Tokyo’s surprising lack of density

Wendell Cox has received his fair share of criticism from this blog, but his post last week about Tokyo’s surprising lack of density is very interesting. Sure, Tokyo’s suburbs are dense enough to be connected by job centers by rail, but the core is almost completely low- and lower-mid-rise, and thus not very dense:

Tokyo does not have intensely dense central areas. The ku area [historic core] has a density of 37,300 per square mile (14,400 per square kilometer). This is well below the densities of Manhattan (69,000 & 27,000) and the ville de Paris (51,000 & 21,000). Only one of the ku (Toshima) exceeds the density of Paris.

And then the suburbs themselves aren’t as compact as they could be:

Further, according to the Japan House and Land Survey of 2008, Tokyo has a large stock of detached houses, by definition lower density. Nearly 45 percent of the Tokyo region’s housing is detached. One-third of the dwellings within 30 kilometers (18 miles) of the core are detached. This figure rises to more than 60 percent outside 30 kilometers from the core and 85 percent between 60 and 70 kilometers (37-43 kilometers) from the core (Figure 2).

Some might see this as a validation of New Urbanism (which is sort of a bastardization of Old Urbanism), whose response to tall building enthusiasts like myself, Ed Glaeser, and Alon Levy is that “dense doesn’t have to mean tall.” And it’s true – Tokyo manages a relatively high density with very few tall buildings.

But there are costs that Tokyo bears for its lack of height and downtown density.

First and foremost are the high housing prices. Imagine New York City if Midtown and the Upper East and West Sides were still tenement neighborhoods, and everyone living and working above the sixth floor was competing for housing and space in the outer boroughs. Narrow the streets and replace the prewars with postwar buildings, and that’s Tokyo.

Tokyo’s high housing costs manifest themselves in many different ways. For one, people cram themselves into tinier and tinier homes, and are forced to endure the noise of their neighbors in a way they wouldn’t be if half of them could be elevated into the sky. Smaller homes are ceteris paribus good more energy efficient, but not when so many of the homes in the suburbs are single-family detached, and thus less energy efficient than slightly larger apartments that aren’t leaking energy from all five exposed sides.

High prices also cause people to live very far from work. Many of them still take the train, but the commute is very long, sapping what is becoming an increasingly precious commodity: time. And some commutes are just impossibly long, limiting job opportunities and flexibility.

And then on an aesthetic note, the density caps lead to ugly (not to mention energy inefficient) buildings. Japanese cities have very few historic areas left compared to European and American ones, but redevelopment is limited by the fact that many urban buildings are built to the zoning envelope, and thus tearing them down and building anew will result in higher rents per square foot, but not more square feet.

In this way, Tokyo is a little like the northern Brooklyn neighborhoods of Williamsburg and Greenpoint: huge demand drives up prices for a hideous housing stock that isn’t being expanded even when it is replenished.

Many people say that Japan’s low-rise skyline is the result of its earthquake-prone geology, but from what I understand, the consensus among engineers nowadays is that skyscrapers aren’t actually more earthquake-prone than lower buildings. Shortly after Japan’s earthquake last year, R. Taggart Murphy at TNR wrote that “skyscrapers, if properly constructed, were actually more structurally stable than the six-to-eight-story office buildings that then constituted Japan’s standard office blocks,” saying that Japan accepted this engineering reality in the ’60s when it allowed the first skyscrapers to go up.

And in fact Japanese seem to be demanding downtown high-rises. Even after the earthquake, Japanese buyers didn’t abandon high-rises like everyone thought they would:

In the months right after last year’s March 11 earthquake, sales of [high-rise] condominiums in Tokyo dropped 30 percent compared to the previous year. Much of the drop was in areas surrounding Tokyo Bay, which is basically landfill. Fears of liquefaction caused potential buyers of tower condos to reconsider, and for a while the media surmised that planned high-rise housing projects might be abandoned.

That didn’t happen. According to real-estate analysts, the earthquake convinced many commuters to move closer to their workplaces, so if another major one strikes they would be able to get home quickly and without the need for public transportation. And the waterfront is within 5 km of the central business district of the capital.

