Coyote Blog

Last year, there were about 3.2 trillion passenger miles driven by urban drivers in cars in the US.  My point about light rail is that we can barely afford it for just a few people, given that we spent $1.3 billion to build a rail line for just 17,000 daily round trip riders in Phoenix.  If it were truly a sustainable technology, it could be applied to all commuters.  But at a national average taxpayer subsidy per light rail passenger mile of about $2**, this means that to roll light rail out to every urban commuter would cost $6.4 trillion a year in government spending, almost half our annual GDP.  If it required the subsidy rates we have in Phoenix per passenger-mile, such a system would cost over $12 trillion  year.  In fact, the numbers would likely be even higher in reality, because light rail in most cities is almost certainly built on the highest populated corridors with the most bang for the buck (though some of the diminishing returns would be offset by network effects).

Light rail only works today because we drain resources from millions of taxpayers to benefit just a few generally middle class commuters.    This is not a model that will scale.

** This includes both service on the debt, which is payment for the original construction costs, as well as annual operating losses.  This subsidy is required essentially forever.  After 20-30 years when the original bonds are paid off, by that time systems generally have to be rebuilt in their entirety   (as folks in places like Washington DC are learning).  There are probably only 5-6 cities in this country that have the urban population densities to make rail systems come even in the ballpark of working financially, and places like Phoenix, Seattle, Houston, Portland and LA are NOT among them.

Let me offer up a definition of sustainability that I think most environmentalists and progressives would accept:

We are acting in a sustainable manner if we are achieving our goals in a way that does not hamper the ability of other people in the world, or of future generations, to achieve their goals.

Most environmentalists and progressives would call light rail lines in US cities a "sustainable" technology because of its notional impact on fuel use and CO2 output (yeah, I know, but we are not going to address those assumptions today).

Let me present one fact, from Federal Transit Administration's 2009 survey of public transit authorities, whose data is linked in various ways here.  Or you can download the summary spreadsheet here.  For all US light rail systems in total:

User fares paid per passenger-mile:           $0.18

Total cost per passenger-mile:                     $2.22

Taxpayer subsidy per passenger-mile:       $2.04

Since I live in Phoenix and the Phoenix light rail system seems to get particular praise as a "success" from light rail supporters, here are the Phoenix light rail numbers;

User fares paid per passenger-mile:          $0.07

Total cost per passenger-mile:                     $3.89

Taxpayer subsidy per passenger-mile:       $3.82

So there, folks, is your sustainable technology.  As I have written before about sustainability, "I do not think that word means what you think it means."

Nationwide, non-users of light rail pay for 92% of its costs.   In Phoenix, non-users pay for 98% of the costs.  Taking the Phoenix system as an example, resources are drained from literally millions of people so that 17,000 or so people can ride it round trip each day.   Using resources from millions of people, and building up debts that will last into the next generation, to support the transit of just a few people, seems to be the antithesis of sustainability.

If there is any common denominator among progressives, it is that they have little respect for how individuals spend their money.  So they might be unmoved by the loss of resources from so many.  So lets just look narrowly at transit, which I presume the do care about.

Before Valley Metro operated a light rail system in Phoenix, they also operated a bus transit system.  This system still requires a subsidy, but it is much lower than the light rail subsidy.  In 2009, the bus subsidy was $0.74 per passenger-mile.  This means that for the same amount of taxpayer funds, Valley Metro can provide 1.0 passenger-mile by train or 5.2 by bus ($3.82/$0.74).   I can guarantee that cities building light rail are not having their budgets quintupled.  So the result is that, as light rail gets built, total transit ridership falls in most cities as rail costs crowd out existing bus services.

Update: Most light rail articles in our local papers, which have been mindless boosters of the system, generally consist of asking riders if they like the system, who inevitably answer "yes!"  This is somehow a proof the system is great.  Well, duh.  I too am likely to be happy with a service where I only pay 2% of the costs.

Update #2:  Last year, there were about 3.2 trillion passenger miles driven by urban drivers in cars in the US.  My point about light rail is that we can barely afford it for just a few people, given that we spent $1.3 billion to build a rail line for about 17,000 daily round trip riders in Phoenix.  If it were truly a sustainable technology, it could be applied to all commuters.  But at a national average taxpayer subsidy per light rail passenger mile of about $2, this means that to roll light rail out to everyone would cost $6.4 trillion a year, almost half our annual GDP.  If it required the subsidy rates we have in Phoenix per passenger-mile, such a system would cost over $12 trillion  year.  In fact, the numbers would likely be even higher in reality, because light rail in most cities is almost certainly built on the highest populated corridors with the most bang for the buck (though some of the diminishing returns would be offset by network effects).

