Urban Returns

Affordable housing densities

Austin's freshly minted Comprehensive Housing Market Survey has lots of interesting tidbits.  One is this map depicting Austin's concentrations of housing affordable to households earning between 51% and 80% of median family income.  (The darker the color, the greater the density of affordable housing.)

I suppose this is no real surprise, but the degree of segregation still is striking.  Most cities have pockets of affordable housing concentrated near the city center.  Inner city neighborhoods have high concentrations of old housing, and old housing tends to be cheap housing.  Mass transit is usually better in central cities, too, and the poor need good mass transit.  And, finally, central neighborhoods are usually quite dense, which means lower land costs per unit.

Central Austin, though, has very little affordable housing.  Even central east Austin, historically a repository of cheap housing, now has only a moderate density of affordable housing.  Affordable housing in Austin instead has been relegated to large, deep pockets in the older suburbs to the north and south; these, incidentally, contain some of the densest census tracts in the city. 

A couple of those who commented on my post on the city's affordable housing investment in the Village on Little Texas pointed out that there is already a lot of affordable housing in the area.  This map confirms the point.  I've marked the site of the Village on Little Texas with a light green arrow.  It is in, or at least on the fringe of, one of the largest pockets of affordable housing in Austin.

It's fair to ask whether the city should invest millions to subsidize a relative handful of affordable units in an area that already has market-rate affordable units.  Particularly such steep subsidies.  The city needs to step back and sort out its priorities.  

My piece at Newgeography

Here.

My thesis:  Central Austin has a strikingly low percentage of families compared to the suburbs.  None of the usual explanations applies -- i.e., central Austin families (except for those in a few neighborhoods) have not abandoned central Austin because of crime, overcrowding, post-industrial blight, bad schools, or the flight of jobs to the suburbs.  The only explanation, I think, is that Austin's suburbs have better, bigger and cheaper housing.

I did not offer a prescription.  Anyone who reads this blog can guess, though.  We need more multi-family in central Austin to relieve the price pressure on single-family homes and to give households more options.  (Multi-family has many other benefits besides these, of course.)  And we need to make it easier to upgrade central Austin's generally older, smaller and inferior housing stock.   The McMansion ordinance did not cause this divide, but it certainly won't help.  (For those who don't know, the McMansion ordinance bans even relatively modest homes on some lots, homes no one would call a McMansion.)

I often experience buyer's remorse after writing a piece like this because I'm afraid my single friends will take it the wrong way.  I don't think families with children are more important or "better" than other households.  I merely think Austin's core must have a good mix of households to remain healthy.  (The suburbs would be healthier, too, if they offered singles or couples something other than inexpensive  housing in compartmentalized pods.)  Balance matters, if for no reason other than allowing us to continue using the expensive assets (like schools) already in place.  I don't think we have a good balance now.

For those interested, here is my previous piece at Newgeography.

Rethinking Federal Housing Policy

Ed Glaeser and Joseph Gyourko's e-book (pdf) on federal housing policy really is worth a read.  

Glaeser and Gyourko argue that federal housing policy wrongly treats every jurisdiction alike.  The country really has three different housing markets.

First, there are markets with high demand but elastic supply -- e.g., Houston and Atlanta.  Housing in these markets closely tracks the cost of construction.  Because developers cannot supply housing more cheaply than this over the long run, it makes no sense for the federal government to intervene directly in these markets -- there simply is no market failure.  (For the poor, Glaeser and Gyourko suggest outright transfers so they can buy more housing, or housing vouchers as a second-best solution). 

The second kind of market is the low-demand market -- Detroit, for instance --where housing can be had for less than the cost of construction.  It is silly to subsidize the construction of new housing in these places.

The third kind of market is the high-demand, inelastically-supplied market -- the West Coast, the Amtrak corridor in the northeast, and a few other markets in the country's interior.  These markets exhibit government-created market failures.  Tight zoning regulations in these markets combined with high demand have raised the cost of housing well above the cost of construction, sometimes by hundreds of thousands of dollars.  New construction permits amount to a tiny fraction of the housing stock.

The market failures in these jurisdictions endure because local governments have no incentive to increase the supply of new housing:  the benefits from new housing go mostly to non-residents, while the costs from polution and congestion -- not to mention the effect of new competition on housing prices -- fall on existing residents.  

