Even though it is a question that frequently arises, comparing income levels or living standards between different countries is more difficult than people think. And when people don’t understand some of the potential snares involved, they can reach erroneous conclusions that lead to bad policy decisions.
This is prompted by a posting on journalist Felix Salmon’s Counterparties blog on Reuters about unrest in Turkey. Describing increasing prosperity under Prime Minister Recep Tayyip Erdogan, it stated that per capita gross domestic product had tripled over the 2003-2013 decade he has been in office. It included a link to a Bloomberg-BusinessWeek article that stated it somewhat differently: “per capita income, now more than $10,000, has nearly tripled.”
The claim that either income or GDP have tripled brought me to a halt. The Turkish economy certainly has prospered. But, while it is not impossible, a tripling of per capita incomes or GDP in 10 years would be rare indeed. To do so, either would have to grow at a compound rate of 11.6 percent a year. And because Turkey is a country that still has a growing population, per capita income, and the GDP figure from which it is derived, would had to have grown even faster to keep up.
A quick Internet search brought me to an official Turkish government website that listed “average annual real GDP growth” of 5 percent from 2002 through 2012. This is about the best among wealthy and industrialized nations but below the 9 percent or more regularly racked up by China and other Asian countries.
Another Turkish government site gave population estimates of 67 million in 2003 and 76 million for 2013. That is an increase of 1.26 percent per year. If the Turkish government is correct that overall real GDP grew by 5 percent and population by 1.26 percent, then it is impossible for real per capita GDP or income to have tripled.
Yet International Monetary Fund and World Bank’s data cites Turkey’s GDP at $232 billion in 2002 and $794 billion in 2012. That is an annual growth rate of 14.6 percent, nearly enough to triple per capita incomes as asserted in the Reuters and Bloomberg articles.
So what the heck is going on? How could Turkey’s government claim GDP growth of 5 percent and these international agencies show nearly triple that? Who is right and who is wrong and why?
This is where we get to the tricky ins and outs of international economic comparisons.
First of all, Turkey’s figures are described as “real.”
They are adjusted for inflation. If you took “nominal” GDP before adjusting for inflation, you would have a much higher number since Turkish consumer prices now are 2.24 times higher than they were a decade ago.
Are the World Bank-IMF numbers then “nominal” ones, unadjusted for inflation? No, not really. These institutions take Turkish official GDP data for each year, adjusted for inflation in Turkey itself, and convert it to U.S. dollars at the average market exchange rate of the Turkish lira vis-a-vis the dollar for each year in question.
That does remove the factor of Turkish inflation, but then introduces the complication of variations in the lira-dollar exchange rate.
That rate has yo-yoed up and down. And it embodies an implicit assumption that the U.S. dollar is an immutable standard, which it is not.
It took 1.51 lira to buy a dollar in 2002, and it took 1.80 lira to buy one in 2012. But in the intervening decade, the rate fluctuated, falling as low as 1.18 lira in January 2008 as the emerging financial crisis sapped confidence in the dollar.
So the price of a dollar in terms of the lira has risen by 19 percent. This depreciation over a decade is low by historical standards.
But it also is anomalous when one considers relative inflation in the two countries. Econ students learn that such differences in inflation rates are one reason why exchange rates vary over time.
Turkey’s CPI is 124 percent higher than a decade ago and that of the United States only 26 percent higher.
This higher inflation implies that the price of a U.S. dollar in Turkish lira should have risen much more than the 19 percent increase that actually occurred. This relative “overvaluation” of the lira means that when Turkish GDP or per capita incomes are converted to U.S. dollars, the increase in Turkey during Erdogan’s tenure is greatly exaggerated.
More generally, converting incomes from any currency to another at prevailing exchange rates can create erroneous impressions. Consider a Dutch family with a constant 40,000 Euro per year income from 1999 to the present. When the euro was introduced in 1999, the income in U.S. dollars would have been $47,600. But this dollar value would have fallen rapidly to $34,106 by October 2000. Then it began to rise again, hitting $63,022 in July 2008. And this week it would be $52,900.
Now for the Dutch family, ignoring their country’s low inflation, there was no change at all in their income. Yet with it translated into U.S. dollars, they seem to have gotten much poorer and then much richer. But this is only an artifact of exchange rate fluctuations.
These fluctuations really confuse some people. In the waning days of the Clinton Administration, when the U.S. dollar was hitting its greatest strength against the euro, conservative U.S. pundits scoffed that there never was anything to learn from Europe because average incomes in many of its nations were just above the U.S. poverty level. But as the value of the U.S. dollar plummeted in the new century, Europeans seemed to be magically getting richer. The pundits had only demonstrated their own ignorance.
There are attempts to calculate exchange rates that reflect “purchasing power parity,” taking into account different price levels for comparable goods in different countries. These provide some insight but are fraught with other problems.
Summing up, Turkish incomes have not risen nearly as much as many U.S. journalists seem to think. More generally, one should be very careful about deriving strong conclusions from superficial comparisons of international income data.