Tokyo doesn’t need to embrace all aspects of the “hypertrophic” city. The streets, for example, can stay narrow. But it’s got to at least give up on the obsession with low- to lower/middle-height buildings if it wants to bring down housing prices, cut commutes, and give its inhabitants more space and privacy.

Come to think of it, a modern Japanese take on Manhattan’s Financial District would be pretty cool!

The High Cost of Free Parking Chapters 5-9

This post follows on the earlier discussion of the first four chapters of The High Cost of Free Parking. Sorry for the delay on this next section. Work has been getting in the way of reading this parking tome. I will plan to post on chapters 10 – 14 next week on Wednesday.

Chapter 5- A Great Planning Disaster

Shoup sets up parking requirements as a great planning disaster. If an individual developer chose to dedicate more of his land to parking than his customers demanded, he would lose money on the margin. If he is a major property owner and somehow made this mistake repeatedly at many properties, we might consider it a disaster. But a planning disaster occurs when no individual loses a lot of money in this type of error, but rather we all lose some. Shoup explains that parking requirements breed demand for more parking. By subsidizing driving, these rules lead more people to become drivers and encourages sprawling development. This in turn creates an increased demand for free parking and leads to higher parking requirements, since many cities base these requirements on the peak number of people who would like to park at a building for free, leading to the parking disaster we have today.

Shoup explains that oftentimes parking requirements are so onerous that they dictate development both in use and in architecture. For example, Los Angeles’ “dingbat” apartments which are apartments built on stilts over driveways were created to fulfill requirements for covered parking. This chapter includes the empirical evidence that I find most persuasive so far, a study of changes in development after Oakland implemented a parking requirement in 1961. For new developments in the two years after the regulation went into effect, residential construction costs increased 18% per unit, housing density decreased by 30%, housing investment decreased by 18%, and land values fell by 33% compared to the four years before the requirement. This is strong evidence that in Oakland, at least, parking requirements, rather than demand for parking, drive parking supply.

Parking requirements also make existing development less flexible. As he explains, the specificity of requirements for each land use mean that buildings can rarely be used for something with higher requirements than current uses. Shoup points out that this obstacle toward reusing existing buildings creates incentives to pursue greenfield development where it’s cheap to build required parking, leaving properties vacant in center cities.

Shoup points out that the costs of parking requirements do not fall evenly on everyone. They act like a regressive tax because low-income people who are less likely to be car-owners and make fewer car trips receive fewer benefits but all of the costs. Additionally, they encourage people to become drivers who, in a world without parking requirements, might prefer to use other transportation modes. Parking redistributes wealth from those with preferences for making frequent, short car trips to those who drive less or don’t drive at all.

Chapter 6 – The Cost of Required Parking Spaces

The title of this chapter is somewhat misleading because it focuses on the cost of parking generally without distinguishing whether or not these spaces would have been built absent parking requirements. Shoup details the cost of constructing parking structures, using data for garages on UCLA campuses, finding that the average cost of each space built from 1961-2002 is $22,500 in 2002 dollars. Unsurprisingly, underground parking is significantly more expensive that above-ground parking. Since parking is usually built underground when land is more valuable, we can see that parking requirements are more expensive in more expensive areas. Using the assumptions of a 4% interest rate and a 40 year amortization period, Shoup calculates that the average parking spot at UCLA costs $127 per month.

Because free parking induces people to drive more than they otherwise would, Shoup also discusses some of the externalities of parking. As more roads induce more driving, so do more parking spaces. He estimates that each parking spot at UCLA has monthly external costs of $73 in traffic congestion and $44 in pollution, for a total of $117 per month. He points out that while in 2002 the average debt-financed parking spot costs UCLA $201 per month and has $117 in externalities, students paid $54 for parking passes.

