Live data from Hacker News

Companies ran an experiment: Pay workers their full salary to work fewer days

npr.org

81–90 of 156 posts

Re: Companies ran an experiment: Pay workers their full salary to work fewer days

#81
post #40

Compensation should be based on output. If hours translates to more output (like an assembly line worker or something) - great. Simply pay hourly and require minimum amount of hours (like 40) with options for "go-getters" to work more if they want. Plenty of jobs (like a lot of tech) should be based on "if you're producing what we need from you (or more if you want) we're happy" regardless of the hours.

I am always fascinated by people who want to earn less, pay more in tax, have less rights, etc.

Curious, why would want to be paid by the hour instead of getting the same pay for less hours as this experiment would want to do? Technically thats a pay raise and better for your quality of life.

Re: Companies ran an experiment: Pay workers their full salary to work fewer days

#82
I think I personally know many people that will simply use the extra time to work another job, thus erasing the intended benefit to the generous company.

They work you less so that you will have better morale and energy and loyalty while you are there. But you use the time to collect a 2nd paycheck, and the gets the same tired employee they would have had if they just worked you fully the old way.

I don't know what could practically be done about that.

It also seems like a 6 month trial where the participants are still the exception not the rule, doesn't really predict what would happen if it became the rule. If most companies adopted this as a norm, surely simply the average salery would just inch down over time until it became essentially necessary and the norm to work 2 jobs, since the 4-day job won't actually be enough by itself except in theory, like how minimun wage is supposed to be enough in theory.

There will always be huslters aggressively trying to maximize every minute, and they will always set, or at least perturb, the going rates for things. I think this experiment can only really exist as a special bubble experiment where the ceos just wanted to do it, and all employees were convinced to voluntarily play along and not just fill the gap with a side job. Maybe there can always be some few companies that can do it as their particular distinctive policy just like there already are different companies with different cultures and norms. But across the board? I think the hustlers and hungry scrappers will just set the going rates.

Re: Companies ran an experiment: Pay workers their full salary to work fewer days

#83
post #50

Earlier quoted context omitted.

> So during those 40 hours, let's say a person delivers 100 work units. If we were to say, from now on you only have to work 32 hours, some people will still deliver 100 work units, but many will deliver an approximation to 80 work units. Many people will automatically adjust. People have a "pace of work" that is like breathing. Perhaps in manual labor, but I don't think this applies to white collar work. The limitin…

Exactly. Setting aside the "metabolic drag" of working in a company (time spent attending mandatory trainings, recurring stand-ups/all-hands, etc), I think everyone working in software has more productivity in their 1st hour than in their 39th. The question is where does the N * f(N) first derivative turn negative. I think it's lower than 32 hours/week for software development myself.

I would agree with that figure.

Personally I assign development to more of a "creative" role than a lot of people tend to. Your productivity can often depend on your inspiration, moreso the more senior/architectural your role is.

I've certainly had moments where I accomplished more at one 3am moment of clarity than an entire week before - and I think most experienced developers can say the same.

Re: Companies ran an experiment: Pay workers their full salary to work fewer days

#84
I've been working four-day weeks for almost a decade now, and it's been amazing. I briefly switched back to a five-day week at some point, but I had enough of that after a few months. The weekends felt way too short and the workweek way too long.

Four-day weeks are just the right balance, though I wonder if I'll like three-day weeks even more.

Re: Companies ran an experiment: Pay workers their full salary to work fewer days

#85
post #18

This won't be valid for everyone, but I'm getting more and more convinced that most (a lot of?) people don't "work" work 40 hours anyway. I'm not even mentally capable of doing 40 intense hours of focussed work. Not saying I don't have other tasks besides coding/thinking in that time, but honestly not for the full slot of time. It just feels to me like some sort of secret we all share and pretend like it's fine, but…

If we "actually work" 32 hours out of 40 (80%), what is the likelihood that if we shift to a 32 hour max week, that the average person will only "actually work" ~26 hours (80% of 32) in this hypothetical future? What if at least part of the 20% is spent doing social activities, water cooler conversations, etc. that enable the remaining 80% to flow more efficiently?

About 0% probability.

They will work more than 80%, less than 100%, say 87%. And that is what you want: keep some off time for conversation, recovery, etc..

