Does Governance Matter?


Charles Kenny is a Senior Fellow at the Center for Global Development and a Schwartz Fellow at the New America Foundation. His latest book is Getting Better: Why Global Development Is Succeeding—and How We Can Improve the World Even More (Basic Books, 2011).

The institutional theory of development appears to have swept all before it. Those traditionalists who argue for the centrality of investment for growth are told that institutions determine its quality and extent. Unreconstructed supporters of the primacy of policies are schooled that governance lies behind good policy selection and strong implementation. All that is left to discuss is “what causes strong institutions,” to which the answer appears to be “a whole load of history.” Indeed, the whole question of the wealth and poverty of nations sometimes appears reduced to an argument between those who favor soil type and those more concerned with disease burden at the time of European imperial expansion a century and a half ago or more.

And if history really matters, the predominant literature suggests that's because institutions are slow to change. A recent paper by Michael Woolcock, Lant Pritchett, and Matt Andrews suggests that at current rates of progress, it would take Zimbabwe 184 years to reach the institutional quality of Singapore as measured on the Kaufmann, Kraay, and Mastruzzi governance effectiveness index—given where it is today and assuming it started at zero at independence. It is hard not to get a little depressed about the prospects for poor countries as a result.

There is a derogative term used by British historians to describe narratives that present the past of rich countries as an inevitable progression toward enlightenment and wealth—it is called Whig History. It might be fair to say that institutional economics has been hit with a dose of the Whiggisms. For a start, it is clear from patterns of economic growth that there must be far more to income change than (glacial) institutional change—countries can get richer without reverse-engineering their history to include a Magna Carta. Second, it appears that some institutions can actually change reasonably fast. Third, countries weakly and strongly governed alike have seen dramatic progress in health, education, reduced violence, and improved respect for human rights.

First off, if governance features related to colonial history matter so much to growth outcomes, why do growth patterns look so random? Bill Easterly, Lant Pritchett, Michael Kremer, and Lawrence Summers found that growth across time was very weakly correlated—so that less than a seventh of relative growth rates across countries persisted across 15-year periods. Long-term historical explanations cannot explain why Latin America grew quite fast until the 1970s and then slowly thereafter—or why China and India reversed that pattern. Or, indeed, why Africa has taken off in the new millennium. These are rather important phenomena unconnected with which countries had a compass and ocean-going ships in 1500.

That's not to say history doesn't matter. Not least, rich countries yesterday are by and large rich countries today. And governance matters, too. But history isn't everything, and weak governance is neither unfixable nor an insurmountable obstacle to progress. It appears that countries can overcome their weak institutional endowment to do very well, thank you. It may be, as Douglas North argues, that the Magna Carta and the Glorious Revolution made Britain rich while absolute monarchy made Spain poor—but even so, and even after the global financial crisis, both are now at the top end of the global league tables of income.

Related to this, there are in fact examples of fairly rapid institutional change. Countries rarely change their constitutions or adopt whole new legal systems (their “grand institutions,” if you will). But they do adopt (and enforce) new laws and regulations that make a big difference to the quality of life of millions. Take the seemingly intractable case of reforming water utilities. Phnom Penh has seen the quality and reach of its water supply extend dramatically with the introduction of market pricing, a decline in nonrevenue water from 72% to 6% and customer-responsive regulation, introduced between 1993 and 2009. Piped water coverage increased over that period from 40% of the city to over 90% in 2009, while water service increased from an average of 10 hours a day to 24/7.

Or what about leakage of the financial variety: the percentage of central government financing for equipment in Ugandan schools that actually reaches the schools has increased from next to nothing to next to everything over the past 10 years after newspapers published how much money the schools were meant to get.

Even the attitudes that underlie institutions can rapidly adjust—surveys suggest that discrimination against Dalits (“untouchables”) in India is rapidly and dramatically on the decline, for example, so that very few lower-caste people in India are still performing their traditional roles as bonded laborers, tanners, or night-soil collectors.

And even if you remain convinced that the grand (immutable, historically determined) institutions of a country determine both income and growth, these same institutions cannot explain progress in the broader quality of life. That's because progress in health and education and civil and political rights has been occurring everywhere, even in countries that have seen little or no income growth. While the most healthy countries are also the richest, progress in income and health is very weakly related—so while East Asia may be the miracle region in terms of economic growth over the past 30 years, it is the Middle East and North Africa, an economic snail, that leads the world when it comes to growth in life expectancy—the regional average climbed from 55 to 70 years between 1975 and 2005.

Indeed, the great majority of countries that have gone backward in terms of per capita GDP have seen progress in the quality of life. In the 12 countries worldwide for which we have data where incomes were lower in 2005 than 1960 (by an average of 27%), life expectancy climbed by an average of 10 years and literacy rates close to doubled. If institutions matter to both income and the broader quality of life, they must at the least be very different institutions, then. And the ones connected to health and education must be improving worldwide at a rapid clip.

Again, institutions do matter to outcomes in broader of quality of life. That only 1% of Chad's nonwage health budget officially allocated to front line clinics actually reaches those clinics has a real impact on the quality of care—and of health—in the country. And improved government provision alongside private sector regulation will become increasingly important as we defeat the most basic global development challenges (infectious disease, getting kids into classes) and move on to more complex areas (noninfectious disease, actually teaching kids in the classroom). But still, there is way more to life than strong governance and—perhaps even better news—governance can change in time frames considerably shorter than the tectonic. So ignore the counsel of despair emanating from development economics departments across the country—development appears to be the default state worldwide.