Chapter 7 – Putting the Cost of Free Parking in Perspective

Because we pay for free parking in the costs of all other goods, we often don’t have a good idea of what it really costs. Shoup cites a study from Mark Delucchi who estimates that the total cost of off-street parking in the United States from 1990-1991 was between $79 billion and $226 billion per year. Drivers during that time paid about $3 billion of that cost  in lots that charge for parking. He also compares parking costs to the cost of roads and vehicles, finding that using conservative estimates of the cost of parking spaces, our stock of parking is worth more than both our stock of vehicles and our stock of roads. He argues that charging for parking would be a more effective way to reduce traffic congestion than congestion tolls.

Chapter 8 – An Allegory: Minimum Telephone Requirements

Shoup proposes a thought experiment about what would happen if we treated phone service the way that we treat parking, requiring those who receive calls to pay for all of the calls they receive free to callers. This policy would increase the demand for telephone service, leading to minimum telephone requirements and even more calling. It’s funny how this chapter feels very dated even though the book was published only seven years ago.

Chapter 9 – Public Parking in Lieu of Private Parking

Shoup suggests that cities can improve their parking policies by allowing developers to pay for spaces in public parking garages rather than requiring parking on site. This policy introduces design flexibility, reduces waste in parking spots for proximate businesses that experience peak parking demand at different times, allows existing buildings to be reused more easily, and allows people to park in one place and leave their car there while they visit multiple businesses.

Cities that offer the option of in-lieu fees for parking offer interesting insight into the costs of parking for developers because when  developers choose to pay the fee rather than build parking, they forgo the value that parking adds to their property. Thus when developers choose to pay in-lieu fees, they are less than [cost of building parking space - capital value of the space].

This chapter includes interesting data on the parking requirements and in-lieu fees charged across cities. Shoup demonstrates that there is essentially no relationship between these two variables. This suggests that city planners do not take into account the opportunity cost of parking when they set requirements, ignoring that demand slopes down.

Thoughts so far:

This section provides some convincing empirical evidence, but I think at times Shoup stretches this evidence to undermine his own case. The Oakland case gets exactly to what I see as the relevant question: How do parking requirements influence development? In a free market, it’s quite possible that some business and residential developments would provide free parking for their patrons, and we need to know how parking requirements increase the supply of free parking.

Shoup relies heavily on data on the cost of parking at UCLA because of its availability. However, I think it would strengthen his case to acknowledge that the UC system might not embody efficiency in development. He points out that students pay less for parking passes than a conservative estimate of the cost of the spaces. However, UCLA students don’t pay market price for many university services.  It’s not clear that parking should be singled out as the one part of the university that taxpayers shouldn’t subsidize.

In chapters 6 and 7, Shoup doesn’t make an attempt to distinguish between the free parking that would be provided absent parking requirements and the total free parking that we have today. He verges on suggesting that it’s wrong for retailers, universities, or housing developers to provide free or below-market-price parking because parking is expensive and suggests parking maximums should be in place rather than parking minimums. I think his case is stronger when he sticks to the problems with parking requirements rather than free parking in general, and I hope the rest of the book includes further studies like the evidence from Oakland.


A quick primer on CBTC and driverless trains

While doing some research for an article about driverless trains, I came across this document by Mircea Georgescu (who most recently worked at Thales [I think?] and whose email I can’t track down! Mircea, if you’re reading this, trimite-mi si mie te rog frumos un email la smithsj@gmail.com!), that’s a sort of primer on CBTC and its application in driverless train operation. The paper is very short as far as these things go, and surprisingly readable, even if Mircea’s English ain’t the best. You can download the PDF here, and here’s the abstract:

Reliable driverless operation requires specific features implemented at system and subsystem levels of the train control system. Communications-Based Train Control (CBTC) is now proven as the best choice for driverless systems due to inherent high levels of safety and reliability with a low life cycle cost. This paper proposes a systematic approach that may be used to determine the most efficient way to fulfil the requirements specific to each customer faced with driverless operation (green field or re-signaling). It also defines “must have” requirements (functionality) to obtain the desired performance and cost. The paper also addresses issues related to the operability, maintainability, and availability of different types of driverless CBTC systems implementations, and the advantages and disadvantages of each solution.

