Greater Greater Washington

Development


Housing is more than supply and demand

In his Kindle book The Gated City, Ryan Avent argues that the limitations governments place on cities through zoning and other policies are holding back the nation's economic growth.


Photo by M.V. Jantzen on Flickr.

Avent is a very smart writer on urban economics, who I interviewed in 2009. The book is very good, and I recommend it to anyone reading this post.

There are two points that I wish would have gotten fleshed out more in the book, and in discussions of housing markets more generally. The first is whether adding more density will, by the laws of supply and demand, make housing more affordable.

This is a recurring argument among writers like Ryan Avent, Matt Yglesias, and Ed Glaeser. They say that the housing market suffers from a supply/demand imbalance. There isn't enough supply, and that's the reason why so many neighborhoods in so many cities are unaffordable. If only we could boost the supply of housing to meet the demand, we could bring down, or at least stabilize, rents.

This idea rests, first and foremost, on the assumption that housing is a commodity. Consider a different commodity marketwheat. If wheat prices are very high, one way to bring them down is to boost production and flood the market with more wheat. This works because wheat actually is a commodity. Every unit of wheat is more or less the same as every other unit, so the price is whatever price the market dictates.

Housing is different. There are many unique types and styles of housing, some of which are more desirable than others. When demand for housing rises in a neighborhood, rents will rise, regardless of the type or quality of the housing. A neighborhood might have century-old rowhouses, 1970s apartments, and brand new luxury buildings. If demand is rising in that neighborhood, rents for all types of these units will rise.

But what if a neighborhood doesn't have any vacant land sitting around waiting to be developed? How do you increase the supply of housing when there's no place to build new housing? Basically, you have to knock something down and replace it with higher density housing.

Let's imagine that a developer is proposing to level some not-so-great '60s-style townhouses in an urban neighborhood. In their place, the developer is going to build a multi-level apartment complex with a gym, pool on the roof, and ground-floor retail. Perhaps the developer is going to knock down 10 low-quality units and replace them with 50 high-quality units, for a net-gain of 40 housing units.

Even though the number of housing units in the neighborhood goes up, it's virtually guaranteed that the market rents for those new units are going to be higher than the rents for the old units. So the folks who might have been able to afford one of the '60s-style townhouses no longer can afford a luxury-apartment in the neighborhood.

From a developer's perspective, this is a no-brainier. The biggest cost in constructing housing is building the structure itself. The cost of making the units look cosmetically luxurious is marginal, but the price that people will pay for a "luxury unit" makes it more than worth it for developers.

For this reason, it's understandable why some people oppose development, and it's not so simple as sitting them down and saying, "Don't you get it? There's not enough supply to meet the demand in your neighborhood, and that's what's driving your rents up! Once we increase the supply of housing, everyone will be better off."

The truth is that increasing the supply of housing units in a single neighborhood might not have its desired neighborhood-level supply/demand effect, because housing isn't necessarily a commodity. But city-wide, and metro area-wide, it might actually accomplish something.

Unless a developer is going to replace 10 low-quality units with 50 low-quality units, there's going to be a change in the structure of the neighborhood housing market that's different from the impact on the metro area housing market.

Avent argues in The Gated City that homeowners often engage in NIMBYism because they are conservative, risk-averse, and are cautious out of fear that the project might fail. But it goes a bit beyond that. Renters often engage in the same behavior, not because of fear it will fail, but because they're afraid the same project might actually succeed.

Rob Pitingolo moved to the DC area in mid-2010 and currently resides on Capitol Hill. He also writes about issues of urbanism, economics, transportation and politics at his blog, Extraordinary Observations

Comments

Add a comment »

re housing:

1. Adding more housing makes the housing market more affordable over extremely long periods of time. (Decades.)

... if it is built in areas with strong demand.

2. It's just that by definition new housing is more expensive than "old" housing, because it is built at current materials, labor, and land costs. (see the discussion of the need for a large stock of old buildings in _Death and Life of Great American Cities_.)

3. So simultaneously, new housing tends to cost more than old housing, but by increasing supply and slaking demand somewhat it reduces demand of extant housing and releases some of the upward pressure on pricing.

4. And if some of the new housing is constructed through the demolition and conversion of "old" housing, then it has mixed effects because the supply of cheaper housing was reduced while the total supply was increased.

5. This is the case for strong urban markets like DC and SF, parts of NYC, Boston, Seattle, West Los Angeles and County, San Diego, etc. and is specific to particular submarkets within a city and metropolitan region (what I call the regional residential landscape).

6. The dynamics of the housing markets in weak real estate markets and in less dense suburban and exurban locations I am less familiar with and don't feel so authoritative commenting on.

7. But the problem in exurban situations is more housing was built than there is demand for...

8. The books are old, but if you want to understand urban housing issues, check out Rolf Goetze. The books are long out of print. But there is a journal article in the Journal of the Am. Institute of Planners, sometime between 1980 and 1982 summarizing his arguments from his three main books.

by Richard Layman on Sep 21, 2011 10:27 am • linkreport

I'd make the case that there's plenty of housing supply in urban areas. It's just in neighborhoods that certain people don't particularly want to live in. Shaw was once such a place. Those who wanted to live in Dupont were priced out and found affordable housing stock in Shaw. Once U Street became popular again, Shaw became expensive. People who wanted to live there were priced out and moved to Ledroit. And on and on and on. Avent is correct in that zoning rules seem locked in a 19th century mindset where commercial and residental are almost completely segregated. There really should be greater flexibility in how residential and commercial units are zoned to allow residents greater access to the former and businesses a wider customer base.

by monkeyrotica on Sep 21, 2011 10:28 am • linkreport

One Tysons developer plans to build only four-to-five story apartments (c. 800 square feet for a one-bedroom) and indicated it would need rents of at least $2200 per month to make a fair profit. The average existing rent in Tysons is $1800 +/- per month. Another plans to build a 400 unit apartment building of c. 30 stories. Several developers don't think the latter will be built for many years because of the very high cost for concrete and steel construction. The Tysons Partnership estimates each landlord will need to bump all residential rents by $375 per month to recover the costs of proffers and other fees/taxes for the typical apartment in Tysons. Tysons Corner will probably not be a source of reasonably priced housing, except for Fairfax County's requirement that there be 20% workforce/affordable housing for rezoned properties. Tysons will offer people a new choice in housing, but it will most likely be an expensive one.

by tmtfairfax on Sep 21, 2011 10:44 am • linkreport

I'm puzzled as to how your example is supposed to demonstrate that "it isn't just supply and demand" or that "housing isn't a commodity." The example demonstrates that there are quality as well as quantity effects to consider--that's fine. That's as true of wheat as it is of housing.

That doesn't change the fact that Phoenix, which in Ryan's book approved of 60,000 new residential units in one year, has the same price level that it did in 2000, whereas San Francisco, which only approved of 5,000 new residential units in the same year, is at a price level 30% higher than it was in 2000.

If your point is that there are quality effects, then OK--true. So what? Ryan's argument still stands.

by Adam on Sep 21, 2011 10:44 am • linkreport

I'd make the case that there's plenty of housing supply in urban areas.

I'd make the case that there's plenty of housing supply in urban areas.

Hmmm. I'm not so sure that's the case. Existing housing stock tends to be single family housing and not very dense. Occasionally in the eastern reaches of Capitol Hill, you'll see a vacant lot or a dilapidated row house razed, and a 2-4 unit condo erected, but the total number of units tends to be the same. I think we've got about another 10 years of "on and on and on", but eventually you run out of neighborhoods. And "bad" neighborhoods only become "marginal" neighborhoods--and hence acceptable to middle-class residents--after the previous marginal neighborhoods become "good".

by oboe on Sep 21, 2011 10:45 am • linkreport

This phenomenon is precisely what inclusionary zoning (IZ) is designed to combat. You demolish 25 low-quality, low-cost units, and build 150 units, at least 25 of which (ideally) will be low-cost. The developer maintains profit (and therefore incentive to invest); less affluent residents get better housing at comparable cost; and upward price pressure on the broader market is relieved, at the margins.

by The AMT on Sep 21, 2011 10:45 am • linkreport

Monkeyrotica -- that segregation is not a 19th century mind-set, but a mid-20th. See: suburbia for the boldest example of strictly segmenting residential (bedroom communities) and business (the cities and CBDs).

