Robert Feenstra is a Professor at the Department of Economics, University of California, Davis, and holder of the C. Bryan Cameron Distinguished Chair in International Economics. He gave a keynote address on ‘The ‘China Shock’ in Trade Reconsidered’  at the GGDC Conference 2017, marking 25 years of the centre. Feenstra works closely with FEB’s Robert Inklaar and Marcel Timmer on producing the Penn World Table. FEB Blog caught up with him to discuss how he has used the Groningen Growth and Development Centre’s datasets in his work.

Q. How have you used GGDC data in your work?

A. I am more of a producer of data for GGDC than a consumer, because I collaborate with Robert Inklaar and Marcel Timmer on the Penn World Table (PWT). I use these data myself when doing work on China, and when computing the quality-adjusted prices in international trade that feed into the calculation of new variables in the PWT database.

Q. Could you describe the reputation of GGDC in the economics research community?

A. The GGDC is well-known, of course, for the historical work of Angus Maddison and for the work by many faculty dealing with the measurement of sectoral productivity across countries. By taking over the PWT the GGDC is extending this work in both historical and productivity comparisons.

We have improved the measurement of real GDP across countries by introducing real GDP on the expenditure side and on the output side. These two distinct concepts are important for research in macroeconomics. Real GDP on the expenditure side is what has traditionally been measured by PWT, reflecting the cost of obtaining final goods. Real GDP on the output side is what you would end up with if you were able to measure the real value-added of each and every sector in the economy, and then add this up across sectors. But it is too difficult to make that calculation for all countries. So we take a short-cut by using countries terms of trade (i.e. quality-adjusted prices of exports and imports) to measure the difference between real GDP on the expenditure side and on the output side, and then basically infer the latter using data on the former and on the terms of trade.

In addition, the new methods used in PWT can be pushed back in time to “rebase” the calculations of Maddison for early periods, as done in recent work. I believe that economic historians will embrace these new calculations.

Q. Is there a particular data point from GGDC data that you remember as illustrating an important trend?

A. According to PWT 9.0, the real GDP of China exceeded that of the United States in 2014. That is more of a milestone than a trend, and it will not be reversed.

Q. What research that has come out of GGDC do you think is particularly valuable?

A. Well, I think that that PWT is very valuable, of course. This dataset was developed in the 1960s by work at the University of Pennsylvania, and it therefore pre-dates the recent trend towards randomized controlled trials (RCT) in development economics. The PWT data is better able to give an overview of the living conditions and productivity across countries. A friend of mine, Steven Radelet, has recently used the PWT data to argue persuasively that the developing world is better off today than at any time in the past.

Further Reading:

• Robert C. Feenstra, Robert Inklaar and Marcel Timmer (2015), “The Next Generation of the Penn World Table”, American Economic Review, vol. 105, No. 10, pp. 3150-82