I would be interested in takes on the Henry George diagnosis. He too started with the question, why do we see more extreme and desperate poverty as economies grow more advanced and wealthy? His answer: because of the unique ability of landowners to take unearned economic surplus in the form of rent, the more economic product is generated, the greater share of it they skim off.

In his view, the core conflict was not labor versus capital, but between labor and capital on the one hand, and landowners on the other, because they simultaneously take the surplus that could be profits or wages, and also constrain the total amount of economic activity that could keep everyone employed.

I don't know that he addressed the issue of capitalists simultaneously being landowners, or striving to become landowners as soon as they're successful, but I don't think it invalidates the premise of them being separate interests that operate in different ways.

I think the idea is that wages are not tied explicitly to productivity - you aren't getting paid based on your output - but that they end up being tied indirectly. Imagine all workers are identical in output terms but they've all got different wages they're prepared to work for. And they're all lined up, cheapest on the left.

I run a business (maybe a coffee shop) and, however big my workforce is already, if I add one more worker to my workforce, I'll make $10 more sales per hour. So should I hire Alice for $1 an hour? Definitely. Should I hire Bob for $1.25? Definitely. And so on all the way to Kevin who wants $9.50. But I won't hire Lionel, because he wants $11 an hour, so if I hire him my profits will go down because he won't bring in enough extra sales to cover his salary. Lionel remains unemployed.

Now I devise a new clever way of selling coffee and the result is that every employee I have can make and sell twice as much coffee - their productivity has gone up. Now I can hire Lionel and Mike and Noreen and so on all the way to Tess (who wants $19.75). So their wage, or rather the wage of the marginal employee, is definitely linked to productivity.

Productivity now meanss usually either better software or much cheaper hardware. In more detail, a team of 50 software-capable people at Google can generate a very large revenue stream with a year's work, compared to what the same engineers could generate improving say Safeway's or Pemex's computer infrastructure.

For the service economy in the US or anyone working outside software, finance, or medicine, the core issue is (I think) that the 21st century doesn't need too many workers because of decades of improved efficiency-- better machines and processes, better organized over decades has added up. Huzzah, we live in an enlightened paradise where, freed from drudgery, we can outgrow our madness. Great returns for capital, and just look at all the selfies that have replaced poetry and painting.

None of this would matter if we had an adequate standard of living floor, and we absolutely didn't let anyone fall beneath it.

One of the mental threads for this post was the ongoing discussion about, "have the Democrats moved left." I found myself thinking about the question of, "has reality changed in such a way that stronger policy responses are required to achieve the same end goal*?"

I don't know for sure, but it seems likely that if one looked at early-2000s inequality (pre-financial crisis) it would be easy to think, "there are cyclical elements to inequality. Those are high right now; if we have a policy response that is sufficient to this level of inequality** that should be enough." But perhaps not.

* For example, "adequate standard of living floor."

** Reader, we didn't.

That is where we get to Moby's comment in the other thread.

My main political belief is that income and wealth need to be distributed exactly as they were in 1980

4: Piketty doesn't talk about cycles. He does, however, acknowledge exceptions to the rule that r>g.

I doubt that, with the passage of time, greater action needs to be taken to achieve the same result. The US, at least, has increasingly abetted wealth inequality in the last 50 years, most obviously with changing tax rates and a permissive attitude about union busting.

I doubt that, with the passage of time, greater action needs to be taken to achieve the same result. The US, at least, has increasingly abetted wealth inequality in the last 50 years, most obviously with changing tax rates and a permissive attitude about union busting.

Help me think through this, because part of my framework for all of these posts are, "if we take believe that the last 150 years have been unique in terms of the amount of economic growth, we should expect that to create politics that are different than what has come before."

It's easy to see the late 60s, early 70s as a high point for high marginal tax rates and powerful unions. That had a lot of benefits, but it was also unpopular. As you say, we've had 50 years since then and the idea that the early 70s should be a political model has not caught on (though marginal tax rates on high _income_ have increased significantly).

