Hacker News Reader

Comment by ed312 | Hacker News Reader

The reality is basic unskilled labor is worth about as much as it has been. At the same time, multiple new industries have sprung up and computer/ai-aided workers are vastly more productive than ever. I think a higher minimum wage could be a good "forcing function" to eliminate more low-skill jobs. The problem is, the general population is always going to have some proportion of people who can not or will not do higher skilled work. Also, normalizing WFH/remote work might help to let folks diffuse to lower cost of living areas.

View on HN

On the contrary, what this means is the same “unskilled” labor is now actually worth more because it is resistant to automation. People doing automation resistant work get to share in the productivity improvements of other sectors, and that is perfectly fair.

But what’s odd about your argument is that the federal minimum wage ought to be $12-13/hour if it had simply been adjusted for inflation since the late 1960s. So by that measure, we value “unskilled” labor even LESS than we did half a century ago.

If we had adjusted minimum wage for productivity growth, it’d be about $21/hour right now.

So I think we ought to have a nationwide federal minimum wage minimum of $12.50-15/hour, adjust it for inflation continually, but also adjust for local cost of living so that the nationwide average (average of regions) minimum wage is around $21/hour, continually adjusted each year for cost of living.

And I am not worried about automation increasing unemployment over the long term. We should have targeted efforts to increase employment rate if there are periods of disemployment from disruptive automation, but we have more jobs now than we did a century or two ago in spite of two centuries of increased automation and mechanization.

continually adjusted each year for cost of living.

One big risk here, or with any cyclically referenced benchmark, is runaway positive feedback. This can be exponential growth, oscillation, or exponentially growing oscillation.

If wages are referenced to CoL, but CoL is affected by wages, you have this risk.

A real example I have encountered: years ago, before an area I was living in experienced hyper demand, rent prices were fairly stable. I was able to get decent places just under $1k. My landlord was pretty chill, but said that they had to raise the rent eventually because they couldn't write off expenses (or maybe it was losses) if they were charging below the area median rent. So every year, landlords would have to increase the rent to the median, which would almost certainly raise the median (unless literally the bottom 50% of prices were exactly the same), and then they have to raise again, etc.

I don’t buy it. Civil servant salaries are adjusted for cost of living, and this isn’t just the lower levels but ALL levels.

Again, this is just excuse.

You can’t fix a lack of housing issue by keeping wages low.

I don’t buy it.

It's math. If f(x)=K * g(x-1), and g(x)=L * f(x-1), you need to be very careful about choosing values for K and L.

It's not math, it's a hypothetical, somewhat absurd model.

There's actual empirical research on minimum wage, and it doesn't support these claims of exponential cost increase: https://noahpinion.substack.com/p/why-15-minimum-wage-is-pre...

That's because minimum wage isn't actively pegged to a cost of living measurement, it's manually adjusted from time to time (ostensibly to be inline with cost of living). That manual adjustment is slow and decoupled enough that there's no runaway feedback loop.

If you legally define the minimum wage as a formula instead of an absolute number, and that formula includes cost of living, you can create a runaway situation.

The fact that the minimum wage isn’t pegged means it IS adjusted: downward over time due to inflation. That has massive societal impact as it reduces wages for folks over time, risking homelessness and other effects of poverty. It also reduces productivity growth. THAT has real risk of runaway effects due to the compounding effects of poverty.

I don’t see why a simple adjustment for inflation or cost of living is bad and risky a priori but allowing it to decline over time is a priori good and not risky.

There is substantial bias in that assumption.

It would be interesting to see a breakdown of the composition of an annual raise for a civil servant. How much (and from what factors) is the raise in pay related to CoL, and how much is it related to productivity? How is that labor actually valued? I think if 80%+ of that raise is due to CoL, that's clear evidence of this feedback loop in action.

The percentage of civil servants in the local workforce probably also has an effect. If only 2% of people are civil servants or otherwise receiving an indexed wage, then other factors may dominate. But if a large number of people are civil servants / receiving an indexed wage, then positive feedback might dominate.

There isn’t a strong feedback effect, though. We know this because there are areas that have large civil servant populations.

Secondly, we’re talking about minimum wage, not all wages.

> One big risk here, or with any cyclically referenced benchmark, is runaway positive feedback. This can be exponential growth, oscillation, or exponentially growing oscillation.

Have we ever witnessed an exponentially growing oscillation in a large scale economic feedback loop like this before? I'd never really considered that concept for macroeconomics.

It’s not really a big risk, here. It’s hypothetical. A scare tactic to argue against increasing the minimum wage to even moderately adjust for for inflation and overall productivity growth (and cost of living growth).

