the poverty of international GDP estimates

UPDATE: My suggested robustness check has been done! Two European researchers, from ICREA-Universitat Pompeu Fabra and the European Central Bank, ran tons of regressions, and report

the PWT 6.2 revision of the PWT 6.1 1960-96 data lead to substantial changes regarding the role of government, international trade, demography, and geography. Overall, our findings suggest that margins of error in the available income data are too large for empirical analysis that is agnostic about model specification.

Antonio Ciccone and Marek Jarocinski, “Determinants of Economic Growth: Will Data Tell?”, September 2009.

HT to Bill Easterly for the pointer.

Domestic prices differ from country to country. Services, especially, are much cheaper in poor countries compared to rich countries. Market exchange rates will thus understate the real GDP of a poor country relative to that of a rich country. Prices are the weights used to add up GDP, so GDP growth rates are also biased when prices differ from country to country. Successive versions of the Penn World Table (PWT), continuing work begun by Irving Kravis, Alan Heston, and Robert Summers (1978), adjust national GDP by measuring it with common international prices, known as purchasing power parity (PPP) prices.

Four economists from distinct institutions (three US universities and the IMF) in joint research examine PWT versions 6.1 and 6.2, “two seemingly minor revisions in the Penn World Table mark 6”, and find huge differences between the two sets of data. This raises doubts concerning the quality of the PWT data, so they urge researchers instead to use national accounts data – at least for comparative analysis of annual growth rates – even though they are not PPP-adjusted.

A puzzling feature is that data – especially for GDP growth but also for the level of GDP and the PPPs – for the same country at the same point in time change across successive versions of the PWT. A stark example of this relates to Equatorial Guinea’s growth rate. According to PWT (version 6.2), it was the second-fastest-growing country among 40 African countries during the two-and-a-half decades beginning in 1975. However, according to the previous version (PWT 6.1), Equatorial Guinea was the slowest-growing country. ….

The rationale for the PWT is to come up with GDP level and growth data that are at common international (the so-called PPP) prices so that the data are comparable across countries. The methodology, however, leads to the construction of GDP growth estimates that are based not on common international prices but on a mixture of international and domestic prices. In this case, it is not obvious that the data are comparable across countries.

Simon Johnson, Will Larson, Chris Papageorgiou and Arvind Subramanian, “Is newer better? The Penn World Table growth estimates”, VoxEU, 7 December 2009.

PWT version 6.3 is now up and running, no doubt with even more changes to PPP-adjusted GDP. Sadly, these data are used widely, and uncritically, to ‘explain’ why some countries grow faster than others. As a robustness check, it might be wise to run growth models on subsequent versions of the PWT. The makers of PWT from the very beginning attached warning labels to the data, but these warnings have largely been ignored. It is so much easier to run regressions than to do the hard work of examining with a critical eye the underlying data and, for that matter, to question the underlying theory – or absence thereof. On the latter, see my 7-part series on “economics as faith” beginning here.

Tags: ,

One Response to “the poverty of international GDP estimates”

  1. You might be interested in two recently published papers. One in the Journal of Development Studies looking at growth rates in Africa.

    Random Growth in Africa? Lessons from an Evaluation of the Growth Evidence on Botswana, Kenya, Tanzania and Zambia, 1965-1995

    http://www.informaworld.com/smpp/content~content=a917873145~db=all~jumptype=rss

    and another in African Affairs having a look at income level estimates in the African Affairs

    Random Growth in Africa? Lessons from an Evaluation of the Growth Evidence on Botswana, Kenya, Tanzania and Zambia, 1965-1995

    http://afraf.oxfordjournals.org/cgi/content/abstract/109/434/77