Federal housing policy -- particularly the mortgage interest deduction -- has different effects depending on the market.  In elastically-supplied markets like Houston or low-demand markets like Detroit, the mortgage interest deduction increases consumer demand and stimulates the construction of more housing.

In high-demand, inelastically-supplied markets, though, the benefits of the mortgage interest deduction flow to existing homeowners.  The mortgage interest deduction paradoxically makes housing in these markets less affordable.   

Glaeser and Gyourko argue that federal housing policy ought to target the broken markets.   The goal, they suggest, should be to incentivize the high-demand, inelastically-supplied markets to allow more housing.

They propose capping the home mortgage interest deduction at $300,000 for households in these markets.  (They would phase out the deduction for current existing mortgages for equitable reasons.)  This cap would generate additional tax revenue.  The federal government would then refund that revenue to the local governments (via the states) provided the local governments made sufficient progress toward increasing their housing supply.  Glaeser and Gyourko essentially propose bribing local jurisdictions with their homeowners' tax dollars. 

Needless to say, I think this is a great idea.  Rich housing markets are becoming increasingly stratified by income.  The lower and middle classes are being run out of these markets by high home prices.  Their allegedly progressive citizenry, embarrassed by the increasing inequality, flail about for solutions.  But they won't allow the one thing that would improve affordability over the long-term.   

Incentivizing new development in these cities would benefit the rest of us, too.  The high-demand cities have high demand because they make their citizens more productive.  Restricting access to them creates a net welfare loss for the country as a whole.  

But Glaeser and Gyourko's proposal will be dead on arrival.   The costs of their proposals would fall on a small, clearly identifiable group -- wealthy homeowners in high-priced markets.  These homeowners would suffer a double-whammy, through the loss of their tax deductions and the loss of home value due to increased supply.  Everyone else would benefit, of course, but those benefits would be diffuse.  Local governments would benefit from the extra revenue, but local governments don't vote for senators or congressmen.  If this proposal were enacted, Barbara Boxer would hear the howls of Marin County homeowners from her seat on the Senate floor.  

I'm not making this filtering stuff up.

Here's how two economists describe the filtering process:    

Filtering.

If you don't understand filtering, you don't understand housing markets.  Because lots of people plainly do not understand filtering, here's my attempt at an explanation:

When property values rise, low-quality housing "filters up" to the high-quality housing sub-market.  The reason is that rising rents encourage landlords to invest more in the property.  When property values fall, high-quality housing "filters down" to the low-quality housing sub-market.  The reason is that falling rents encourage landlords to invest less in property.  The key in either case is that old housing costs more to maintain than new housing.

Here's the longer version:

Housing naturally deteriorates as it ages.  Stuff breaks.  Stuff -- expensive stuff like the roof -- has to be replaced.  Minor wear and tear accumulates over the years until the place is downright seedy.

Seedy, that is, unless the owner regularly invests in maintenance.  Every homeowner knows that.  It would be nice if the mortgage payment and property tax covered the cost, but they don't.  Those occasional, "extraordinary" expenses turn out not to be extraordinary at all.  And they are far more frequent for older home.  Thus, as a house or apartment ages, it takes ever more money to keep it in good shape. 

Every landlord with an old property confronts the same decision:  Should I spend the money to keep the property in top shape, or should I let it go?

Keeping the property in top shape allows the landlord to charge more rent.  But it also costs the landlord more in maintenance and periodic renovation, especially for old apartments.  What the landlord decides to do depends on the current market, his estimate of future demand, and about a million other things.  But we know one thing for sure:  A landlord is less likely to maintain an apartment in tip-top shape when the rent for tip-top apartments is falling.  Declining rents mean a declining return on the maintenance investment.  That spells less investment .

Without constant maintenance, apartments deteriorate from "tip top" to "slightly dilapidated."  Renters, sensible people that they are, are not willing to pay as much for bad quality.  Rents fall, and the apartments "filter down" into the pool of affordable housing.  

If you want more affordable apartments, build more tip-top apartments.  Increasing the supply of high-quality apartments lowers the rent for high-quality apartments, all else being equal.  Falling rents for the good units encourage landlords to let the older ones slide into the affordable sub-market.

On the other hand, if you want to raise the rents for older properties, then discourage the construction of new apartments.  Restricting supply will raise the rents for high-quality housing, including older, well-maintained properties.  Higher rents will encourage landlords to spend more on maintenance and renovation to move their properties from the low-quality sub-market pool to the higher-quality sub-market.

