Long post, probably only of interest to me. But you have made this claim before:
the people who live in cities are richer or more accurately more productive on average.
And I agree with this. In a way. Where I disagree, is in our definition of productive. If we agree that more GDP is more productive, this is absolutely true. And I agree that GDP is a useful measure of productivity. In particular when comparing dissimilar systems.
Where I disagree, is when we try to break out productivity within a system, using the same currency, in dissimilar environments. Let me try to break this out with examples (entirely fictitious samples, with numbers chosen for ease of math.)
Al is a plumber in Manhattan. He has a small apartment that he rents for $4/mo. He makes 160k/year. A typical days work looks like: walk to subway, take subway 3 stops, walk to work. (0.5 hours) Get truck from shop, pick up service tickets, head out. (0.5 hours) Drive 17 blocks (0.5 hours) circle blocks, looking for parking close to worksite. (0.5 hours) find parking several blocks away. Grab toolbox, a few materials, and walk to site, clear doorman/security get to leaky faucet (1 hour) turn off water, open valve, diagnose problem, realize that the right washer is back in the truck. (0.5 hours) go to truck, grab parts, go back to site, install, close up, get paperwork signed, and back to the truck. (2 hours). We are 5 hours into the day, but maybe he has time to go clear a drain a few miles away. Let's give him the benefit of the doubt, and maybe he can fix the leak, and a drain, in the same day.
Brian is a plumber in suburban Columbus Ohio. He rents a 3/2 house in Columbus for 2k/mo. Brian makes 80k/year. He drives to work (0.5 hours) Get truck from shop, pick up service tickets, head out. (0.5 hours) Drive 17 blocks (0.25 hours) find parking at site.(0.0 hours) Grab toolbox, a few materials, get to leaky faucet (0.1 hour) turn off water, open valve, diagnose problem, realize that the right washer is back in the truck. (0.5 hours) go to truck, grab parts, go back to site, install, close up, get paperwork signed, and back to the truck. (1.5 hours). We are less than 3 hours into the day, but maybe he has time to go clear a drain a few miles away. Let's give him the benefit of the doubt, and maybe he can fix the leak, and a drain, and yet another drain, in the same day.
Charles, used to be a plumber. Then he started his own shop. Now he employs Brian, and 5 other plumbers, plus Wilma, who runs the shop, orders parts, answers the phone, and does the bids, to keep the place going. Charles makes 240k/year. Charles now plays golf, and drinks all day.
Who is most productive?
Measuring by GDP, Charles is half again as productive as Al, who is twice as productive as Brian.
But if we measure productivity in working nodes in plumbing systems, Charles is entirely unproductive, and Brian is half again as productive as Al.
So how do we account for this? There seems to be a disconnect between GDP and productivity.
Cost of living. It much more expensive to live in Manhattan. Are groceries more expensive? Yes, but not significantly. Are utilities more expensive? Again, yes, but not significantly. But Real estate is more expensive. By a lot. So much so, that Al pays 4k/mo, for an apartment that couldn't be rented in Columbus. On order to get a plumber to come work in an area with such a high COL, he has to make twice as much as Brian. And Al's boss isn't cheap, he has to make twice what Charles makes. Is Al's boss twice as productive as Charles?
So what are we to make of this? Higher real estate costs translates to higher GDP, and higher wages. Perhaps, what we need to do is pump up asset prices, to raise cost of living, to raise wages.
And as we have seen over the last decade, this works. We have the magic wand to fix the economy. We call it QE!
But there's always a catch. The number of plumbing nodes in plumbing systems, is not expanding at the rate that our measure of productivity is expanding.
Maybe it's just a plumbing issue. Let's try a new example. One at the other end of the income and education spectrum.
Dave is a badass lawyer in SF. Top of his class at Stanford, Dave makes 1.5M per year. Dave works 80-100 hours a week, and bills for 160. Dave bought a condo across the street from the office for 3 million.
Edith is a lawyer in Columbus Ohio. She did well at Ohio State, and she makes $150k/year. She works 40 hour weeks, and bills 60. Edith has a newer mcmansion in a newer development that she bought for 600k.
Dave makes 10 times what Edith does. He also works 2-2.5 times as many hours, and bills over twice what she does. Is Dave 10 times as productive as Edith? Is he only twice as productive? Dave paid 5 times what Edith did for real estate. Are we better off with a lawyer spending 3 million for housing, or 600k? How much better is the whole economy, if we encourage more Dave's, rather than more Edith's? More Al's, Brian's, or Charles'?
And that is why I object to the use of GDP in this way. The increase in GDP over actual increase in productive assets is an artifact of how we measure GDP. Our productivity hasn't increased, we are merely diluting the unit of measure.
Yes, higher costs of living are associated with higher incomes. But how much of that higher income goes to higher costs. And is making a landlord richer the goal of our economic policy? Should it be? Are we all richer because the landlord is richer? (Full disclosure, I am a landlord.
If we only use GDP as our measure, conveniently, we don't need to ask. But I submit that urban workers are less productive due to logistical considerations, but capital is concentrated in urban areas, and as assets appreciate, this distorts our unit of measure.
TL;DR: high urban incomes seem to get split between higher taxes, and higher housing. Productivity goes down, and costs go up, in urban environments.
Bonus points: Now think about how those differing ideas of costs and income affect choice of shirt color.
More bonus points: How does this income distribution affect urbanization/traffic/development/real estate costs? Are these reinforcing feedback loops, or balancing feedback loops? And what would change that?
Well, maybe that is too many bonus points...
But when I get to talking about simple economics, this is sort of what I am thinking about. It doesn't seem as simple when I try to write about it, though. Sorry about that.
It's probably hard to tell, but I am working on improving my writing for clarity.