e-drivers, e-cigarettes

by lewis500 on June 3, 2013, no comments

The FHWA has released a statement about driverless vehicles and the New York Times has the story here. The FHWA is being pretty liberal right now. In the future I expect less tolerance, especially once representatives of those who drive vehicles professionally start getting concerned. Teamsters, livery associations, etc.

Even without pressure groups, driverless vehicles face the same challenge as e-cigarettes: vague unease. Although there is no evidence that electronic cigarettes hurt people, nor any theoretical reason why water vapor with nicotine in it would hurt your body more than gum or patches; and although there are hundreds of thousands of people who have quit smoking by using e-cigarettes, authorities have been rushing to give e-cigarettes the same restrictions as the flaming tar-sticks that are actual cigarettes. Why? The root cause is magical thinking:

amongst the Azande[6] tribe members rub crocodile teeth on banana plants to invoke a fruitful crop. Because crocodile teeth are curved (like bananas) and grow back if they fall out, the Azande observe this similarity and want to impart this capacity of regeneration to their bananas.

Because e-cigarettes are designed to imitate real cigarettes, it’s no surprise they give us the same heeby jeebies. But no one will admit this publicly.  The heavy-smoking nation of France is moving to ban e-cigarettes in public places for reasons that include the following:

Another worry they cite is that the electronic alternative will increase the general temptation to smoke, including enticing those who have quit to start again, or that smokers may use them alongside rather than instead of regular cigarettes.

“This is no ordinary product because it encourages mimicking and could promote taking up smoking,” said Touraine, who announced her plans at a news conference.

Maybe that’s true. Maybe it’s not. But we have no reason to think it’s true, and it doesn’t even make sense: “Now that I have tried a safe nicotine product, I will switch to an alternative that delivers as much nicotine but also kills me, costs more, and makes me smell bad.” It’s as though vodka were a gateway drug to premium gasoline.

But this is par for the course: reason nearly always arrives after the emotions have decided and has to mop up. Even if e-cigarettes turn out not to be completely harmless, we know that cigarettes are massively lethal, so rational authorities would be tolerant if not enthusiastic about explosive growth of a heretofore harmless alternative. But because reason is not involved in creating such opinions, it is ill-equipped to overturn established prejudice, and bad policies can persist indefinitely. Medical outsourcing, gay marriage, deregulation of professions and narrow banking are other domains where small downsides loom disproportionately terrifying due to unfamiliarity. I fear this will happen to driverless vehicles as soon as a few salient accidents happen.

Here in Berkeley, my neighbours stopped a new eight-story apartment building. They said that it would take away too much sun. It would seem the pinnacle of sun-greed for a Berkeleyite, who lives in near-constant sunlight in a city with only two tall buildings, to worry over losing some of it…especially when the city faces a housing affordability crisis. But it’s really just that some human beings were able to stop something new from happening, and so they did.



agglomeration or substitution?

by lewis500 on May 28, 2013, no comments

Karl Smith poses a theory about China’s absurdly high rates of investment:

I think the Chinese government is…willing to lose money on its capital investments in order to provide jobs for people moving to the city. This is a smart move if you think cities produce agglomeration effects.

With apologies to the less wonkish, China is using physical capital as a loss leader in order to grow cities that will produce network effects will in turn foster the human capital that really makes a country rich.

This begs a question about what counts as an agglomeration effect. We could think of a ‘true’ agglomeration effect as arising from better matching of workers to jobs, the easier flow of non-patented ideas, economies of scale, or the greater varieties of products made viable by niche markets in large cities with diverse tastes.

But there is another agglomeration effect that is more relevant to China’s situation: substitution of market production for household production. This will show up as economic growth even if there is no material improvement. For a given level of material well-being, a rural Chinese household probably transacts less in the market.

