In his recent book Shut Out, Kevin Erdmann, a finance expert and visiting fellow at the Mercatus Center at George Mason University, has two main messages. The first, which is not controversial among economists, is that restrictions on residential construction in coastal California and the urban Northeast have constrained supply so much that housing in those areas is virtually unaffordable for people in the lower- and middle-income classes. His other message is more controversial, that the financial crisis last decade was not due to a housing bubble but, rather, to bad policy decisions based on the idea that there had been a bubble. Whereas I was already convinced of his first point, I, like the majority of economists, was skeptical of his second. But because of all the data and reasoning he brings to the issue, I now find myself at least 90% convinced.
Probably because his second point is the more controversial, Erdmann spends about the first half of the book making that case. At times his narrative gets bogged down and his language is often sloppy. For example, he uses the word “shortage” to refer to a situation where demand increases but supply doesn’t. Economists, however, tend to reserve that word for situations where the price fails to clear the market such that quantity demanded exceeds the quantity supplied. The good news is that he often saves the day with pithy, clever quotes that sum up his message. Also, the more than 100 graphs he uses in the book seem like overkill, but that is better than underkill.
Types of cities / Erdmann makes his case by looking at the diverse characteristics of U.S. cities rather than lumping them all together, and by studying changes in housing prices and rents over time. He focuses on the 20 largest U.S. metropolitan areas and divides them into four categories: Closed Access cities, Contagion cities, Open Access cities, and Uncategorized cities. The five Closed Access cities are New York City, Los Angeles, Boston, San Francisco (including San Jose), and San Diego. In those cities, local and state governments have imposed strong restrictions on construction.
Erdmann seems a little vague about when those restrictions got really tight. His narrative suggests that it was in the 1990s, but there’s no index to help one look for a clear answer; he did confirm in an email to me that he dates it to 1995. In those cities, housing starts, even in economic expansions, have been low, incomes have been high, rents have been high (and rising) even relative to incomes, and there were large rates of out-migration of households with low incomes.
The above are the opening 4 paragraphs of David R. Henderson, “Was There a Housing Bubble Last Decade?” Regulation, Winter 2019/2020, pp. 63-65. Read the whole thing. [Scroll down about 60 percent of the way.]
Thanks to Jeff Hummel for improving a previous draft and to Kevin Erdmann for promptly answering the questions I emailed him.
READER COMMENTS
Phil H
Dec 12 2019 at 8:03pm
“which is not controversial among economists, is that restrictions on residential construction in coastal California and the urban Northeast have constrained supply so much that housing in those areas is virtually unaffordable for people in the lower- and middle-income classes”
It should be controversial, because it’s a deeply parochial and politicized viewpoint. How do these economists account for the fact that housing in Chinese cities, where they throw up high-rises with gay abandon, is also utterly unaffordable for anyone on a normal income? Alternatively, compare rents in high-rise Manhattan with low-rise parts of the same city. Does density make things cheaper? This seems to be a laughable oversimplification.
Jon Murphy
Dec 12 2019 at 8:08pm
The effects of price controls on quantity supplied and demanded are neither parochial nor politicized. Indeed, as your own examples point out, they are confirmed in reality: price controls lead to shortages.
Don Boudreaux
Dec 13 2019 at 2:11pm
Phil H:
The position that David correctly says is uncontroversial among economists might be mistaken or not, but it is certainly not as you describe it – namely, as “a deeply parochial and politicized viewpoint.” All David – and Erdmann, and any other respectable economist you care to name – is saying is that, ceteris paribus, the less is the supply of something, the less will be that something’s availability and the greater will be that something’s market value.
Do you doubt that a massive drought causes the availability of food to fall and food prices to rise? Do you doubt that fires in gasoline refineries that cause the availability of gasoline to fall and cause the price of gasoline to rise? Do you doubt – do you think it to be a claim “deeply politicized and parochial” – that when wars disrupt the exploration and drilling for oil in the middle east that one result is a fall in the global supply of oil and higher oil and fuel prices?
