Both income and wealth inequality have been on the minds of voters, politicians and academics for decades, especially as evidence has accumulated that inequality is not only high but, for the most part, rising. And along with the obvious consequences in terms of the growing gaps in living standards among groups – for example, between urban and rural Americans, and college and non-college graduates – the indirect impact of inequality has been deeply troubling.

It has likely undermined the pace of economic growth by slowing investment in human capital and by reducing economic mobility. Moreover, there is evidence that inequality makes economic growth more fragile and episodic. And while it’s a bit of a stretch, some analysts even see the political pressures to reduce inequality at the root of the housing crash in 2008 because the government actively encouraged homeownership among lower-income households unable to reliably make mortgage payments.

Even as rising economic inequality has been a part of our landscape for a number of decades, the U.S. Supreme Court’s 2010 decision in Citizens United v. the Federal Election Commission reinvigorated debates around inequality and the role of money in politics. The Court’s decision, effectively giving carte blanche to corporations and unions to make unlimited contributions to election campaigns, raised the question of whether it was possible for a country to sustain a thriving democracy in the face of massive concentrations of wealth. But despite the public focus on this issue, researchers have yet to reach a consensus about the relationship between inequality and democracy.

Framing the Issue

We speculate that this lack of consensus in the academic literature follows from a number of issues. First, the measure of inequality used in the research varies from study to study, with capital’s share of income, land-holding inequality and the Gini index of income inequality all being used. Each of these measures has advantages and disadvantages.

For example, although the widespread availability of the Gini index of income inequality facilitates its use in cross-country comparisons, it may be a misleading proxy for wealth inequality. By the same token, while land ownership inequality as a proxy marks an improvement in that it directly assays the concentration of holdings of an important asset class (especially in low-income, agrarian economies), it doesn’t fly for highly industrialized countries like the United States where other sources of wealth – housing, tangible and intangible capital – dominate. And while as scholars we would prefer to use a rigorous, inclusive measure of wealth inequality such as a Gini index of wealth, few countries outside of the OECD systematically collect the data on household wealth needed to create such an index. This is where our own research fits in.

Billionaires as Proxies

In 1982, Forbes began curating a list of the 400 richest Americans, introducing the phrase “the Forbes 400” to the popular lexicon. Five years later, the magazine set out on the even more ambitious task of compiling a list of billionaires from around the world. We’ve used these estimates to generate three proxy measures of wealth inequality. For each country, we added up the wealth of all the billionaires on the list in a specific year and created three proxies for wealth concentration by dividing it by (a) the country’s GDP, (b) its physical capital stock or (c) its population.

Although the use of billionaire wealth as a proxy may seem a bit arbitrary, this measure correlates well with the conventional (but far more limited) data on wealth inequality that exists for about 25 countries – in particular, the share of wealth held by the top decile of the population and the Gini coefficient of wealth.

Moreover, the use of this measure of inequality has three advantages over the measures that have been employed elsewhere. First, billionaire wealth is a far more comprehensive gauge of wealth than holdings of land. Second, by virtue of the fact that our measure looks exclusively at billionaires, we focus on the effects of concentration of wealth at the very top of the distribution. This makes sense because it is where most of the recent action has been.

There is one final advantage of using these novel data. Aggregate measures of inequality, such as the Gini index (whether of income or wealth), do not permit us to analyze what drives changes in inequality in the first place, even as prior research has shown that differences in the source of economic inequality can matter significantly for the outcomes we care about. However, we can get at this issue with our novel billionaire data.

We classified each Forbes billionaire by the criteria of political connections and entrepreneurial status. More precisely, we classified each as either politically connected or not, and as either self-made or primarily the beneficiary of inherited wealth. (The latter category includes second-generation owners actively involved in managing their families’ businesses, like Mukesh Ambani of India or Charles Koch of the United States.) We then estimated the impact of the size of each on the level of democracy to allow for the possibility that they have different impacts.

The task of classifying individuals into politically connected versus unconnected is challenging – it is difficult to find billionaires who are not courted by politicians and whose businesses aren’t significantly affected by government actions. The question we ask ourselves, though, is whether the political connections existed prior to the individual becoming a billionaire and whether there was good reason to believe that these connections materially contributed to the individual’s subsequent success.

In the end, we used a very conservative standard in classifying individuals as politically connected, only assigning billionaires to this group when it was clear that their wealth was a direct product of political influence. For example, we did not classify individuals as politically connected simply because they have been involved in politics themselves, absent further evidence that such involvement in politics led to favors from those in government.

Thus, we did not classify JB Pritzker, the billionaire governor of Illinois, as politically connected because his wealth results from his family’s 85 percent ownership of Hyatt Hotels, which far predates any elected office that Pritzker has held. By that same token, we did not classify Joe Ricketts, the founder of TD Ameritrade, as politically connected. While his son, Pete Ricketts, has had a successful career in Nebraska politics, the elder Ricketts’ fortunes in the financial services industry are not linked to it.

