The new economy and the measurement of GDP growth

In connection with a surge in the 'new economy' that is statistically difficult to measure, doubts have been cast both on European growth and its comparability with that of the United States. This article explores measurement problems in the French case. The data at current prices seem to be reliable, but information by product is becoming increasingly difficult to establish. Contrary to widely held belief, the differences in the application of the famous "hedonic methods" have only a small impact on the French data. On the other hand, there seems to be a difference in the treatment of data between France and the United States, or rather between several European countries and the United States, regarding the distinction between final consumption and intermediate consumption of IT products. The United States national accounts record more gross fixed capital formation in software, ceteris paribus, and this automatically produces a higher measure of GDP in recent years. The difference may be the result of different industrial processes, but it cannot be ruled out that it may be merely the result of applying a different statistical convention. In that case, one might then speak of a comparability bias. The method most widely used in Europe -- and which therefore preserves intra-European comparability -- maintains consistency with the results of private accounting, whereas the American method diverges from it. The use of "net domestic product" instead of the usual "gross domestic product" improves comparability with the United States. Taking net domestic product reduces the growth differential between France and the United States in 1999 by half a point.