As a simple example, consider a household with a husband farmer and housewife in a rural area. The housewife turns some of the farm’s output into three meals per day and cares for their child. Now suppose the couple moves into the city, where the husband earns a wage equivalent to his farm output. The wife gets a full-time job and uses her earnings to buy three meals per day and to put the kid in daycare (I have no idea if Chinese people actually use much day care but it didn’t seem realistic she spent all day on cooking.). Now the couple is about as well off as before, but the household’s GDP has risen enormously.

I think this story, though very simple and unrealistic in a thousand ways, tells us a lot about China’s growth and even provides a clue as to why the government tries so hard to move people into cities.

The wife’s move from household to market production has afforded the state an opportunity to claim some cut of the family’s consumption. She works and spends, and the government may tax her paycheck, employer’s profits, and purchases. If the family is taxed exactly the average rate of all Chinese households, government revenue’s as a share of measured Chinese GDP has not risen. But revenues have expanded in absolute terms, which is what matters, to some degree, to the state.

This explanation is more plausible than Karl Smith’s, because it does not rely on Chinese bureaucrats understanding agglomeration economies. It merely relies on a vague intuition about job creation and the connection between jobs and state revenues. It also posits self-interest for bureaucrats rather than improvement in the Chinese people’s material well-being. The Chinese state is not very concerned with its people’s well-being relative to competing goals like national glory and bureaucratic reach. It’s a country with high speed rail but no Social Security.

An even more plausible explanation, though, is simply that the Chinese state is captured by construction interests. If you take Brent Flyvberg seriously, you’ll have admitted that Western government often get captured by such parties and build absurd things. China might just be an extreme example where the construction interests have also seized the banking sector. Add to this the observation, not well known in western countries, that some of China’s elite are deeply embarrassed by rural Chinese customs.

Here is the world’s longest bridge over water, Jiaozhou Bay Bridge, and it only serves 10,000 vehicles per day:


by lewis500 on May 27, 2013, no comments

As an avid blogosphere consumer, I’ve become intrigued by what policies are perceived as exercises of government power. What counts as coercion? Recently, the Antiplanner argued in favor of allowing food stamp recipients to spend food stamps on junk food:

It would be one thing if NCPPR were arguing that SNAP is a waste of money that should be eliminated. But its argument that government should only allow food stamps to be spent on healthy food effectively admits that NCPPR believes the government is capable of defining what is healthy. And if government can do that, why shouldn’t it extend its wisdom to everyone, instead of just SNAP recipients?

I couldn’t disagree more. In his last sentence, O’Toole seems to think that, by admitting the government is capable of deciding what counts as junk food, one necessarily admits the government should keep everyone from eating junk food. The conclusion (government should decide what everyone eats) does not follow from the premise (government should not buy junk food for poor people).

O’Toole’s leap is really symptomatic of a deeper intellectual failure in America’s conservative/libertarian politics: intransitive preferences. On issue after issue, America’s right wing has fought tooth and nail for Pareto-inferior middle ground. The confusions are explainable, however, when you consider that individual humans reason through symbols and heuristics rather than rules.

Small government advocates should really be interested in two things: (a) tax rates and (b) deadweight losses of regulation. Otherwise they should be orthogonal to policy…or perhaps utilitarian, preferring the state that maximizes welfare given X taxation and Y regulation.

Many libertarians, however, like O’Toole, focus on a tertiary aspect of policy: salient purposefulness. They cringe at top-down planning. But every policy accords with some plan, though, which begets the great irony of our policy era: if you successfully make the case against planning but fail in your quest to strip the government of a popular responsibility, all you’ve accomplished is to commit the government to a very bad plan…one that is probably planned by suppliers. Worst of all, for the libertarian, the outcome is inevitably profligate. I will give some examples

  • Given a food stamp program, libertarians rally for coverage of junk food. Bryan Caplan, the uberlibertarian, asks his fellows, “Is it ‘government rationing of food’ if you can’t buy cigarettes with food stamps?” (link). For a given public expenditure on food stamps, it seems this simply lowers human welfare and does not change the price of the program.
  • Given Medicare exists, libertarians oppose conscious decision-making about which treatments it covers. So Medicare grows unsustainably by spending money on unproven treatments and begets higher taxes, since, being perhaps the most popular program ever, it will never be eliminated. Again, Bryan Caplan lays it out:

Unlike most opponents of Medicare, I think that restricted reimbursements and so-called “death panels” are great ideas.  If the government is paying the bills, saying “We’ll pay for whatever you want” or “We’ll pay whatever it takes to save you” is highly imprudent.