This general conclusion about the effect of changes in the supply of X on the availability (and, hence, on the affordability) of X doesn’t change simply because a reduction in supply is engineered by the state for reasons ostensibly, or even actually, lovely.
Does anyone doubt that land-use restrictions cause land-use to be restricted? What is so “parochial and politicized” about recognizing that restricting the use of land decreases the supply of those land-uses made more difficult by said restrictions? Have you an alternative prediction about what happens to the availability and market value of X when the supply of X decreases?
As for your other points, as Scott Sumner frequently – and wisely – notes: never reason from a price change. The fact that the supply of urban housing in China is rising while the price of that housing keeps it out-of-reach for most Chinese people does nothing to discredit David’s point. Ditto for your observations about housing in NYC: that city, recall, has rent control throughout it – a policy that reduces the supply of rental housing throughout the entire city, including those parts where density is relatively low. (Actually, the way that NYC’s rent control is set up, it has a far larger negative impact on the supply of rental housing for lower- and middle-income people than for upper-income people. But this part of the story – which also further nullifies your NYC-criticism of David’s point – is a tale for telling at another time.)
Jon Murphy
Dec 12 2019 at 8:12pm
Here’s the thing: Henderson’s point is about the effects of price controls. You’re discussing shifts in supply and demand. Different analysis.
David Henderson
Dec 12 2019 at 10:55pm
Actually, Jon, I’m not discussing price controls. I’m discussing supply restrictions. What I have in mind is the work by Glaeser and Gyourko.
Jon Murphy
Dec 13 2019 at 9:37am
Yeah I realized that this morning when I re-read. I was just jumping on to correct my error when I read your comment.
Can’t even blame lack of sleep for this mistake!
P Burgos
Dec 12 2019 at 10:05pm
China has a weird housing market in many of its major cities. People pay a ton of money to buy condos when they could rent an equivalent apartment for much lower monthly payments. For example, the condo I rent is valued at around half a million dollars, but it costs only $1k a month to rent.
David Henderson
Dec 12 2019 at 10:56pm
I thought it was clear that both Erdmann and I are discussing the housing market in coastal California and in the northeastern United States, not in China.
Matthias Görgens
Dec 13 2019 at 6:36am
Density is a way to offer more supply where there’s more demand.
You would have to look at the impact of additional density in the same location (or a reduction in density) to make the kind of claims you make.
Otherwise, by just comparing place with different density, we don’t know whether demand is held constant.
Phil H
Dec 13 2019 at 11:37am
Hi, Matthias. I think you’re agreeing with me? My point is precisely that changes in supply/density can affect demand as well, so there is simply no guarantee that increasing the supply/density will lead to lower prices.
Jon Murphy
Dec 13 2019 at 12:05pm
Woah, be careful here. This is incorrect. Changes in supply cannot affect demand. It affects quantity demanded, not demand itself. Matthis’ point is that you are reasoning from a price change.
The following is a question I ask in my Econ 201 and 202 classes (into to macro and micro):
The proper answer is Joe is incorrect. Joe is reasoning from a price change. He is noticing a shift in the demand curve (an increase in demand). You are making the same mistake as Joe. The NYC and California examples are examples of restricted supply (supply curve not shifting because of regulations). The China example you posit is of demand curves shifting (increase in demand). Either way, you get an increase in price. But simply from that increase in price, one cannot reason (as you do) that one explanation is incorrect and the other is correct; an increase in price can be caused by a decrease in supply or an increase in demand.
Phil H
Dec 14 2019 at 9:54pm
“one cannot reason (as you do) that one explanation is incorrect and the other is correct”
You have the shape of the argument backwards.
Henderson/Erdmann claim: “restrictions…constrained supply so much that housing in those areas is virtually unaffordable”
That is the Henderson/Erdmann claim that the high price is the result of restrictions. Their claim is that the price change is caused by changes in supply – a clear example of “reasoning from a price change”.
My point is precisely what you said: price changes can be caused by changes in supply and/or by changes in demand.
