HCoFP: chapter 7: parking costs in perspective

* We know office buildings cost a lot, parking are is often bigger than the floor area of the building it serves, shouldn't be surprising parking costs a lot.
* Considering curbside parking + off-street parking, land and capital devoted to parking may exceed the land devoted to travel. (Certainly true on most of my streets, with two lanes of parking and one lane of travel.)
* Mark Delucchi of UC Davis has the biggest evaluation of the total cost of vehicle use in the US, including monetary and non-monetary (pollution) costs. For 1990 he estimated the total annualized capital and operating cost of off-street parking at $79-226 billion/year. A lot lower than the end of my previous post but perhaps curbside parking outnumbers off-street.
* He notes most parking costs are bundled with other goods; drivers pay 1-4% directly.
* Estimated the cost of roads (including curb parking) at $98-177 billion. Comparable to his off-street parking lot estimate; total of $177-403 is still a lot less than the $720 billion I had. Though that's 1990 to 2002; $1990 GDP was only $6 trillion.

(EDIT: I should read endnotes before finishing these posts. Shoup notes that Delucchi's estimation of the amount of parking spaces was very very conservative, with 1.06 non-residential off-street parking space per car as his *high* estimate. A different estimate had 4.5x that number.)

* A 36 foot wide residential street might have 2 10-foot travel lanes and 2 8-foot parking lanes; parking is 44% of the road use.
* US Dept. of Commerce estimates roads are 36% of all state and local public infrastructure, a category including schools, sewers, water, parks...
* Parked vehicles don't pay gas taxes, a major funding source for roads.
* Subsidy for off-street parking was 1.2-3.7% of GDP, and 4-11 cents a mile, vs. a driving cost (oil, tires, maintenance) of 8 cents a mile. 1990 dollars, again.
* Removing the subsidies would affect driving like increasing the gas tax between $1.27-3.74 a gallon. That's 2002 dollars.
* Commerce estimates for 1997: $5507 capital value per vehicle (they cost more up front, but depreciate), $6542 capital value of road per vehicle (including lots of cheap crappy rural roads) (construction value only, not land value). Also, 208 million vehicles; my 160m was low.
* If there's 3 spaces per car, and only $4000 per space (lowballing it, given the previous chapter), that's $12,000 of parking per car. More than twice the value of the cars...
* 1997 parking receipts were $6.6 billion, vs. $90 billion in fuel tax, vehicle tax, and tolls.
* We can also compare new costs: average cost of a 2001 car is $21,440, vs. $15-30,000 for parking spaces.
* Free parking at work reduces the cost of auto commuting by 71%. At $127/month -- a perfectly reasonable estimate for gargage parking -- and 22 workdays, that's $5.77/day in parking subsidy if the employee doesn't have to pay it. Compare to transit costs of $4-5 day for expensive systems. (Granted, transit is subsidized too.)
* Average of 1.3 gallons gasoline per commute, so subsidy is like $4.44 per gallon. Removing subsidy would probably have more effect than raising the gas tax; more expensive gas -> buying more efficient cars.
* Parking cost is constant given the length of a trip, so free parking encourages more short trips.
* Model studies suggest parking fees would be at least as effective as congestion tolls in reducing congestion (and speeding up buses.)
* It's technologically a lot simpler to charge for parking than to charge for congested driving, especially if you don't have freeway or bridge bottlenecks for toll gates.
* In 2001 the amortized cost of owning and operating a car was $4.38 a day. Compare to any reasonable gut estimate of fair parking fees...
* Total area to park cars, at 3 spaces per car, but excluding maneuvering space, would be 4950 square miles or the size of Connecticut. (230e6 vehicles * 200 foot2/space * 3 spaces).

* So, parking has huge costs, comparable to or bigger than the cost of buying and running vehicles, building the roads, or both combined. But drivers hardly every pay the cost directly. Which means we're all paying it, through higher prices for things.

HCoFP: chapter 9, in-lieu fees

(Chap 8 is a two page analogy about what if making phone calls was free and buildings had to provide as many phone lines as occupants, so all could make a phone call at the same time.)