We applied a similar principle of not classifying large wealthy donors to election campaigns as politically connected absent further evidence that those contributions formed the basis for the donors’ gains in wealth. For instance, some of the most prolific contributors in U.S. politics over the past decade have been Sheldon Adelson, Michael Bloomberg and George Soros. Adelson and his wife contributed a total of $480 million to Republicans over the past 10 years. Yet even critics of the now-deceased Adelson would be hardpressed to attribute his wealth to his political connections. The New York Times noted in Adelson’s obituary that his rise began in 1979 when he and four partners started Comdex, the Las Vegas computer trade show, which ended up becoming “the nation’s top computer exhibition.” Likewise, Bloomberg’s and Soros’ wealth can be attributed to the success of their commercial ventures – not to political influence.

In contrast, the countries rife with politically connected billionaires range from more obvious candidates such as Russia and Malaysia, where natural resource billionaires exploited connections to political elites, to countries such as Italy, where media magnates and car manufacturers used connections to politicians to secure great advantages. Also included in the category of politically connected billionaires are the individuals running the South Korean business conglomerates, the chaebols, as well as the billionaires who emerged in Indonesia during the autocratic regime of General Suharto, when ties to the regime were all but indispensable for securing business licenses.

We recognize that our classification of billionaires as politically connected or not is somewhat subjective. But so are most measures of corruption, including the widely used Transparency International’s Corruption Perceptions Index that “measures how corrupt each country’s public sector is perceived to be, according to experts and businesspeople.” As an additional check, though, we examined whether our measure of politically connected wealth inequality, defined as the sum of the wealth of all politically connected billionaires adjusted for GDP, correlated with well-known proxies for corruption. For instance, we found that countries that have a higher fraction of their wealth controlled by politically connected billionaires were rated more corrupt by the International Country Risk Guide of the University of Maryland.

Note, too, that the five countries with the highest levels of politically connected billionaire wealth as a percentage of GDP (Georgia, Kazakhstan, Lebanon, Russia and Ukraine) had a median rank of 126 out of 180 countries on the aforementioned Corruption Perceptions Index in 2019. (The higher the ranking, the more corrupt.) In contrast, for the four countries with no politically connected billionaires (Canada, Netherlands, Sweden and Switzerland), their median rank on the Corruption Perceptions Index was a squeaky-clean 6 out of 180. Thus, while the classification of billionaires into politically connected and unconnected is admittedly subjective, it does correlate with widely used measures of corruption.

What’s a Democracy?

One other issue worth mentioning before previewing our results: There is no hard and fast definition of democracy. At a bare minimum, for a country to be considered a democracy, its citizens should be able to choose their leaders in free and fair elections. This narrow definition is in line with how the noted political scientist, Robert Dahl, conceived of democracy in his seminal work, Polyarchy.

But perhaps free and fair elections, along with freedom of association and expression, are not all it takes for a country to be defined as a democracy. When talking about democracy, what many of us are really thinking of is liberal democracy – a democracy that protects minority rights and civil liberties. This broader concept can be seen in numerous definitions – for example, the definition of democracy provided by Freedom House, whose democracy index includes a measure of Political Rights and Civil Liberties, as well as the Liberal Democracy index of the Varieties of Democracy (V-Dem) project. And, in principle, one could expand the concept even further to include the quality of institutions such as the rule of law.

While most of us living in liberal democracies like the U.S. and UK might find a narrow definition of democracy unsatisfactory, it is unclear whether an illiberal state that allows for universal franchise and truly free elections should be classified as democratic or undemocratic. One could make arguments in favor of either position.

For the purposes of our study, we are agnostic. But we lean in favor of either a very limited, Dahl-like definition or a more expansive one that very clearly and unambiguously defines the concept – like the V-Dem Egalitarian Democracy index. To put a stake in the ground, we chose not to use the most comprehensive definitions of democracy out there because conceiving of democracy so broadly complicates the issue of measurement. A narrower definition is better suited because it strips out the influence of other institutional factors and focuses on the impact of inequality on a country’s inclination and ability to conduct the free and fair elections that are indispensable to democracy.

So, What Did We Conclude?

With those definitional issues out of the way, here’s the approach we used. We assembled data for 149 countries for the years 1987- 2017, and then examined the impact of our measure of wealth inequality (and its subcomponents) using three different measures of democracy: one from Polity, a second from the V-Dem project, and a third, the Continuous Machine Learning index of democracy, from Klaus Grundler and Tommy Krieger. Extending our analysis through 2017 allowed us to examine the beginning of the rise of populist regimes across several democracies and the beginning of a “democratic recession,” or “third wave of autocratization.”

Because it takes time for large institutional changes to occur, we allowed for a fiveyear cause-and-effect lag between wealth inequality and measures of democracy. We first recorded democracy scores in 1992, as Forbes first compiled a global list of billionaires in 1987.

Using these measures in a statistical correlation model, we found that, after controlling for other influences, no relationship could be found between inherited wealth inequality and democracy. However, we did find robust evidence of a negative relationship between politically connected wealth inequality and democracy across the globe using all of the measures from the V-Dem project as well as the Continuous Machine Learning index. In other words, the higher the proportion of billionaire wealth in a country that is politically connected, the lower the level of democracy in that country.

