the United States; democracies the age of senior citizens, such as France, Germany, and Japan; and even older liberal democracies: this includes Australia, Ireland, New Zealand, and the United Kingdom. The outlier by far is Switzerland, which the RoW classification categorizes as a liberal democracy without interruption since the middle of the 19th century.

Democracy is a recent achievement regardless of the measure used

As emphasized earlier in this article, researchers come to different conclusions about whether a country should be considered democratic or not.

This becomes clear when we compare the RoW classification — which we just looked at — with the classification by Carles Boix, Michael Miller, and Sebastian Rosato (BMR). One example of where the classifications disagree is India. RoW considers the country to have been an electoral autocracy in 1975 and 1976, and since 2019. This is because its classification emphasizes freedoms of association and expression in addition to the freedom and fairness of elections. V-Dem’s experts deem these to have declined during those years.19 The BMR classification disagrees with this assessment and considers India to have been continuously democratic since 1950.  

But, overall, a comparison of sources shows broad agreement, and we see that the BMR classification agrees that democracy is a recent achievement in most countries.

The interactive map and bar chart again display the age of democracy in each country and their frequencies by year.

The data shows that in 2020, the BMR classification considers more countries to have been minimally democratic than in RoW’s classification: 118 of the 193 countries covered were considered to have been electoral democracies. This includes small countries which RoW does not cover, such as Andorra and Saint Lucia, plus countries such as Hungary and the aforementioned India which RoW does cover, but recently classifies as electoral autocracies.

Among the democracies classified as such by BMR, those aged one or two generations are especially common. This includes countries categorized the same based on RoW, such as Indonesia, Mexico, and South Africa; and it includes some democracies RoW considers to be younger, such as the Dominican Republic and Ukraine.

While the measures for some countries disagree about whether they are democracies, and for how long they have been so, the measures agree in their assessment of many others. And importantly, they agree that few countries that are democratic are older than two generations.20

Ultimately, regardless of the measure used, the data shows that many people have experienced non-democratic rule in their lifetimes.

In almost all countries, non-democratic rule is far from a distant memory

Electoral and liberal democratic political rights are recent achievements in the countries where citizens have them.

Countries such as the United Kingdom, in which democracy is older than almost all of its citizens, are rare; and even countries such as Spain have democracies which are younger than many of its older citizens. 

The experience of young people living in democracies in Western Europe or North America is therefore a fairly unique one: it is not representative of other people’s experiences — not of older citizens in their own country, or of people in other parts of the world.

Democracy’s spread is largely a recent achievement, and dictatorship is far from a distant memory for people in almost all countries. It instead is either still with them, and a part of their daily life; or it is not forgotten, and part of their own or their older relatives’ memories.

Democratization

Countries with better education in the past tend to be more democratic today

A long-standing theory in political science argues that education is a key determinant of the emergence and sustainability of democracy, because it promotes political participation at the individual level and fosters a collective sense of civic duty.

And so, under this theory, we should expect that education correlates positively with measures of democratization in subsequent years. This is precisely what we see in the data and chart below. Those countries that had higher average education levels in 1970 are also those countries that are more democratic today (you can read more about measures of education level in our entry on Global Rise of Education).

Again, these data should be interpreted carefully. They merely show a correlation and do not imply a causal link. That is, these data do not imply that education leads to democracy.

Nonetheless, academic work does suggest that there is such a causal link. Even after controlling for many other country characteristics, the positive relationship between education and democracy holds (see, for example, Lutz, Crespo-Cuaresma, and Abbasi‐Shavazi 201021).

Living conditions in democratic countries

Democratic countries tend to be richer

This chart plots each country’s GDP per capita against the Liberal Democracy index.

All highly democratic countries have relatively high GDPs per capita. Only few countries that are highly undemocratic are as rich. Those countries either export large quantities of fossil fuels or are small city states.

People in democracies tend to be healthier

This chart plots each country’s child mortality against the Liberal Democracy index to illustrate the association between health and democracy.