A handful (he found 47) of cities worldwide allow developers to pay fees in-lieu of providing parking spaces. Some even require it, or partially so. This has a lot of advantages: old buildings can go to new uses more flexibly; shared parking can be more efficient (big buildings are more efficient; different time uses can be combined); visitors are more likely to park once and walk around a district, vs. driving between individual businesses; consolidated parking in public buildings can make for a nicer streetscape than scattered lots, especially if the city puts retail in the first floor.

Fees are usually less than the cost of building parking spaces, generally because city planners chickened out when they saw the actual cost. "The fee would be 'too high' if the city charged the full cost." Of course, sometimes parkers have to pay something so costs are re-couped that way.

In Beverly Hills, developers were willing to pay up to $53,000 to get out of having to provide a parking space. $16,000 is a more usual 2002 average. Combining fees per space with spaces per office building area allows us to calculate the virtual parking impact fee per square foot, and compare it with actual impact fees for stuff like transit, schools, sewers, etc.

Average parking impact fee 2002: $46 per square foot.
Average total of all other impact fees: $1.86

San Francisco has the highest transit impact fee in the US, $5/sqft. So the average parking impact fee (via in-lieu) is 9 times the highest transit fee.

Of course this is still a modest reform of a broken system; it's still requiring developers and thus the people who buy or rent from them to pay for the cost of parking, subsidizing the decision of other people to drive. Radical reform would be to require neither fee nor spaces, and let markets provide the needed parking at market prices... kind of sad that getting out of the way and letting markets work, for a fairly fluid commodity markets are entirely capable of providing (unlike health care or education, arguably) is 'radical'.


Incidentally, the tables showed parking requirements as high as 22 per 1000 sqft for Beverly Hills restaurants or 28 for Palm Springs Cabaret.
For office buildings, the in-lieu US cities required an average of 2.8 spaces, non-US cities 1.7.

Just what we needed, more beef drugs!

New drug Zilmax causes cattle to put on more pounds of muscle, less fat, is taking over the industry.
http://www.slate.com/articles/health_and_science/food/2013/02/zilmax_the_cattle_growth_drug_that_s_making_beef_more_like_chicken.html

Chicken white meat is fatless protein. Pork has become leaner, "the other white meat". Now beef. At this rate you might as well eat tofu...

Much longer article from last year using the issue to talk about the corruption of public agricultural scientists.
http://chronicle.com/article/As-Beef-Cattle-Become/131480/

HCoFP: chapter 10, Reduce demand

I ran across this "total cost of car ownership" calculator.
Nothing special about it, except I noticed that there's no question or input box for parking costs. You'd have to remember and stick them in with "other fees". Seems... illustrative.

***

You can also let developers reduce demand rather than providing parking. Like providing free transit passes to employees or tenants. One study found "ride-alone" share dropping from 76% to 60% and transit use going from 11 to 27%; parking was reduced 19%.

Many commuters won't ride transit even when free, so employers can bulk buy passes from agencies on the cheap; Santa Clara sells Eco Passes for $5-80 a *year*, vs. $420/year for individual purchases. And the Eco Pass includes a free taxi ride home for transit users, in the event of illness, emergency, or unscheduled overtime.

Conservatively: if every 6 passes at $80 each saves you one parking space at $26,000 in construction cost or in-lieu fees (Mountain View low end), that's $480/year to save $26,000 in capital cost, not counting lower garage operating costs as well. $480 is 1.8% of $26,000, so it's a great deal if your interest rate is higher than 1.8%. Which it is.

***

Passes in general are subject to adverse selection: heavy transit users tend to be the ones buying them, driving up the cost, a la insurance death spirals. Per capita cost of providing transit passes to an an entire urban population -- or making it zero-fare -- might be a fair bit lower.

For developers, passes could be more attractive: they're a benefit tied to the development, vs. public parking funded by in-lieu fees, which helps competitors as well. Also the cost varies with the number of people, vs. the fixed cost of parking capacity.