You don't want 100% (very short hours) and you don't want too low a percentage (long hours). There is some balance in between and no slippery slope.

Re: Companies ran an experiment: Pay workers their full salary to work fewer days

#86
post #18

Earlier quoted context omitted.

If we "actually work" 32 hours out of 40 (80%), what is the likelihood that if we shift to a 32 hour max week, that the average person will only "actually work" ~26 hours (80% of 32) in this hypothetical future? What if at least part of the 20% is spent doing social activities, water cooler conversations, etc. that enable the remaining 80% to flow more efficiently?

Exactly. The whole modern trend of working less hours will somehow equate to more work being done is ridiculous.

I don't think anyone is necessarily saying more work will get done although that's possible I guess, my understanding is that in a humanitarian sense cutting down expected working hours would increase employee satisfaction/happiness without affecting productivity much if at all and that's the main draw of the movement.

Re: Companies ran an experiment: Pay workers their full salary to work fewer days

#87
post #54
post #44

At my previous startup we tried this with engineering and actually saw a productivity _increase_ from 4-day work-week.

I will say the morale improvement on my team when we switched to 4/10s was worth it alone. We couldn't swing the reduction in hours unfortunately.

This is a bit different from a 32 hour week.

But yes you will see productivity go UP with 4 10s. That's a full days worth of break and lunch interruptions that isn't there.

I switched to 4 10s a couple years back and it's nice, but the days are quite long so anything else you plan to do probably isn't happening. Like gym time, which is going to cost long term.

Re: Companies ran an experiment: Pay workers their full salary to work fewer days

#88

Earlier quoted context omitted.

Because the workers produce as much (or more) in that arrangement.

more is never enough in the capitalist system

If you have two equivalent competing companies, both paying the same salaries to the same number of employees, and one company accomplishes more by asking their workers to 4 day weeks, then that company will be more successful and can put the other one out of business.

Thus the key measure that this study would need to demonstrate would be not that the productivity/worker-hour ratio goes up but that also the productivity/worker-dollar goes up. If the latter is true, a lot of companies are going to take a close look at this because more is more.

Re: Companies ran an experiment: Pay workers their full salary to work fewer days

#89
post #18

Earlier quoted context omitted.

If we "actually work" 32 hours out of 40 (80%), what is the likelihood that if we shift to a 32 hour max week, that the average person will only "actually work" ~26 hours (80% of 32) in this hypothetical future? What if at least part of the 20% is spent doing social activities, water cooler conversations, etc. that enable the remaining 80% to flow more efficiently?

Exactly. The whole modern trend of working less hours will somehow equate to more work being done is ridiculous.

The only thing that's ridiculous is a simplistic sweeping statement like that. For one thing, no one even said "more work" necessarily, although that is absolutely perfectly possible and valid, but better work, and better overall performance as in morale, loyalty, effectiveness (better imagination and problem solving), longevity (less turnover, more return on training investment, better institutional knowledge retention), etc.

Re: Companies ran an experiment: Pay workers their full salary to work fewer days

#90
post #76

Earlier quoted context omitted.

Another fascinating thing about teacher claims of working hours is how technology apparently hasn't moved the needle at all. Lots of schools have moved to more and more online assignments that can be in large part automatically graded yet the time spend grading remains constant, or so they claim.

That's probably because IT busywork replaces manual busywork. Before: you take a stack of papers, and go through each one with a red pen. Now: you log on a system, wait, click on something, wait, click on something else, wait... the stupid tests are auto-graded, but chances are there is still something you have to grade student-by-student, and that's now slower to do. Plus, obviously, the usual annoyances (got to upd…

"you log on a system, wait, click on something, wait, click on something else, wait..."

You know, if someone wants to write The Next Great JS framework, give me one that above all else prioritizes latency and expert-level usage. There's no fundamental reason we can't have most of the nice things from the modern graphical web, and get to the legendary efficiency of those text-based consoles... but it will take some work and thought. (For example, you're going to need to insert yourself in between the user and the browser's concept of events, so you can buffer up commands the user is typing while their target hasn't quite loaded yet. This is one of the fundamental reasons why GUIs are less efficient than TUIs, though by no means the only one.)

This framework won't take over the world, but it sure would save a lot of people a lot of time.

Sooooo tiiiiirrrrreeeeed of programs that take seconds to do every... little... thing!

Post reply on HN