Could wage decoupling be because of health insurance?

From Scott Alexander:

There are a few lines of argument that suggest it’s not true.

First, wage growth has been worst for the lowest-paid workers. But the lowest-paid workers don’t usually get insurance at all.

Second, the numbers don’t really add up. Median household income in 1973 was about $48,000 in today’s dollars. Since then, productivity has increased by between 70% and 140% (EVERYBODY DISAGREES ON THIS NUMBER), so if median income had kept pace with productivity it should be between $82,000 and $115,000. Instead, it is $59,000. So there are between $23,000 and $67,000 of missing income to explain.

The average health insurance policy costs about $7000 per individual or $20000 per family, of which employers pay $6000 and $14000 respectively. But as mentioned above, many people do not have employer-paid insurance at all, so the average per person cost is less than that. Usually only one member of a household will pay for family insurance, even if both members work; sometimes only one member of a household will buy insurance at all. So the average cost of insurance to a company per employee is well below the $6000 to $14000 number. If we round it off to $6000 per person, that only explains a quarter of the lowest estimate of the productivity gap, and less than a tenth of the highest estimate. So it’s unlikely that this is the main cause.

I don’t agree with all of his framing (are there different deflators floating around in those estimates?  Scott does discuss that later in the post), but those points are worth considering nonetheless.  On Scott’s broader points (not discussed in my excerpt), I think he is underemphasizing the possibility that productivity may be measuring better than it really performed, and thus there is not so much decoupling at all.

For the pointer I thank Benjamin Cole.

Comments

Millenials are foolish to waste their money on avocado toasts, Starbucks, and new IPhones every year and they wonder where all the money went. I got married when I was 23. My first house at 25. My first boat at 30. When will these young socialists learn the value of a dollar? Never. That's why socialism never works.

Agreed. People who live their life irresponsibility expect other people to pay for their problems. Wrong. You get what you deserve.

Owning a boat is a waste of money, given that they can be rented for so cheap and that they sail on average 50 hours per year.

Presumably the owner of a boat decides how often he sails it.

Not sure what the average hours tells me other than there are many rich people who own boats for a status symbol.

And exactly how much did that first house cost you? Upper-middle class carreers were obtainable without a collage degree, and you weren't competing in a global labor market. Now we have the largest generational transfer in the history of the world, and it's flowing the wrong way. No more socialism, but keep the government off of your social security and Medicaid, right? Sorry to have to be the one to break it to you but you're entering your senility.

"flowing the wrong way"
There are a number of factors you don't seem to know. 1. Starting in the mid-50's the federal government transferred a lot of money ($4 trillion) from the SS fund to their coffers. 2. Some 50 years ago and continuing through today the federal government has loosened the laws around SS such that huge numbers of recipients are getting money from the system who put nothing into the system. This includes huge numbers of immigrants. The system was torpedoed exactly for the purpose of dividing the young from the old for exactly the result you exemplify. Wake up, you are being used.

Over the past decades, pay has stagnated (or declined), housing prices have increased roughly 3x (adjusted for inflation), and people are now crushed by college (and often medical) debt. But sure, the problem is people spending money on coffee...

"Over the past decades, pay has stagnated (or declined)"

Nope

"Over the past decades, housing prices have increased roughly 3x (adjusted for inflation)"

Nope

"Over the past decades, people are now... [investing in college]"

And?

You forgot to add "...And get off my lawn!"

Every generation seems to the think subsequent generations are lazy, crazy, immoral, have bad music, etc.

I don't even think millennials are particularly socialist. Recall the socialism of the baby boomers; didn't amount to much, did it? Young people generally grow out of their idealism.

+1. Lotta cheap talk flying around nowadays.

Here we go with grumpy old fogeyism as politics. "You kids get off my lawn!"

First house at 25? What decade was this in, and what was your job at the time? How much did it pay? Did your job require education?

Tyler, what's your take on why wages haven't kept pace with productivity? As Scott shows, this isn't an uniquely American phenomenon. All modern economies have this problem where labor keeps its end of the bargain with productivity but just doesn't see a proportionate share in the rewards. Should we be concerned more with monopsony bargaining power of business or with the inflationary lack of cost discipline in healthcare? Without addressing this, expect deeply illiberal populism or socialism here on out.

You’re losing so much information by aggregating this data that it’s almost meaningless.

You need to choose a few industries where productivity is unambiguous, measurable, and at the employee level. Machining? Mining?

Then you can track whether actual increases in productivity over time match the wage gains.

This isn’t much more useful than measuring angels dancing on pins.

