INTRODUCTION

The idea that the poor – through their vote – represent a threat to democracy and property has inspired research on regime change for decades. From Lipset (1959, 31) to Boix (2011), scholars have expected growth, equality, and democracy to run together. Although it has gone largely unchallenged over decades, the notion that the poor threaten property and democracy is fundamentally mistaken on both empirical and theoretical grounds.

The roots and persistence of this error lie with a failure to properly connect social scientists' standard quantitative measure of income inequality – the Gini coefficient – to equally standard sociological understandings of class structure, in terms of the relative sizes and incomes of different social groups: the incumbent autocratic elite (which we assumes includes large landowners if they exist), rising (relative) economic elites, including the bourgeoisie and in many cases industrial workers, and the poor.

In this chapter, we show that properly connecting Gini coefficients to social structures supports our contention that competition over regime change tends to occur between relative economic elites. We start from the observation that income inequality is typically very low in preindustrial societies. Historically, the onset of sustained economic growth – a secular shift from agricultural to nonagricultural sectors – tends to generate a rapid increase in income inequality. Simon Kuznets (1955) noted this phenomenon decades ago, but students of regime change have not teased out this correlation's political implications.

The key theoretical point is that the combination of growth and inequality in a developing autocracy does not imply a growing redistributive threat from the median voter. As Kuznets noted, economic development has the effect of increasing intergroup variation in incomes, which is what the Gini coefficient actually measures. That is, inequality does not increase simply because the rich are further distancing themselves from everyone else, as Lipset and others appear to assume, but because of the emergence and growth of the bourgeois, middle, and working classes. Members of these groups tend to earn far more than the future median voter, who in nearly all historical cases tends to remain relatively poor. And because they have far more to lose, members of these rising economic groups are also increasingly likely to mobilize to press for political reforms.