Doing Business 2011: 8 years, 11 indicators, and a new 5-year cumulative score

This page in:

DB2011 The new Doing Business 2011: Making a Difference for Entrepreneurs report has just been published, the 8th in the Doing Business series. Let me quickly dispense with the most-watched metric, the 10 economies that have improved the most on the overall Ease of Doing Business index during the last year. This year it's Kazakhstan, Rwanda, Peru, Vietnam, Cape Verde, Tajikistan, Zambia, Hungary, Grenada, and Brunei Darussalam.

Of course, whether an economy manages to pull off a series of regulatory reforms in a single year requires a bit of luck. While policymakers can often cut the days to open a business with relative ease, it's rather more difficult (and more time consuming), say, to increase the percentage of the population covered by a credit bureau. So I was pleased to see that the Doing Business team has drawn on their longitudinal dataset to construct a measure of cumulative change for the past 5 years—the "DB change score". (Click on the figure below to see a bigger version.)

Db big

Following this metric, we get a somewhat different group of economies that have consistently reformed their business regulatory environment: Georgia, Rwanda, Belarus, Burkina Faso, Saudi Arabia, Mali, Kyrgyz Republic, Croatia, and Kazakhstan. There's some overlap between the one-year measure and the five-year measure, but the 5-year measure features a larger contingent from Eastern Europe and Central Asia, reflecting the fact that this region has consistently been the most aggressive in pursuing regulatory reform.

I suspect that this year, as in past years, some folks will reasonably ask some version of the following questions: "If (country X) has done so well in the Doing Business report, why has (country X) suffered so much during the financial crisis? And why has (country Y) done so well when it hasn't reformed its business regulatory environment?" Put very simply, the Doing Business reports don't measure everything that matters for economic growth, nor do they aim to. Things like proximity to markets, the stability of financial systems, the rise and fall of commodity prices, etc., all have an impact as well.

That said, I would be quick to point out that noone really knows exactly in what order which ingredients have to be mixed together to achieve economic growth. Even the Commission on Growth and Development came to this conclusion. Nevertheless, we do have bits and pieces of evidence that point to the importance of reforming the business environment. The evidence is strongest on the impact of business entry reforms (i.e. cutting the costs and procedures to open a business). For instance, one study looked at the impact of the introduction of one-stop shops in Mexico and found that the number of registered business increased by 5% after the reform. This short Viewpoint provides a summary of the evidence in this area.

How can researchers make progress on these issues? In my view, there is a gap that needs to be bridged between two schools of research: the body of econometric research on business environment that tries to make generalizations based on the average impact of reforms, and the Ricardo Hausmann/Dani Rodrik "growth diagnostics" school of thought that tries to identify the binding constraints in particular economies. In some cases, the binding constraint may be found in the business environment, and policymakers are right to expend limited political capital on these types of reforms. In other cases, other aspects of the economy may be holding growth back, e.g. a repressed banking system or low educational attainment. It would be helpful to provide policymakers advice on how to identify when the business environment is the key binding constraint. Together with companion datasets like Investing Across Borders, Enterprise Surveys, and Women, Business and the Law, this should be within reach.


Ryan Hahn

Operations Officer

Join the Conversation

The content of this field is kept private and will not be shown publicly
Remaining characters: 1000