To the whole article, though, one very effective way to include housing at all price points in a neighborhood is through inclusionary zoning. Avent would probably be against this, as it is government regulation, but it's the kind that has been shown to work quite well in the places it's been implemented wisely. Basically, developers get to trade off density (more, meaning more tenants, meaning more revenue for very little additional construction cost) for restricting the rent (and eligible income levels) in a portion of the units. There was a great article in the NY Times a few weeks ago about several "luxury" buildings in Manhattan that have been built this way, with the affordable units indistinguishable from the market-rate. In fact, since it makes the development eligible for tax credits, it's one of the few ways new multifamily housing is being built in the current tight-credit markets.

by Elle on Sep 21, 2011 10:47 am • linkreport

wrt monkeyrotica's point, my retort to the work of Prof. Hyla about Shaw and how it is "resistant to gentrification" is that it isn't resistant, just lagging, because the green line wasn't fully operational, and the properties in other areas had to be snagged first, before people would be willing to consider living there.

wrt tmtfairfax's point. I could have illustrated my point 1 with this article on the construction costs of various types of housing, single family detached, attached, and multiunit.

http://places.designobserver.com/media/pdf/Explaining_Res_803.pdf

DK if the cost index has changed since.

by Richard Layman on Sep 21, 2011 10:55 am • linkreport

I'd argue that IZ and other forms of rent stabilisation actually INCREASE rents, not decrease them. Think of it this way: say you have to make 20% of your development affordable. That automatically means that the 80% at market value must be priced higher in order to overcome the shortfall in revenue from that mandatory provision. The same goes for rent stabilisation within certain apartments based on duration of living there (like NYC), where a landlord must increase rents to compensate for lower-than-needed revenue from rents that have been artificially lowered. In essence, other residents are paying more to subsidise others' housing. While I understand that housing is expensive, and I know that all too well, we do a disservice to everyone by distorting the market. Yes, luxury housing will remain above nominal values, but eventually, if landlords and developers want to make a profit when there is a saturation of a similar housing type (in this case, luxury condos or apartments), they will eventually have to lower their rates, but it's not apparent initially.

by Phil on Sep 21, 2011 10:56 am • linkreport

It's just that by definition new housing is more expensive than "old" housing, because it is built at current materials, labor, and land costs.

This isn't exactly correct. Rather, in a idealized world, the marginal cost of a new housing unit should equal the construction cost of a new housing unit.

The comparison to housing as a commodity does work in the aggregate, but it's not supposed to work in comparing individual houses or even metro areas. Housing isn't a commodity, but neither are tons of other products. Here's an example.

Fancy suits aren't commodities like wheat. There's a lot of variability between business casual and tuxedo. But simply not being a commodity doesn't mean supply and demand models break down and collapse.

(Sorry for the double comment, I accidentally put this in the wrong GGW post)

by WRD on Sep 21, 2011 11:17 am • linkreport

Phil -- the theory of IZ is that you have the ability to provide significant density bonuses, which covers the additional costs of selling units below market cost.

In general, most jurisdictions have the ability to grant density bonuses. For DC, this capability while not nonexistent, is pretty minimal by comparison, so in DC, you're right, IZ leads to higher prices on the market rate units, probably.

2. Another thing I forgot to mention in my first response is that the housing supply at the submarket level (e.g., "Capitol Hill" or "Downtown Silver Spring") can often be extemely homogeneous. Adding supply by adding different types of housing (e.g., apartments, condos, flats, etc.) does change the market and can reduce pressure on costs and rising housing prices.

Again, it takes a long time for this to work through the market given that properties are constructed at different times with different cost considerations.

But it does behoove one to look at various submarkets by type of housing (single family detached, attached, multiunit) and owning vs. renting, to be able to answer the question definitively.

by Richard Layman on Sep 21, 2011 11:19 am • linkreport

Partial response to tmt fairfax. In tysons especially where new development is taking place on solely commercial corridors and now will be more mixed use rents are going to be high but those high rents aren't replacing lower rents through demolishment or whatever either. So I don't expect it to be reasonable in the short term but that doesn't discount it in the long term (though with its metro accessibility it is likely to remain high though this would have effects on other parts of the county/region)

by Canaan on Sep 21, 2011 11:37 am • linkreport

WRD -- I don't get your point. Yes the marginal cost of a new housing unit is theoretically equal to the cost of construction of a new housing unit. The issue is that the marginal construction cost of a new housing unit is > the cost of an extant comparable housing unit constructed earlier, when inputs costs were less then, compared to today. (And this is irrespective of the other issue of calculating-generating profit over a total development vs. an individual unit.)

by Richard Layman on Sep 21, 2011 11:58 am • linkreport

1. re not a commodity. yes its true that other goods and services have complex quality distinctions and are not "commodities" (wheat itself happens to be relatively complex comapared to corn - wheat has different gluten levels, protein levels, etc - corn, at least feed corn which is most of the market, is relatively homogeneous, at least it was pre GMO) The issue with housing are the high costs of moving - both the actual physical move, the search costs, and the loss of the non market provided externalities - all the neighbors you know, etc. If for some reason I need to switch from buying black suits to buying blue suits, its no big deal. If my (lets say rental) house Clarendon is torn down for a hirise, and I have to move to Fairfax, not only is that a big chunk of money and time, but it may take me years to recreate in Fairfax the community I had in Clarendon. Thats a very rational reason for resistance to change. Its ALSO an argument for IZ.

Richard points out that in places without height limits, IZ financed through density bonuses should be easier. Well guess what, here in NoVa, where, IIUC, Arlington and I think Fairfax do just that, theres ideological opposition. "OMG! Poor people are getting luxury amenities! Now those lazy bums will have no incentive to work!" Opposition to affordable housing programs, INCLUDING those based on density bonuses is now, IIUC, the official position of the Fairfax County GOP.

by AWalkerInTheCity on Sep 21, 2011 12:05 pm • linkreport

Steel and concrete construction with underground parking in a booming amenity rich neighborhood should not be expected to be of the low rent variety. That is the most expensive type of housing stock to build.

But if enough of that "luxury" stock can be built it can satisfy a great deal of currently unmet demand. Then the older apartment stock or new "stick built" housing in adjacent neighborhoods have a better chance of having stable pricing rather than rising swiftly with the tide. Continue to build hi-rises in Noma/Navy Yard but encourage stick built in Fort Totten, Anacostia, Trindidad and Deanwood...

by Jason on Sep 21, 2011 12:12 pm • linkreport

Rob is conflating a couple of different geographies here.

Housing most certainly behaves like a commodity at the regional level - Adam cited the stats from The Gated City about the PHX and SF metro areas and their new housing units. You can't expect for one to add 60,000 and another to only add 5,000 and not see some constraints of supply.

Rob attempts to rebut this with a neighborhood-level example, but that's not really disproving Avent's point. The neighborhood-by-neighborhood price dynamics are much more complicated. As TheAMT notes, policies like IZ are meant to try and address that on a case by case, project by project basis. We can argue if they're successful or not, but that's the goal. That doesn't change the fundamental need at the functional economic area of the 'city' - which is the region.

Richard is exactly right about the need to differentiate the submarkets, both in geography and in terms of unit type.

I guess the point of all of this is that none of the arguments Rob presents really addresses the core of Ryan Avent's argument. You can't mistake neighborhood-level change with the broader patterns.

I'd also note that you don't have to knock everything down to add more supply. Accessory dwelling units, English Basements, alley dwellings, etc - all of those are nice and easy ways to add units without fundamentally altering the physical landscape of an established neighborhood.

by Alex B. on Sep 21, 2011 12:13 pm • linkreport

I guess the point of all of this is that none of the arguments Rob presents really addresses the core of Ryan Avent's argument. You can't mistake neighborhood-level change with the broader patterns.

Alex, I'm with you 90% here. I should have done a better job fleshing out the point that city-wide and metro-wide, markets behaves much differently.

But the key is this: people oppose development because it's in their neighborhood, not because it's in their metro area. Unless you're developing greenfields, the only way you're going to change the metro area market is by simultaneously changing one (or some) of the neighborhood markets.

by Rob P on Sep 21, 2011 12:22 pm • linkreport

Rob P: The city which defines the context of your question, Washington, is growing and there is a need for more places to live. In housing, demand changes much faster than supply.

You could put some of your supply-vs-demand questions to rest if you consider the opposite situation: a city with a contracting population and negative demand for housing, such as Detroit, where a family-sized house costs $20,000.

by goldfish on Sep 21, 2011 12:50 pm • linkreport

Rob, I think it's simpler than that (and hence why I don't buy your distinction): people oppose change, period. It's not a rational decision, but it is (at least in their mind) a risk-averse one which is really the fundamental issue of NIMBYism that Avent mentions.

