Salaries aren't about what someone "deserves" or "should earn".

Those in control will try to capture as much of the return as possible. How much value the worker captures is based on their relative power (ability to move to a higher paying employer, scarcity of skillset, laws such as minimum wage, etc).

> Or should the innovators, firms, and customers all get less, and instead my wages should get bigger?

In almost all of the cases the "innovators" are themselves workers whose share of the outcome has been dropping. And the "customers" have never gotten a piece of the profits; we are already past the point where reduced prices would have happened (competition) in this system.

And I think that by "firms" you really mean some combination of executives and investors/shareholders. That is where the gains have been centralized. Do you really want to argue that management and investors deserve to have more of the gains? What have they done that makes them so much more valuable than similar groups in bygone days?

If using autocad does indeed make you more productive, then your 8 hours per day of labor generates more value for your employer.

Where the benefits of that end up is one of the most fundamental questions of politics. As you note, there are arguments for it to flow to any combination of several different groups. Deciding how much goes to each group is what politics is all about, in the end.

In practice what happens is that on average the tool-user's wages go up slightly but most of the jobs in the field are eliminated, and the resulting large profit mostly goes to managers and financiers.

Looking at cohorts of workers, automation tools like AutoCad reduce the expertise needed and commodify the job, which reduces leverage and thus per capita wages.

Consider the taxi or unskilled worker examples here:

https://mitsloan.mit.edu/ideas-made-to-matter/a-new-look-how...

All to say, as an individual, individuals rationally try to maximize leverage in a negotiation to capture the increased output. But what happens to classes of workers in different jobs varies. My take is that capital appears to be gaining leverage over wages back from many classes of workers, even if some limited classes of expertise are stable or gaining.

Productivity is just aggregate output of the economy divided by the number of people-hours worked. You can argue about if that's a useful thing to measure or if the measurements themselves are accurate or if you should capture more of the output, but at root it's very simplistic. If you can use AutoCAD to generate more drawings than using paper which you (or your firm) sells for the same price per drawing, then your productivity did go up. Is that meaningful? Less certain.

I think it is pretty clear that knowledge workers are internalizing the gains that newer and newer tools (and higher and higher rungs on the global economic ladder) have provided them, while at the same time capturing less of the overall gains that are being had.

Basically it's taking fewer bodies to create more value, and that value creation is so intense that even workers getting "ripped off" feel like they are making bank.

More productive companies earn more revenue and can offer more compensation to try to get better workers. That's why software engineering pays more than many other occupations. It's not because they're nice.

But for this to work, employers have to believe that hiring better workers matters.

> Should I capture the increased output

You do capture the increased output by benefiting from a society where the cost to build safe buildings has drastically reduced.

Just because you don't get an immediate financial benefit doesn't mean you haven't benefitted from the increased output.

And many laborers have retirement accounts and pension funds that are also capital owners, so they benefit from increases in capital too.

r>0 is not the problem, r>g is the problem, and that one's a lot less morally ambiguous.

> Should I capture the increased output

yes

Thats a good list of questions here’s another good thought provoking line of thinking:

As someone trading labour for a wage should I adjust my productivity to match the tools I’m using? That is to say if I’m using CAD should I bother using the tool to raise my productivity? Or should I just match my old hand drafting productivity rates? Should I attempt to raise my productivity rates with these new tools to meet or exceed the best rates from my coworkers?

What can we do to align my interests with those of my employer?

> Increases in labor productivity is a curious thing to think about. Do I deserve more wages for using AutoCad instead of drafting paper?

Please...

> But it's also true that in spite of this, my output is higher. Should I capture the increased output or should the innovators of the tools? What about the firms that invest in procuring these tools and production technology? Should the customers capture the increased output through lower prices? Or should the innovators, firms, and customers all get less, and instead my wages should get bigger?

No, only your boss deserves to gain from you productivity increase, that's obvious.

yes, but wage growth has not tracked productivity: https://fred.stlouisfed.org/graph/?g=1X020

household expenses have been increasing without commensurate wage growth, resulting in lower savings: https://fred.stlouisfed.org/series/PSAVERT

2% is average. 1-1.5% is considered a slump, while anything over 2.5% is considered a boom. for instance, the post-ww2 boom (1947-1972) averaged 2.9%. at that rate of growth, a country's total output per worker doubles in roughly ~25 years.

Is this inflation adjusted?

One cause for lack of productivity gains is Baumol's cost disease, which generally affects organizations involving N different people (for N>1) where for one or more reasons N cannot be meaningfully reduced, if at all. Orchestras are the canonical example.

There's a variant of this, however, in activities that are done essentially by 1 person (as is true for most of the examples you mention in your last paragraph). You can improve their individual productivity - more pipes fixed, more joints welded, more patients well-attended to (*) - but in the end you cannot get rid of the individual doing the work in the way automated manufacturing has.

(*) even with a nurse though, this starts to break down for activities where time is a critical part of whatever is being done. Sometimes caring well for a person is primarily a matter of spending time with them, and this is certainly true for teaching as well. In such cases, you cannot make the person "more productive" no matter what technologies you might provide them with.

Construction notably has had productivity losses since the 80s afaik.