Therefore their claim does not stand up to scrutiny. My suggestion is that if these restrictions were loosened, it is very possible that housing prices in the two markets would in fact increase further (because of more demand).
Phil H
Dec 14 2019 at 10:01pm
I just spotted Erdmann’s response below, where he makes the same point as me:
“It is possible that more density would increase the value of urban housing…then the price would rise”
He then has a more sophisticated argument about rent, which I haven’t yet understood! So I’m happy to go away and do more homework. But the supply and demand/”arguing from a price change” argument makes clear: simply stating that supply restrictions caused a price change is, as Jon likes to make the comparison, sub-101 economics.
P Burgos
Dec 13 2019 at 8:54am
Does Erdmann show that it is regulations causing the lack of housing starts, as opposed to a lack of developable greenfield land? My understanding is that even without taking regulations into consideration, some land is easier (more profitable) to build on. Then of course throw in that greenfield development is less costly than infill or tear down development even without taking into account regulations. And then even a modicum of regulations on infill and tear down developments might serve to make those developments much more expensive.
That is to say, I suspect that there is way more at play in making cities unaffordable than just regulation, and that a lack of developable land due to geographic factors is especially relevant to Californian cities that are located in a fairly narrow strip of land between the coast and mountains. Also, my impression when looking at Zillow was that there are a lot of housing units for sale in Metro NYC priced around $300k, which isn’t an astronomical price. My impression was that metro Boston was more expensive, but nothing like California. And if it were just regulations in California, why does Sacramento have housing prices that aren’t too far off from someplace like Durham, NC?
David Henderson
Dec 13 2019 at 11:43am
You write:
It has been over 6 weeks since I’ve read it, but I’m pretty sure he doesn’t show that. He doesn’t need to, though, because Glaeser and Gyourko have shown it with a simple, but very clever, methodology. If you can’t find their piece in Regulation magazine, tell me and I’ll find the link. It’s a short version of their NBER study, but having read both the Regulation piece and the NBER study thoroughly about 6 years ago, I was impressed with how the Regulation piece got even the nuances right.
You write:
Sacramento is not in coastal California. The restrictions on building are much tougher in San Francisco and on the Peninsula than in Sacramento.
Steve
Dec 13 2019 at 5:30pm
Count me in as one who’s not buying that the answer to high housing costs is to create more density. In fact, empirical evidence seems to suggest the opposite. In every country I can think of, the most dense cities are the ones that are least affordable. To suggest that more density is the secret to affordable housing just doesn’t work for me.
I totally understand that a standard supply/demand curve analysis that suggests that additional supply would reduce the price (and it’s possibly true in the short term), but that assumes that you’re just traveling down the demand curve, not actually moving it. Additional density brings jobs, restaurants, things to do, people to meet etc. and that increases the desire of many people to live there.
Of course, you could just increase density to such a magnitude that the living conditions become intolerable and everyone wants out. That may work.
Kevin Erdmann
Dec 13 2019 at 6:25pm
It’s a water vs. diamonds issue.
No value + high scarcity = no economic activity
high value + no scarcity = consumer surplus
high value +high scarcity = high prices
It is possible that more density would increase the value of urban housing, but it does not follow that it would increase the price. In a context where price simply reflected unencumbered costs, then the price would rise with marginal increases in the cost of building at higher densities. But, in those 5 cities, the price currently reflects artificial scarcity that is enforced through political obstruction. As you point out, however, even if high prices were the result of the natural cost of building and the added value of density, then adding density would still be adding value.
More importantly than all of that, though, is the realization that high prices in those key cities are and were highly correlated with rising rents, which means that the conventional narratives that favored economic penance for an unsustainable price bubble have been horribly misguided.
Vivian Darkbloom
Dec 16 2019 at 3:40am
I understand that housing restrictions can put upward pressure on prices, but what is missing from your narrative (I have not read the book) is why prices suddenly declined. For example, it appears that the peak in housing prices in SF was March 2006 (before the official start date of the recession). What caused housing prices in SF to decline? Surely SF did not loosen regulations so that there was a sudden increase in supply. Nor do I see any significant change in mortgage interest rates. What is your explanation?