What's missing is the alternative: Some people oppose local development because it will raise costs, or because they won't benefit from the increase in aggregate supply. That might be true. But if the development doesn't happen and the neighborhood remains in demand, their rents will rise anyway. That's what is commonly missing from a NIMBY mindset. It misses on the dynamism of any urban place by assuming that the status quo can be maintained, and that not changing will maintain an equilibrium. That's a false assumption.

To your larger point, if you're trying to say that Avent's argument that more housing on the regional level is necessary won't be a convincing argument to a local NIMBY, I don't disagree. I don't think Avent would, either. You're not likely to convince NIMBYs on a case-by-case basis. The problem isn't really one of individuals who are opposed, but one of policy. If those opposed can shape policy (as they have in our history), then we get restrictive policies. Avent's work aims to inform the policy makers to try and add some relief valves into the system to provide that regional perpsective to help balance out those local concerns.

Think of something like a zoning budget, for example: http://thinkprogress.org/yglesias/2011/04/27/200745/balancing-the-zoning-budget/ You don't argue to the local, neighborhood opposition to adopt a zoning budget, you go to the city and convince them with your data on aggregate.

As for addressing neighborhood-level concerns, you'd also need a broader set of tools to manage change (that's where something like IZ comes in, and the devil's in the details) - but we have to fight back against the idea that stopping development counts as management of change.

by Alex B. on Sep 21, 2011 12:59 pm • linkreport

I agree with commenter Adam above. If you start discussing things in a supply and demand paradigm, then at some level you are implicitly assuming that the goods are homogenous ("goods" in this case being units of housing). If you want to make the point that the goods are not homogenous, then you can't fault the supply and demand paradigm itself for not taking that into account.

Now once you've pointed out that units of housing are not homogeneous, the next step is to extend/incorporate the assumption of heterogeneous goods into the supply and demand framework, which would yield a useful model for analytical use. It's not enough in policy circles to just point out problems and not try to solve them.

by gogurt on Sep 21, 2011 1:08 pm • linkreport

There is a classic externality argument hiding in here, somewhere.

by charlie on Sep 21, 2011 1:14 pm • linkreport

Alex:

There is one very rational side to NIMBYism. It's true that it is not in renters' interest for prices to rise, but I wonder how many NIMBY folk are renters (I ask this genuinely, as I have no idea).

The case for NIMBYism for an owner is clear. Not only is construction a nuisance, and population density in their neighborhood means more noise and such, but they have a direct stake in restricting supply so that prices will go up. Simply put, if the price of their home goes up, they profit from it.

by Adam on Sep 21, 2011 1:31 pm • linkreport

Alex B -- one of the points I make about community planning is that it is set up to fail because the "city" planners in theory have a set of objectives concerning "citywide" issues, while residents in a particular area are focused on the "neighborhood" issues.

While planners are supposed to address both, residents are not. Furthermore, mostly city planners do a piss poor job articulating and defining citywide issues in the context of neighborhood issues. So when citywide and neighborhood concerns conflict, it can be difficult to come to consensus and the residents believe that they are being screwed.

This comes up with regard to new development all the time, such as housing. Especially because the time horizons people are thinking about are so different. The planners think in terms of decades, the residents in terms of "right now" or "since I've lived here."

It's worse when the citywide plans and objectives are inadequately defined or unfamiliar, such as a decent housing policy for a city, and the idea that neighborhoods to be resilient need a variety of housing types.

E.g., wrt DC, I argue that rather than immediately jump into zoning revision following the Comp Plan approval process in DC, that instead OP should have gone on a deep and wide road show throughout the city to explain the citywide elements of the Comp Plan as well as the area elements, and if not to build consensus about it, at least awareness.

by Richard Layman on Sep 21, 2011 1:34 pm • linkreport

@Richard Layman

The issue is that the marginal construction cost of a new housing unit is > the cost of an extant comparable housing unit constructed earlier, when inputs costs were less then, compared to today. (And this is irrespective of the other issue of calculating-generating profit over a total development vs. an individual unit.)

My point is that this simply isn't the case in many cities around the country. The model of "house = marginal cost of construction of a new unit" holds up pretty well in places like Dallas, Houston, and Phoenix most famously. But it also holds up well in other places, too, like Charlotte or Atlanta.

As another aside, let's not conflate real and nominal construction prices. While the inputs were cheaper in nominal terms years ago, the real cost may have increased, decreased, or stayed roughly the same.

@AWalkerintheCity

Certainly my metaphor of a suit wasn't perfect but hey, it was off the top of my head. Your example of moving costs is a great example of a transaction cost. Certainly the housing market has high transaction costs. Some are inherent (like the ones you listed). Others are created by the government (mortgage interest deduction, to name one that's not the zoning code).

Housing isn't the only market to have high transaction costs. It doesn't suspend supply and demand.

@gogurt

the next step is to extend/incorporate the assumption of heterogeneous goods into the supply and demand framework

But it already does include heterogeneous goods! Who ever looks at the market for Blue Ford Mustangs or Dole bananas from El Salvador? Market researchers, not micro-economists. This is why we're discussing the "housing" market together, not the rental market or the single-family homes market in isolation. The idea of substitution is handled by supply and demand models.

by WRD on Sep 21, 2011 1:40 pm • linkreport

WRD -- oh, I see your point now. I was writing about DC mostly. But obviously what you write is the basis of why "the national housing market is not recovering." In most markets the overhang of an overbuilt market has depressed prices below new construction costs. Not to mention overbuilding (in terms of the number of households that can the mortgage for an owned house) and the impact on supply.

And now there is more construction of multiunit apartments in a number of markets as a result.

Another response has been to decrease size of units to cut costs somewhat, but in most markets production single family housing builders are screwed.

by Richard Layman on Sep 21, 2011 2:33 pm • linkreport

@ alex

"But if the development doesn't happen and the neighborhood remains in demand, their rents will rise anyway. That's what is commonly missing from a NIMBY mindset. It misses on the dynamism of any urban place by assuming that the status quo can be maintained, and that not changing will maintain an equilibrium. That's a false assumption."

Well its also often that the renters in an area of increasing desirability oppose not only the new high density residential, but other changes that increase desirability. Even when an area increases in desirability, and rents for older housing increase, it often will still be more possible for them to afford to stay then if their units are torn down to create new units. It thus may STILL be rational for them to oppose new hirise development. as another poster mentioned, the incentives are different for owners - who may WANT to see prices and rents for older units rise, without the nuisance effects of new high density construction.

@WRD I would not suggest that supply and demand do not funtion in housing markets. They do, and can explain a great deal in terms of pricing. If "supply and demand functioning" means supply and demand determining price thats one thing. If 'supply and demand functioning' means that the market price is A. pareto optimal and even B. optimal for all market participants thats not the case I believe. Not just the transaction costs, but the externalities in housing, can make resistance to change thoroughly rational. And by externalities I am NOT referring so much to traditional nuisance effects, as to the social benefits of living in certain communities. Unless you are going to pay your new neighbors to hang with you on the street corner, there are things that you can't buy - because you cannot buy them, there disruption by market forces is a negative externality of market forces (of course there are ways people try to buy "community" but the results of that are mixed, IMO)

by AWalkerInTheCity on Sep 21, 2011 2:46 pm • linkreport

It's not that "If I restrict housing in my neighborhood, then the value of my house appreciates". That's a classist argumentative statement that does nothing to further the discussion

It's that I chose a house with a certain density characteristic, and if you change the characteristic of the neighborhood, then I have to move. Why should I have to move so that you can make more money on high density housing and/or someone else can benefit from being close to amenities that have developed from the existing density (amenities that will gradually disappear as the density changes)?

Why do all the young 20 something year old writers on this blog seem to push collectivist solutions to their personal economic problems?

by ahk on Sep 21, 2011 2:47 pm • linkreport

@Richard Layman

Actually, there hasn't been a boom in housing construction.

Yglesias says it's more of a boom in price of land.

by WRD on Sep 21, 2011 2:49 pm • linkreport

It's that I chose a house with a certain density characteristic, and if you change the characteristic of the neighborhood, then I have to move.

Says who? If municipal governments adopted Ryan Avent's policies, then the housing supply would expand rapidly to meet housing demand, so prices would not rise and you wouldn't have to move.