I return to an old favorite topic of mine this week, government subsidies for business relocation, in my column at Forbes.com.  An excerpt:

To see this clearer, lets take the example of Major League Baseball (MLB).  We all know that cities and states have for years been massively subsidizing new baseball stadiums for billionaire team owners.  Let’s for a minute say this never happened – that somehow, the mayors of the 50 largest cities got together in 1960 and made a no-stadium-subsidy pledge.  Would baseball still exist?  Sure!  Teams like the Giants have proven that baseball can work financially in a private park, and baseball thrived for years with private parks.  But would baseball be in the same cities?  Well, without subsidies, baseball would likely be in the largest cities, like New York and LA and Chicago, which is exactly where they are now.  The odd city here or there might be different, e.g. Tampa Bay might never have gotten a team, but that might in retrospect have been a good thing.

The net effect in baseball is the same as it is in every other industry:  Relocation subsidies, when everyone is playing the game, do nothing to substantially affect the location of jobs and businesses, but rather just transfer taxpayer money to business owners and workers.

I would have sworn I hated the whole Duran Duran catalog, but I was watching this video (because a friend of mine did the costumes, including the wedding dress) and I must say its not a bad song.  So my apologies to Duran Duran.

By the way, my friend said that I would be surprised how many women later wanted to get married in exactly this dress.

Also, speaking of music videos, if you have never seen the Hugh Grant / Drew Barrymore movie Music and Lyrics, the plot is an utterly predictable romantic comedy (with Hugh Grant playing the same guy he always plays) but in the opening scene they nail the 1980's pop music video genre dead on.  I can't embed it but the video is here. You can miss the rest of the movie, but the spoof is genius.

This story is from the WSJ, and gets extra bonus points for including the poster-boy of the corporate state, Jeffrey Immelt

Treasury and OMB singled out an 845-megawatt wind farm that the Energy Department had guaranteed in Oregon called Shepherds Flat, a $1.9 billion installation of 338 General Electric turbines. Combining the stimulus and other federal and state subsidies, the total taxpayer cost is about $1.2 billion, while sponsors GE and Caithness Energy LLC had invested equity of merely about 11%. The memo also notes the wind farm could sell power at "above-market rates" because of Oregon's renewable portfolio standard mandate, which requires utilities to buy a certain annual amount of wind, solar, etc.

But then GE said it was considering "going to the private market for financing out of frustration with the review process." Anything but that. The memo dryly observes that "the alternative of private financing would not make the project financially non-viable."

Oh, and while Shepherds Flat might result in about 18 million fewer tons of carbon through 2033, "reductions would have to be valued at nearly $130 per ton CO2 for the climate benefits to equal the subsidies (more than 6 times the primary estimate used by the government in evaluating rules)."

So here we have the government already paying for 65% of a project that doesn't even meet its normal cost-benefit test, and then the White House has to referee when one of the largest corporations in the world (GE) importunes the Administration to move faster by threatening to find a private financial substitute like any other business. Remind us again why taxpayers should pay for this kind of corporate welfare?

First, the moment GE said that this could be financed privately, the Feds should have said "then what the f*ck are you talking to us for? Get out of here."  By the way, privately probably does not mean privately -- it probably means going to private banks or investors who in turn access many of the same taxpayer funds.

Second, its amazing that the threat to finance this privately rather than sponging off taxpayer funds is treated as a threat by the Obama Administration.  They desperately want to "take credit" for the project and can only do so by spending our money  (this is the same impulse that propels politicians who have never given a dime to charity to want to spend taxpayer money in order to be called "caring.")

Kevin Drum wonders whether the proposed $700 million bid by Farmers Insurance for naming rights on a prospective LA NFL stadium makes any economic sense.   It is an interesting question.  I wrote:

This has always struck me as one of those agency problems, where the executive's incentives are different from the shareholders. Executives get a ton of benefits personally from this -- higher profile for the company which improves their profile and marketability, they get a prime box for the games, parties, etc.

Before the audience here slips into a round of corporate executive bashing, my sense is that the same perverse incentives are working for municipal leaders who have a mismatch with taxpayer interests when they shove huge amounts of taxpayer funds to owners in stadium deals (deals which economists speak with one voice on -- they never pay off for the community in full). One of the dirty secrets of these deals is that they generally include a sort of kickback in the form of boxes and club seats for the Mayor and city council's use (and sometimes multiple boxes for leaders of other government agencies in the town).

This chart, from a book by Branko Milanovic via Carpe Diem reinforces a point about income distribution I make all the time -  for all we talk about income distribution in this country, our poorest 20% would be middle class in many countries of the world.  While I would love to see our poor doing even better, it begs the question of whether distribution or absolute prosperity is more important.

Just to give you a feel for reading the chart, the US's lowest ventile, or bottom 5%, have income that would put them in the 68th percentile worldwide.    Our poorest 20% (the first 4 ventiles) would be upper middle class or better in Brazil, China, and India.

When comparing to European social democracies, it turns out that while the US's income distribution is wider, that is almost entirely due to the top end being higher.  The poorest 10% make about the same as the poorest 10% in Europe, and I would argue that this analysis (from a leftish think tank) actually underestimates a quality of life advantage for American poor, who come out higher even than the middle class in Europe on things like living space and appliance ownership.

