You see this headline a lot: 鈥淴 will lead to $Y in lost work hours.鈥�

The most recent version came out this week. You may have seen it on social media, in claims that 鈥渆clipse fever鈥� could cost employers almost $700 million in the U.S. Presumably, it would have cost about one tenth of that in Canada, and then reduced to about 垄79 on the dollar.

It鈥檚 hogwash, of course.

These stories are nonsensical clickbait, designed to throw up a huge number that shocks you.

But beyond getting you to click, I鈥檓 fascincated by the assumptions behind these stories.

First, the numbers are bunk.

The 鈥渁lmost $700 million鈥� figure comes from the idea that, potentially, every single worker in the U.S. would go outside for 20 minutes to watch the eclipse.

Obviously, some people aren鈥檛 going to do that. Surgeons aren鈥檛 going to skip out in the middle of removing an appendix. Bus drivers aren鈥檛 pulling over and getting out while their passengers fume. Some people just weren鈥檛 interested in the eclipse at all. (And for a few of us, watching the eclipse, and interviewing others watching it, was part of our job!)

So on the surface, this is ridiculous.

Below that, there鈥檚 a level in which it assumes that work is continuous, and that any given minute of the work day is interchangeable with any other.

That hasn鈥檛 been true for most of human history, and it certainly isn鈥檛 true now.

If you work in a factory, a meat packing plant, or a call centre, then perhaps you do arrive, begin work, and do the same thing continuously for eight hours.

Most jobs aren鈥檛 like that.

Most jobs have ebbs and flows. Whether it鈥檚 customer service or IT, whether you鈥檙e working in a law office or on a farm, you鈥檙e going to have minutes, hours, days, in which nothing much is happening. And then you鈥檒l have minutes, hours, days that are all furious, frantic work.

The difference between working in retail on a quiet Wednesday morning in June and working retail on December 23 are stark.

So the articles are wrong about work, for the most part. Then why write them?

The obvious reason is that it鈥檚 free publicity for the human resources consulting firm that tossed out the figure for the eclipse. They do this on a regular basis, it appears, as do a few other companies.

The other reason is that it plays into a deep-seated myth. This is the idea that any time taken away from 鈥渨ork鈥� is time 鈥渟tolen鈥� from employers.

There are two ways of looking at employment.

One is that employment is a contract between employer and employee to achieve a certain goal or task, within certain parameters. Move that cargo from X to Y. Write code that accomplishes such-and-such task. Make 500 widgets by Wednesday.

If the worker accomplishes that goal, within the time, then they鈥檝e succeeded, right? Take an extra 10 minutes on your lunch hour? Go see the eclipse? Head out early to see your kid鈥檚 soccer game? That鈥檚 fine, if your work is squared away, right?

Wrong, according to the second view of work.

This is the idea that the employer owns the worker鈥檚 time. You鈥檙e supposed to be in the office from 8:30 to 4:30? Then you鈥檇 better be Adding Shareholder Value that entire time, buddy. You made your 500 widgets by Tuesday? Then keep making widgets. Next round, we鈥檒l increase the number you have to make by Wednesday.

In highly structured environments, like the factories where this idea first took root, it鈥檚 possible to actually enforce this kind of regime.

In most jobs, it鈥檚 just not.

It also ignores the fact that the reverse is often true: employers often steal time from their workers. Unpaid overtime is a serious problem in many industries. The fact that ever-growing numbers of workers are more or less permanently 鈥渙n call鈥� thanks to smartphones also eats into hours outside of work.

We measure work with precise units 鈥� dollars and hours. But 鈥渨ork鈥� is an imprecise thing. Effort and time are not equal for all tasks, nor should they be. And trying to hammer everyone into a rigid system of endless, minutely-measured toil, is not going to make for happy, productive employees.

If you want to see a real drop in productivity, try treating workers like machines.