High density transit oriented development (TOD) especially would benefit from this, since parking is more expensive in denser areas. This awakens me to an oddity of the TOD by Wellington station that I found last year: yeah, there's six story apartment buildings with retail under them, but to get to the train station you have to go through a big-ass parking structure and then down a long walkway. And looking at Google Maps, the garage is a solid block taking most of its, er, block, while the other buildings are thinner Ls that make way for yet more parking spaces on the ground, dominating the area.

California TOD is apparently pretty ineffective, with most employers offering parking not passes.

Universities can arrange for ID cards to be passes (IU did.) Stadiums could pay to have tickets good for free transit on the day of the game -- especially useful given how few days a year stadiums are in use; rather silly to be paying for a peak parking structure that's empty over 300 days a year. Hotels could offer passes to guests, both getting out of parking and reducing rental car expenses. And like in-lieu parking fees, this isn't just for new buildings, but can allow conversion of old ones and infill of old parking.

***

Another tactic: California law requires many employers to offer parking cash-outs to employees, taking cash instead of free parking. This can reduce driving to work by 11%. He says the offer costs $24/year per employee because of the money not spent on providing parking, but doesn't spell out what the actual cash offer to the employee is.

BTW, I like his apparently numerical conservatism. Having spent an earlier chapter establishing parking costs of $15-30K, he'll then use $10,000 for an example.

While searching on cash-outs I found this: " Employers offer employees subsidized parking because these benefits are usually untaxed. A typical employee must earn $1,500 or more in pre-tax income to pay for a parking space that costs their employer only $1,000 to provide". Also has a table; cash-out ranged from $18 to $126/month.

***

Car sharing is another option, and in some San Francisco City CarShare program, 29% of the members had gottern rid of a car within two years, though 8% had bought one. Cities could allow a developer to reduce spaces if a shared car is provided.

Next, part 2: Cruising for Free Parking.

HCoFP: chapter 11, Cruising

Chapter opens with a quote about paying for parking being like going to a prostitute. "Why should I pay when, if I apply myself, maybe I can get it for free?" -- George Constanza, Seinfeld character. Pun on "cruising".

Then an analogy to ocean fishing, and the depletion of common property resources. Ah, we're moving from the costs of mandatory off-street parking to the costs of curbside parking. People search for free or cheap curbside parking to save themselves money, while costing everyone else in congestion and pollution -- and often costing themselves in time.

Anecdote about desperately looking for parking in 1920s Connecticut, back when there were traffic officers at every corner.

A 1975 Colorado Supreme Court decision used the word "autoists". Never seen that before.

The first cruising study, in 1927 downtown Detroit, found 19% of traffic was cruising at one point, 34% at another, between 2 and 6pm.

In 1933 DC, cruising was 19% of trip time, and reduced overall speed from 14.2 mph to 8.5 mph.

In 1960 downtown New Haven, cruising was at least 17% of total miles traveled.

Cruising congestion particularly hits buses, making them slower and more irregular, as they're using the same lanes, and the cruisers are going particularly slowly as they look for spaces. IME it affects bicyclists as well... Bad buses encourage driving, which makes the congestion and buses even worse...

In 1977 Freiburg it was estimated that *74%* of the traffic was cruising.

Cruises of more than 5 minutes are common, or even more than 10 minutes.

1985 Cambridge Harvard Square: average of 11.5 minutes in cruise time, range of 2-25 minutes, average distance traveled 1.27. Estimate that 30% of traffic was cruising.

***

Mobile parking! When one person drives round and around while someone else runs and errand. At 12 cents a mile and 10 miles an hour, this can be $1.20 an hour, cheaper than lots of parking. Also 'live' parking, where a driver remains in a stationary car in a spot that would be illegal to simply park in. Provo sends zoning inspectors in pairs, one to inspect a property and one to tend the car. Given the cost of labor that's pretty expensive.

***

If average search time is 3 minutes per downtown space, and a space sees 10 cars a day, it's generating 30 minutes of cruise time a day. At 10 mph, 5 miles of cruising, just for that space. At 33 spaces a block, 165 miles per day per block, or 60,000 miles traveled per year -- more than two trips around the Earth.