I think there's some appeal to that claim; I'm convinced that the answer is as simple as, "more unions, higher marginal tax rates," but I'm also curious if I'm just talking myself out of the best answer. I am completely on board with higher inheritance taxes but, again, that's unpopular and would require significant changes in the tax system (to get rid of family foundations).

Continuing 7.

I think unions have value in creating improvements for workers at a specific workplace, and that is reason enough to support unions.

I think that if you want unions to serve as a counter-weight to economy wide trends you need the assumption that the companies that make the highest profits also employ large numbers of people. If you are concerned about oversized FIRE* industries, I don't know how much unions help to restrain them, and we also don't know much about tech sector unions yet.

* Finance, Insurance, Real Estate

8: As a systemic solution, unions are about counterbalancing the power of capital, not about direct benefits for organized workers or indirect benefits for non-organized ones.

Without even needing to get into what happens in Washington, CEO salaries are a good example: a powerful, widespread labor movement will put direct downward pressure on CEO salaries in the industries they've organized, which in turn puts downward pressure on CEO salaries economy-wide. It's certainly not the only factor affecting those salaries, but it's a real one that's missing right now.

Somehow it's never occurred to me before that the US construction industry suffers quite badly from Baumol's Cost Disease: productivity sucks by pretty much every measure and lags behind more or less every other industry*, but wages have nonetheless increased alongside more productive industries. And since society can't respond by requiring fewer plumbers, we're stuck with this inefficient sector chewing up potentially productive capital.

*I read something a few months ago that tried to dispute this, but was left unconvinced. I think the best you can spin it is that certain sectors of US construction (highways, mostly) are middle of the pack for productivity if you measure them just so.

2 is a nice illustration.

To the first part of Heebie's question, the answer is definitely "all of the above", although economists often want to tease out exactly what's happening so you don't confuse a country increasing output by working longer hours with one that has innovated in some way.

The famous productivity story of our time is that computers for a long time did not increase productivity. This was a great mystery, and people spent most of the '90s arguing about it. Then, for whatever combination of reasons*, there absolutely was an increase, but it's turned out to be kind of a one-time thing. Ubiquitous smartphones certainly seem like they should have lead to another jump, but if it's happened, it doesn't show up in the numbers.

*I don't think there's any consensus on exactly what done it, but AFAICT everyone agrees that it was a combo of ubiquitous internet and crossing some threshold where enough people were using them for what they could do better/differently rather than treating them like glorified typewriters. But there are/were probably other factors as well.

I think it was because the offices were full of managers who printed all of their emails and then one day, they all retired.

CEO salaries are a good example: a powerful, widespread labor movement will put direct downward pressure on CEO salaries in the industries they've organized, which in turn puts downward pressure on CEO salaries economy-wide. It's certainly not the only factor affecting those salaries, but it's a real one that's missing right now.

I like that story, but is it true? Looking at a list of countries with high CEO pay, it's a mix of countries with limited worker power (The US, India, China) and countries in which I would expect significant worker/union power (Germany, Spain, Netherlands, Switzerland).

Consider an alternative hypothesis. this report on CEO pay mentions

In their studies of tax returns from 1979 to 2005, Bakija, Cole, and Heim (2010, 2012) establish that the increases in income among the top 1% and top 0.1% of households were disproportionately driven by households headed by someone who was either a nonfinancial-sector "executive" (including managers and supervisors, hereafter referred to as "nonfinance executives") or a financial-sector worker (executive or otherwise). Forty-four percent of the growth of the top 0.1%'s income share and 36% of the top 1%'s income share accrued to households headed by nonfinance executives; another 23% for each group accrued to households headed by financial-sector workers (some portion of which were executives).

What if you think that CEO pay is likely to increase most in countries with a large financial sector (because there is some comparison of salaries between executives and financial sector workers (or, now, tech workers)?

I'm not arguing that high CEO pay is good (or that a large financial sector is good). I just think about the William Gibson line, "The future is already here. It's just not evenly distributed yet." You might wish for a smaller finance sector, but it's also worth thinking about, "does the present tell us something about what the political fights of the future will be?"