Kevin Erdmann
Dec 16 2019 at 2:38pm
That’s a great question, and it is something I will address more directly in the follow up book that should come out some time next year. The key is that, generally, home prices reflected fundamentals, so while everyone was waiting for prices to “correct”, the housing market was deeply contracting in every way but falling prices for nearly two years before prices finally relented. We were deep into a housing bust by mid 2007, but most would not be satisfied until prices declined.
By 2006 and 2007, both RGDP and NGDP growth were marginally in recessionary territory, but it didn’t show up in unemployment because the Contagion cities were the places where NGDP growth trends were most severe, and sharp downshifts in employment growth there were associated with declining in-migration rather than unemployment. Furthermore, those local NGDP shocks were lessened by borrowing from the significant home equity that Closed Access and Contagion homes had accumulated.
It is not controversial to suggest that home equity borrowing was boosting consumption at the time. But, in the “bubble” mindset, that just looked like more recklessness that had to be snuffed out. In hindsight, pre-2006 borrowing had been associated with rising residential investment and the spike in migration out of the Closed Access cities. But that wasn’t a credit bubble, it was a bubble in an inferior good – homes in places that were substituting for the Closed Access cities. That was already a bubble in compromise. By the end of 2005, though, things were going downhill. The lending in 2006-2008 happened with collapsing residential investment, first time homebuying, etc. Conventional wisdom said that the Fed had to strangle the money supply to make up for that reckless borrowing. In fact, the borrowing was making up for extremely tight monetary policy, which would have led to recession by 2006 without it.
In general, lending has been given far to much causal importance. It was a lagging factor. Mortgage growth lagged local spikes in home prices, and mortgages outstanding were still increasing into 2008, even after an extended period of declining housing activity. Sentiment, money supply, migration shocks, etc. are more important elements in the story. Where debt is important is (1) delaying the realization of tight monetary policy in 2006 and 2007, and (2) the persistent and relentless suffocation of working class housing markets from 2009 onward that created a second, worse housing bust unrelated to the pre-2006 building activity. The scale of that late bust and the universal lack of interest in stopping it has been a public policy tragedy of historic proportions.
Thaomas
Dec 14 2019 at 7:20am
What kicked off the 2008 financial crisis is pretty irrelevant. A Fed actually committed (and therefore know to be committed) to carrying out its mandate to keep prices stable (rising steadily at some target rate = PL targeting) and preventing unemployment of resources (as proxied by the unemployment of labor rate) would have avoided a the 2008-2016 recession.
As for the residential NIMBY problem, this is largely a result of failure to correctly price urban street and road use for congestion externalities.
David Seltzer
Dec 14 2019 at 5:40pm
I lived in NYC from 1980 till 1990. The pernicious effects of rent control and rent stabilization in the NYC engendered housing “bubbles” across the Hudson to Hoboken and Jersey City. Under NYC’s pricing regimes rents were perennially below market equilibrium rates. New construction slowed and property owners often deferred maintenance. Real estate developers partnered with investors and private equity firms. They began purchasing six and eight-unit apartment buildings in Hudson County and converted them to condominiums. They bought out tenants and rehabbed the units. Local banks were lending at rates averaging 12%. We were glad to pay it. By 1988, 5000 converted units came online. In 1988, Our group purchased a six-family apartment building at 114 Monroe Street in Hoboken for $140,000. We converted six units to condos and sold each for an average of $70,000. As the NYC housing market became more expensive, people moved to Hoboken, now referred to as the sixth borough. Now the “bubble” part. The City of Hoboken and Hudson County saddled condo owners, builders and rental property owners with regulations not unlike those in NYC. The economic impact, a 506 square foot condo at 114 Monroe Street is currently priced at $385,000, an increase of 10.9%, compounded annually. The national average for similar units is between two and three percent.
David Seltzer
Dec 14 2019 at 5:54pm
Mea Culpa. I miscalculated the compounded rate of increase. The rate is 5.653%. Still about twice the national average for similar units over the same time horizon.
Comments are closed.