Latest Issue
Vol 8, Issue 3, September 2017 Global Policy’s September 2017 issue contains, among others, research articles on the BRICS’ new development bank, the G20 and climate change, the human right to health, and digital diplomacy. It has a special section on ‘Policy Instruments for Innovation, Investment and Global Trade’

Paul Collier - Implications of Egypt for Autocrats

Paul Collier - 13th February 2011

Two decades ago the ‘end of history’ supposedly arrived: democracy was the only conceivable future. Instead, during the past decade China’s success rehabilitated autocracy. Now the uprisings in Tunisia and Egypt appear to have restored the primacy of democracy: even the most entrenched autocrats are revealed as vulnerable to the mass coordination of protest enabled by the information revolution.

Yet Egypt does not imply that the ‘end of history’ thesis was right. Rather, it suggests that autocracy is too crude a category. Judged by their ability to deliver economic growth, the key difference between autocracy and democracy is not their averages but their dispersions: the performance of autocracies varies massively. There are several potential explanations for this variation. Lacking legitimate mechanisms for the transfer of power, autocracies might be more prone to instability. Lacking checks and balances, they might be more prone to large swings in policies. However, one obvious possibility is that because autocracy hands unconstrained power to a few individuals, growth is affected by their particular ambitions and capacities.

In a well-known study, Jones and Olken posed the question ‘do leaders matter’. Using a global sample which included both autocracies and democracies, they tested whether changes of leader altered the growth rate. They excluded those leadership changes in which causality might have been reversed, with performance determining the duration of power. Hence, they confined their analysis to those changes which were due to the natural death of the leader while in office. They found that such changes of leadership significantly affected the growth rate (for good or bad): the person of the leader indeed matters. In our latest work, Anke Hoeffler and I have revisited this result, distinguishing between those changes of leader which occurred in the context of a genuine democracy (defined by clean, contested elections), and other polities which were either openly autocratic or, effectively autocratic because of electoral manipulation. We found that the Jones and Olken result was entirely driven by those changes of leadership that occurred in autocracies. In these polities the person of the leader substantially affected the growth rate, whereas in the genuine democracies leadership change had no significant effect. The obvious interpretation of this result is that the discipline imposed by democracy pinions leaders to a relatively narrow range of policies regardless of their personal whims.

An implication is that autocracy is too crude a category. Such regimes must be assessed according to the ambitions and capacities of their leadership: the key issue is whether they are Pharaonic or Platonic. In a Pharaonic autocracy the leadership is predatory, sustaining its power through patronage financed by distortionary policies: Egypt has been such an autocracy. In contrast, the Chinese leadership is a modern approximation of Plato’s ideal government: one composed of philosopher guardians. Admittedly, while Plato regarded philosophy as the essential skill set, China is ruled by highly qualified technocrats. While not philosophers, the objectives of the Chinese leadership are manifestly not predatory: they have publicly asserted that to maintain social peace the economy must grow by at least 8 percent per year. The evidence from the street in Cairo and Tunis suggests that this assessment is right: open access to information has made the cocktail of unaccountable government and despairing poverty intolerable. Yet sustained rapid growth is incompatible with the predatory policies that Pharaonic autocracies need for the patronage which is their rationale.


The implication of the uprisings is not that the ‘end of history’ was right after all, but that the Pharaonic autocracies which predominate in both the Middle East and Africa face a stark choice. Twitter and the internet have decisively tilted the balance of power from autocrats to the street: if repression could not save Mubarak, few can rest secure. By its nature the power of the street depends upon coordination which cannot be sustained. This may tempt the Pharaonic autocrats to hang on by making temporary concessions. But such tactics would backfire. The street knows only too well its inability to sustain protest, but this gives it no option but to press to the point where change is irreversible. Once it has neutered the regime’s capacity for violence this is its only rational strategy.

Autocrats now have a pressing need to pre-empt the street. Their stark choice is thus to hand over power either to democrats or technocrats. As autocrats wake up to this new reality, expect the rise of the technocrat. Many autocrats will find effective technocracy beyond their reach, but the very attempt will be beneficial.

This article originally appeared in Social Europe Journal.