A thought about worker productivity and compensation which perhaps illustrates my perspective on this.

I'm a computer programmer. I work for a small company, and have been in the same job for almost 18 years. I like the stability, and I'm doing fine, but the company I work for has definitely not seen the sort of outsized returns that are associated with "the tech industry" and doesn't have a business model that lends itself to scaling in that way.

Lets say I make three times as much as a barista and one quarter as much as the average employee at Netflix (and that average employee has much less experience than I do).

Whatever you think about my productivity compared to a barista, it's pretty obvious that the Netflix employee is more productive than I am. If my company needed to pay me half as much as the average Netflix employee they would go out of business, because there isn't the money to support that. Netflix is much better equipped to turn programmer hours into money than we are, and that's reflected in their compensation!

What happens at that point. If there are a bunch of companies offering Netflix salaries, eventually I leave my job for one of them and my existing company loses a lot of experience (and eventually folds if they can't continue to hire people). If there aren't many companies that can afford to pay that much you get the status quo -- a few tens of thousands of people working for extremely productive companies making obscene salaries and a lot of computer programmers making much less.

I think that's a predictable outcome; the future is not evenly distributed, and not every company is going to be equally good at making money. But, I'm not quite sure how larger unions change that picture. Larger unions would be great for any number of reasons, but I don't think they'll change the fact that there's just a lot more money to be made by Netflix (or Apple, or Facebook) than there is working on our software.

Again, I'm not saying that anyone should feel sorry for me. I don't see any way that our business could pay me (or the owners) obscene amounts of money, but it's still good work, and I'm glad to be doing it.

But, as I say in the OP, I worry that an economy that has a lot of technological change will generates more of those gaps -- the overall economy becomes wealthier but it isn't evenly distributed.

That's why I'm skeptical that the primary thing that matters is, "changing tax rates and a permissive attitude about union busting" but I'm open to being convinced that I'm wrong.

11: Saiselgy has written about the fact that US construction of roads and other transportation infrastructure like rail is wildly expensive compared to other well off countries. I wonder if it's BS, b/c a lot of his stuff is too just so, but I'd be interested to learn more.

It definitely is true that US infrastructure costs are way higher than in other comparable countries, but the reasons aren't entirely clear. The phenomenon JRoth refers to makes sense as one component. I think another is likely that engineers in the US have done a better job creating barriers to entry that keep their pay high, the same way doctors have.

14, 15: Now that a lot more YIMBYs understand that the building code drives building cost almost as much as zoning code, I spend my time on Twitter explaining to infrastructure nerds that US construction is overpriced shit in every single sector, and they need to be less myopic about causes (most of them know nothing about construction outside of infra projects, so they think the problems must be infra-specific, so you end up with endless ink spilled over subway station mezzanines).

As I said in the thread the other day, I really don't have a solution (because I think the causes are interlinked and deep and somewhat unknowable), but we need to see the problem before we can solve it. It is almost certainly the case that the US has specific shortcomings in infra policy & procurement, but you could wipe it all clean and have it run by a consortium of international best practitioners, and they'd still run up against the poorly trained US workforce and its abysmal quality control.

You've seen me nail down subfloor.

The real treasure is the equity you build along the way.

The people analyzing transit-construction costs love to harp on the number of people involved - for example, US construction crews running TBMs (not that there are really *that* many of them) seem to involve 3x as many humans as comparable projects in other parts of the world. This counts as a "productivity" issue of sorts, but it's not super clear where the difference comes from.

19: I think that the US never really transitioned away from labor-intensive practices, thanks in some part to undocumented labor. However, I don't see any evidence that density of such labor has any correlation with less efficient practices--or rather, that the scarcity of it correlates with more efficient ones. To use the example I know best, Pittsburgh has the lowest density of Hispanic people and low education immigrants in the country (I don't think our South Asian professors sling much drywall), but there is absolutely no more innovation here than anywhere else. To some extent, Amish framing crews make up the difference, but this is itself a recent innovation, so to speak.