This is pretty basic supply and demand.  But some people just don't believe it.  I think there are three reasons.

First, apartment renovation usually coincides with new housing construction, and people mistake this association for cause and effect.  Condos and apartments get built during periods of rising rents.  Everyone understands that.  But the same rising rents give landlords an incentive to invest in renovations and upgrades.  We thus should expect new construction and apartment renovations to go hand in hand.  These are symptoms of the same market forces, not cause and effect.

Second, people mistakenly use what is happening to their apartment to extrapolate to the entire market.  The filtering process does not guarantee that every older property will become more affordable.  On the contrary, as rents rise, some old properties will be torn down and replaced with new, more expensive units.  The old units never get a chance to filter down.  But this is simply an example of the "filtering up" that occurs when rents rise.

The remedy for "filtering up" is more housing, not less.  If we halt new construction in order to preserve the existing stock of low-quality units, we guarantee that more of the low-quality units will eventually filter up.  We cannot predict exactly which units will filter up and which will remain affordable.  We just know that there will be more of the former and fewer of the latter.

Finally, neighborhood gentrification can blur things.  Gentrification is a complex process.  It typically occurs because other neighborhoods have become more expensive, pushing buyers and renters to what otherwise would be a less desirable neighborhood.  The increased demand raises rents, which spurs more renovation.  In other words, rising rents in one neighborhood can cause units in another to filter up.

Gentrification can also be spurred by falling crime rates, improved amenities (such as new parks or a decommissioned power plant), and by new investment in the area.  Sometimes it's just chance.  It is rare, though, that adding new housing causes a rise in rents.  In fact, new housing would not be built unless rents were already rising.  New housing is thus generally a symptom of gentrification rather than the cause.

Related posts:

1. Are luxury condos downtown good for Austin's housing market?  (Sept. 11, 2006)
2. Those high-priced condos will be great for families with children. (May 28, 2007)
3. Freeing up space for others. (June 7, 2008).

Housing + Transportation Affordability Index

Check out this totally cool site that maps housing plus transportation costs (plus a whole bunch of other statistics, like average block size) by census block group.  They use census data to estimate the average home and transporation costs as a percentage of median income. The map's creators' point is that those suburban homes are a whole lot more expensive when you factor in transportation  costs. 

I'm fuzzy on their methodology for estimating transportation costs.  I spot-checked the block groups around my neighborhood, and found big differences in transportation costs that I'm skeptical exist.  And bear in mind that housing + transportation costs may take a huge bite out of, say, a Pflugerville household's income, but they would probably be a mere nibble compared to the bite of a central Austin mortgage payment.

Still, this is the coolest thing I've seen in a while.  (I admit that I probably have a warped sense of "cool.")

H/t Smart Growth around America.

Condo conversions

One argument frequently trotted out against zoning for greater density is that it encourages the redevelopment of low-income rental.  But it is silly to believe that zoning can maintain the stock of affordable housing.  When rents lag behind home prices, owners will simply convert the apartments into condos. 

I'm aware of three condo conversions on South Lamar, one on Thornton (around the corner from my house), one off Oltorf, and one on South Congress.  There apparently are many more:

Last year, 824 apartments were converted to condos in Austin, from the downtown area to the suburbs. This year, 1,167 more units are expected to be converted in the Austin area, said Robin Davis, manager of Austin Investor Interests LLC, which tracks the apartment market.

That's a lot of low- to moderate-rental taken out of the market.  On a positive note, the converted condos are a lot cheaper than the new stuff being built.

I suppose Austin could take the San Diego/San Francisco/Berkeley approach of limiting condo conversions.  A saner approach would be to permit small infill projects that blend in with the neighborhood.  And lots of dense multi-family projects on transit corridors.  The stuff built today will be the affordable housing 20 or 30 years down the road.   

Interior vs. exterior space

We live in a small neighborhood of "urban homes" -- 1600 to 2400 sq. ft. homes on small (3500 to 4500 sq. ft.) lots.  We live close to downtown and have a large home by central Austin standards.  But to get this at Austin's prices, we had to give up a real yard.  That was a good trade for us.  We knew we wouldn't use a yard very much, and it would be a headache to keep up.  We care more about the interior space.   

Not everyone shares our preference.  That's fine with me.  But my neighbors obviously do.  And I've always thought there were probably lots of others who would make the same trade if they could.