More to the point, the fact that a neighborhood had certain characteristics when you moved there does not give you the right to tell others what they can do to that neighborhood. Either you own the land or you don't; if you don't, then it isn't your place to tell other people what they can do with their property.

by Adam on Sep 21, 2011 2:52 pm • linkreport

@ahk

Why do all the young 20 something year old writers on this blog seem to push collectivist solutions to their personal economic problems?

Can't let this one go. To carry your example forward, the houses can't be made more dense unless your own neighbors sell the land to developers in a voluntary transaction. The developers are, absent collectivist interventions, free to develop the land as they see fit.

No one is forcing you to move at all. In fact, no one promised the density or characteristics of your neighborhood would stay constant. It's a risk you have to run, absent collectivist intervention.

So I'd turn your question around and ask: Why do people like you (whoever you are) demand collectivist protection from the free market?

by WRD on Sep 21, 2011 2:56 pm • linkreport

"...I chose a house with a certain density characteristic, and if you change the characteristic of the neighborhood, then I have to move. Why should I have to move ... ?"

Because you don't have property rights on land you don't own. So go west, young man... go west.

For everything else, WRD already said it better.

by Andrew in DC on Sep 21, 2011 4:51 pm • linkreport

I don't interpret that data the way you do WRD. Post 2008 real estate crash isn't the period I'm referring to, at least on a national basis.

by Richard Layman on Sep 21, 2011 5:04 pm • linkreport

you don't have property rights on land you don't own

Property rights are created by government. The practical effect of zoning and historic districts is to give owners of single-family homes property rights over their neighbors' property. Whether this is a good thing or a bad thing is a partly empirical and partly value-related question, and decisions about property rights should be made through the political process. But there's no point to arguing (except in courts of law which base their decisions on legal fictions) that transfer of property rights to neighbors is illegitimate -
that horse has left the barn.

I chose a house with a certain density characteristic

Most zoning ordinances forbid you to convert a one-family to a two-family house, even if the change is invisible from the outside. Zoning is about keeping out people as much as or more than it's about keeping out buildings.

by Ben Ross on Sep 21, 2011 5:38 pm • linkreport

Oops!

by Ben Ross on Sep 21, 2011 5:39 pm • linkreport

@Richard Layman--

That graph is showing that housing construction during the housing boom of 2002-2008 wasn't high by historical standards. You can even ignore everything after 2008 and still see the housing crisis wasn't preceded by a huge increase in the number of homes constructed. This is especially true given population growth and home formation over that time period.

by WRD on Sep 21, 2011 7:34 pm • linkreport

Not high by "historical standards" because the US housing market supra-overproduced in the 1970s and 1980s. See the points by Rolf Goetze. His basic point is that more housing was produced than needed, which is what led to the large scale abandonment of center city housing.

by Richard Layman on Sep 23, 2011 7:07 am • linkreport

No one is forcing you to move at all. In fact, no one promised the density or characteristics of your neighborhood would stay constant. It's a risk you have to run, absent collectivist intervention.

No one is forcing you to move. However, it will be necessary to move if you want to live in a neighborhood like the one you had before the developers changed it.

I agree that change is a constant and that the only way to have guaranteed control over a piece of land is to buy it yourself. (Unless the government condemns it, of course.) It doesn't follow, however, that people will therefore accept all changes without protest. Or even that they ought to.

Also, some people are more enchanted with the virtues of the free market than other people.

by Miriam on Sep 23, 2011 9:06 am • linkreport

"In fact, no one promised the density or characteristics of your neighborhood would stay constant. It's a risk you have to run, absent collectivist intervention. "

Absent government intervention. The use of the term "collectivist" is a libertarian scare word. The government engages in zoning activity to prevent nuisances imposed on property holders. It seems reasonable to expect it to similarly use zoning to prevent the kinds of nuisances imposed by changes in density, UNLESS said changes are manifestly in the broader public interest. I happen to think that in many instances today the social need for higher density overrides the otherwise reasonable expectation of homeowners that zoning will protect prevailing patterns - but to label that expectation itself "collectivist" seems to me to go too far, and to lead to a set of ideological that would block out far more desirable interventions than prevailing pattern zoning.

by AWalkerInTheCity on Sep 23, 2011 2:24 pm • linkreport

"Why do people like you (whoever you are) demand collectivist protection from the free market"

because in a situation with such manifest externalities, there is no assurance that the free market will result in an optimatal solution.

Why do property holders demand that the govt protect them from burglars? Because it leaves them better off, and is consistent with societal goals.

by AWalkerInTheCity on Sep 23, 2011 2:28 pm • linkreport

I'm not saying there is zero role for government in the housing market. I was a little sarcastic because ahk's comment was hypocritical. (He praised "collectivist intervention" while simultaneously criticized people praising "collectivist intervention.")

because in a situation with such manifest externalities, there is no assurance that the free market will result in an optimatal solution.

I've grown skeptical of any internet argument that relies on externalities. Externalities do exist, all the time. But there are good methods for the market to address externalities.

Coase Theorem describes how markets address externalities. Market participants are free to bargain with developers and pay them to change building characteristics.

Further, there remains another problem: government failures. Externalities may exist, but there is no promise the government's solution will be better than just living with the externality.

To relate back to housing in a brief summary:

While housing externalities do exist, homeowners, homeowner associations, and others are free to purchase properties themselves so as to eliminate "development risk." Additionally, they can negotiate with developers to get developers to change their plans. Lastly, the government's solution (zoning and other regulations) often leave everybody worse off. At the margin, we should roll back zoning protections to neighborhoods and move toward a more market-based system without completely dismantling the institutions in place now.

by WRD on Sep 24, 2011 1:56 pm • linkreport

WRD,

How do you propose to internalize the negative externalities of higher density housing and higher density urban development in general (more congestion, more crowding, more noise, more pollution, more litter, less privacy, less greenspace, etc.) using coasian bargaining?

by Bertie on Sep 24, 2011 4:02 pm • linkreport

Why do people like you (whoever you are) demand collectivist protection from the free market?

Because the free market often produces outcomes that reduce welfare. That's why we have laws that set aside land for recreation (e.g., public parks). And laws that protect land from development on environmental grounds. And laws that protect land or buildings from being developed or destroyed on historical or cultural grounds. And laws that set health and safety standards for building construction. And laws that prevent forms of land use that we have decided are incompatible (e.g., building a boiler factory in the middle of a quiet residential neighborhood). And so on.

by Bertie on Sep 24, 2011 4:20 pm • linkreport

Coase theorem requires transaction costs to be low. I doubt this is the case when you are talking about many small participants in the market trying to negotiate with a larger, professional developer.

by Michael Perkins on Sep 24, 2011 6:02 pm • linkreport

@Michael Perkins--

I think the fact that we observe citizen's groups are so successful at organizing and influencing the process now is a strong piece of evidence transaction costs aren't that high.

@Bertie--

Remember, I'm not advocating the Ron Paul position of no zoning here at all, just a rollback. Coase's original paper was about one business paying another to do (or refrain from doing) something analogous.

But for illustration purposes, let's imagine an apartment building in Dupont. Neighbors don't want it taller and do want more parking spaces. You know, hypothetically.

If the neighbors feel so strongly the hypothetical development should have more parking and fewer housing units, the neighbors should be forced to pay actual dollars to compensate.

These changes are real, dollar losses. To the developer, who can't make as much money. To renters, who suffer higher rents. To the neighbors themselves, even. But none of the hypothetical NIMBYs pay anything for their preferences.

I don't really believe forcing neighbors to pay is a realistic solution, of course. Realistically, I think the proper solution is a different legal standard applicable to developments. I hope the different standard ends up resulting in outcomes that are closer to what would happen in the "free market."

by WRD on Sep 24, 2011 6:29 pm • linkreport

If the neighbors feel so strongly the hypothetical development should have more parking and fewer housing units, the neighbors should be forced to pay actual dollars to compensate.

The new arrivals are imposing costs on everyone who already lives there -- more congestion, more crowding, more noise, less privacy, etc. -- that are not reflected in the market prices the new arrivals pay. That's why we limit new arrivals through various density-restricting zoning laws. If you want to eliminate these restrictions, you need to internalize the costs through a "density tax" or somesuch. People are not generally willing to sit back and allow other people to damage the quality of life in their community without demanding some form of compensation in return.

by Bertie on Sep 24, 2011 7:26 pm • linkreport

But they do pay for those! The newcomer sits in the same traffic the old residents do. The newcomers hear the same noise, and have the same level of privacy the existing residents have.

by WRD on Sep 24, 2011 8:50 pm • linkreport

The newcomer increases congestion ("traffic"), noise, pollution, etc. for everyone else. That's why these costs are negative externalities.

by Bertie on Sep 25, 2011 1:25 am • linkreport

Newcomers don't pay the marginal cost of the increased density...but neither do existing residents, either.

