The second quarter’s total nonfarm productivity came in lower than expected at -0.5% (an increase of 0.4% was anticipated). Meanwhile, unit labour costs in the quarter were a little above what was anticipated at 2.0% (highest since fourth quarter 2014), where a reading of 1.8% was expected. Productivity in the fourth quarter is now estimated to have fallen by 2.4%. Unit labour costs remain an incredibly volatile index and subject to fairly large revisions, hence any near-term changes here should be taken with a pinch of salt. On a 12-month rate-of-change basis, productivity fell by 0.4%, versus an unchanged reading in the previous quarter, while manufacturing productivity was a little better, in that it rose by 0.9%. Unit labour costs for this sector rose by 3.1%, after a 5.9% decline in the first quarter. On a 12-month rate-of-change basis, unit labour costs were 1.7% for manufacturing and 2.1% higher for nonfarm business.
There clearly has not been much improvement in the second quarter following already dismal productivity statistics in the previous two quarters. Even on a longer-term trend rate, productivity remains incredibly muted. For example, since 2005, productivity growth has been 1.8% per annum lower than it was in the previous decade, which is a very significant slowdown. This slowdown has been apparent not just in the United States but also in other advanced and developing nations. Part of the reason (we believe) is simply down to the supply glut of labour, which up to now has been apparent globally with the emergence of the Chinese and other workforces, along with the opening up of the formerly communist states. This excess supply of cheap labour has likely offset the need for companies to invest in productivity enhancing technologies due to rising unit labour costs. More recently, however, the slowdown seems strange in the face of developments such as the internet and more powerful digital technology. Some, such as Harvard’s Martin Feldstein, have argued it is due to faulty measurement, and productivity is being underreported because the inflation statistics are not adequately capturing quality improvements, which means that real output is being underreported. Even if this is the case, their estimates would still not account for all of the slowdown. This presents a dilemma for the Fed. If productivity really is so dismal, it raises the possibility of inflation rising sooner than later; if it is actually stronger than reported (as Greenspan believed in the late 1990s), the bank can possibly afford to keep rates lower for longer. For now, it seems to be taking them with due caution.
For a copy of this report or to subscribe to the Economics Weekly or Economic Indicators reports, please contact your William Blair representative.
Richard de Chazal, CFA is a London-based macroeconomist covering the U.S. economy and financial markets.