  • Given the state monopoly on money, libertarians want tight money. Why? There is absolutely no philosophical justification. Every monetary stance is equally synthetic. Krugman and some other left wingers would point to sadism. But a simpler explanation suffices: expansionary policy in a recession is conducted with the express purpose of relieving the unemployed and ruined, so it feels more ‘interventiony.’ I believe if there were corporate debtors prisons, and the Fed adjusted policy with the express purpose of keeping these institutions afloat, then Ron Paul would argue for loose money under the same conditions that now stoke his deflationist paranoia. When behavioral economics is more widely understood, in fact, we will even explain historic FOMC policy this way, as Krugman has tested the waters with in one of my favorite of his posts (The Focal Point Fed). Interestingly, loving tight money will beget a much larger state as private individuals fail to find jobs and free market arguments fail the layman’s common sense.

Conservatives, on the other hand, go astray in their own ways, mistaking a different tertiary attribute for small government: profit salience. Dr. Steven Landsburg recently used a term “psychic harm” to talk about why people hold policy preferences over things that don’t personally affect them (although he used it in a strange context), and I would say conservatives derive “psychic benefit” or perhaps “existence value” from mental images of profit-taking, and this begets larger government (even apart from crony capitalism):

  • Conservatives attack the means rather than the ends. Paul Ryan had an alternative to traditional Medicare that included two bold changes: (i) vouchers for private insurance and (ii) capped expenditure increases lower than Obama’s forecast. His program will never ever occur because of (i); but only (ii) impacts the total tax bill. Here’s a feasible small-government alternative proposal: {‘traditional Medicare structure’ + ‘capped expenditure increases lower than Obama’s forecast’ }. But I think Ryan was wedded to privatization…the image of someone, somewhere along the line distributing dividends, because profits are a sign of ‘marketocity.’ Therefore, America will have unadulterated Obamacare.
  • Conservatives waste money on privatization. Privatization often saves money but sometimes doesn’t. It is an empirical matter. In episode of This American Life, Oklahoma state legislator Joe Eddins explains that privatization proved enchanting to Republicans when he secretly brought universal kindergarten to Oklahoma (transcript):  

Joe Eddins
I probably sat down at their desk with each Republican and showed them two or three things in the bill that I thought they ought to know.
Alex Blumberg
And what were those things?
Joe Eddins
Well, I showed them where we were keeping four year olds out of kindergarten, we’re saving enormous amounts of money, you can contract with private providers, and they loved that. That’s all I said.

The money-saving came entirely from changing a loophole for kindergarten coverage, but Republicans seemed very excited about the fact of private providers even when that had nothing to do with the cost savings. Likewise, in another episode of This American Life (link):

So did the city really save any money here by outsourcing, by privatizing Roland’s job? We asked the city of Colorado Springs this question over and over again. And they hemmed and they delayed. They couldn’t find a number. Then they said it’s tough to calculate….

What I learned, though, from talking to the people in Colorado Springs is that for a lot of them these calculations don’t really matter. They don’t care if privatizing actually saves the government money, so long as the government is doing less.

Pause for a moment and reconsider the preferences embodied in the sentence: “They don’t care if privatizing actually saves the government money, so long as the government is doing less.” Why don’t they care about the lower tax rates afforded by cost savings? And why would they possibly care who’s providing the same service for a given tax bill? [Robert Smith is paraphrasing what he heard from a swatch of the community, but I believe he is a credible journalist (bio) who does a good job on Planet Money. ]

It strikes me as relatively easy to maintain public employment in Colorado Springs. Public unions could rebrand themselves as corporations that pay dividends to shareholders on a per-hour basis. By a coincidence the shareholders are also the employees who fulfill government contracts, but this is an internal matter no more of interest to the public than executive compensation at a landscaping firm. In fact, go ahead and make 100% of the employees executives and managers. I think Taco Bell did this for a while. In no time, you’s see tearful defenses of employee pay and ‘dividends’ from Colorado’s most ardent conservatives.