I think the tacit assumption for many is that if a factory is automated, it will hire less workers. The workers that remain are more productive with modernized tools and people would expect that their wages should rise in proportion to their increased productivity. What is happening instead, is that those fewer workers' wages are rising less than the proportional increase in productivity, and the owners of the machinery are investing in more automation or taking the gains as profit, thus increasing wealth concentration.

That’s certainly the assumption of what is happening. What id like to see is some evidence on a micro level.

Choose an industry where employee productivity is obvious and measurable, and track it through time with panel data. Machining, mining or sales would be the obvious ones to me.

Otherwise the aggregation is giving me a whiff of composition fallacy.

Just talk to a few people working these jobs. I have and it's pretty consistent. A few new machines get put in. A few rounds of layoffs where the highest paid usually first to go. Usually no new recruits but if there are, it's a few young workers (saves on wages and health costs!). The math becomes clear. Higher productivity, fewer jobs, less pay. The average voter is not blind to this reality and will vote accordingly. The fact that data at the macro-level is just more confirmation of what everybody already knows. The only question now is what marginalized group will get scapegoated and beat up by the next populist we elect to office.

> The workers that remain are more productive with modernized tools and people would expect that their wages should rise in proportion to their increased productivity<

But why would that flow to the workers if the owner had to buy the new, expensive machine? Anymore, buying a new machine to keep you competitive isn't a luxury, it's a necessity. And as is the nature of these new machines, the operator requires increasingly less skill to operate.

A machinist to make a finial required a lot of skill 70 years ago. The guy that puts the brass blank into the $30,000 machine today and presses the button does not.

Why on earth would you expect the unskilled button pusher to earn what 10 machinists earned? Yes, he's "making" as many finials in a day as the 10 machinists. But really the machine is doing all the work.

Now, on the flip side, I'm an electrical engineer. I can design and prototype in one day what took 6 engineers a year to do in the 1970. And the tools to do that are free. My skills today compared to the 1970 engineer are very similar, but most likely much broader. In other words, not as deep in certain areas, but much broader in all areas.

The gains from my productivity will flow to my employer, who will share them with me because he recognizes I'm very skilled an integral to his success.

Could it be because as the population ages more capital is invested at avoiding the use of labor?

Was labor capturing all the productivity gains before? Maybe capital got better at grabbing its share.

I wouldn't expect this to be true, because as we get more capital, capital's bargaining power goes down.

So at least one of my two theories is wrong.

There's no "bargain" - no prior arrangement that if workers made more they'd get paid more, nor any reason to think that the workers made the change themselves instead of, for example, the capitalists investing in technology to get more product out of the same level of employee effort.

The price of labor is dictated by supply and demand. If productivity goes up, you demand fewer workers, and the price of those workers will drop. The only reason pay should go up is if the increased productivity causes the firm to drop the price, triggering a even greater surge in demand, forcing the firm to hire even more workers than previously. The demand increase from price reduction would have to significantly outpace the productivity increase to cause a labor demand to push up the labor price, which is not going to be generally true as markets will saturate at some point for most products.

Put this in the monopsony column:

"In many rural communities, a farmer raising animals for slaughter has the 'choice' of selling to only one slaughterhouse ... The farmer has very little power in this transaction, and this asymmetry is a key factor in why 71 percent of American chicken farmers live below the poverty line. "

https://www.theamericanconservative.com/articles/to-revive-rural-america-we-must-fix-our-broken-food-system/

American chicken farmers!!

Workers don't just pay for medical care through employer-linked insurance premiums. To put it another way, how much has Medicare and Medicaid grown in the last 50 years?

Scott Alexander has done yeoman work in pulling together a lot of data.

In some regards, it would be a surprise if wages didn't stagnate after 1972 . You had the huge Baby Boom into the labor market, magnified by higher female labor participation rates. For better or worse, the southern border was more or less open to immigrant labor.

And we have a Federal Reserve that is absolutely obsessed with inflation and tight labor markets (try reading the Beige Books sometime).

However, the screw may be turning. The baby boomers are going through the python, and the southern border appears to be getting more difficult to cross.

The Fed may find itself trapped by circumstances and unable to raise interest rates for fear of causing another Great Recession.. The Fed may have to tolerate a level of real wage growth for several years.

Interestingly, in Japan workers entering the workforce are seeing real wage gains for the first time in two generations.

Let us hope that Americans can share the same prospect.

