Hacker News Reader

Comment by ctoth | Hacker News Reader

> There is no economic rule that says that riveting should pay more than taking care of the elderly or food delivery.

There is, in fact. Baumol's Cost Disease, and it's a real bitch.

Manufacturing has compounding productivity gains - one worker today produces vastly more than one worker in 1950. Elderly care doesn't. You can't make a caregiver 10x more productive through better tooling. The productivity ceiling is fixed by the nature of the work.

Wages across sectors compete for workers. So service-sector wages only rise when they have to bid workers away from high-productivity sectors. This predicts: service wages are borrowed from productivity gains elsewhere in the economy.

This doesn't mean we shouldn't want caregivers paid well. It means wanting isn't a mechanism. The mechanism is productivity growth in tradeable sectors pulling wages up economy-wide.

View on HN

> It means wanting isn't a mechanism.

Well we still have demand in the equation. If demand for service workers grows, so will their compensation. (And so will the cost for those services.) So the possibility is there.

People with more disposable income (the high-productivity ones I guess) demand more services. The question will be whether that demand will grow sufficiently to raise the compensation to where we want it to be.

What I also don't know is how we will respond to service jobs being automated. "Premium" service usually shuns automation. Will we have fewer fast food workers and more massage therapists?

And then you get to the reason why everyone complains childcare is expensive.

I thought about addressing this in the original post, but I want to address part of your response:

> one worker today produces vastly more than one worker in 1950. Elderly care doesn't

Indeed, this is a core component of Capital vs Labor issue I mentioned. Productivity increases, workers generate more value, but Capital's share of the results increases nonetheless. The change in this relationship is why I would expect a reinvigorated manufacturing sector to be dominated by Bad Jobs instead of the Good Jobs everybody hopes for.

(There's also a measurement issue in comparing those two. The benefits of e.g. caregiving are measured by the ability of other workers to be productive. For ex if a high-wage worker has to stay home to care for someone, then they are not doing high-wage work, which reduces the overall GDP potential of the country.)

What you’re overlooking is that productivity doesn’t increase uniformly. The low paying service sector jobs aren’t any more productive today than they were 50 years ago.

You're overlooking the main thrust of what I am saying from my top comment to the one you replied to, which is the relationship between Labor and Capital:

> Productivity increases, workers generate more value, but Capital's share of the results increases nonetheless.

To your point, this is even more true for the jobs where productivity has increased the most. Agriculture is vastly more productive today, and the share of profit captured by Capital increases every year. The main point is that productivity gains do not really matter to Labor because the benefits of those gains rarely accrue to Labor.

Put another way: there was a value capture split operant in 1950s manufacturing that contributed to factory jobs being Good Jobs. That split hasn't been operative for decades. In all likelihood, re-shoring factory jobs today would just create crappy sweatshop jobs.

Thank you, I learnt a new term today.

Reply on news.ycombinator.com