Basically he traces the origin of the "fallacy" back to "a hodgepodge of borrowed working-class slang, middle-class prejudice, and archaic economic doctrine" from David Schloss in 1891, and shows it had nothing to do with the 8-hour day effort at the time. The "lump-of-labor fallacy" is essentially a strawman trotted out by opponents of lower hours as a rhetorical device, similar to how the waterfall model never existed before Agile came along as "the alternative". (See e.g. https://pragtob.wordpress.com/2012/03/02/why-waterfall-was-a...)
Then he uses Chapman's theory of hours and some estimates from Denison to show that most likely the 35 hour work week would result in lower unit labor costs, increased demand, and a net gain in productivity overall from the redistribution of working time.
Then he gets into a discussion of fixed costs, and basically says "the government can subsidize the transition by restructuring taxes".
I'm not saying his arguments are sound, although they don't seem obviously wrong, but the chapter is certainly a lot better than the original link, which just summarizes the strawman conclusion but not the evidence.
The assumptions get shown when thinking about why many countries are raising retirement ages, or what happens to work when pandemics occur, or just how fungible people are between different jobs, or why we allow immigration.
The article says that the author thinks the “lump of labour fallacy” is wrong, but doesn’t provide any hint as to why they think there actually is a lump of labour (except to read their other work).
A very sloppy editorial in The Economist on the topic: https://archive.ph/attgx
[minor edits]
https://en.wikipedia.org/wiki/Lump_of_labour_fallacy
> In economics, the lump of labour fallacy is the misconception that there is a fixed amount of work—a lump of labour—to be done within an economy which can be distributed to create more or fewer jobs.
This paper - funnily enough written by the article author - does attempt to tell you what Chapman argued. https://papers.ssrn.com/sol3/papers.cfm?abstract_id=1125543#....
Assuming his paraphrase is accurate, Chapman argued that workers would act against their interests by picking higher pay over shorter hours, and employers would respond in kind by poaching employees from rivals by offering higher salaries for longer hours. It's also implied that this is somehow also against the corporation's long term interest but without a proper explanation of why; but perhaps the original writing offered a less muddled concept of the "optimal" amount of working hours (for whom, on what basis?) than the paraphrase.
That provides some theoretical justification for state limits on reducing working hours in the interests of preserving employee health, something which happened in many industries a year after Chapman made his argument (a bunch of industries where people worked an average of 57 hours saw a mandatory 48 hour maximum working week; though I suspect arguments that UK mining and manufacturing workers didn't have much choice over schedules was a larger factor in that decision than the assumption they were working those hours out of greed)
But I'm not sure this theory which doesn't have a whole lot of empirical support particularly conflicts with what mainstream economists believe[1] or gives us any reason to support the author's contention that the "optimal" maximum working week is 35 hours.
[1]few economists will argue for the benefits of a 57 hour work week, and other theories in labour economics acknowledge certain scenarios in which workers will become overly fixated on the size of their pay package over longer term considerations
1 - There exists an optimum amount of working hours that maximize the society's wealth.
2 - A competitive labor market will maximize the hours worked, completely ignoring the optimal value.
He cites Javon (from the famous "paradox" that the "lump of labor fallacy" fallacy misinterprets) for item 1, but does not supply a reference.
Do you believe the quantity of work that there is to be done in an economy is fixed and the work readily substitutable, so if HN dropped down to 10 hour weeks the result would be unemployed factory workers all finding jobs writing backend code, not less software development work being done (and probably 10 hour work week devs buying less stuff from factories)?
Unless you do, the lump of labour fallacy is a fallacy.
Maybe save the snark until you have at least a high school level understanding of the subject matter?
There is so so much in mainstream economics that "feels rooted in fact," aka, lies or more accurately, overly-relied-on-unprovable-theories, the work of trying to debunk them can make one forget which one you're talking about today.
His style would do well in journalism today.