Hacker News Reader

Comment by armchairhacker | Hacker News Reader

How is "productivity" being measured? The only source I can see is https://twitter.com/GregDaco/status/1654126098794565639, where "productivity" is a quantity, is there a economics-specific definition?

Also:

> U.S. productivity plunged 2.7% in the first quarter of this year compared to last year, EY found. That’s a 0.9% year-over-year drop. Concurrently, quarter-over-quarter output grew slightly (0.2%), and hours worked grew 3%. That means people are working longer hours and barely putting out more products, because they just aren’t as productive as they used to be.

Is 0.9% significant enough to not be a rounding error (especially considering "productivity" is a very broad measurement). I want to hear devil's advocate for how this shows anything.

The rest of the article is speculation from a chief economist and 2 CEOs that this is caused by WFH and increased job-hopping.

View on HN

On that chart, labor productivity is defined as output per labor hour.

You can think of it as company revenue per hour worked.

This is a standard definition: https://www.bls.gov/k12/productivity-101/content/what-is-pro...

I don't know that it's a very useful metric for businesses that are capital intensive. For example, say you automate your factory and have only one employee left, who is in charge of turning the machines off each day. That person's labor productivity would be off the chart, but it's not because they've become more 'productive' in the regular sense of the word.

The definition says “output per hour worked”. Company revenue would depend less on individual workers and more on strategic decisions from upper-level than, and target audience spending above all.

But how do you measure worker output? Even “units produced” can be altered by higher-level strategy (e.g. less efficient production). And some aspects software development don’t have units produced at all, like how much “progress” someone makes on an app (LOC doesn’t count)

  The definition says “output per hour worked”

When you measure this at the level of the whole economy, you can think of the numerator as being the country's GDP, and the numerator as the number of hours worked by all employees.

  But how do you measure worker output?

Great question, and that is exactly my point: in many modern businesses (and hence in the economy as a whole) 'labor productivity' is not a measure of worker output.

Imagine an economy in which everything has been automated, and everyone is unemployed and living on welfare, except the one guy to turn the machines on and off. He is paid $100k for a not-very-demanding job.

Now you calculate labor productivity, and it's a gazillion dollars per labor hour, the highest it's ever been. At the same time the (ex-)workers are as unproductive as they have ever been, because their skills are no longer applicable to any useful activity.

Consider the times in your career when you were most productive (in the everyday sense of the word). Do those coincide with the times when you worked at a company that high the highest profit per employee-hour? Maybe, maybe not.

That’s by definition more productive. But it’s not just a definitional quirk. Your example is one where economists would indeed consider labor productivity to have risen dramatically because of automation.

Yes, but my point is that Labor Productivity measures two things:

A) how efficient individual people are at doing their jobs, and

B) how much progress we have made replacing people with software and machines.

(A) aligns with people's everyday understanding of the word 'productivity', but (B) has a larger influence on changes in the measure.

This sounds more like hopium to me. I'd sure like some of what they are smoking, but that doesn't make the reality any better.

You would think these people would be educated enough to know how supply and demand actually works, and how if inflation goes up, you are paying your workers less. Completely disingenuous.

You get what you pay for, and arbitrarily abusing your staff by changing the metric goalposts every few months to justify churn to pound pennies breaks trust, as much if not more than making them RTO during the pandemic, risking their lives on a job role that can be done completely remotely because you believe workers work best in the office when you really just want to justify the outlays you're stuck with having an office.

I know people who died because of that.

> How is "productivity" being measured?

It's measured the same as any other quantity with political implications: However the author of the article in question wants it to be measured.

Reply on news.ycombinator.com