And also in many developing nation, a democratic government is relatively weak in power and easy to succumb to some traditional consensus such as sex-discrimination or even class discrimination. They have to comply with people’s ‘tradition’ or ‘culture’, whether it is enlightened or not.
Thus like the paradoxes we find in india and other developing nation, albeit their people are freer in theory, but their social equality is a way worse than country like cuba or china because democratic government can not interfere people’s opionion on sex, religion and castes.

Think back on Peter the Great and Meiji emperor, you will find the main enemy of a reformer is often his own people and culture. In many great country, its modernization procedure is actually a battle between the enlightend idea and the nation’s tradition and the vested interests of the old regional lords. A democratic government is just too weak in such a transitional period.

How, history has produced both benigh dictators and self-serving democratical leaders.

Many heroic figure in backward nation are dictators in nation. Such as Peter the Great of russia. Deng Xiaoping of china. William I of prussia, Meiji emperor of Japan. They are dictator by you definition, albeit their top agenda is to push forward reformationm or modernization which leads to the renaissance of their people and nation.

there certainly are multiple equilibria: a dictator certainly will try to maximize his gains, were I refer to agency theory; either the agent’s optimization can come at the cost of the principles (voters). However, if institutions are weak, the principle’s uncertainty will not foster a good democracy. Dicators sometimes take so much, the people lose. Yet, under different circumstances, the principle is so uncertainy that the consequences of democracy can be worse than under a dictator. If the agent (dictator) cares only about his gains, in the short run principles can be hurt, yet might gain in the future. For example, the dictator benefits from growth in output, such that he can reap more. At the same time, people benefit from output gains, more than when they are active in a chaotic democracy (can the chinese show initiatives if the government does not determine investment)
And is he not more likely to exploit economic resources for his own short-term gain, than for long-term benefit of his population?”

As you say, I agree that good leadership is better, but this is a tautology, not an argument. Given that we have dictators (China? Russia? CR Congo?), can it be true that this is better than democratic society? I belief under circumstances, the answer is a definte yes. fortunately, I argue this is a transition path to good democracy, not a way of good dictatorship.

quote=dwwp
“growth causes institutional improvement, I am not familiar with the study by Gleaser. However, any attempt to suggest causation from growth to institution building must be treated with some skepticism.”

Actually, the nice thing about instruments (such as applied by Acemoglu to which you refer) is that we can determine granger-causality, which comes very close to our concept of causality (e.g. see my study on FDI determinants in China
http://iaes.confex.com/iaes/Boston68/webprogram/Paper3526.html

)

again: under circumstances, growth causes institutional improvement, as in Glaeser. Actually, the colonial history of Africa is not so hopeful when it comes to democracy, therefore again, dictators can be good (Mugabe vs Bonga, Congo?)

please also see my blog:
dutcheconomist

http://dutcheconomist.blogspot.com/2009/08/article-democracy-and-growth.html

There is good reason to believe that both growth and institutional strength might be linked to colonial history (particularly relevant to Africa). See for instance Acemoglu et. al (2001, 2002, 2005) and Luis Angels (Income Inequality & Colonialism, 2005).

But why do you think a dictator would do any better at creating those prerequisites? Would a dictator really create institutions for democracy that essentially make him redundant? And is he not more likely to exploit economic resources for his own short-term gain, than for long-term benefit of his population?

Yes, African countries may require a different recipe for development. But what they really need is good leadership, not good dictatorship.

According to Lipset a society first needs the “necessary” ingredients for democracy, which are supported by education and a strong middle class. These things are missing in African countries, where corruption is the rule, informal employment thrives and people do not trust others in general. How much harm can a dictator do if he can encourage economic growth, after which democracy will follow by itself anyway? For example, Gleaser and others (2004) conclude that “economic growth and human capital accumulation cause institutional improvement, rather than the other way around.”