Here's empirical evidence that our preference for interior over exterior space may not be that weird.  (h/t Matthew Kahn.)  These UCLA anthropologists followed 24 Los Angeles families around for four days to study how they use their house space.    Most of the them ignored their yards:   

Although the back yard is a purported center of family leisure, enjoyment, and privacy, the tracking data from Families 1 to 24 reveal limited uses of back-of-home spaces by family members, despite the fact that every sample included many weekend daylight hours and some afternoon and evening daylight hours, and the weather was generally mild and pleasant enough to be outside on most days. The most salient trend in the data is that 13 of the 24 families did not spend any leisure time (neither kids nor parents) in their back yards during the four days per family available for review. . . . In quite a few of these cases, no family member so much as stepped into the back yard. Sporadic activities in other cases were confined to non-leisure chores such as taking out trash or briefly feeding dogs or washing off chairs.

The yards seemed to function mainly as status symbols.  On the other hand, most of the families had converted their garages into storage rooms or extra living space.

The fact that most households in this sample––and millions visible throughout the U.S.—have converted their garages to spaces not focused on car storage signals a changing need of middle-class families. Families living in average-sized homes (1500–2000 sq ft), as most of these are in our sample, simply do not have enough living and storage space for all of their possessions, and they value garages more for these purposes than for housing cars.

It sounds like many of these families would be willing to trade exterior space for interior space, too.  Like I said, if you want a house with a big yard, that's fine with me.  I think Austin could use more neighborhoods like mine, though; there's probably plenty of demand, even among families with children.

Irreversible unaffordability

We are stuck with pricey single-family homes in central Austin.

Rents are volatile; they can go up or they can go down.

Condo prices tend to be volatile, too, for reasons I don't understand.

Single-family home prices are sticky, though.  Even when a housing market collapses, single-family home prices don't tend to fall, at least not in proportion to the collapse.  Sales activity just dries up.  Homeowners hold onto homes they otherwise would have sold.  Buyers wait for the trough.

Aside from an out-and-out collapse, it is rare to see single-family home prices trend downward.  If you do, you know that calls for new city leadership are just around the corner. 

This is worth remembering when you scan the listings for, say, South Austin, and see a 1,800-square-foot home priced at $400,000 or more.  No matter how much the supply of single-family housing grows, that $400,000 home will not get any cheaper, and I don't pretend otherwise.

Still, there is a very good reason for letting the supply half-way keep up with demand:  a growing supply can help hold price increases in check.  We don't have to let things get worse.

Take South Austin.  Homes fetched an average of $210-$215/sq. ft. in South Austin (MLS 6 & 7) in 2006.  Prices per square foot rose by more than 12% over the year before.  That may be what we can expect for the next few years, unless the economy falters or we allow more (= denser) single-family construction.

Look at what this means for a typical 1,500 square foot house.  At the ppsf's listed above, this house was worth $315,000-$330,000 in 2006.  Let's call it $300,000, because the average ppsf tends to skew above the median.  $300,000 is a lot for a 1,500 square foot home, I know.  But just wait.

Suppose home prices increase by 4% per year for the next five years, roughly the historical rate relative to inflation.  Our $300,000 home will be worth $365,000.

Now take the same home and assume the price increases by 10% per year for the next five years.  Price:  $483,000, more than $100,000 more expensive.  Plug in last year's rate and you get  $528,000.   (Historically, that's extremely high, but more moderate than the boom years in California and the Northeast.)  Whatever price we end up with, you can bet it will be sticky, too.

Some people have expressed concern about a potential glut of luxury condos downtown.  I think the worry is unfounded.  If we don't reel in single-family home inflation, people will be moving downtown for the bargains.

Increasing the supply of housing really does affect prices

Here's a good illustration of the fact that the rules of supply and demand apply even to housing.  (Thanks to the Crossland Team.)  Check out the second chart.  Between 2001 and 2003, home prices stabilized (or stagnated, depending on your point of view) because of a surplus of housing.  The oversupply directly affected the rental market:  Average rents dropped for four straight years, and only recently have begun to rise again.

The market provided affordable housing in Austin for decades.   It can still.  Just let the market meet demand.  Reduce minimum lot sizes.  Permit more dense development like row houses and town homes.  Open residential neighborhoods to small multi-family developments -- four-plexes or six-plexes.  That's the solution to our affordability crisis, not affordable housing quotas.