Thought for later: theory of the firm —> theory of the job: occupational disaggregation limited by transaction costs.

— Ryan Avent (@ryanavent) April 5, 2013

Much as I’ve written about the height limit, I’m not sure additional posts add value. Since the debate is relatively live now and potentially meaningful, given policy movement in Congress, I’ll give it one more go. Two quick responses to recent urbanist takes.

First, Henry Grabar reckons my calculation that the limit could cost the District upwards of  a billion dollars a year is overstated. Here, as a reminder, is the calculation in question:

The average commercial rent in downtown Washington is $94 per square foot. Using the 1998 [22%] shadow-tax figure, we can guess that perhaps $21 of that rental cost is attributable to regulatory restrictions, which means that the total annual regulatory cost in downtown alone is around $1.4 billion: well above even implausibly high guesstimates of the aesthetic benefits of the height limit.

I want to emphasize that this was meant to be an informative but essentially back-of-the-envelope bit of arithmetic. Still, to the extent that the figure is in error, it is most likely an understatement of the true costs of the limit.

Grabar wonders whether the shadow tax figure, which is computed as a percentage of the total value of property, ought to be applied to rents. The answer is that it depends on what you’re trying to figure out. If one were interested in the precise incidence of the regulatory burden — its distribution between renters and owners — then my calculation would be problematic. But I’m merely aiming to compute the total cost of the  limit. In fact, some of the “tax” wouldn’t be passed through to renters, but its cost would still be there, paid by District property owners, and is still quite real.

The 22% figure is, however, from 1998. At that time, the District was still losing people and downtown was not approaching its physical construction limits. The actual tax is almost certainly quite a bit higher now, which means that my calculation is conservative.

Grabar notes that the 22% figure is an estimate of the regulatory cost of all land-use regulations, and not simply the height limit. True enough, but my calculation covered just the downtown area, in which the height limit is the binding constraint. Virtually all of the cost of the “tax” in the downtown area is associated with the height limit. 

But there are plenty of other places in the District where the height limit is also binding or soon will be: the “Golden Triangle”, Capitol Hill, and NoMa, not to mention an increasing share of other major commercial/residential corridors like Connecticut Avenue and U Street. I’m leaving quite a lot of affected real estate out of my calculation, suggesting that the true cost of the limit to the District as a whole is larger than my figure.

So no, my cost estimate, of $1.4 billion a year, is not at all a fair estimate of the real cost of the height limit. In fact, it’s almost certainly far larger.

On to the second issue. In response to Kaid Benfield’s “urbanist case” for the height limit, David Schleicher asks six pointed questions of limit defenders. Dan Malouff gives his answers, but all but ignores the most critical of Schleicher’s questions: what sort of effect do urbanists think the dispersal of economic activities has on agglomeration benefits?

This is a critically important issue for urbanists to wrestle with. Cities are much more than just economic units. But without an economic basis, modern cities are destined to wither. That basis — the glue that holds cities together — is the set of benefits that emerge from close clustering among economic actors. That’s why people pay so much to be in prosperous cities (and why those cities are prosperous in the first place) and it’s why it is so costly to adopt rules that make it harder for people to cluster.

In the absence of an answer to that question, it may be instructive to answer questions that Malouff himself asks. Like:

What about the current extents of downtown DC make you think it is the perfect geography in which to cluster office development?

But see, this is not at all the argument that “demand-side urbanists” like Schleicher and myself are making. We’re fairly agnostic about where development ought to be. If firms weren’t interested in locating downtown I certainly wouldn’t be trying to force them to move there. 

Instead, we’re observing data that suggest firms very much want to be downtown — or, more accurately, near lots of other firms — and we’re concluding that it is therefore costly to keep them from locating there. I’m certainly not opposed to development in other neighborhoods of the District. Like Schleicher, I am perplexed at the seeming desire among urbanists like Malouff to push offices full of lawyers into those neighborhoods, and at the apparent insensitivity to the fact that pushing such activities into those neighborhoods will raise the cost of real estate in those places and crowd out other economic activities. The height limit is the enemy of the sort of economic diversity Jane Jacobs prized. Limiting the supply of downtown office space raises the cost of downtown office space and of all office space that represents a halfway decent substitute for a location in downtown. As costs rise, only the firms that care most about being in the area — and this is going to be government-sensitive business, because the government isn’t going anywhere — will pay the soaring rents. That, in turn, chokes off every other enterprise that isn’t connected to the business of K Street.

Next, Malouff asks:

Do you accept that there are reasons some people like the height limit which cannot be captured in traditional cost-benefit models?