Perhaps more importantly than income inequality, income mobility remains high in this country. More on income inequality concerns here.

I am in the process of completing a home mortgage.  The process has become awful again, not as bad as it was in the early 90's but harder and more frustrating than in the mid-2000's.  There is one set of rules, and if one's situation does not fit those rules, good freaking luck.  Right now, for example, getting a home mortgage when one is self-employed, even in a pretty large business with a decade of history, is really hard.

So I like this proposal

At the moment there is nobody doing conforming mortgages except Fannie and Freddie. Indeed there is almost nobody doing mortgages of any kind except Fannie and Freddie. If the free market wants the business they can have it. (They just don't want it at this sort of interest rate spread - and I don't blame them.)

All the government need to do is tell Frannie to raise their price a little each quarter. Currently they charge 20-25bps for guaranteeing mortgages. (The free market won't take credit risk at that price.) So it is entirely open to the FHFA (and hence the Treasury) to tell Fannie and Freddie to raise their prices by 5bps. The government will get paid better for the risk they are taking (and what free market ideologue will disagree with that) and the private sector can compete if they want to.

I doubt the free market will. But then in a quarter or two Frannie can raise their pricing by another 5 bps. And a quarter or two later Frannie can raise by another 5bps.

At some stage you will get to a level where the private sector chooses to compete. Frannie should not set its price competitively though. In another quarter they should raise the price another 5bps. And in another quarter they should raise again.

By the way, this is a classic example of not learning from your last mistake.  That spread is absurdly low.  I wouldn't guarantee my best friend's loan for 20bp.  Would you take on the default risk of a $100,000 mortgage for $200 a year?

I have hypothesized that immigration may have an effect on median income -- not because it is bad for the economy per se but just form the fact of adding millions of new people in the bottom quartiles would tend to shift the median downwards (just the pure math of the thing).

I have only given them a quick glance at this point, but Tyler Cowen links several studies that tend to say my hypothesis is false.

Also, here is an interesting discussion about the median income stagnation hypothesis itself and how sensitive it is to the end point, pointing out in particular that most folks start their analysis from a point within Nixon's wage and price controls, which skew the data - shifting the start point forward even a couple of years makes most of the stagnation go away.  He builds on a post by Brink Lindsey showing historic median income growth over a longer time frame, implying the aberration may have been the boom of the 1950s and 1960s rather than the lower growth rates of today.

I would normally find this a fascinating debate if it weren't for the dark cloud behind it that half the folks arguing the point wish to use it as a justification for further reductions in economic liberties.

Human ingenuity keeps finding more oil and gas but we are close to running out of IP addresses, at least in the old IPv4 system, which all of your are probably using right now.  This does not mean the world will shut down - already, for example, all the computers in your home probably share a single IP address to the outside world, and for many of you that IP address is dynamically assigned by your Internet provider to further save addresses.  Many web sites on the same server will share an IP address (which is actually a good reason not to used shared hosting, because if one of the other accounts on your server is a bad actor, your IP address can effectively get banned from sites and networks trying to ban that other person on your server).

However, a new system is in place, but as with many standards transitions the details are tricky.  It will be interesting to see how this mostly free-market transition goes in comparison to government enforced transitions (e.g. television broadcast standards).

The following will probably just demonstrate my total ignorance of networking protocols, but I am not sure why IPv6 couldn't be written in a way that the extra bytes would just be ignored by IPv4 systems.  It could be assumed that all IPv4 addresses of the form www.xxx.yyy.zzz map to www.xxx.yyy.zzz.000.000 in IPv6, but this may be wildly simplifying what is going on.

The reason I bring this us is because I have always thought the way black and white TV was transitioned to color was particularly clever.  They could have broadcast color with three signals of Red, Green, and Blue levels, and then black and white TVs would have to be thrown out - they wouldn't show anything meaningful with that signal.  Instead, though, they mapped color with a three part system of an absolute brightness signal for each pixel, plus two color signals.  If you are familiar with Photoshop, when you choose a color, you can enter the color as three numbers R-G-B for the intensity of each color or as Hue-Saturation-Brightness.  While not the same as the TV system, it is similar in that it has a pixel brightness component, plus to color components.  (my memory is that in the TV system, it is brightness plus two colors and the third color -- blue, I think -- is arrived at by subtraction from the total brightness minus the two other colors.)

Here is the trick - the signal which was just the pixel brightness component is essentially identical to the old black and white TV signal -- after all, a black and white signal is just the relative brightness of each pixel.  So they took a black and white signal and then added bandwidth so that there was more information if one had a color set.  Both technologies, old and new, worked from the same signal.

I suppose the problem with this is that I am thinking of routers like telephones.   Most folks know that if we dial more than 10 digits, the extras are just ignored.  My guess is that routers are more finicky and precise than this, and they can't just ignore the fact the IP address they are getting are too long.  But I still would imagine there could be a simple hardware hack to cheaply strip off the last part of a longer IP address so that older IPv4 infrastructure could still work in an IPv6 world.  Or is this hopelessly misinformed and naive?