It is very frustrating to have any kind of productive discussion when the slightest hint of anything unexpected is negatively labeled and associated with the entirety of some diametrically opposed position that was never actually raised.

I apologize. But it's frustrating when people express opposition to even a minimum wage level from the 1960s with sort of outlandish hypotheticals. There's a bunch of pretty good empirical research out there on the effects of minimum wage increases, and none of them suggest this sort of "exponential" increasing effect. https://noahpinion.substack.com/p/why-15-minimum-wage-is-pre...

The thing is, you assumed what I believe. I'm actually in favor of higher minimum wages and indexing to costs. But the feedback-loop nature of the economic system, or any feedback system (where future output numbers of the system depend on previous outputs), must be acknowledged and accounted for, or else you will, mathematically, get exponential growth and/or oscillation.

I also have a lot of family who currently live just fine making less than $15/hr. Their employers (small businesses) would have to raise prices (or go out of business, or both), which would raise costs, which would raise the CoL index, which would raise future minimum wages in a feedback cycle, if there is not some compensating factor in how wages and prices are set.

> Their employers (small businesses) would have to raise prices

If the market would bear it and they could increase prices profitably, why wouldn’t they just do it today, regardless of wages?

Because then they would lose profits. If labor becomes more expensive, maximal profit would be found at producing less and charging more, but overall profit would decrease.

...you’re ignoring demand effects. Higher wages means more aggregate demand. Empirical research on minimum wage increases shows that this basically fully compensates for hypothetical disemployment effects.

I'm sort of extrapolating from the very little I know of control theory and feedback systems. I've seen others mention that economics and control theory could benefit from collaboration.

Maybe the boom/bust business cycle, coupled with inflation, is exactly this phenomenon? Maybe social media frenzy is also a feedback phenomenon.

What do you hold as baseline? Baked in to your reply is a theme of prices go up over time (inflation). To take a concrete example: has a person employed to stock shelves at a store meaningfully increased their productivity in the last 25 years? What if, instead of increasing wages to match "cost of living", we made it easier for people to change jobs or move to an area with a more favorable cost/quality of life ratio?

> minimum wage ought to be $12-13/hour if it had simply been adjusted for inflation

The federal minimum wage in 1960 was $1.00 [1].

Adjusted for inflation, $1.00 in 1960 is equivalent to $8.86 in 2021 [2].

[1] https://www.cnn.com/interactive/2019/business/us-minimum-wag...

[2] https://www.dollartimes.com/inflation/items/1960-united-stat...

> At the same time, multiple new industries have sprung up and computer/ai-aided workers are vastly more productive than ever.

I used to believe this but the reality appears completely the opposite. If the US economy was vastly more productive then we should see capital investment in areas with huge unmet needs. Take healthcare as an example, one of the shining beacons of AI-assisted efficiency, a colossal failure to actually address healthcare needs cheaply and effectively. Real estate is another; here we are in a world of telecommuting and automated brick-laying machines - and prices are sky-rocketing in many areas despite apparent "urban exoduses". We have high quality printing for cheap, yet the art market has reached such peaks that it has spawned a market in owning an exclusive "token" that links you (the purchaser) to an artwork in an essentially ineffective way except as a speculative asset. Exclusivity is the name of the game, and those who afford to pay for it will pay through the nose.

Unskilled labour is cheap because there are a lot of skilled people also competing for it. They are competing for those low wages because the success of monetary policy has led to a situation where there is a greater return selling exclusivity and limitation than there is actually producing products to sell to everyone. Capital investment has become so overrun with cash we treat it as an expense (subscription, everything-as-a-service, etc) or as a bet in the hopes we hit a unicorn (PACS, bitcoin, etc).

The slow march back to fiscal policy will, hopefully, deflate a lot of this inefficiency. One might hope that we end up in the fertile lands between the territories of the fiscal and monetary hawks, but life isn't like that.

How many people would Google actually need to supply humanity with targeted ads?

I suspect that ultimately, most are just there to out-innovate would be upstarts that might, perhaps, take some of their pie. But this doesn't have to go on forever, there already is the precedent of Google not dying at all from utterly failing with their Facebook counter. And that's just one example. I believe that the "productivity crisis" is actually much deeper than it seems because so many parts are glossed over by zero sum games taking up some of the overcapacity.

If double click/search is a true monopoly google may not need any further work to sustain itself and could push new costs to their customers.

I won’t venture a head count figure for google without R&D but it’s probably a tiny fraction of current head count.

Reply on news.ycombinator.com