I don't see this as a convincing economic argument to restrict density.

by WRD on Sep 25, 2011 1:35 pm • linkreport

The newcomers cause an increase in density and hence an increase in negative externalities. To limit those costs, people support laws that limit increases in density. If you want to get rid of those laws, you're going to have to find some other way of dealing with the costs. And not just any way, but a way that is politically and economically viable. Simply ignoring the costs and telling people that you think they should accept whatever increase in density the free market may produce isn't likely to get you very far.

by Bertie on Sep 25, 2011 3:31 pm • linkreport

Remember, I'm not advocating for "simply ignoring the costs and telling people that you think they should accept whatever increase in density the free market may produce."

I'm trying to apply the following process: get the economics right first, then think about how to get there using politics second.

So economically speaking, newcomers don't actually pay the marginal cost of their moving into a neighborhood. However these people aren't just materializing out of nowhere. They live somewhere else, where they are imposing these same costs on a different neighborhood.

Also, existing area residents also don't pay the marginal cost of the externalities they cause.

Let's also think of this another way. The "newcomer" is causing a given amount of pollution, noise, and other externalities right now. (They're moving from somewhere, right?) Is their move to a denser area reducing the overall amount of negative externalities?

More concretely: is it better to have a newcomer move to Dupont Circle or Anne Arundel County? Is it better to have a new resident in NYC or Houston?

by WRD on Sep 25, 2011 3:50 pm • linkreport

But you ARE ignoring the costs. You're saying that communities should get rid of zoning laws that restrict density and that the residents of those communities should just endure the resulting increases in congestion, crowding, pollution, noise, etc.

Again, I don't understand why you seriously think people are just going to roll over and agree to eliminate laws that protect aspects of their community they value highly -- mobility, privacy, cleanliness, peace and quiet, and so on.

by Bertie on Sep 25, 2011 4:27 pm • linkreport

It would be interesting to see Mr. Avent reply to this. Wonder why he hasn't left a note here?

by Geoffrey Hatchard on Sep 25, 2011 10:42 pm • linkreport

Correct me if I'm wrong here, but we're having another version of a baseline debate. What is the "baseline" to which a change should be measured?

You're arguing the newcomer imposes new costs on his new neighborhood. I'm arguing that the newcomer is also relieving his old community of those costs and that the net effect on costs to "society" is ambiguous (but probably positive).

This helps illuminate an interesting corollary: if a newcomer is imposing a marginal negative externality on his new neighborhood by moving in, is he also imposing a marginal positive externality on his old neighborhood by moving away?

You're also making a point about distribution: who pays the new costs. An important discussion but beyond the scope of what I can address here.

by WRD on Sep 25, 2011 10:43 pm • linkreport

@WRD: Is the newcomer also bestowing a positive externality by moving to the new neighborhood?

*Tax revenue, which allows public services and/or lower tax rates on you
*Increased patronage of local shops and stores, allowing a greater variety of viable businesses
*Increased patronage of public transit, allowing more frequent or extensive network given a certain level of public support

New people coming to a neighborhood aren't always a net negative. Some of the new people could end up being good friends?

by Michael Perkins on Sep 26, 2011 9:10 am • linkreport

@Michael Perkins--

I was hinting at that, in response to Bertie's comment above, but it's nice to see it stated explicitly. These whole externality debates are way more complicated than they seem at first!

by WRD on Sep 26, 2011 11:04 am • linkreport

you guys are conflating two issues

1. How do you assure the optimal outcome and
2. How do you distribute the benefits

As was stated above, the Coase theorem requires small numbers. because of the free rider problem. Why should I kick in to bribe the developer to build smaller, when my neighbors will do so anyway? Thats why we have govt, to deal with free rider problems like that. (As for negotiating that occurs now, that has little or nothing to do with Coase - thats made possible BECAUSE zoning and other state interventions gives neighbors bargainging power without having to put together a bribe fund)

As to govt failures - well yes they exist. But private negotiations as a supplement can help lessen them (for example the developer bribing the neighbors to support zoning waivers) Also, democratic discussion (like GGW arguing for density) can address those.

As for who gets the money - personally I see nothing holy in the rights of EITHER the neighbors, or the land owners of land for development(the ultimate benefiariaries of rights for developers) I think all property rights are arbitary, and that we establish them for functional reasons. Id just as soon see the surplus go to broader public interests, via the tax system (call me a Henry Georgist, if you must)

But again, Mr Coase's basic insight, IIRC, was that the question of getting the optimal physical result is conceptually distinct from the allocation of benefits. That it doesnt matter WHO has the property rights, given that SOMEONE has the property rights (abstracting of course from the transaction costs problem)

by AWalkerInTheCity on Sep 26, 2011 3:49 pm • linkreport

" I'm arguing that the newcomer is also relieving his old community of those costs and that the net effect on costs to "society" is ambiguous (but probably positive"

If you ignore the differences between different communities, and assume that, pardon, density is a commodity with equal external costs everywhere (like say sulfur dioxide) than OF COURSE zoning to preserve prevailing development patters is going to come out sub optimal.

The entire logic of prevailing development patterns is that another apartment building in an area of apartment buildings has DIFFERENT externalities than an apartment building in a neighborhood of SFH's. Because of different preferences (combined with transaction costs of moving that make keeping the status quo optimal) , different infrastructure, etc, etc, etc. Now you can debate the specifics of that, but when you ASSUME that density is a commodity with no geographic specificity - you have assumed away the entire problem

by AWalkerInTheCity on Sep 26, 2011 3:56 pm • linkreport

Coase theorem doesn't require small numbers, just manageable transaction costs. I said above but will repeat: if transaction costs are so high, why do we see neighbors organize so effectively now? This isn't dispositive, but it is some concrete evidence that transaction costs are manageable.

There are large free-rider problems in development issues but I think they run both ways. And let's remember, this "free market" scenario I laid out is totally unfeasible and I do not suggest it is sound policy. In some ways I've set up a straw man that I'm only really half defending.

If you ignore the differences between different communities, and assume that, pardon, density is a commodity with equal external costs everywhere (like say sulfur dioxide) than OF COURSE zoning to preserve prevailing development patters is going to come out sub optimal.

Let's go to what I said: "I'm arguing that the newcomer is also relieving his old community of those costs and that the net effect on costs to 'society' is ambiguous." Perhaps I should have included the parenthetical in a separate sentence to be more clear.

By the parenthetical, I'm trying to say that I haven't seen is a convincing argument that we face net negative externalities by zoning more densely in reality. I actually haven't read even a clear yet unconvincing argument along those lines, so if you know of any, please post some links!

What I have seen is a lot of "This imposes costs on us!!!!!!!" arguments. True, but to make an externality argument about costs to society, we need to look at the alternative.

I'm also suggesting the trade-off we face, generally speaking, is between: (1) the less environmentally friendly development in a exurb versus (2) denser city development. By analogy, when we keep a resident out of Dupont, we put him in Anne Arundel County, causing more negative externalities than existed in the first place.

The entire logic of prevailing development patterns is that another apartment building in an area of apartment buildings has DIFFERENT externalities than an apartment building in a neighborhood of SFH's.

I'm glad you brought this up because it is a key point often lost by people like me. Across-the-board denser zoning isn't a great thing. Rather there's a status quo bias I'm trying to illustrate. The difficult part, as I see it, is how to go from SFHs to townhouses, townhouses to apartments, apartments to high-rises. It's a gradual process, certainly, but has a jerkiness partially because the zoning code has blanket categories.

Also, thanks for continuing to engage. I think I'm learning here.

by WRD on Sep 26, 2011 4:30 pm • linkreport

"By the parenthetical, I'm trying to say that I haven't seen is a convincing argument that we face net negative externalities by zoning more densely in reality. I actually haven't read even a clear yet unconvincing argument along those lines, so if you know of any, please post some links"

"Zoning more densely" can mean allowing, oh, the redevelopment of the Potomac Yard area in Alexandria to be 30 story buildings instead of 20 story buildings (with a much higher FAR) OR it can mean allowing a 20 story building on an existing block of 3 story townhouses.