I don’t really know what to make of this. I certainly accept that some people derive some benefits from the height limit that are difficult to value. But to say that there are some attributes of the limit that simply defy all attempts at considerations of trade-offs is to make a mockery of this debate and urbanism and lots of other things as well. The dollar value of the aesthetic benefits of the viewsheds afforded by the limit are hard to figure, I’ll admit. “Difficult to value” does not necessarily imply that they are of a very high value, it’s worth pointing out. It also doesn’t protect them from assessments of the sense of various trade-offs. I assume that if someone credibly threatened to drop a nuclear weapon on New York if Washington didn’t disrupt its viewsheds Malouff would be all in favor of throwing up buildings around the Washington Monument as quickly as possible. Even if we can’t know with certainty what the market value of a city-wide aesthetic is, we, as sensible people, can think about whether it makes sense to protect certain aspects of a city’s design at extraordinary economic cost to its residents.

I fully accept that some Washingtonians and tourists derive aesthetic benefit from the height limit — maybe even a very large benefit. But the simple difficulty of assessing the precise monetary value of that benefit does not immunize the height limit and its aesthetic benefits from inclusion in cost-benefit analysis. There is a point at which residents will think to themselves, “How much are we willing to suffer to keep this?” And the answer, “I can’t tell you exactly how much, but I know it’s less than what we’re dealing with” is not only valid but is most likely the right one under the present circumstances.

Finally, he asks:

Instead of repealing the height limit, would you accept modifying it to permit taller buildings only at specific and limited locations? If so, how might you go about determining those locations?

I’m not one to let the perfect be the enemy of the good. I’m all for any modification to the limit that makes it easier for people to satisfy market demand. Even a small increase in the limit will have large, tangible benefits for the city.

Ultimately, however, this debate is about more than just the height limit. My hope is to convince urbanists that the economics of cities matters a great deal, and that understanding those economics is critical in ensuring that planning doesn’t systematically undermine the cities urbanists want to help. That’s the point I hope comes across in this discussion.

The debate over the Height Limit is back, and earlier today twitter delivered unto me a piece from Atlantic Cities making “the urbanist case” for keeping the limit. I let loose an intemperate flurry of tweets expressing my frustration with the piece. As I reflect on the brief twitter conversation that followed, I worry that my frustration is being misunderstood. 

I would obviously love to be able to convince most people that my policy position on the height limit is the right one. But the pure fact of disagreement with my view is not particularly vexing. I’m used to having people disagree with me and to occasionally being shown to be wrong, in part or in whole. So while the classification of sub-categories of urbanism is interesting and useful, it’s not the source of my frustration with and feelings of alienation from “urbanism”.

What frustrates me is that many urbanists seem to be having an entirely different conversation. I am used to participating in, and I’d say I identify with, a community of wonky journalists — Economist colleagues, econobloggers, and wonks general — who may be motivated in their arguments by ideological bent but who, when confronted with a tricky question, instinctually head for public data or search the latest working papers. Published research isn’t the beginning and end of most policy questions, of course. But it provides a framework through which disagreeing parties interested in finding the right answer can get closer to the truth.

It is therefore bewildering to be confronted by the sort of loopy logic in the Atlantic Cities piece. Better, in the piece, appears to mean little more than “closer to the author’s conception of good”. We learn that, “parts of the region that are already relatively dense, such as downtown Washington, are fine as they are”. This seems to mean that downtown density is high enough that marginal gains in “environmental indicators” from density are no longer large (enough?), and such that taller buildings might require more energy-intensive materials to construct. Where does one begin to respond to such an argument! Why are those two variables the critical ones — so critical, in fact, that it’s acceptable not only to disregard literature on the economic costs of the supply limits but to act as if it doesn’t exist? 

And to argue that:

Building height has little to do with affordability. The argument that a limit on building height restricts housing supply and thus leads to higher prices is essentially the same argument made against Portland’s urban growth boundary. In both cases, it’s hogwash: if affordability were closely related to building height and density, New York City and San Francisco would be the two most affordable big cities in America.

Is just extraordinarily fallacious. I’m not sure I could construct a paragraph that more aggressively ignores the content of Avent-Glaeser-Yglesias arguments or more egregiously misattributes causation. 