Be careful, your data on Japan might be fake news. The Abe administration had been trying to engineer inflation for a long while now but just recently got caught lying about wage gains dating all the way back to the beginning of his term 7 years ago.

https://www.upi.com/Top_News/World-News/2019/01/25/Japan-misreported-wage-growth-for-seven-years-probe-shows/4761548441604/

Add on: Also, the US, we might as well talk about two Americas: Not rich and poor, but those who live in NYC, Boston and along the West Coast, and the rest. An employee fording housing costs in the aforementioned regions has a whole different take on living standards.

Add on, add on: Scott Alexander posits health insurance costs $7,000 for an individual and $20,000 for a family of four. Egads!

Given the realities of housing and health insurance, and stagnant wages, I am surprised that Bernie Sanders is not serving his fourth term in the White House. And yes, I am aware of the 22nd amendment.

You mean he's not honest Abe?! Sorry ....... couldn't resist.

"better than it really did" -> "better than it did earlier"

'but those points are worth considering nonetheless'

No, it is the data worth considering, because a proven way to ignore 'points' is to rely on data.

Yeah, if only there was only one way to produce data and only one way to interpret all the data. What a utopia that would be.

Not all industries experience the same growth in labor productivity.

Wouldn’t the most honest investigation dig in a little further at the micro level and try and find the true rate of labor productivity let’s say in health care, engineering, computer programming and retail and then track the wage growth or lack there of?

I mean how much honestly different is a retail job at Walmart in Bentonville, Arkansas in the 60’s compared to today? I mean your average Walmart retail person is 70 percent more productive than they were 60 years ago?

I'm as libertarian as any on a good day, but honest libertarians have to admit at least the theoretical possibility of "winner take all" dynamics and ZMP workers, with monopoly leading to gross market failure leading to clearly sub-optimal social outcomes.

This data doesn't come out of the blue. We've had over 2 decades of similar findings showing a broad consilience, if not detailed agreement. We are heading into a future of the few and the many with unprecedented division between them. This may be "just" under strict libertarian construction, but it sure as hell isn't "safe".

Matthew 13:12

Shhh, you're making too much sense. Don't you know that when evidence-based conclusions conflict with deeply held priors it must mean that the evidence is wrong?
"... productivity may be measuring better than it really performed..."
I mean, seriously? IOW "I really, really don't want to see reality here, so I'm going to believe that even the most conservative estimates may really be significantly overestimating the growth in productivity." Phew! Good thing I can do the intellectual acrobatics necessary to cling to my world view.

Is median household income a good measure? Hasn't the average household size been declining over time? If so, median household income understates the wage gains.

Median household composition has changed dramatically over 40+ years, making income comparisons largely meaningless. In addition to the declining household sizes you mentioned there is;
- greater number of Hispanic households (from less than 5% to nearly 15% of Census figures) , Hispanics households have median incomes well below the median U.S. household and larger household sizes than U.S. median
- the percentage of two-income households has actually peaked because of delayed marriage/cohabitation and increased numbers of retirees within the population

Doesn't increased productivity lower the price of goods and therefore make them more affordable? Does income matter as much as purchasing power? And if people are able to buy more, won't they be less likely to demand more money?

It's the libertarians fault. Well, many do. Why? The rise of libertarianism coincided with (or was promoted by) the rise of Reaganism, income tax cuts for the wealthy and payroll tax increases for the non-wealthy, the preference of capital over labor, the demonization of both government and labor unions, the racialization of politics, the shift of production to places (including China as well as the South) with the lowest labor costs and the de-industrialization of America, and the reliance on rising asset prices for prosperity. So here we are, with a broken body politic, racial polarization in politics, and an ascendant populism, populism on the right that is repressive and populism on the left that is socialist, both anti-liberal. During the rise of right-wing populism I wrote that the real danger would be a concurrent rise of left-wing populism, the latter at the time highly unlikely in my view given the Democrats' and Obama's mainly liberal views (liberal in the classic sense). Sure, it may be unfair to blame libertarianism for the current state of affairs, especially now that libertarianism is all but disappearing except in a few eccentric quarters. But here's the irony: libertarianism holds the key to rescuing us from our current predicament, but it must be a new (neo) libertarianism, one that recognizes reality rather than one that worships theory (and markets), that has broad appeal to what is essentially a liberal (in the classic sense) nation. Sorry, but writing books that promote current sacrifices so future generations have a better life and praise monopoly power won't do it.

Health care is the most consistently rising price in the consumer index. That alone makes it a candidate for the most inefficient, prices are unpredictable. Medical is also consistently rising meaning that government is likely consistently demanding more than supply.

It seems like the best interpretation of Scott's data is that health insurance costs are not the *sole* driver of decoupling. There may be different causes at different income levels, too. Impressed immigration at the low end, increased regulatory benefits (OSHA, EEOC) at middle levels, with health insurance in the mix toward the higher end.

