Skip to content
 

What is Russia’s GDP per capita?

$7,600 (World Bank 2007)

$9,100 (World Bank 2007)

$14,700 (PPP adjusted, World Bank 2007)

$4,500 (World Bank 2006)

$7600 or $14,400 (gross national income: “Atlas method” or “purchasing power parity,” World Bank 2007)

$12,600 (IMF 2008), $9,100 (World Bank 2007), or $12,500 (CIA 2008)

$2,637 in 2000 US dollars (World Bank 2007); that’s $3,200 in 2007 dollars

$2,621 (World Bank 2006) or $8,600 (IMF)

Sure, I realize these statistics cannot be calculated exactly, and, sure, I realize there are definitional issues within a country and choices to be made when converting to other currencies. Still, there’s a lot of variation here!

At the very least, this is a good example for a statistics, economics, or political science class to illustrate the difficulties of measurement.

P.S. See here (scroll down to item 3) for why we’ve been looking this up.

8 Comments

  1. Hmmm… if you go back through previous issues of the Bank's World Development Report, you'll see that there is a huge difference between Russian per-captia income at nominal prices and at purchasing power parity, and the latter seems a lot more stable than the former, though both change to what seems to me an astonishing extent.

    2009: 7560 (nominal) vs. 14400 (PPP)

    2008, p. 335:5780 vs. 11630

    2007, p. 289: 4460 vs. 10640

    2006, p. 293: 3410 vs. 9620

    2005, p. 257: 2610 vs. 8920

  2. Daniel says:

    But you're reporting GDP and GNP at different instances – and you have it in terms of PPP, nominal dollars (I guess?) and current U.S. dollars.

    Of course they're going to be different – those are all very different variables. The U.S. would have huge variation on those different calculation methods as well.

    Is there really that much variation in a specific measure – ie, nominal GDP in rubles and the Russian population? If not, then there shouldn't be that much variation in Russian GDP/capita, should there?

    But if you compare apples to oranges, of course it looks different! Look at US nominal vs. real vs. PPP GDP. They vary considerably. It's not a measurement problem – it's that these things are conceptually different.

  3. Phil says:

    As you probably know, but some readers may not, there are at least two standard ways of summarizing GDP. One of them is just to use currency conversion to convert GDPs onto a common scale (often US dollars). If one dollar buys ten Burzukas, and the per capita GDP of Burzukistan is 30,000 Burzukas, then we say that the per capita GDP of Burzukistan is $3000. Simple, and what else could it be?

    Well, it could be something else: if we want to compare lifestyles, then we don't just care how many Burzukas people have, and how many dollars they could buy with them, we also care what a Burzukistani can buy with his Burzukas: how many Burzukas does it cost to buy a house, to buy a loaf of bread, to buy a car, etc. So the other way to put GDPs on a common scale is to estimate how many dollars someone in the U.S. would need, in order to have the same lifestyle as the average Burzukistani. This is called "purchasing power parity" normalization, and it can change things a lot. There is wiggle room in exactly how the normalization is done, because there's not a single normalization factor that applies to the entire economy: food might cost only a few Buruzkas while computers cost a boatload of them, for example. But pppGDP per capita is certainly better at gauging "quality of life" than is raw GDP. For the very small amount of work I've done involving GDPs, my economically educated colleagues always told me to use pppGDP.

    At a minimum, never mix the two (don't use GDP for some countries and pppGDP for others). That would be bad.

  4. Antonio says:

    That's not surprising, though. What is surprising is people claiming they are doing causal inference using cross country data! Think about these proxies for corruption or property rights based on "expert" opinion that people in economics and political science use!

  5. bccheah says:

    There are measurement issues:

    1. Converting to U.S. Dollars – do we use current exchange rates and what does "current" mean (usually averaged over a year)? Or do we use PPP exchange rates? Economists tend to prefer PPP. (Discussion here: http://internationalecon.com/Finance/Fch30/F30-7….

    2. Gross National Income and Gross National Income are two different things. (Discussion here: http://www.oecdobserver.org/news/fullstory.php/ai

    3. Possible data revisions should also be expected. The source for cross country income comparison is usually the World Bank World Development Indicators database using PPP. Another source is the Summers-Heston Penn World Tables. http://pwt.econ.upenn.edu/. I haven't looked at the PWT in a while but it may not have the same coverage as WDI.

    I would guess that the ones on the low side are current exchange rates and the PPP exchange rates are on the higher side.

    Then there are also philosophical issues – happiness index, human development index, you name it – they've all been tried.

    Then there are also cross country comparisons of GDP growth rates? Should these be based on local currency measures (e.g. dinars) rather than those based on U.S. Dollar? Most would say that the former matters more than the growth rates based on U.S. Dollar measures.

  6. Andrew Gelman says:

    To the first two commenters: Yes, I understand that there are different definitions. And of course the numbers are different. But (a) I'm surprised that they're so different, and (b) I'm surprised that a quick trawl on the net gave us 10 different values!

  7. bccheah says:

    Are you surprised that they're so different because the definitions of GDP and GNI are so different or are you surprised that GDP (or GNI) estimates they are so different because exchange rates can vary so much within a given year?

    There are courses or at least workshops on these measurement issues. Check the Penn World tables website. The folks there do a great job trying to make sure that we're comparing apples with apples.

    A little nit picking here – in the estimate where you translated $2637 in 2000 dollars to 2007 dollars using US inflation rates, would there not be another estimate which converts first to Russian rubles then apply inflation rates of the Russian Federation and then converting that back to 2007 dollars? I would also say that estimates (1), (3) and (5) are pretty much in agreement so it really isn't 10 different estimates.

  8. Andrew Gelman says:

    Bccheah: I was surprised the numbers varied so much. I can accept that the PPP adjustment can make a big difference, and I know that exchange rates fluctuate–but varying from $2600 to $14000? That's enough variation that I don't know what to do with these numbers at all–except to be careful to get all the country numbers from the same source.