But anyway, it's fairly typical for any but the smallest job to feature a lead carpenter, an assistant, and a laborer, but the laborer is doing pure scutwork. To speak broadly, the skilled guys do sloppy work, and the laborer cleans up, literally and figuratively. In Germany, a similar job might be 2 more-skilled guys using a small crane and working more carefully/skillfully/efficiently. Some chunk of that is training (DE has trade schools, the US has an informal apprentice system except for union carpenters, who mostly work commercial), but IMO it's more cultural. Wasting material and keeping a sloppy job site aren't viewed as sins here. Good carpenters* aren't like that, of course, but they command premium prices for premium work: they're not really a model for better construction as such.

*using them as a stand-in; AFAICT everything I'm saying applies to everyone in the industry

Making what would be $14/hour adjusted for inflation, I was labor on a framing crew in 1992. If you have a house, all the workers have peed in your basement before the cement floor went in.

AIMHMHB that is far from the most disturbing thing that has happened to my basement.

13: Technological change isn't something new -- as you point out, there have been radical changes in technology over 150 years. I haven't got the numbers at my fingertips, but surely the increases in productivity in the postwar decades were larger than they have been for the last 50 years.

Unions and taxation don't solve all problems, but I'm not sure how you see them as being nonresponsive to the issues you raise. Empowering workers will, as a rule, give them a bigger slice of the pie. Higher marginal tax rates, all things being equal, result in less income for wealthy people. Estate taxes are fine things, as would be a wealth tax.

Are these solutions popular? I think so, but in the "democratic" US, they have been rolled back, so maybe not. But that's a separate issue. Would they work to decrease inequality? Of course they would. (And of course they did.) I'm not seeing any counter-argument.

22: did you get the house?! We call it closing here, but I don't know if that's the British term.

20:

Re: Union trades. Most of that work seems to be in the city and not in the suburbs. I only know electricians. That work is episodic in a way that somebody working for a large electrician company doing commercial work in the suburbs isn't. A job ends and you get laid off, but you still have access to health insurance. The trade unions seem very different from unions at big companies.

The German comparison is kind of interesting to me. Tim tells me that in Germany (very old Pharma company) the chemists just sit in their offices and don't do any experiments, but they have technicians for that, whereas here non group leader PhDs and MSc s will do experiments. The Germans will be super hierarchical about who does what and it can really slow things down.

They also outsource some compounds to China. That's often fast, but by most measures of "productivity" it's not efficient. They out 2-3 times as many chemists on a project and work them 14 hours a day 7 days a week. However, frequently what they ship isn't very pure. And the synthesis steps they provide aren't really reproducible.

I just realized that I'm an idiot and that I got the name of Brad DeLong's book wrong in the OP.

Would on of the FPP be willing to correct it to _Slouching Towards Utopia_?

23: I don't think your answer was non-responsive, I just think that it is only part of the puzzle and that you are overstating how successful that solution would be. Longer response in a bit.

24.1: not yet, we were hoping to close in two weeks but it may slip by another week or two.

19, 20, etc remind me of this great set of posts about how WWII-era US and postwar Japan rapidly improved shipbuilding efficiency (shipbuilding, by their argument, being in between assembly-line manufacturers and custom-built construction, and this requires different organizational techniques).
https://constructionphysics.substack.com/p/lessons-from-shipbuilding-productivity

The basement remark was not referring to the future house, which had Blzck Sabbath in the basement, but the previous house, which had a bleeding fugitive in the coal cellar.

Did he pay part of the rent?

I think that inequality is linked to consolidation and enterprise size (which means it is tied to market size and antitrust law).

Imagine that you have 10 banks, each independent and with a single branch. Each bank has a president who gets paid 100K.

Imagine that the banks consolidate. Now you have one bank, with 10 branches. Instead of 10 bank presidents making 100K, there are now 10 vice presidents, each responsible for a branch, who each get 90K. But there is now a bank CEO, who necessarily gets paid more than the vice presidents. Perhaps 150K (100K from paying the vice presidents less and 50K captured from decreased costs / increased profits arising from consolidation).

