Why is productivity growth stalling?

There’s been a flurry of press about stalling productivity growth in the West over the past few years. The usual explanations from economists tend to revolve around low levels of capital investment, poor measurement of certain new forms of innovation, or simply stalling levels of innovation.

I’d like to point out a few more possibilities that have received less coverage. The actuality is likely some combination of many of these factors.

a) The deployment of innovation matters.

A new product may have the potential to influence productivity, but may take a long time to actually be used widely. Electricity, for instance, allowed vast growth in productivity during the course of the 20th century, but took decades to show up in productivity growth, since not everywhere had access to it initially (many places in the world still do not have reliable access to electricity, which is a known factor in their productivity levels), and additionally there were further innovations required to make it useful.

The example currently given by economists is the internet, which now has close to global coverage, but which may require decades of further innovation before we see its full potential.

Additive (i.e. 3D) printing is another good example; while many manufacturers are now using this approach to some extent, there is still vast unfulfilled potential for 3D printing to work with new materials, produce at higher speed, print with greater precision, or simply print at lower cost (cost could be a factor all of its own in this list, since innovative technologies tend to initially be expensive, which limits their deployment).

b) Different sorts of innovation have different real effects on productivity.

Electricity allowed manufacturers to do more; the internet allowed people to have access to the information they required to do their job immediately; the newest upgrade to your smartphone likely won’t do much more than make you feel happy.

Additionally, some sorts of innovation may actually be detrimental to growth. It isn’t unusual for companies to complain about staff playing games during business hours, for instance.

The advent of self-driving vehicles will likely lead to the loss of jobs in areas such as taxi drivers and trucking. This may have some positive effect for companies that currently hire many workers in that area, but will likely have longer-term negative productivity effects for the economy as a whole, at least until those workers have retrained for other jobs (remember, productivity is “per capita”).

c) Demand matters.

A new process may allow manufacturers to produce more product per day, per worker, but at the same time they still need to sell that product. If the demand isn’t there, there’s little incentive to increase output.

d) There may be an education gap that prevents the full usage of some types of innovation to their ultimate fulfillment (note that this theme is controversial).

The example I’ve been giving lately is quantum computing. I recently tested out IBM’s quantum solution, and quickly discovered that it requires a PhD in physics to actually program. Higher-level languages for quantum devices will eventually allow more people to use them, but for now their usage is restricted more by the difficulty in creating new quantum algorithms, than by the availability of quantum computers (which are expensive, but are actually available today, inexpensively, on a time-share basis).

e) Sentiment matters as well.

If workers do not feel like active participants in the benefits of economic growth, there will be little incentive for them to increase productivity. If wage growth picks up again, this may change.