I personally tend to support (given appropriate infrastructure,etc) the former. Prevailing form zoning is designed to prevent the latter. Indeed doing the latter can undermine "good" density by distributing it where it can't create the critical mass for amenities and facilities, etc. I can think of places in the suburbs (more in Florida than Virginia) where they plopped a half dozen THs in a SFH nabe - the SFH owners didnt like it, and it DID not create a walkable area, or support transit, etc. Indeed lowering prevailing density can ALSO reduce QOL (look up some pictures of parts of Detroit, or Camden, NJ, to see that). The assumption (which I assume is supported in the zoning and planning lit, but which seems intuitive to me) is that there are benefits,NOT to lower density in the abstract, but benefits both visual, and in terms of infrastructure use, to relatively homogeneous densities (and with variation in density within reason, and planned out intelligently - think Manhattan with THs on the side streets and large apt buildings on the avenues and the wider side streets)

How to get from one to the other - in some places, you can't. The costs of transition simply mean that history makes the transformation suboptimal - so a former rail yard near old town Alex will end up much denser than Old Town alex itself ever will. Sometimes its possible through state action, usually done to transform an otherwise blighted area (this isnt as popular at it once was, and historically was not used to increase density) Sometimes its possible for a developer to do this by buying up an entire neighborhood and transforming it all of a piece - see MetroWest, which WAS a suburban SFH subdivision near the Vienna metro, and which was completely bought out by Pulte, I think, and is being rebuilt as TOD.

by AWalkerInTheCity on Sep 26, 2011 4:52 pm • linkreport

I'm trying to say that I haven't seen is a convincing argument that we face net negative externalities by zoning more densely in reality.

Even if you could demonstrate that eliminating zoning laws would increase welfare in the aggregate (and you can't), that wouldn't make the negative externalities of higher density any less real for the people who would have to endure them. People are not going to give up laws that protect aspects of their community they value, like mobility and privacy and quiet. I don't know why you keep ignoring this basic point. To the extent that people genuinely prefer high density, the political process and the market will deliver high density. But most people don't want high density. That's why it's relatively rare. Absent a huge shift in preferences, this isn't going to change. You can keep throwing around feel-good buzzwords like "sustainability" and "efficiency" as much as you like, but unless you can persuade people that living in small, dense housing and giving up car travel for public transit is genuinely in their best interests they'll just keep ignoring you.

by Bertie on Sep 26, 2011 5:04 pm • linkreport

@Bertie--

I never suggested eliminating zoning laws. I have repeatedly rejected that as a possible OR desirable change.

To the extent that people genuinely prefer high density, the political process and the market will deliver high density.

I think you need to support this with some evidence. There are tons of things people want that the political system can't deliver in a timely manner. But even if you're right, is it possible we're seeing the beginnings of that process right now? Who knows. This has point has moved from a policy to a political debate, so I'm well outside my (very limited) comfort zone.

But most people don't want high density.

But a ton of people DO! Prices are higher in cities, which reflects people's desire to live there! That's a critical, central part of what Ryan Avent's book argues!

I'd also like to hear what you have to say about the positive externalities of moving issue discussed above...

by WRD on Sep 26, 2011 6:04 pm • linkreport

I think you need to support this with some evidence. There are tons of things people want that the political system can't deliver in a timely manner. But even if you're right, is it possible we're seeing the beginnings of that process right now?

Well, the 2010 Census shows that suburbs continued to grow faster than central cities over the most recent decade, and that people have continued to migrate out of dense old transit-oriented cities in the northeast and into sprawly new car-oriented cities in the south and west. New York City, for example, lost over a million domestic migrants to other parts of the country between 2000 and 2010. So I'd say, no, we're not seeing any sign of a back-to-the-city movement. There have been isolated pockets of increasing density in the gentrifying areas of some inner cities, but the overall trend continues to favor low-density suburbs and car-based lifestyles.

by Bertie on Sep 26, 2011 7:04 pm • linkreport

Prices are higher in cities, which reflects people's desire to live there!

You can't infer anything about relative demand from price comparisons alone. Housing prices tend to rise with density because land costs are higher (higher density by definition means that more people are competing for each square foot) and construction costs are higher (it costs more to build vertically than horizontally).

I'd also like to hear what you have to say about the positive externalities of moving issue discussed above...

I'm sure that Avent is correct that increasing density CAN produce positive externalities, but beyond a certain point the costs outweigh the benefits, which is why people support limits.

by Bertie on Sep 26, 2011 7:16 pm • linkreport

Bertie, your representation of development is completely backwards.

Yes, multistory buildings are more expensive and financially riskier, even though you overemphasize the degree.

However, no one would build an apartment building at an arbitrary size and then charge a rent based on the cost. No, the developer designs the building based upon anticipated rents or sales.

They will only take on more risk if they know people will be willing to pay rents that cover the cost of construction and financing. Demand first, higher costs second.

After the mortgage is paid off, the rents can go down significantly - but they won't if there is demand demand at those prices.

Similarly, the cost due to competitiveness is the direct result of demand by residents. You can only compare density against price at a regional scale, where transportation, amenities, and location are averaged over a large area. At a local or neighborhood scale, other factors dominate.

by Neil Flanagan on Sep 26, 2011 11:10 pm • linkreport

@Bertie--

I will respond more tomorrow, but here's a first crack;

You can't infer anything about relative demand from price comparisons alone.

This is an old aphorism in economics ("Never reason from a price change"). But it's not on the price comparison alone, it's from knowledge of the supply constraint. It's from looking at different cities (Houston v. NYC) and from different areas within a metro area (Dupont v. Anne Arundel County).

housing prices tend to rise with density because land costs are higher

This contradicts what I excerpted earlier. Why are land costs higher? Because developers value the higher rent streams from that land. Because people want to live there...right?

by WRD on Sep 27, 2011 12:15 am • linkreport

@WRD

Naw, nobody goes there anymore. It's too crowded.

by Alex B. on Sep 27, 2011 9:14 am • linkreport

"I think you need to support this with some evidence. There are tons of things people want that the political system can't deliver in a timely manner. But even if you're right, is it possible we're seeing the beginnings of that process right now? Who knows. This has point has moved from a policy to a political debate, so I'm well outside my (very limited) comfort zone."

while theres much thats dysfunctional in our political system, one thing that seems positive to me that HAS occured here in our region is widespread support for TOD. Not just in DC (where there is a bit of a backlash) or in Arlington (moving right along with no significant backlash) or Alexandria (supportive despite specific concerns wrt Old Town) but also in MoCo, in Fairfax, and even, mirabile dictu, in Loudoun. The county boards pretty much all see the writing on the wall, and want to leverage their transit assets to get a piece of the TOD positive market.

To the point that I worry there is SO much TOD in the pipeline, that we may see prices for it fall when those projects come on stream, leading to possible disillusion with TOD.

by AWalkerInTheCity on Sep 27, 2011 11:40 am • linkreport

I think Ryan Avent is trying to push back on the idea that our political system is (or should be) delivering this type of change. He is pushing for a government that is more responsive to market forces. He is pushing against a government that causes market forces. I agree with him.

To address Bertie's "suburbs growing" argument, it's pretty easy to see why this is occurring. There's very little capacity growth in cities, so people are moving in response to price signals that are driven by zoning policies he doesn't like. I'm sure someone on GGW has written about this in the last few months.

by WRD on Sep 27, 2011 3:09 pm • linkreport

"I think Ryan Avent is trying to push back on the idea that our political system is (or should be) delivering this type of change. He is pushing for a government that is more responsive to market forces. He is pushing against a government that causes market forces. I agree with him"

But part of the argument for that is that govt is inherently rigid and gums up smart growth. IF that were the case, it MIGHT be worth tossing out the many benefits of zoning (which ARE driven by real externalities). But since local govts ARE (in part due to revenue incentives) responsive to the need for TOD, it seems unwise to abandon the benefits of zoning (and no, the Coase theorem type actions will NOT be a suitable substitute) Why throw out the zoning baby with the the antidensity bathwater, when you have a way to toss out the bathwater while keeping the baby?

Unless you have an ideological reason to oppose the baby that has nothing to do with the bathwater.

by AWalkerInTheCity on Sep 27, 2011 4:33 pm • linkreport

However, no one would build an apartment building at an arbitrary size and then charge a rent based on the cost. No, the developer designs the building based upon anticipated rents or sales.

