The productivity enigma (part 1)
Global growth is picking up again, but the recovery from the Great Recession has been unusually weak. There are no signs of a catch-up that might have put activity back onto the pre-shock trend path and swiftly repair the economic and social damage wrought by the crisis. Some see this unusually sluggish recovery as a “temporary” change of pace, with the world economy locked into a period of soft growth until the legacy of past financial excess can be cleaned up. The end of the debt super-cycle has apparently spurred indebted economic agents to straighten out their balance sheets simultaneously and for an extended period of time, leading to a chronic lack of demand and hence of growth. But the problem may be even deeper. The transition to a less leveraged growth model could actually conceal a secular stagnation resulting from slow or declining demographic growth and feeble productivity gains. From this perspective, the pre-crisis credit boom was merely a stopgap period, when demand was artificially inflated by an equally imaginary rise in asset wealth.
Long-term growth determinants
Growth can be fuelled over the long term by an extensive accumulation of the factors of production so as to raise the level of potential output. In particular, the labour supply is correlated with the size of the work force, which in turn depends on demographic trends. In this respect, the data from the United Nations are incontrovertible: with the exception of Africa, population growth is slowing everywhere. This is unlikely to automatically produce additional demand in terms of spending on consumption or new homes, which would stimulate growth by bringing young, dynamic workers into the work force.
Thus our ability to generate growth over the long term and more intensively depends on the efficiency of production processes and thus the way in which the factors of production can be combined more or less effectively. These productivity gains are inextricably linked to innovation and technological progress. It may seem illogical to talk about economic stagnation at a time when a digital tidal wave has triggered a secular period of innovation which, judging by previous industrial revolutions, is likely to have ramifications for productivity and growth.
And yet there is no getting away from the numbers: the productivity slowdown is both a deep-seated trend that began in the 1970s and a global phenomenon affecting developed and emerging countries alike. That trend saw a single five-year interlude in the late 1990s, in line with the boom in information and telecoms technologies, before slowing once again and actually deepening after the crisis.
Measuring the intangible
Proponents of secular innovation believe that the productivity decline is merely a statistical illusion and that its evolution is constantly underestimated because the effects of the digital revolution are hard to capture. This raises the issue of GDP measurement, especially the need to construct price indices that give an accurate breakdown of value and volume. To estimate real production, and hence productivity, it is necessary to adjust for movements in prices, the “market” value of goods and services produced. However, the relative weightings of the two bases are out of date. This poses a problem if new and more competitive products spread quickly because this movement is captured belatedly, meaning that inflation is overestimated and traded volumes minimised. And the qualitative changes factored into goods and services prices also seem to be mis-measured, especially at times when technical progress accelerates. For instance, it is easy to image how hard it was for national accounting offices to take into account the exponential growth in computing power and speed at a time when component prices were plummeting, or to gauge the progress made in healthcare – and thus in labour productivity – due to the rapid spread of new generic drugs. The explosive growth in free online services or smartphone apps, financed through advertising, is also contributing to the debate on the relevance of GDP calculations that do not include things with no value. Nonetheless, although these services enhance consumer satisfaction, they do not lead to an increase in produced wealth, which is related to wellbeing rather than to productivity. More generally, in service economies it is hard to find reliable productivity indicators, particularly for measuring quality effects, because these yardsticks were originally designed to capture physical quantities.
That said, attempts to overcome these drawbacks indicate that measurement problems alone are insufficient to explain the production gap that has widened in the past decade, estimated at around $3 trillion in the US, assuming a productivity rate equivalent to the pre-crisis rate. Neither do they provide any insights into the post-crisis trend break, given that such errors have persisted over time.
Since measurement problems alone are not enough to elucidate the trend decline in productivity, next week we will look at some other causal factors.