For some decades now, economists have assembled data measuring “economic freedom” in order to create an index that can be tracked over time. The idea of an index of economic freedom is to measure the extent to which individuals are free to pursue their self-interest and how secure are the fruits of this pursuit. Some 400 articles have been written using this index.
A substantial share of those articles deals with the relation between prosperity and economic freedom. Most of these articles conclude that there is a positive relationship between growth and economic freedom: freer countries are richer countries and countries that increase economic freedom speed up growth.
Many scholars have expressed skepticism towards these measures arguing that it was hard to “put a number” on institutional quality. Thus, any empirical finding derived from using this variable are deemed to be suspicious by these skeptics.
However, there is an important point that the skeptics fail to realize that should increase their willingness to accept the results. The claim that growth and economic freedom are positively related is a conservative one as the data is biased against finding a positive relationship. This is because growth is a concern shared by all regimes, free and unfree.
Dictatorships or authoritarian regimes (which also tend to be economically unfree) often put a lot of pressure on their bureaucrats to develop the economy and increase output. More often than not, bureaucrats will be rewarded for achieving the objectives set by the central government. Punishments are also handed out for failing to meet out the objectives. This is a potent incentive to “fudge the data”.
To best consider the role of incentives that may induce biases in the production of data, Cuba’s demographic statistics is a good example. Cuban physicians have to meet targets of infant deaths that are set by the central government. Because the regime uses infant mortality rates to vaunt their regimes, physicians who fail to achieve the centrally-defined targets are punished. This causes them to reclassify certain deaths that occur after births (known as early neonatal deaths) as pre-birth deaths (known as late fetal deaths) which are not included in the infant mortality rate. This artificially deflates the infant mortality rate and improves life expectancy at birth. Because the government cares so much about this particular outcome, there are incentives to make the statistics say what the government wants them to say.
The same happens to estimates of the size of the economy. In numerous countries with authoritarian governments, the central government defines target rates of economic growth. Local bureaucrats who face punishments for not meeting targets (and potential rewards in case of success) thus have an incentive to use methodologies that are bound to overstate the size (and growth) of the economy. This is notably the case in China where regional party leaders exert considerable pressure in the process of data production regarding local economies. When adjustments are made to account for evident manipulation, China’s rate of economic growth between 2010 and 2016 is 1.8 percentage lower than the official numbers.
To measure the extent of the data fudging for all dictatorships, Luiz Martinez of the University of Chicago found an ingenious solution. Thanks to the many satellites orbiting the planet, we can measure artificial lighting at night. This light is strongly correlated with levels of economic development.
However, unlike government-produced measures of economic development and activity, that data cannot be manipulated. Martinez assumed that the light data was a true representation of economic activity which could be used to measure the extent of the data trafficking by type of political regime.
Martinez found that the numerous dictatorships of the planet overestimate economic growth by factors of 1.15 to 1.3. As way of example, this means that when Chinese authorities report an economic growth rate of 6.6%, the true growth rate is closer to 5.1%.
Why is all this relevant to the empirical finding regarding the relation between prosperity and economic freedom? Because authoritarian regimes also tend to be less economically free. There is thus a correlation between economically unfree and upwardly biased GDP estimates. This indicates that the data is biased against finding any positive effect of economic freedom on growth.
This is an important bias that makes it all the more surprising that the empirical literature tends to find that economic freedom is associated with development. This bias ought to make all skeptics more marginally inclined to embrace the finding that more economic freedom is linked with increasing prosperity.
Vincent Geloso, senior fellow at AIER, is an assistant professor of economics at King’s University College. He obtained a PhD in Economic History from the London School of Economics.
This article was sourced from AIER.org