  • Conservatives dislike government prizes but love patents, even though patents are a form of regulation with a massive deadweight loss. When Democrats tried to cut Medicare spending on drugs, Dr. Greg Mankiw said: “The Dems will likely give us lower drug prices and less research into new drugs. Good news if you plan to be sick soon. Bad news if you plan to be sick in the more distant future.” It would seem he has a preference for long-term R&D over deadweight losses from higher taxes used to finance Part D. But when Dr. Joseph Stiglitz wrote in favor of government prizes for new drugs, Dr. Mankiw said, “ He makes some good points, but remember that prizes have to be financed, presumably with taxes. The distortionary effects of higher tax rates could be worse than the distortionary effect of the temporary monopolies created by the patent system.” What’s consistent in both statements is a concern for drug companies’ profits.

under control

by lewis500 on May 26, 2013, no comments

This summer I’m working at a startup that uses control theory to prevent bus bunching.Their first pilot is running in San Sebastian, and right now the team is at a transit conference in Geneva hawking their wares. Anyway, I made these promo videos to explain:

and an animation:

like a phoenix from the ashes

by lewis500 on May 26, 2013, no comments


Last week I had a post about why my roommate, who is a talented computer programmer, only works a few hours per week. It turns out that a much better explanation is that he is a terrible computer programmer because he accidentally deleted priceroads.com. Now I’m back in action but it will take a while to set up everything the way it was before.

economists get what they want

by lewis500 on May 16, 2013, no comments

Over at Marginal Revolution, Tyler Cowen says:

Voters are getting more or less what they want, which is some spending restraint, mostly holding the line on taxes, not too much trust in government as a way of moving forward, and a love of entitlements. One can find that objectionable, and indeed I do across a number of fronts, but there you go. We are not going to elect a new people anytime soon, and in this odd sense you can see all the recent political gridlock as reasonably democratic, more so than its critics would like to admit (I know I’ll generate a bunch of criticisms citing poll data about how Americans really want this, that, or the other but I’ll hold my ground on this one).

Cowen has bemoaned the trend of declining public investment and rising entitlements. He says this in line with voter preferences. I take a very different tack. This is exactly what we would expect from a government run by economists over the last 30 years.

For most of the 20th century, government had more public provision and more regulatory cross-subsidy than it does now. Economists hated this situation because it wasn’t pareto optimal. Government housing created hellholes so bad as to inspire movies like Candyman, so economists pushed for less spending and more direct aid. Price controls and protectionism (regulation), like the Interstate Commerce Commission and tariffs, created such yawning deadweight losses that economists pushed for free enterprise and direct aid for the ‘losers.’ These ‘swaps’ were better off for nearly everyone.

But both swaps will lower investment as a share of public spending.


The first change: swapping {public provision} for {private provision + vouchers} moves plant off of the government’s balance sheet. Whereas building a 30-story, squalid din of vice was a public investment, the reasonably clean housing built to satisfy demand from Section 8, affordable housing tax credits and other transfers are not. The ‘investment’ shows up on landlords’ and affordable housing NGOs’ balance sheets.


glorious public investment of America’s, forward-thinking golden age


affordable housing in our happy-go-lucky modern era


My own field, transportation, represents another area where public investment is frequently wasteful. Many experts agree. Dr. David Levinson (link):