Now image that ten such banks consolidate. Now you have 100 branch managers, making 80K, 10 regional managers, each making 90K, 2 vice presidents, each making 150K, and a CEO making 300K.

The larger an organization, the more profits are available for the leadership to skim off or divert. And the easier it becomes to depress wages for lower-level workers. So as firm size increases in an economy, you will necessarily see greater inequality in the economy as a whole.

Since the 1970s, the US has seen a massive amount of consolidation.

Honestly, I don't get how banks work these days. There's like 300 of them and I don't understand why they pay rent on so much space. I've heard it's to get deposits, but even now they want to pay like .05% interest.

I assume they are just recycling money from the fed into loans. But do you need buildings in prime business areas for that? As far as I've ever seen, the borrowing happens at real estate closings, car dealerships, and the like.

Here is an article (from the Economist of all sources) lamenting the increasing consolidation of industries and the effects of such consolidation.

https://www.economist.com/briefing/2016/03/26/too-much-of-a-good-thing

Being the Economist, increases in wages are mentioned as a bad thing...

Did he pay part of the rent?

He did not. I can't find it in the archives but he had been down there for some time before we noticed the bloody handprint on the door and investigated.

35 Wasn't there a plague pit under your kitchen floor too?

That was a soup tureen.

36: no, that was the house of a friend of mine. (My flat was on the first floor, and while I didn't get on very well with the woman downstairs, I think that description of her living room might be a little harsh.)

A friend of the Selkie's lived in a house in Glasgow on a street locally known as Murder Row because every house in the street but one had been the scene of a murder at some point. It wasn't a long street, but still.

The occupants of the last house must have been pretty tense about the whole thing.

The occupants of the last house must have been pretty tense about the whole thing.

I chuckled at that (also, if you're willing, I'd repeat the request in 25 to ask you to correct the error in the OP).

13: Technological change isn't something new -- as you point out, there have been radical changes in technology over 150 years. I haven't got the numbers at my fingertips, but surely the increases in productivity in the postwar decades were larger than they have been for the last 50 years.

Coming back to this, because this gives me a chance to explain my thinking behind this post. According to Our World In Data* the global market economy increased from $5T in 1913 to $110T in 2015 (22x increase). Obviously that represents a combination of population growth, increases in productivity, and people moving from subsistence to market economies. But let's say that productivity increased by close to 10x**.

I think it's unlikely that productivity will increase 10x over the next 100 years (see, for example, this recent vox article), but I don't think that chance is 0 and, on average, I spend a fair amount of time in my life considering the possibility that human life on earth will be worse in 100 years than it is now, and 0% of time thinking about the possibility that it will be much, much wealthier. So I'm trying to at least consider the question of, "if there's a real chance of another 10x increase in productivity, what would that mean?" I have no idea how that would translate to the experience of daily life, but I do wonder about how that will shape politics.

As I said in the OP, one reason why I don't spend much time thinking about the, "what if the world is much richer" scenario is that I would have thought it would make the politics easier, but I'm not so sure. Because it represents a lot of change, I think that inevitably creates political problems. As I said in the last post, the question of, "what is required to create a politically sustainable coalition for equitable growth?" is a really difficult question.

I don't doubt that unions and high marginal tax rates are good pieces of the answer, but I'm not sure that they are either a good model for dealing with massive social change or a good way to build a politically sustainable coalition. I'd love to be wrong on that; I'd be very happy if simple tools were up to the challenge.

One problem, obviously (and the title of the post) is that one person's income is another person's cost. Is Baumol's cost disease a good thing or a bad thing? If we want people to make a living wage then we want to see increases in salary even in industries which have below-average productivity gains. But, as JRoth says, that makes construction (and education) much more expensive and, as a society, we want to have more construction and education. I'm not trying to say, "health care and education will bankrupt the government" I think a wealthy society can and should continue to afford them, I'm just noting that there is both an economic and political dimension to the problem and that unions are an imperfect solution to both (in that they don't by themselves, offer a way to equalize between high-productivity and low-productivity sectors, and public sector unions (which represent areas in which Baumol's cost disease is a major concern) have political challenges).

