Automation And Inflation: An Uncertain Linkage

  • Automation And Inflation: An Uncertain Linkage
  • House Prices Remain A Global Concern
  • A New Reservoir of U.S. Labor Supply

In many respects, economists are a little unusual. We think in odd ways, and we arrange data into odd patterns. We find it hard to reach conclusions without significant equivocation. We have difficulty explaining to others exactly what it is that we do.

Given the irregularity endemic to the vocation, I was therefore surprised to see a recent study suggesting economists were likely to be replaced by computers in the coming years. How will a machine be able create the kind of confusion that a human economist can? I guess time will tell.

The topic of technological advance was very much on the minds of the Federal Reserve and the Bank of England this week. (The former raised rates, while the latter opted to wait.) Automation may be altering the way that inflation evolves, causing frustration among central banks seeking to achieve their price level targets.

New equipment and new processes are taking root across industries, with the aim of increasing efficiency. And the possibilities for future transformation are impressive. But while the promise of technology is clear, its effect on present performance is much less so.

There is little doubt that firms around the world are innovating aggressively. The number of industrial robots in service has grown geometrically, while investments in artificial intelligence (AI) are expected to follow a similar path in the years ahead.

These developments represent the intersection of opportunity and necessity. Smarter machines can handle more intricate processing and reduce production times. “Bots,” programs that allow computers to do routine tasks, are having a similar impact on services. (Bots have gotten a bit of a bad name because of their role in election influencing, but Siri and Alexa have been generating more positive interactions with humans.)

The necessity of technology is rooted in aging populations. Slowing labor force growth places a premium on finding new ways to keep up output. As one might expect, the countries facing the most significant demographic challenges have made greater strides into digital assistance. China, among others, is moving aggressively in this space.

AI takes things a step further, enabling robots and bots to learn from experience. As an example, AI is being used to program autonomous vehicles; its ability to synthesize huge amounts of data and calibrate the instructions used to direct cars goes well beyond human capabilities.

The overall trends in automation are both exciting and concerning. It is truly amazing to contemplate what is being done and what lies ahead. But robots (especially sentient ones) also create a certain degree of apprehension. A recent film of robots learning to open doors was both amusing and a little chilling. (It reminded me of the scene from “Jurassic Park” where velociraptors opened doors.)

As we discussed in last summer’s essay “Automation and Anxiety,” this is far from the first time that machines have made inroads into the global economy. Past periods of technological advance created the same kind of mixed feelings that are present today, but they eventually paved the way to higher output, employment and standards of living. For instance, the rise of spreadsheets reduced the need for bookkeepers, but created more work for financial analysts.

Advanced technology has been cited as a governor on the price level, in two main ways. Firstly, digital approaches can increase output and reduce costs. And secondly, the possibility that computers will take on a wide array of functions currently executed by people may lead today’s workers to hesitate before requesting higher wages. Cashiers who press for higher pay might accelerate the adoption of kiosks to take their places.

All of this makes intuitive sense, but the data are not yet aligned with theory. In the past decade, the increasing penetration of robots has coincided with a steep fall in the rate of productivity growth. This is exactly the opposite of what one might expect. Furthermore, people spend a substantial amount of their incomes on services like housing and medical care, which automation is finding more difficult to penetrate. Prices in these areas continue to escalate at a more normal pace.