I didn't say anything about design or size. I described the reasons why high-density housing tends to be more expensive to supply, and hence more expensive to buy, than low-density housing. This has nothing to do with the relative level of demand for each type of housing. The fact that high-density housing is more expensive than low-density housing does not mean that there is more demand for high-density housing, or that there's a shortage of high-density housing, or that high-density housing is "undersupplied." It's simply a consequence of the fact that dense housing costs more to supply.

This contradicts what I excerpted earlier. Why are land costs higher? Because developers value the higher rent streams from that land.

No, land costs are higher for high-density housing because, as I said, there are more people competing for each square foot of land. More demand per unit of supply means a higher unit price. But this does not mean there is more demand for high-density housing than for low-density housing. If 100,000 people live on 10 square miles of land in the urban core (10,000 people per square mile), and 1,000,000 people live on 1,000 square miles of land in the suburbs (1,000 people per square mile), then land will be more expensive in the urban core because there are ten times as many people competing for each square foot as there are in the suburbs, even though ten times as many people live in the suburbs as live in the core.

by Bertie on Sep 27, 2011 6:51 pm • linkreport

The fact that high-density housing is more expensive than low-density housing does not mean that there is more demand for high-density housing, or that there's a shortage of high-density housing, or that high-density housing is "undersupplied." It's simply a consequence of the fact that dense housing costs more to supply.

First, I think you mean more dollars competing for each square foot of land, not more people. The number of people competing for each unit doesn't matter. Their relative willingness to pay does.

Second, we come back to why are there lots of dollars chasing the upzoned land? Because lots of people want to live there! (A demand-side factor)

Your theory might be true if the price of housing was equal to its construction costs. It isn't.

The higher density housing sells at a premium over and above its relatively more expensive marginal cost. There is abundant empirical evidence suggesting the higher costs in supply-restricted markets are due to a combination of high demand and restricted supply.

Your theory also fails to explain the relatively low cost of land in some areas in DC, for example. You can build densely in some high-crime areas, but no one does. Some land, even zoned for high density, simply doesn't command high prices. That's because of demand-side factors.

by WRD on Sep 27, 2011 7:39 pm • linkreport

Second, we come back to why are there lots of dollars chasing the upzoned land? Because lots of people want to live there! (A demand-side factor)

I just explained why this isn't true. Higher density means more demand PER UNIT AREA OF LAND. Not more demand in total.

Your theory might be true if the price of housing was equal to its construction costs.

For the reason I explained, land prices tend to rise with density. In addition to higher land costs, dense development also tends to have higher construction costs. Dense housing is more expensive because of both higher land costs AND higher construction costs. In the suburbs, land is cheap because there are fewer people per acre to bid up the price. And construction costs are low because most construction is low rise and can take advantage of inexpensive materials, labor and equipment. In dense urban cores, land is expensive because there is much less land per person. And construction costs are high because high-rise construction requires more expensive materials, labor, and equipment. There are other factors that tend to increase the price of dense housing too. In multi-unit buildings, a substantial fraction of the building has to be used for common areas (lobbies, hallways, stairwells, elevator shafts) needed to access the private dwellings. So buyers have to pay extra to build and operate (light, heat, cool) these common areas in addition to their own private living space. In contrast, essentially every square foot of a single-family home is private space. Also, tall buildings are subject to more stringent building codes and safety regulations, to reduce the risk of a fire in one unit spreading to other units, and to facilitate evacuation in the event of fire or some other emergency. These things also tend to increase costs.

by Bertie on Sep 28, 2011 2:34 am • linkreport

This is like a chicken-egg argument. A supply-side vs. demand-side debate. I think it comes down to the question of why do cities even exist? Bertie seems to be saying that they are just there (inferring here), and as a result of the density and sheer number of people crammed in a small land area (or the zoning constraints, perhaps?), that is the major determining factor in price.

Sure, density constraints and layers of regulation/requirements can influence property characteristics- (lobbies, hallways, stairwells, etc), but I don't think that this is driving the market price. Demand is! Generally, people want to live in cities! Yes, they have to pay more for the infrastructure costs in high density areas, but in making their market decisions they take that into account. Per your example, why are there more people living in the urban core, and hence competing for that land? I think it's because they want to be there (Demand).

High-density housing tends to be more expensive to supply, and hence more expensive to buy, than low-density housing. This has nothing to do with the relative level of demand for each type of housing.

How does it not have anything to do with relative demand for each type of housing? The density differences don't exist in a vacuum. Developers make decisions on whether to build and what to build based first on demand. Agree?

by TimW on Sep 28, 2011 10:30 am • linkreport

Per your example, why are there more people living in the urban core, and hence competing for that land?

There aren't more people living in the urban core. In my example, there are ten times as many people living in the suburbs as living in the urban core. But land prices are still higher in the urban core because there are more people per unit area of land. I don't know why you and WRD are having such a hard time understanding the difference between demand per unit of supply and total demand.

How does it not have anything to do with relative demand for each type of housing?

Because it's dictated by the cost of supply. Since it costs more to supply high-density housing, builders must charge a higher price to buyers to cover their costs.

by Bertie on Sep 28, 2011 8:27 pm • linkreport

If 100,000 people live on 10 square miles of land in the urban core (10,000 people per square mile), and 1,000,000 people live on 1,000 square miles of land in the suburbs (1,000 people per square mile), then land will be more expensive in the urban core because there are ten times as many people competing for each square foot as there are in the suburbs, even though ten times as many people live in the suburbs as live in the core.
You also said: “The fact that high-density housing is more expensive than low-density housing does not mean that there is more demand for high-density housing, or that there's a shortage of high-density housing, or that high-density housing is "undersupplied." It's simply a consequence of the fact that dense housing costs more to supply.

I'm not Paul Krugman so I'm not as good and concise at explaining things. Please forgive me if I misunderstand what you are arguing (it isn’t perfectly clear to me) or this rambles on a bit. If you only read one part of this, read the last paragraph.
I'm going to divide all the numbers in your example by 10 and restate the situation:
(a) 10,000 people on 1 sqm (10,000 to 1)
v.
(b) 100,000 people on 100 sqm (1,000 to 1).

First, the number of people competing for each square foot is irrelevant. What matters is the reservation prices of the bidders. If the 10,000 bidders in (a) value the land at $2, the land will sell for that or less. Having a ton of people is neither a necessary nor a sufficient condition to cause the difference you identify. Remember, “Quantity actually bought and sold is whichever is less: quantity demanded or quantity supplied.” (And supply and demand are simultaneously determined)

In your scenario, housing prices must be higher in Capitol Heights than Fairfax County because Capitol Heights is denser than Fairfax. Fairfax is (b) and Capitol Heights is (a). The analysis fails to consider why there are people competing for the land in the first place. Take another extreme: construction costs are higher in rural Alaska than DC but housing prices are higher in DC. Why is that? Your general rule seems to be that higher density must result in higher prices because more people are chasing a fixed amount of land. This isn’t the case at all precisely because the supply of housing can expand.

You’re talking about the price of land while I’m talking about the price of housing units. While land is fixed, housing units are not fixed. The whole point of this debate is to illustrate what would happen without such a strict zoning code.

If the housing units in (a) are selling at $10 but cost $5 because more people (really, dollars) are chasing those units, developers have an excellent situation. They can add an additional housing unit costing $5 and sell it for $10. Aggregate this effect across the market and you end up with two results. First, it increases density and second it puts downward pressure on housing costs. In a free market situation, this will continue until housing costs are driven down or construction costs are driven up. Do you disagree with this? But the research and data tell us the difference between construction costs can only possibly explain a very small difference between the housing price differentials. See the other paper I linked to plus this one.

I know this probably isn't very helpful. If you break down your position a little more simply, perhaps I can do the same.

by WRD on Sep 29, 2011 12:52 pm • linkreport

First, the number of people competing for each square foot is irrelevant.

No, it is entirely relevant. The issue I am discussing here is the effect of density on land prices. Obviously, many other variables also affect land prices. But the effect of DENSITY is to increase the price. Higher density by definition means that there are more people competing for a given area of land. More demand per unit of supply means a higher unit price. That is why the price of land tends to rise with density. Hence, the price of housing tends to rise with density. This has nothing to do with the total amount of demand for different types of housing. It simply follows from the definition of density and the economic principles governing the relationship between supply, demand and price.

