Andy Warhol supposedly said, "In the future, everyone will be world-famous for 15 minutes." He
actually didn't, but never mind: It well describes the Internet. The latest fleeting star is Jose Luis Yangali Garay, a Peruvian who posted to LinkedIn a colorful video of extreme-poverty rates in Latin America since 1979. In five days, the video has been reposted at least 439 times.
Unfortunately, there is less to it than meets the eye. Garay hasn’t provided the source for his statistics, or even explained how he defines extreme poverty, despite roughly a dozen questions from readers. “Extreme poverty” is usually defined as life on $2.15 per day. For international comparisons, the World Bank computes these amounts using purchasing power parity, which adjusts for changes in the exchange rate. The reason is that foreign exchange rates, as the Mexicans could tell us, often swoop and swoon for reasons that have nothing to do with the structure of the economy. The “extreme-poverty rate”—the share of the population living on less than $2.15—is a statement about economic structure.
The World Bank’s current estimates of
extreme-poverty rates are the gold standard. And Garay’s estimates vary wildly
from them.
An example is Venezuela. Garay’s video
shows continuous extreme-poverty rates from 1980 through 2021, ending with 68%,
more than two-thirds of the population.
But as you can see in Figure 1 below, the World Bank has few estimates
for Venezuela, and none beyond 2006. This may be because of problems in
calculating 2017 international dollars (that is, dollars that have the same
purchasing power everywhere, using 2017 prices) for Venezuela, where data from the
government became increasingly unreliable until President Nicolás Maduro,
enraged by the high poverty rates reported (and they were high in reality),
ordered the statistical agency in 2015 to stop calculating them. (Don’t know
what to do in a crisis? Shoot the messenger!) Since then, a consortium of
universities in Venezuela has calculated poverty statistics through an annual
survey of almost 10,000 residents.
The Survey of Living Conditions (ENCOVI) is a serious effort. But there are problems in reconciling the estimates of two vastly different statistical organizations, one public and one private. In the video, you will see that the reported extreme-poverty rate jumps from 14% or 15% in 2014 (one cannot ascertain the exact number from the video) to roughly 40% in 2015. Is this due to a deteriorating economy, suffering from low global oil prices and from government mismanagement that stoked an inflation rate in 2018 of 65,000%? Or did the change in surveys play a role?
The World Bank does not report average income, adjusted for inflation, for Venezuela from 2015 forward, but the International Monetary Fund does: Real GDP fell 6.2% in 2015. This loss is severe but not likely to be enough to cause the extreme-poverty rate to almost triple in one year. So the conversion in surveys may have hastened the rise. However, in the long run the economy surely accounted for most of the increase reported in extreme poverty: After 2015, GDP fell faster and faster, diving to a breathtaking minus 30%—nearly a third of the economy—in 2020, as shown in Figure 2.
Figure 1
Data source: World Bank
Data source: IMF
Wait, there's worse. Venezuela estimates a poverty line, below which people are considered poor. The poverty line varies from nation to nation; in Venezuela, it encompasses those too poor to buy basic food. Because the poverty line depends on the nation, it is not useful for the kind of international comparisons that Garay attempts. That’s why researchers apply the $2.15 criterion to all nations. Garay does not seem to realize this. But it is not clear whether he uses poverty lines for other nations as well; certainly, his numbers do not line up with the poverty lines reported by the World Bank. It is not clear where he gets his data, period.
So, what do we actually know about extreme
poverty in Latin America? Surprisingly, the rates are below global averages,
although they are not declining as rapidly, as Figure 3 shows.
Figure 3
Data source: World Bank
For the five largest countries in Latin America, rates are volatile—because they are sensitive to average income—but generally declining, as Figure 4 shows. The exception is Argentina, where extreme poverty peaked in 2002, when the government was in the throes of a financial crisis. Rates rose sharply in Colombia and mildly in Peru in the pandemic year of 2020. Remarkably, they fell in Brazil. There is little data for Mexico.
Garay has a talent for visuals that may someday benefit scholars of poverty. But at present, his indifference to statistics so pollutes his video that it does, I regret to say, more harm than good.
–Leon
Taylor, Baltimore tayloralmaty@gmail.com