The Cost of Productivity

The US has a productivity problem, and it isn’t what you think. The problem isn’t that productivity has stagnated or is growing too slowly. The problem is too many people and organizations believe that productivity is a (or even the) key factor for an effective economy and the ability to improve our standard of living.

Productivity is measured by total outputs divided by total inputs. The Bureau of Labor Statistics’ (BLS) non-farm labor productivity data is a common reference point for economic productivity. This measure relies on the number of hours worked per unit of output, meaning that people are the inputs.

In this model, automation increases labor productivity by reducing the number of hours (e.g., people) needed to achieve the same level of output. Additionally, getting people to work harder or more efficiently without an increase in the hours worked increases productivity. This means that productivity is a function of getting people to work harder or reducing the need for them to work at all.

For instance, a grocery store that installs self-checkout kiosks increases productivity because they need fewer checkers. Replace nearly all the checkers with self-checkout, which some stores have done, the measure of productivity can skyrocket.

Amazon infamously tracks productivity of every warehouse employee to the minute. They know the exact number of minutes each worker is idle and use the collected data to aggressively manage their staff. This level of tracking and productivity measurement allows Amazon to maximize productivity. But it also has resulted in employees not taking lunch or bathroom breaks and high attrition rates. UPS uses similar productivity management practices whereby they track the movement of their trucks and the number of deliveries each day. What value is created by managing productivity so closely and aggressively? And who benefits from whatever value is created?

Increased productivity is meant to improve the standard of living – which ostensibly means our quality of life – as represented by “the level of income, comforts, and services available…to a society or a location”. It is hard to see how the practices described above, which are all intended to improve labor productivity, contribute to an improved standard of living for the people performing those jobs.

Bureau of Labor Statistics' reporting shows that productivity is increasing nearly twice as fast as compensation. And the Pew Research Center demonstrates median income growth of less than 1% per year from 1970 to 2018, with just 0.3% income growth between 2000 and 2018. At the same time, the share of income and wealth grew substantially for upper income households with a nearly equal and corresponding decline for middle income families. Again, steady increases in productivity haven’t translated into improvements in the standard of living for most people when measured by income and wealth.

In his February 7^th, 2023, article in the New York Times, Paul Krugman highlights the challenges with using productivity to measure economic success. He points to a decline in productivity for the period 1973-1979 compared with 1947-1973, with productivity dropping to a 1.2% annual gain, down from 2.8%. Some of the decline is attributed to increased government regulations which slowed construction and manufacturing because companies needed to spend more time on compliance.

By the traditional measure of productivity – increased productivity is the key to an improved standard of living – the standard of living might be understood to have declined because productivity dropped nearly 60%. But during that time, Krugman notes, regulations implemented by the Occupational Safety and Health Administration (OSHA) in 1971 resulted in a 72% decline in the incidence of nonfatal occupational injuries and illnesses. And the EPA passed the Clean Air Act in the early 1970s which has resulted in obvious, tangible improvements in air quality country wide. So, while productivity decreased, it is very difficult to argue that a drop in workplace injuries and major reductions in air pollution don’t represent an improved standard of living or contribute significantly to our quality of life.

People don’t exist to support the economy – the economy exists to support people as Krugman notes at the end of his article: “[T]he broader lesson is that measured productivity isn’t the only thing that matters. What, after all, is the economy for? The goal is to improve people’s lives…We could have a bigger economy if we were willing to have filthy air and a lot more injured workers, but that’s not a trade-off we want to make.”

Improving labor productivity can have upsides, but those upsides must be shared with everyone for them to have real value. For instance, improved productivity could allow employees to work a 32-hour week without a change in compensation. In that case, quality of life is materially improved, providing a reason to care about labor productivity.

We must think about how changes in labor productivity impact the economy, the community, and the environment rather than looking at productivity in isolation. Otherwise, we are merely looking at a metric without knowing if real value is being created, and who it is being created for beyond the usual suspects.

It is time to find a new measure of success.