Many urbanists have a view on what cities ought to look like. There isn’t necessarily anything wrong with that. But we should be willing to subject those views to empirical scrutiny. If urbanists aren’t willing to grapple with reasonable and rigorous assessments of the costs from restrictions like the height limit, and either explain in an empirically sound way why those assessments are wrong or why they’re right but nonetheless worth accepting, then they aren’t advocating for public policy. They’re just playing SimCity. Because the point of statistical analysis, however antiseptic and divorced from the view at street level, is precisely to make sure that the people living in those cities don’t get screwed thanks to hare-brained conceptions of what a city should be.

It’s not the disagreement that bothers me. It’s the manner of disagreement.

Trying out some new mental models. Let’s see if this works:

A recession happens when an economy finds itself producing less than it used to. Now, one reason an economy might end up producing less than it used to might be that it has suddenly found itself with less of one of the critical ingredients to production. A foreign country could bomb all of its critical infrastructure. A drought could wipe out critical food crops. A hurricane could destroy a bunch of refinery capacity. Smallpox could kill every tenth worker. Or, more prosaically, a steady rise in marginal tax rates could convince lots of people to work less.

This is what we’d call a supply-side contraction, and it’s pretty intuitive. You have less of something important, the price of that something important goes up, and everyone feels poorer and furthermore understands why they feel poorer (mostly). Demand-side contractions aren’t intuitive. We want to talk about them in supply-side terms: one might argue as if there is a limited stock of prosperity such that an economy that irresponsibly uses too much in one period has less in later periods, or one might simply look around and attribute weakness in the economy to this or that unsatisfying feature of the economy, like undereducated workers or potholed roads or an idiot Congress. If we had more of a good thing, the economy would be doing better. That might be true, but it also misdiagnoses the initial problem. But maybe there is a way to talk about demand-side recessions like supply-side recessions.

Money is a critical ingredient to production. So what if a bunch of money was somehow destroyed?

Well, that’s easy enough to answer. There would be less money to go around. If more money couldn’t be found, then some of the activity that depended on money would have to stop. Unless, of course, ways could be found to do the same amount of activity with less money. The market would try to solve that problem by delivering a rise in the price of money, which means a fall in the price of everything else. To the extent that the market couldn’t engineer a fast-enough fall in the price of everything else, the destruction of money would lead to less production.

But markets are forward looking, right? When the odds of a disruptive war in the Middle East rise, the price of oil rises, even if there is no current disruption in the supply of oil. People anticipate the possibility of a supply shock, and react in the present by hoarding oil. That raises prices in the present, bringing part of the impact of a future supply shock into the present.

We might experience exactly the same thing with money. If people begin to suspect that money will be worth more in the future than they’d previously believed, either because they anticipate that some will be destroyed or simply that the supply of it won’t grow as quickly, then they may hoard money in the present, causing its price to rise in the present and transferring part of the impact of the supply shock into the present. 

Of course, a government that has a sufficiently large reserve of the item that may be in short supply can intervene to smooth out the shock. It could reassure markets that whenever a supply shock seemed to loom and the price of the item looked set to rise above a level dangerous to the economy, that it could and would sell its reserve stocks — of oil, perhaps, from the strategic reserve — into the market, holding the price below the dangerous threshold. Credibility would depend on the size of the reserve; ideally, the government would invent a means to costlessly produce more oil.

One could imagine, though, a really bizarre situation in which people became excessively fearful. They might become so convinced that the price of the critical commodity were going to rise that when the government released more of it into the market, they all just gobbled it up and sat on it, waiting for the eventual appreciation rather than using it, allowing the economy to contract despite an abundance of the stuff the shortage of which initially caused the panic. It’s hard to imagine people falling into a fit of madness like that, but it could be possible. In that case, I think, the government wouldn’t just want to keep parceling out its reserves; they’d just end up being hoarded as well. It would be better to focus on changing people’s minds, explaining, maybe, that they would flood the streets with the stuff until its price began falling steadily, if that’s what it took. Anything to get some people to have doubts and start running down their hoards.

In a serious pinch, I suppose the government could just take its strategic reserves directly to the factories sitting idle for want of it and give it to them on the condition that they get their factories running again, to produce things for the government, maybe, or just to make piles of stuff.

The hard question, I suppose, would be to figure out just what “money” it is that dries up in a monetary supply shock. Is it the stock of paper bills and coin? Is it that plus demand deposits? Is it the stock of credit? Is it unimportant to know the precise forms in which it’s drying up and enough to observe a rise in the price relative to what was expected? Or is it, maybe, just all the money that changes hands in whatever form? We keep track of that, I think.

Is this a helpful mental model? I don’t know. It seems like it could be useful to be able to tell people that we’re in a recession because we suffered a surprising shortage of money, but not to worry because emergency supplies are being flown in as we speak…

Why not tumblr?