No great surprise there. However, there is one nuance worth noting. The negative relationship was only evident in the estimates for countries that do not allow capital to flow freely across their borders. This result supports the idea that the vulnerability of assets to confiscation is what makes inequality particularly harmful for democracy.

The novelty of our results stems from the fact that simply knowing that a country has a high level of wealth inequality is not enough to draw inferences about its impact on democracy. We need to know whether the inequality comes about because of the natural play of (relatively free) market forces or because elites have been able to exploit political connections.

In hindsight, these results make sense to us in terms of political dynamics. Politically connected billionaires are likely to oppose political change because the replacement of the current regime will threaten the advantages that they enjoy in the economic realm, such as monopoly rights or exclusive licenses.

Comparing three countries with very different political systems – the United States, Indonesia and Russia – is useful in establishing the three largest European countries – France, Germany and the UK – averaged just 5.5 percent of GDP. Nevertheless, the United States scored as well on the Polity democracy index as alongside Germany and the United Kingdom, and 1 point higher than France. Although hardly dispositive, these observations are consistent with the argument that it is politically connected wealth inequality that matters more for democracy, rather than overall wealth inequality.

In contrast to the United States, all of the billionaires who show up on the Forbes list for Indonesia in 1987 and 1992 are classified as politically connected because they benefited substantially from their ties to the autocratic, notoriously corrupt Suharto regime. For example, the list included Liem Sioe Liong and, according to his 1987 profile for the Forbes list:

[Liem] developed close ties with an ambitious young lieutenant colonel named Suharto. By the time Suharto came to power in 1965, he and Liem were already business partners. In 1968 Suharto granted one of Liem’s companies a monopoly for importing cloves. Monopolies in flour and cement, as well as open credit lines from state banks, soon followed.

Although a handful of politically unconnected individuals did make Indonesia’s billionaire list in 1996, connected billionaires continued to dominate. Politically connected wealth peaked at over 7 percent of GDP in 1996 on the eve of the Asian financial crisis. But the subsequent ouster of President Suharto in 1998 reset the scene.

All of the billionaires classified as politically connected in 1996 dropped out of the Forbes list in 2002 and 2007. Although some of them re-emerged in 2012, politically connected wealth inequality clocked in at less than 0.6 percent of GDP – less than a tenth of what it had been just about 15 years earlier. In line with our thesis, Indonesia’s V-Dem scores jumped from about 0.05 in 1996 to about 0.5 in 2002 and remained relatively stable during the rest of the sample period.

Then there’s Russia, which offers yet another variant of the relationship between wealth inequality and democracy. Russia’s post-Soviet development includes two distinct cohorts of politically connected billionaires. The first wave, which joined the Forbes list in 2002, included individuals like Mikhail Khodorkovsky of Yukos Oil and Roman Abramovich of Gazprom Neft (formerly Sibneft), who were allies and cronies of President Boris Yeltsin. Bernard Black, Reinier Kraakman and Anna Tarassova described it well when they wrote that Russia ended up “selling control of its largest enterprises cheaply to crooks, who transferred their skimming talents to the enterprises they acquired, and used their wealth to further corrupt the government and block reforms that might constrain their actions.”

But by 2012, several of the billionaires joining the Forbes list had ties to Vladimir Putin. Hence over the entire period we study, politically connected wealth inequality is high, with the proportion of billionaire wealth in that category measuring 75 percent or higher – among the very highest in the world. As expected, V-Dem scores stay low for Russia during the whole sample period. Thus, even though there is a change in the political guard in Russia, unlike Indonesia, the transition does not bring about democracy. In fact, V-Dem and Polity scores all decline in the 21st century.

As it turns out, rather than a new outsider economic elite challenging the incumbents, a new politically connected elite appears on the scene that owes their fortunes to Putin and has strong incentives (financial and existential) to support the autocrat. One can think of Russia as an extreme case of political capitalism, where the political elite helped the economic elite with massive state contracts and sham privatizations, and the economic elite returned those favors to their political benefactors.

On Further Consideration

Our statistically based results are striking, if not entirely surprising. As with the case of the United States, great inequality of wealth can sometimes coexist, albeit uneasily, with democracy. That’s because the wealth is not primarily a consequence of privileges derived from government intervention. While some political parties may favor business interests more than others, the very wealthy are not threatened (or as threatened) by fair elections. By contrast, in countries where great wealth is dependent on political privilege – everything from monopoly rights to guaranteed government contracts to exclusive rights to import key goods – democracy can be an intolerable risk to rich individuals.

Thinking in terms of public policy responses, we speculate that solving the challenges of democratic breakdown or democratic backsliding is unlikely to involve introducing new taxes on wealth or selective enforcement of anti-monopoly laws. Instead, putting measures in place that would thwart the use of political power to enrich certain businesses at the expense of others may reduce politically connected inequality and with it, improve the durability and legitimacy of democratic regimes.

At the risk of repeating ourselves, how inequality manifests has considerable impact on its potential to damage democratic institutions. Getting rich building a better mousetrap is relatively benign; getting rich by, say, holding the exclusive license to certify meat fit to eat by Muslims in Egypt is not. In the debate over the impact of great wealth on society, nuance matters.