But the research and data tell us the difference between construction costs can only possibly explain a very small difference between the housing price differentials.

This just isn't true, either. And the paper you link to doesn't say what you think it says. Indeed, Ed Glaeser, one of the authors of that paper, cites construction costs of $300-400 per square foot for a hi-rise apartment or condo building in New York vs. only about $80 per square foot for single-family homes in the suburbs of Houston. So construction cost differences alone mean that it costs something like 5 times as much to supply each square foot of housing in NYC as in the Houston suburbs. How is that "a very small difference?" And of course land costs are also much higher in New York than Houston, because New York is so much denser. That increases the cost of supplying housing in NYC even more.

by Bertie on Sep 29, 2011 6:36 pm • linkreport

Let me see if I can step in and summarize the point WRD is trying to make:

1. Land price per acre can be higher, but land price PER DWELLING UNIT will be lower. With large enough densities, this is almost certainly the case. Ed Glaeser's paper makes exactly this point.

2. Density does not cause high land prices. Rather, density is a response to high land prices which are driven by demand to occupy the land. If you block density through zoning, both the demand and the high land price will still remain.

3. Construction costs may be higher (not necessarily though), but WRD's point is that there is a significant gap in NYC between Land+Construction Costs per dwelling unit and the actual market price. Glaeser cites Chicago by comparison where high-rise units (with similar construction costs) cost half as much as NYC. The gap between the two is essentially the cost of regulation. Which in turn suggests that if you could build more units in NYC, there is demand for more people to live there.

Another point: it's simply false that construction costs will always be higher. With concrete high rises, yes. With stick framed low rises, no. Yes you have to pay for additional common areas with a multi-unit, but you also pay for proportionally less roof, foundation, exterior walls, mechanical systems, etc. per unit.

by SRL on Sep 30, 2011 7:31 am • linkreport

1. Land price per acre can be higher, but land price PER DWELLING UNIT will be lower.

Land costs per dwelling unit will be lower only if the dwelling units use less land (to the degree needed to more than offset the higher unit price of land) . That generally means smaller dwellings and/or taller buildings. So you're comparing apples to oranges. A 500 sq ft high-rise condo isn't equivalent to a 2,000 sq ft detached house with a yard and garage simply because they're both "dwelling units."

2. Density does not cause high land prices. Rather, density is a response to high land prices which are driven by demand to occupy the land.

Higher density is both an effect and a cause of higher land prices. Older urban deveopment tends to be denser because transportation used to be much slower, so people couldn't travel as far. Newer urban development, especially development in places where most people are wealthy enough to afford a car, tends to be less dense because transportation is faster. So people can travel longer distances in a given amount of time and there is little incentive to conserve land by building densely.

3. Construction costs may be higher (not necessarily though), but WRD's point is that there is a significant gap in NYC between Land+Construction Costs per dwelling unit and the actual market price. Glaeser cites Chicago by comparison where high-rise units (with similar construction costs) cost half as much as NYC. The gap between the two is essentially the cost of regulation.

There are many factors that influence market prices. You certainly can't attribute the difference between the cost of supply and the market price to whatever particular set of regulations it is that you wish to eliminate. I seriously doubt you oppose all regulations that have the effect of making housing more expensive, such as building codes and laws that set aside land for public uses like parks.

Another point: it's simply false that construction costs will always be higher. With concrete high rises, yes. With stick framed low rises, no. Yes you have to pay for additional common areas with a multi-unit, but you also pay for proportionally less roof, foundation, exterior walls, mechanical systems, etc. per unit.

I don't claim that construction costs "always" increase with density. They GENERALLY increase with density, for the reasons I explained: more expensive materials, labor and equipment; stricter building codes; and the need to devote a substantial fraction of multi-unit buildings to common areas. These costs greatly exceed the costs of additional roofing or exterior walls in low-density buildings. A typical one- or two-story suburban home is constructed with a wood or block frame, with drywall on the inside and stucco or siding on the outside. Roofs are laid with concrete tiles or shingles. These materials are all very cheap. Little skilled labor is required. No special equipment is required. Plumbing and electrical layouts are simple. A single design can be mass produced. This is why a detached house in the Houston suburbs costs only about $80 per square foot to build, while a New York high-rise apartment costs $300-400 per square foot to build.

The point of all this is that the mere fact that high-density housing is more expensive -- often, much more expensive -- than low-density housing DOES NOT MEAN that there's more demand for high-density housing, or that there's a "shortage" of high-density housing, or that high-density housing is "undersupplied."

by Bertie on Sep 30, 2011 10:58 pm • linkreport

You certainly can't attribute the difference between the cost of supply and the market price to whatever particular set of regulations it is that you wish to eliminate. I seriously doubt you oppose all regulations that have the effect of making housing more expensive, such as building codes and laws that set aside land for public uses like parks.

The Glaeser/Gyourko paper actually did just that in more depth. Others have as well. It's not just look at prices and conclude, there's more rigor to it. They're trying to quantify to what extent regulations cause the higher prices. They also addressed the building code issue. If it has an effect, it is very small and not statistically significant.

Density again:

Land is priced based on the cash flows that come from it. Land in Iowa, for example, commands a high price precisely because it isn't dense. It commands a high price because there's cash flows that come from growing corn on it. Land in New York is priced highly because there's cash flows that come from renters.

Now add in government. They cap the number of renters on the land in New York. What happens to the price of land?

With no demand growth, the price goes down! Owners lose option value and as a ceiling on rent is imposed.

With robust demand growth, renters compete back and forth for rent slots. Real estate players compete back and forth for ownership rights to the rent streams.

The point of all this is that the mere fact that high-density housing is more expensive -- often, much more expensive -- than low-density housing DOES NOT MEAN that there's more demand for high-density housing, or that there's a "shortage" of high-density housing, or that high-density housing is "undersupplied."

Is housing "undersupplied"? Yes, compared to what would happen without the government regulation. Without the (non-market, government caused) cap on housing units, there would be more housing units. So relative to this baseline, housing is undersupplied. People would be willing build and pay for more units but are prevented by government regulations.

However if you take the supply limit as a given, housing isn't undersupplied. The price of housing adjusts upward freely and this price adjustment "destroys demand."

by WRD on Oct 1, 2011 11:03 pm • linkreport

The Glaeser/Gyourko paper actually did just that in more depth.

No it doesn't. Glaeser certainly does not claim that market price - cost of supply = cost of regulation. Nor does he provide an estimate of the cost of whatever set of regulations it is that you oppose.

They also addressed the building code issue. If it has an effect, it is very small and not statistically significant.

No, they don't claim that either. Building codes have a huge effect on the price of housing. If developers were free to build housing as cheaply as possible, without having to comply with the various codes that set minimum standards for structural integrity, electrical and plumbing work, fire suppression, emergency evacuation, accessibility for the disabled, and so on, the cost of construction would be much lower. Dense housing costs more to construct precisely because it is so much more expensive to meet these minimum standards for tall and multi-unit buildings than it is for single-family homes. As I said, Glaeser himself cites a difference in construction costs of a factor of four or more ($300-400 vs. $80) for high-rise housing in New York City vs. single-family housing in the suburbs. How you can seriously claim that $300 vs. $80 is a "a very small difference" I have no idea.

Land is priced based on the cash flows that come from it.

As I said, there are many factors that influence the price of land. One of those factors is the amount of demand per square foot of supply. Since higher density by definition means more people competing for each square foot, higher density generally means higher land prices. And higher land prices generally means higher housing prices.

by Bertie on Oct 2, 2011 12:49 pm • linkreport

Density does not cause high land prices. Rather, density is a response to high land prices which are driven by demand to occupy the land.

Oh, thank God in Heaven! I thought I was going to have to type this again for the 1,382,328th time!

:)

by oboe on Jan 8, 2012 1:50 pm • linkreport

Add a Comment

Name: (will be displayed on the comments page)

Email: (must be your real address, but will be kept private)

URL: (optional, will be displayed)

Your comment:

By submitting a comment, you agree to abide by our comment policy.
Notify me of followup comments via email. (You can also subscribe without commenting.)
Save my name and email address on this computer so I don't have to enter it next time, and so I don't have to answer the anti-spam map challenge question in the future.

or

Support Us

How can our region be greater?

DC Maryland Virginia Arlington Alexandria Montgomery Prince George's Fairfax Charles Prince William Loudoun Howard Anne Arundel Frederick Tysons Corner Baltimore Falls Church Fairfax City
CC BY-NC