Is that a better big-picture summary?

* It's also worth reading their charts on economic inequality.

** Of particular interest, this chart shows which countries have grown most quickly over the 66 year period 1950-2016. They show the global average being a 4.4x increase in GDP per capita and if growth continues to compound at that rate, that would be 10x over 100 years (4.4x in 66 years * 2.2x in 33 years = 9.7X in 99 years).

Looking forward to reading the link in 28 at my leisure.

Oh, and wholly endorse 31.

As has gotten somewhat memory-holed, Warren talked this up a lot ca 2016, and Clinton incorporated more aggressive antitrust in her platform, including rhetoric that got specifically at the issues in 31. That was one reason I got genuinely excited for her. I really think the issues in 31 are root cause of many of our problems, including especially the languishing of the hinterlands, where every decent-sized city hosted at least one substantial regional bank, and now many of them don't have anything more than a local bank and a bunch of branches from out of town.

It's not about "buy local" sentiment, it's about the concrete effects of pulling the top level decisionmakers (and their wealth) out of all those places. Even if it didn't lead to "streamlining" in the form of shutting down offices and plants (and it does), it would still lead to more decisions being made by people with zero local investment of any kind.

@43

It's not about "buy local" sentiment, it's about the concrete effects of pulling the top level decisionmakers (and their wealth) out of all those places. Even if it didn't lead to "streamlining" in the form of shutting down offices and plants (and it does), it would still lead to more decisions being made by people with zero local investment of any kind.

This is an excellent point. Thank you.

I can see this leading to an almost colonial relationship between the urban center and the rural periphery.

It's not about "buy local" sentiment, it's about the concrete effects of pulling the top level decisionmakers (and their wealth) out of all those places. Even if it didn't lead to "streamlining" in the form of shutting down offices and plants (and it does), it would still lead to more decisions being made by people with zero local investment of any kind.

I agree with this and want to tie it back into the OP (and my personal experience in 13). Part of why my individual productivity is lower than it might be is because my job contains a mix of high-value work (programming things I'm familiar with) and lower value work (customer support, or going through our test server and figuring out what old files can be deleted to make room). There's an economic efficiency argument that we'd be better off if the organization was larger and I could do fewer of the low-value tasks and we could hire somebody less experienced and cheaper to do them.

But there's a human cost to that as well. First of all, doing customer support (for example) isn't purely inefficient, it informs my work and gives me more insights into how people are actually using the product. Similarly, if we hired somebody to do more of the routine tasks that creates a hierarchy -- the economic logic is that I should be trying to hand off routine work to them, but that creates a structure that would reward me for treating them as place to dump things I don't want to do (see, for example, how Congressional staff are treated).

The greater the gradient between high-value and medium-value tasks the stronger that logic grows (of course google should pay for in-house dry cleaning because if it frees up 20 minutes a week from the programmers it's economically valuable to them), but there's a tension between the economic logic and human dignity.

I think some of the same logic affects the relationship between superstar cities and the rest of the country. NYC (or Seattle, or Boston) can correctly say, "it will be better for the state as a whole if the state focuses on solving our problems --- because we drive the state economy -- and everyone else adjusts.

And, going beyond my own perspective, a big part of what I value about working in a small company is that there's very little distance in the hierarchy between the "top level decision makers" and everybody else. Not only does the head of the company have an office two doors down from mine, he's talking to a lot of the same people I am.

I've probably egregiously violated the analogy ban, but I do think there's a parallel between hierarchical distance and geographical distance.

I have thoughts about the decision makers and leaders of the rural periphery, but they are kind of inchoate.

40: or the occupants of the last house had committed all the murders.

The Inspector Barnaby school of detection, just wait because every new victim is another suspect cleared.