Competitive Chic

What is competitiveness? The World Economic Forum thinks it knows. Every year, it publishes the Global Competitiveness Report, which purports to measure "the set of institutions, policies, and factors that determine the level of productivity of a country." Understandably, countries eager to attract investment and tout their economic progress take notice of the report. But is the WEF actually onto something?

Judging by the global media’s reaction to the 2013-14 report, which came out last month, the WEF has caught lightning in a bottle. The Wall Street Journal mocked Argentina for its slide in the rankings, and the Indian Express bemoaned India’s lowest-ever position in the competitiveness table. African countries including Namibia and Ghana were dissected in depth by their local outlets, and even in Bhutan, where the government ostensibly pursues "gross national happiness" rather than material living standards, the report made national news.

This attention undoubtedly pleases the WEF, whose intended audience includes leaders from government, business, and civil society. "We are aiming to have all of these groups be interested in the results, because all of them have a stake in the prosperity of their economies," WEF chief economist Jennifer Blanke said in an interview last week. "We want it to be used. We want it to be a reference document." In fact, Blanke added, some governments already use the report this way. "It particularly focuses attention among the policymakers," she said. "There’s a number of countries that integrate it into their own strategies."

Clearly, a lot of people think competitiveness is an important goal. But what exactly is the WEF measuring? It uses a complex formula devised under the leadership of Xavier Sala-i-Martin, a professor of economics at Columbia University, to weigh dozens of factors, from the extent of marketing in business to Internet access in schools. For all of its complexity, however, the WEF’s definition of competitiveness sounds rather like the definition of "total factor productivity," or TFP, a term that economists have used for decades to refer to aspects of growth not explained by changes in capital and labor.

I asked Blanke if the WEF’s concept of competitiveness was indeed the same as TFP. "Yes, basically," she said. "I mean, how well are you doing things." But later in the interview, she gave a slightly different spin. "We’re not only interested in TFP," she said. "The things that we’ve seen that matter for driving prosperity over time are numerous and interlinked. We’re not so dogmatic about ensuring that it is mapping any particular indicator, because we’re interested in economic development. I don’t think we want to get more specific than that."

It all sounded a bit nebulous to me, and I said so. "Nebulous, perhaps," Blanke responded, "but that’s why we had to define an entirely new concept which we’re calling competitiveness."

New or not, there is a fairly straightforward way to test whether this concept is indeed an important factor driving economic growth in the same way as TFP. In the Neoclassical Growth Model, taught to students of macroeconomics around the globe, TFP has a very special role. All countries in the model are growing toward limits in their material living standards. To make progress, the countries give their workers more capital, which makes the workers more productive. Poorer countries tend to grow faster, since even a tiny bit of capital can go a long way for workers who start with very little.

In the model, TFP captures everything that goes into production besides capital and labor: technology, economic institutions, exchange of ideas, etc. So all other things equal, countries with similar TFP tend to converge on the same living standards, with the poorer ones catching up to the richer ones. Economists say countries that converge like this are in the same "convergence club." To get into a new club with a higher limit for living standards, a country needs to raise its TFP.

If the WEF’s notion of competitiveness has similar relevance to growth, then we should be able to use it to divide countries into convergence clubs. In each convergence club, per capita income should be negatively correlated with economic growth.

To check this, I compared the WEF’s 2011-12 competitiveness scores for 139 countries and regions to the International Monetary Fund’s 2011 estimates of economic growth (adjusted for changes in prices) and per capita income (adjusted for differences in purchasing power). The IMF doesn’t keep figures for Puerto Rico, so I dropped it.

Going down the rankings in order, there were 35 groups of 10 countries that differed by at most 0.1 points on the WEF’s overall metric for competitiveness, which ranged from 2.87 for Chad to 5.74 for Switzerland. Among these groups of 10, some of which overlapped, the average correlation between growth and per capita income was -0.46. That was a pretty good sign for convergence, considering that the correlation among all the countries in the sample was just -0.13.

For a more refined selection of convergence clubs, I looked at the WEF’s three main subindexes: basic requirements, efficiency enhancers, and innovation and sophistication factors. This time I chose groups where no country differed from another by more than 0.5 points in any of the subindexes. I managed to find three groups of eight countries. In each group, the countries were also within 0.16 points of each other in the overall metric.

In the group containing Armenia, Egypt, El Salvador, Georgia, Greece, Lebanon, Moldova, and Serbia, the correlation between growth and per capita incomes was -0.92. This was driven to a great degree by Greece’s extraordinary downturn; even if Greece had grown by a couple of percentage points, though, the correlation still would have been about -0.55. In a group of wealthier countries — Australia, Austria, Belgium France, Malaysia, Norway, Qatar, and Saudi Arabia — the correlation was strongly positive at 0.58, a sign of divergence rather than convergence. The poorer group of East Timor, Ivory Coast, Lesotho, Madagascar, Mali, Mozambique, Swaziland, and Zimbabwe showed a correlation of 0.22, positive and divergent again.

Taken as a whole, this rudimentary evidence neither supports nor negates a central role for the WEF’s concept of competitiveness in convergence. The fact that convergence is not present at a more granular level suggests that the results for the overall metric may be partly due to chance. To be sure, the metric is highly correlated with living standards, and economists have indeed found causal relationships between some of its components and living standards. Yet the metric itself seems to have a connection that is, well, nebulous. That may be exactly what the WEF intended, but it may also give pause to those leaders who would use the report as a reference.