By the way, the article references another written by Mircea Georgescu and Firth Whitwam called “Moving to Full Automatic Operations,” whose citation is “IEEE Hong Kong 2005.” Anyone know where I could get my hands on this? smithsj@gmail.com, as always!

Where is the Canadian real estate bubble going to hit hardest?

We’ve been hearing for a while now about a coming crash in Canadian property values, and it’s really reached a fever pitch lately – seems like denying a Toronto bubble, at least, is pretty rare.

What’s interesting to me, though, is how different the bubble seems to be from the American one about five years ago. In the US, urban real estate definitely took a dive – tons of people went bankrupt, cranes got taken down – but ultimately it recovered much more quickly than the suburbs, and especially exurbs and sprawling Sunbelt cities.

What from what I can tell about Canada, the overvaluation is focused on urban properties, epitomized by the glassy blue towers going up in Vancouver and especially Toronto, and to a lesser extent Calgary. The Vancouver market has cooled and all the worry now is about Toronto, where sales volumes are still up from last year, but I’m not hearing too much worry out of Vancouver, even months after sales supposedly started cooling.

Anyway, the worry in Toronto is really rising. In a very widely-circulated Financial Post opinion piece last week, Diane Francis advocated placing restrictions on foreign (read: Asian) buyers. Despite the jingoism it was an interesting piece, but this chart is more interesting:

Complicating things is the mining energy boom in central Canada, which is also being felt in North Dakota and nearby states.

As you can tell, this is less of an informational post than a post calling for information. Canadians – what do you think is going to happen? Where is the bubble going to hit hardest, who’s going to recover first, and who’s never going to recover? I want your opinion! (Yes, yours!)

Or is there in fact no bubble at all, and all we’re seeing is that Canadians are falling (back) in love with their largest cities? I one said, half jokingly, that the unanimity in agreement that Canada’s in a bubble is the only thing that makes me wonder whether maybe it’s not! (After all, there was a lot of talk about us being in a tech bubble in the run-up to Facebook’s IPO, but that talk seems to have died down.)

Or is the bubble real, and yes, centered on cities, but because of a fundamental rise the demand for urban living, the excess stock will quickly be eaten back up? The US has experienced a boom in renting since our bubble popped, and if the same happens in Canada, then their overbuilt condo units in multifamily buildings will be much easier to convert to rentals than vice-versa, as Las Vegas and Phoenix have to do.

And then where does the rest of the country fit into all this? What about the parts of Canada that aren’t Toronto, aren’t Vancouver, aren’t Calgary, and aren’t energy-rich? Are there any parts of Alberta that aren’t sitting on liquid gold – how are they doing? And I haven’t heard a peep out of Montreal – is that because it’s too regulated to feel anything, up or down? (Kind of like a drugged-up human!)

Canadians – tell me what the hell is going on in your country’s real estate market!

(As always, if you wanna email me with confidential comments, please do! My address is smithsj at gmail.)

Transpo bill gridlock staves off federal transit regulation

There are two general attitudes among urbanists towards the transportation omnibus bill that Congress has been struggling to pass in recent years (?). Some, like Streetsblogs and a number of political advocacy groups, hope for swift passage because of the bill’s transit spending. Others, like Cap’n Transit, balk at all the highway spending, and cheer on the gridlock.

And here’s one other reason to be on Cap’n Transit’s side: no new bill means no federal regulation of rapid transit.

Right now, the federal government only has the power to regulate safety on rail lines that feed into the national mainline network, and could therefore, at least in theory, run into a freight train. This includes all intercity trains (Amtrak and possibly All Aboard Florida), commuter trains (Metro-North, Caltrain, etc.), and the occasional light rail line using an older right-of-way that’s still connected to the national network (e.g., New Jersey’s River Line). Self-contained “rapid transit” networks – subways, elevated trains, and new light rail and streetcar lines – are beyond the feds’ reach.

