The Productivity Puzzle

An increasingly hotly debated topic with respect to the medium-term outlook for both the economy and markets is the increasing signs of a trend fall-back in productivity growth in DM economies. Shortly put, lower productivity growth leads to lower economic growth, increased solvency risks in entitlement (social security, health and pension) schemes, less earnings growth potential, a higher risk of “secular stagnation”, lower interest rates (for even longer) and limited room for a further compression in risk premiums in financial markets. For investors the medium-term consequence of weak productivity growth is therefore very straightforward as it creates less investment opportunities, lower return expectations and more systemic risks. Not really something to look forward to.

At the same time, however, investors need to realise it all follows from an assumption - structurally lower productivity growth - that is surrounding by significant uncertainty itself. Moreover, it is not necessarily in line with either historical experience or current observations on technological progress in many sectors in the economy. Historical patterns in productivity growth show large and sometimes longlasting swings, but mainly reveal unpredictability rather than easily identified regime-shifts into either structural accelerations or declines.

Looking at the evolution of US productivity growth, for example, it is clear that the trend has been sliding down indeed over the last 10-years (see above). At the same time, it has less dramatic than sometimes suggested at is still averaged 1.5% growth per year, compared to a 2% growth rates over the for last 55 years. Moreover, it followed on a 10-year boom period between 1995 and 2005 when productivity grew 3.1% on average per year. It might well be that some of the recent disappointments was partially a “payback” for the advances made in the decade before it and basically reflected some sort of reversion to the mean.

The evolution of the level of productivity (see below) does provide some support for this analysis. After a period of disappointing progress in the ‘80s and early ‘90s, productivity levels moved above trend in the early 2000’s after the IT boom. Thereafter, it actually stayed pretty high until the Great Financial Crisis (GFC) and made yet another jump higher in the corporate cost-cutting drive following GFC. Only from 2010 onwards a clear stagnation emerged, but that does not seem something out of the ordinary in the longer term perspective of productivity dynamics. Actually, this type of evolution seem relatively easy to understand against the backdrop the remarkable saga’s of globalisation and the digital revolution in the 1990’s and the fall-out of the GFC over the last decade.

So, historical patterns suggest that a reversion to the (very long-term) mean seems the most unbiased prior for a forecast of medium-term productivity trends. Maybe “this time is different” in productivity space, but for each example of low willingness to invest by the corporate sector or lack of new breakthrough technologies there at least an equal amount of examples of exiting new “digital” business models or fascinating healthcare and (alternative) energy innovations.

Everyday observation in our personal and professional lives bring clear observations of new applications of digital communication and intensified use of smartphones, tablets, apps, cloud technology and big data analysis. Moreover, innovations in the fields of biotech, healthcare and alternative energy lead to accelerating new opportunities and non-linear declines in costs (for example of batteries and solar energy). After globalising production chains through outsourcing in the 1990’s, digitalization is now sharply reducing transportation and educational costs through locally accessible production (3D printing) or downloadable reports, papers, training animations and video (or live feeds). Today, there are more people or connected to the internet through a mobile phone than people in possession of a tooth-brush.

The connectivity of ideas that this creates might just as well be an unprecedented source of intellectual innovation that drives a new boom in productivity over the next decade.

It cannot be excluded that all these somewhat eclectic observations will prove to be digital delusions or just examples of technological innovation than effective economic applications (which is needed to have them raising productivity), but they certainly help to keep us sceptical of the view that the disappointing trend in productivity growth in recent years is the best guidance for its future evolution.