Isn't the typical claim around productivity about capital's productivity. Income growth for real people explained by increases in human capital. Stagnant income by suggesting it was capital investments by the business that accounted for most of the productivity increase so gains go to capital owners?

If there is a different lens on this relationship maybe 2 holds. If not then I don't really see that line of argument sheds too much light.

"wage growth has been worst for the lowest-paid workers."

Well, yes.

More correct to say that wage growth has been lowest for those industries with the lowest worker productivity.

Food service is a great example. According to BLS, long-term productivity within this sector is virtually non-existent (0.4 percent per annum from 1987 to 2017). Yet, food service has some of the highest measured unit labor costs. Not surprisingly, food service is the largest sector in terms of employees earning at or near minimum wage.

SA is mostly wrong on point 1. Mostly wrong because while low-wage workers are less likely to have health insurance, when they do have health insurance (which is most of them), a much larger percentage of their compensation comes as health insurance. See Gary Burtless at Brookings (Chart 4): https://www.brookings.edu/opinions/income-growth-and-income-inequality-the-facts-may-surprise-you/ .

I appreciate Scott's efforts but in this case don't find it all very engaging, existentially speaking.

First, "productivity" and even "wages" seem like hard concepts to get a real grip on across 45 years and across an entire economy. Maybe a few specific industries would yield more reliable measures.

But even worse, why 1973? Why not 1968 or 1978? Or some starting point just before or after the Internet Revolution?

Most people working today were not even alive in 1973, or were minors or just starting out in adult life. And who cares what some adut worker was experiencing 45 years ago? I was around in '73 and having a fine time but even I don't care about it now, economically speaking. So much has changed all arond us in terms of "quality of life.". It's up to each individual to measure their own well-being and proceed accordingly from there.

1973 may also present cherry-picking problems -- The economy as I recall went into the toilet from about '76 to '82. Skirts got really long all of a sudden and we had disco. Prog-rock died out. Thank God Reagan and metal saved the 80s.

1973 was 46 years ago. 46 years 1973 was 1927.

Would ANYONE in 1973 be seriously trying to argue that the typical American hadn't enjoyed massive material progress since 1927?

"measuring better than it really performed" How many papers, posts, etc., are about how things are not what they seem, often because the tools for measurement are iffy? Or, again, attacking common sense with paradoxical results? You can sound like an economist just by using the phrases ( in a self -important manner, like me ) Things are not as they appear & My results are paradoxical.

>if median income had kept pace with productivity

What kind of idiot thinks income should match productivity? Besides Scott Alexander, I mean?

Because productivity and income increases did track together for a good long time? And they should: more productivity means more income. Shouldn't we expect to see that?

Are a lot more people free-lancing or "consulting" than in the '70s or '80s? How do you compare the productivity -- or wages -- of employees who are in the workplace 40 hours a week vs. people brought in on a project basis?

"So the average cost of insurance to a company per employee is well below the $6000 to $14000 number."

That's not true unless your firm is so large you can self-insure and you can only insure your employee and not their family.

Small firm owner here: talked to Kaiser for "company healthcare" and the rates were EXACTLY the same as buying individual policies. There was zero cost savings. In fact, its far cheaper to have employees buy on the exchanges where they get subsidies from Uncle Sam.

The biggest problem I see is the constant switching back and forth between median and average measures. Productivity as measured by GDP per hours worked is a measure of average. Median wage is of course a median measure.

What if the biggest explanation of increasong income disparity is increasing productivity disparity? This would produce something like what we observe

There are a few lines of argument that suggest it’s not true.

First, wage growth has been worst for the lowest-paid workers. But the lowest-paid workers don’t usually get insurance at all.

Second, the numbers don’t really add up. Median household income in 1973 was about $48,000 in today’s dollars. Since then, productivity has increased by between 70% and 140% (EVERYBODY DISAGREES ON THIS NUMBER), so if median income had kept pace with productivity it should be between $82,000 and $115,000. Instead, it is $59,000. So there are between $23,000 and $67,000 of missing income to explain.

The average health insurance policy costs about $7000 per individual or $20000 per family, of which employers pay $6000 and $14000 respectively. But as mentioned above, many people do not have employer-paid insurance at all, so the average per person cost is less than that. Usually only one member of a household will pay for family insurance, even if both members work; sometimes only one member of a household will buy insurance at all. So the average cost of insurance to a company per employee is well below the $6000 to $14000 number. If we round it off to $6000 per person, that only explains a quarter of the lowest estimate of the productivity gap, and less than a tenth of the highest estimate. So it’s un

Comments for this post are closed