All highly democratic countries have relatively low child mortality rates. Very few less democratic countries have child mortality rates that are as low. Most undemocratic countries have much higher mortality rates.

Democracies are better at protecting human rights

The right to vote and determine who holds political power is in itself a fundamental right. And this right is, by definition, upheld and protected by all democracies.

But of course, there are many other human rights. Are democracies also better at protecting these other human rights?

As noted in our entry on human rights, it is difficult to measure the degree to which human rights are protected. In our opinion, the best available measure is the Human Rights Score developed by Fariss (2014).22

The Human Rights Score focuses on the protection of the physical integrity of citizens. In particular, it takes into account torture, government killing, political imprisonment, extrajudicial executions, mass killings and disappearances.

The chart plots each country’s Human Rights Score against its score on the Liberal Democracy index. There is a clear positive correlation: more democratic countries tend to protect human rights more.

Mulligan, Gil, and Sala-i-Martin (2004)23 investigate the link between democratic rule and the human rights protection in a sample of 121 countries controlling for other important variables. The authors find that countries that are more democratic are less likely to execute citizens, regulate religion, or censor the press.

Does democratization impact education?

Above, we mentioned that improved education might cause greater democratization. Now, is there also a reverse causal effect? That is, does democratization lead to improved education? Once again, this is a tricky question for social science, because we need to distinguish between the two arrows of causation.

Evidence that democratization leads to better education
Gallego (2010)24 presents the most careful analysis that we are aware of. It presents evidence that democracy has indeed had a causal effect on primary-school enrollment.25

Other papers deal with the issue of possible reverse causality in a simpler fashion and use lagged observations of democracy as a possible determinant for the level of education. For example, Baum and Lake (2001) find that democratization increased secondary-school enrollment.26

Also, Acemoglu, Naidu, Restrepo, and Robinson (2015)27 find that democracy is associated with an increase in secondary schooling.

We now briefly discuss several channels through which democratization might improve education:

Electoral competition in democracies increases the incentive to abolish school fees
Harding and Stasavage (2014)28 find that democratization has a positive effect on primary education. Their explanation is that electoral competition in democracies incentivizes politicians to abolish primary-school fees. They argue that democratization has a much smaller effect on the provision of school inputs and consequently the quality of schooling — the reason is that such actions are harder to monitor and would thus provide politicians with a smaller advantage in electoral competition.

Democratization increases educational spending
Stasavage (2005)29 finds that the 1990s shift to multiparty competition in African countries increased total educational spending as a percentage of GDP.
Ansell (2010)30 studies 100 countries over 40 years and finds that democratization increases both total educational spending as a share of GDP and as a share of the government budget.

Evidence that democracy improves teacher–student ratios
Naidu (2011)31 studies the effects of the 19th-century disenfranchisement of black citizens in the US South through poll taxes and literacy tests. He finds that this reversal of democracy “reduced the teacher-child ratio in black schools by 10–23%, with no significant effects on white teacher-child ratios.”

Democracy improved local politics in China and lead to more educated politicians
Martinez-Bravo et al. (2012)32 study the gradual introduction of local elections in China. In particular, they exploit the staggered timing of the introduction of village elections as a natural experiment. They “find that elections significantly increase public goods expenditure, the increase corresponds to demand and is paralleled by an increase in public goods provision and local taxes.” This is consistent with some of the results we’ve already discussed, including increased public education in villages with more children. Overall total public goods investment increased by 27 percent – this increase in public expenditures was funded by villagers and was accompanied by an increase in the local taxes paid by villagers.

The introduction of elections also reduced inequality. This was achieved partly through (a) land redistribution from elite-controlled enterprises to farming households; and (b) increased irrigation and hence improved agricultural productivity that is likely to “disproportionately benefit poorer households”.

Martinez-Bravo et al. (2012) also find that the introduction of elections was followed by the increased turnover of village chairmen increased. Moreover, the village chairmen were now less likely to be Communist Party members and, more importantly, were better educated.