To many legislators, the Fort Totten crash on DC’s Red Line in 2009, operated by WMATA, was evidence that federal regulation is needed. (WMATA’s MetroRail is actually one of the most technologically advanced systems in America – or at least it was, until after the crash when they turned the ATO off, which drives the train while the operator opens and closes the doors.) There was a big outcry about it right after the crash and a few times since then, and the debate seems to be coming up again.

But despite the liberal leanings of most transit enthusiasts, you’d be hard pressed to find one who thinks that federal regulation will do WMATA – an admittedly heinous agency that needs to be reined in – any good. The main reason to be suspicious is the incredibly poor job that the Federal Railroad Administration (FRA) has done in overseeing mainline rail safety, which has been in its regulatory portfolio for decades.

In the words of Eric McCaughrin, the FRA is “regulating passenger rail out of existence” with its insistence that trains be bulked up to survive crashes. Instead, he suggests that instead safety regulators should focus on preventing crashes with technology like they do in Europe and Asia – for example by installing automatic train protection and operation systems, which, at least outside of DC, have very good safety records. The FRA’s idea of safety, he contends, drives up costs, power consumption, and track wear-and-tear, while driving down reliability and performance (namely, acceleration and deceleration).

The FRA (or whatever body would be charged with the new regulatory tasks) may not make exactly the same mistakes with subways as they did with mainline trains, but many are fearful that they’ll screw up in some other way, such as not keeping up with future technological advances.

Democrats are likely to follow the president’s lead on the matter, who proposed expanding federal oversight to rapid transit and light rail back in 2009. Most Republicans are against giving more regulatory authority on this matter to the feds, though their opposition seems to be based in ideology. I would love to be proven wrong, but I doubt any of them are actually aware of the FRA’s regulatory misdeeds.

In any case, the issue is tied up with the larger highway bill which is of course mired in its own controversies. So luckily for those leery of federal oversight in this matter, Politico says we probably won’t see the feds regulating rapid transit this year:

But with House and Senate negotiators still far apart on the bill, many are predicting another extension of current policy. That would mean no changes in the transit safety structure.

I do have to take issue with Politico’s headline, though: “Transit safety still lags.” It’s not safety that lags – in fact, rapid transit has an impecable safety record, even taking into account the deaths at the hands of the the fools at WMATA. Rather, it’s federal regulation over safety that’s lagging. The Democrats argue that the two are the same thing, but most Republicans and transit advocates clearly don’t see it that way.

Libertarians at the Reason Foundation oppose latest California parking minimum reform bill

From Baruch Feisenbaum, who’s the Reason Foundation’s transportation analyst (disclaimer: I did an internship at Reason magazine a few years ago), surprising agreement with the American Planning Association’s California branch on the parking minimum reform bill (or at least, it surprised me):

The proposed bill has both positives and negatives. The positives include introducing a market-based approach to parking, allowing local governments to set higher standards if it is appropriate for the community, granting certain exemptions to the law including rent control and deed-restricted housing and using substantially more quantitative standards than the old ITE approach. (Under the ITE standards, there were multiple categories for each business using insufficient data points and low r-squared values. For example, adult entertainment had multiple categories. The nude dancing category had separate subcategories for different types of nude dancing including fully nude, partially nude, etc.)

However, there are significant problems with the bill that outweigh its positives. First, the bill sets a statewide standard. California is one of the largest, most diverse states in the country. What is effective in San Francisco may not work in Truckee, CA.

He also takes issue with the fact that the bill was sponsored by the California Infill Builders Association (disclaimer: I’m friendly with Mott Smith, who runs the group), which obviously stands to gain from reduced minimum parking requirements:

The bill is sponsored by the California Infill Builders Association. The association is a trade group working to increase infill housing. As parking spaces cost money, for developers to be able to build these apartments/houses they need something in return. The something could be lower parking standards. Parking should be priced and I understand the desire for infill housing. However, the bill would be best originating from someone without a stake in the game. Such legislation can then be reviewed by a university researcher, the California Legislative Analyst’s Office, and another independent party. The only professional transportation group that has weighted in, APA, is not a strong supporter of the bill.

