Demonstrators hold a banner featuring a painting of Pablo Picaso’s “The Guernica” (back) and a cut-out of former Spanish dictator Francisco Franco (C), during a protest against the Spanish government’s new anti-protest security law in Barcelona December 20, 2014. REUTERS/Gustau Nacarino

Joshua Tucker: One of our regular features here at The Monkey Cage is summaries from political scientists of recently published research. We have arranged for articles that are featured in this series to be “ungated” and made freely available to the public for a period of time following the post on The Monkey Cage. The current post is from UCLA political scientist Daniel Treisman describing research from his article “Income, Democracy, and Leader Turnover” in the American Journal of Political Science and will be available for free download  for the next 30 days.

******

Does economic development cause countries to become more democratic? A vast literature says yes. Except for a few petrostates, mostly in the Persian Gulf, almost all the richest countries have responsive and accountable governments.

Yet whenever consensus on this seems about to emerge, objections surface. Influential dissenters point out exceptions and propose alternative theories. Perhaps other factors predispose certain countries both to advance economically and forge democratic institutions, without one causing the other. And what to make of the countries that grow rapidly for years without any hint of political liberalization? What about Spain under Generalisimo Franco, Indonesia under President Suharto, Russia under Vladimir Putin?

In a recent article I suggest a reason why this debate refuses to die. Economic development does lead to greater democracy but not in a smooth, incremental way. At certain times, a country’s income matters a lot for its political evolution; in other periods, income’s influence is muted. What opens such windows is leadership succession. As authoritarian states become richer, they do tend to become more democratic—but the impact of development is concentrated in the early years of new authoritarian leaders.

Take Spain. Under Franco, who seized power in 1939, the country metamorphosed from a rural backwater into the world’s eleventh largest industrial economy. By the time Franco died of old age in 1975, GDP per capita had quadrupled and the number of telephones had increased by more than 250 times. But the political regime remained a brutal and arbitrary despotism.

Yet, within a few years, Spain had shot to the top of the democracy ratings. Historians see a clear link between the country’s multifaceted economic and social modernization in the 1960s and its political transformation in the late 1970s. Economic development set the stage for democracy. But its impact was felt only after the generalisimo left the scene.

This graph—which tells Spain’s story—shows why we may miss the impact of development if we look for only short-run relationships. It plots Spain’s Polity score—a rating which ranges from -10 (“pure dictatorship”) to +10 (“pure democracy”)—along with the country’s GDP per capita (in 1990 dollars). Do the two appear to be related? Looking only at annual changes we would not see much link. In fact, the two lines only move in the same direction for a few years after 1975. Most of the time the Polity score remains flat, seemingly unresponsive to the country’s steadily rising income.

Figure: Daniel Treisman/The Monkey Cage
Figure: Daniel Treisman/The Monkey Cage

Within the tenure of a given dictator such as Franco, increases in income do not prompt democratization. However, in periods in which the leader changed, the link between income and political liberalization is strong. Since leaders always change if you wait long enough, the income-democracy relationship emerges more strongly in statistical analysis in the medium run—periods of 10-20 years—than if one looks at annual or even five-year blocks of time.

Franco is not the only dictator to have held back the tide for a while. The following table lists all 21 leaders of non-democracies between 1875 and 2004 under whom GDP per capita topped $6,000 (at 1990 prices). These dictators were no fans of free government: on average, political institutions became slightly less democratic under them. Yet the pressures they bottled up exploded after they departed. In three quarters of the cases, the decade after the dictator left office saw political liberalization – and often a dramatic breakthrough. On average, these countries’ Polity scores jumped over that decade by more than eight points on the 21-point scale.

What is it about long-lasting dictators that breaks the link between development and democracy? Do dictators change over time, becoming more reactionary the longer they survive in office? Or is it only (initially) reactionary dictators that survive—a mechanism of selection. I found evidence for both, but stronger support for the selection effect. Authoritarian leaders are more likely to liberalize early in their terms than later on. But those that survive for a long time show a lower than average propensity to liberalize even in their early years. Such “survivors” are more likely to have no higher education, to have come of age before their country modernized, and to head monarchies, one-party states, or personalistic dictatorships, rather than military juntas.

If economic development makes it harder for a dictator to pass on his regime intact to a son or associate, why do dictators allow their countries to develop economically?  (Of course, some don’t—but many do.) They might consider an advanced economy necessary for military defense, but that is probably not the only reason.

I provide evidence for another. Paradoxically, high economic growth and the high national income it—ultimately—causes have counterposed effects: the first entrenches the dictator, while the second undermines the dictatorship. Vigorous growth—which boosts household incomes and government revenues—enhances the incumbent’s survival odds. Thus, dictators have a personal interest in growth. But over time that same growth changes society and the ruling elite in ways that make it more likely the regime will collapse after the dictator is gone. What’s good for the dictator is not so good for his dynasty.

Franco presided over dramatic economic development. But after he died, the Francoist regime fell apart. Indonesia’s “New Order” did not outlast Suharto. In Russia, Putin’s period of dominance has coincided with an increase in per capita income (at PPP, in 2011 dollars) from about $12,000 to almost $24,000, according to the World Bank. Will his country fit the pattern?

We will have to wait to see what happens after Putin leaves the Kremlin. Of course, turnover in economically developed autocracies does not always lead to political liberalization: it did not in Singapore, where Lee Kuan Yew handed over to Goh Chok Tong in 1990, when GDP per capita was already about $14,000. Clearly, some dictators in richer countries have managed to select successors who were as reactionary, skillful, or lucky as they were.

Despite its remarkable growth record, China has only recently entered the income range at which continued authoritarianism is surprising. The country’s so-far smooth transition to the leadership of Xi Jinping might indicate that the Chinese Communist Party has crafted a succession mechanism that is more effective at picking regime-preserving leaders. But, as the former top Chinese official Zhou Enlai said in a different context, it is too early to say.