Finally, he sides with APA California:

This bill presents a big government solution to a government created problem. Free market solutions should operate outside of big government meddling. If the government is restricting free-market parking pricing (which it is) the bill should end subsidies to automobile drivers and developers. All transportation modes should operate on an equal plane. Creating a subsidy to counter another subsidy is both expensive and counterintuitive.

Assemblyman Nancy Skinner deserves credit for trying to encourage market-based pricing. And APA California could be more open to the concept. However, this bill is a big government state-imposed solution that may or may not offer exemptions, does not consider the different geographies of the state and does not separate existing from current transit service. As written this bill needs substantial changes. If the author’s amendment passes, it still has some significant aspects that need fine tuning before it should be considered on a statewide basis.

Stam Staley, a senior research fellow at the Reason Foundation, said on Twitter that he agrees with Baruch.

First of all, regarding the fact that developers are sponsoring the group, as long as they’re being honest about their involvement with the bill, I see no problem with it. The APA’s members obviously also have a stake in the legislation: the state of California is trying to reign in their powers to regulate land use. Does Baruch have a problem with the fact that airline deregulation was spurred by lawsuits by Southwest?

But more importantly, I think there’s a slippery slope in saying that state governments should not infringe on their localities’ abilities to infringe upon private property rights. Libertarians often advocate devolving government to the lowest level possible, but there are certain instances – this one included, in my opinion – where decentralization and libertarianism clash. I tend to come down on the side of libertarianism, local autonomy be damned, but it looks like Baruch and Sam have come to a different conclusion here.

But how far are they willing to take that principle? Do Baruch and Sam also oppose state-level laws that emerged after the Supreme Court’s Kelo ruling that limit localities’ abilities to take land by eminent domain for private use? What about Massachusetts’ 1995 state referendum that ended Boston, Cambridge, and Brookline’s locally-enacted rent control laws? Does lowest-level-governance always trump free markets, or do they draw the line somewhere?

I should also say that I’m a bit disturbed by the fact that Baruch thinks that parking minimums are okay so long as they’re based on something more sophisticated than the standard ITE numbers. Is he basically saying that bureaucrats telling developers how much parking they must include on their private property is okay, so long as the bureaucrats are using a good model?

Ending rent control may not lower prices for non-regulated units

That’s one takeaway from a paper sent to me by one of its co-authors, Andy Garin, at MIT, on the effects of the end of rent control in Massachusetts in 1995 on property values in Cambridge. Fascinating topic, and much thanks to Andy for sending it to me – it’s always nice when other people write my blog posts for me!

Andy assures me that they “went through great pains to make sure our results do, in fact, have a causal interpretation, and meaningful,” but as always, I don’t have the statistical background to fisk its methods, so feel free to go at it in the comments.

Here’s the abstract of the working paper, available on NBER, called “Housing Market Spillovers: Evidence from the End of Rent Control in Cambridge Massachusetts”:

Understanding potential spillovers from the attributes and actions of neighborhood residents onto the value of surrounding properties and neighborhoods is central to both the theory of urban economics and the development of efficient housing policy. This paper measures the capitalization of housing market spillovers by studying the sudden and largely unanticipated 1995 elimination of stringent rent controls in Cambridge, Massachusetts that had previously muted landlords’ investment incentives and altered the assignment of residents to locations. Pooling administrative data on the assessed values of each residential property and the prices and characteristics of all residential transactions between 1988 and 2005, we find that rent control’s removal produced large, positive, and robust spillovers onto the price of never-controlled housing from nearby decontrolled units. Elimination of rent control added about $1.8 billion to the value of Cambridge’s housing stock between 1994 and 2004, equal to nearly a quarter of total Cambridge residential price appreciation in this period. Positive spillovers to never-controlled properties account for more half of the induced price appreciation. Residential investments can explain only a small fraction of the total.

Here’s an earlier (?) version of the paper. If you’re sufficiently motivated, you can find one of their email addresses online and ask for a copy of the latest version, and I’m sure they’d be happy to oblige.