If you don’t trust the poor with money (and this seems to be society’s attitude, since we don’t actually give them money directly), give them transportation vouchers (e.g. top up their per use transit pass with $X per week, or give them a monthly pass, or some other mechanism). This can be discreet, so no one else would know who received vouchers, everyone would use the same pay-as-you-go card. It accomplishes the appropriate ends, without burdening the transit system with this welfare function. It also allows freedom to be spent on taxi or rental car as needed (if in cash-equivalent form), rather than just fixed-route transit. Just because some people are poor does not mean they don’t have other transportation needs.


an easier agenda

by lewis500 on May 14, 2013, no comments

(a post only marginally about transportation)

What is the Republican Party’s focus? Just ask, “Which causes capture the marginal returns on the party’s investments of political capital?” The GOP has often gambled political capital on uncertain drives for policies that help rich people, like Social Security privatization, while ignoring bets that would shrink government but don’t help rich people, like easing occupational licensing.

What would be an easier, yet distinctly conservative, agenda for the party?

Matt Yglesias highlights one alternative (link):

But this debate leaves off the table a third option that ought to be considered more frequently in America—cut spending in order to cut taxes on the masses.

Like maybe instead of subsidized health insurance premiums and college tuition we should give people money. Money can be used to pay college tuition. But it can also be used to pay for a vacation. Colleges would have to make the case that their educational offerings are more valuable than a great vacation. Conservatives, more broadly, could challenge liberals to a real debate about the value proposition of government services.

I agree, but here Matt only talks about the composition of the pie, not its size. I’d go further: Republicans should revel in the supply side economics of the middle class. Middle class labor force participation rates are falling, especially for young people and working-class males.

To date, both Republicans and Democrats have advertised strictly demand side solutions. While Democrats increasingly demand more government hiring (David Brooks opines) and Republicans demand tax cuts for ‘job creators,’ both sides have essentially taken the supply of middle class workers for granted. Republicans have completely forgotten the Reagan-era intellectual vocabulary of elastic labor supply (when top marginal tax rates reached the 70%’s) but preserved the concern for rich people’s well-being.

And in a perverse Telephone Game played over decades, this heuristic of asking “What’s in it for the rich?” has led the party to seek tax increases on the middle class. That’s right, the party of supply side economics wants to raise taxes on the portion of the labor force whose supply is shrinking and cut taxes on that portion composed of workaholics. Mitt Romney complained about the 47 percent. Paul Ryan’s budget plan includes big tax increases on the middle class. Even Louisiana’s Rhode Scholar Governor Bobby Jindal has devastated his approval ratings for a charge at the windmill of raising sales taxes for the benefit of ‘job creators’ in a state famous for poverty and homicide, and his doomed crusade was not a cautionary tale but rather an inspiration to a whole Children’s Crusade of suicidal Republican state leaders (evidence).

This agenda makes sense if you’re running a Randian social justice movement to unburden an overworked elite and punish the free-loading masses, but it just has nothing to do with the size of the economic pie, the size of government, utility, or supply side economics. Some Republicans might find it sobering to contemplate the degree of our degeneracy, but Americans simply do not find the question of whether top rates reach 39.6% or 35% to be the litmus test of the just society. It chills the blood.

a radical whom no one that drinks Milwaukee's Best  would ever support

a Wisconsin radical whom no one that drinks Milwaukee’s Best would ever support; also, an apparently terrible boss

What policies could Republicans champion to raise labor force participation? A bunch of nectarines sagging about a foot off the ground:

  • Payroll tax cuts. If the return on middle class labor rises we will see middle class people supply more labor. Ross Douthat (link): “In an era of mass unemployment, mediocre wage growth and weak mobility from the bottom of the income ladder, it makes no sense to finance our retirement system with a tax that falls directly on wages and hiring and imposes particular burdens on small business and the working class.” Last year Republicans did exactly the opposite, using payroll tax extensions as a hostage (link).
  • Reorient infrastructure spending toward accessibility and away from geographic equity and awesomness, under the advertised logic that easier access to jobs will draw marginals into the labor force. Reihan Salam at National Review (link): “My sense is that the best way to increase accessibility is to focus on implementing peak road-user fees and using the resulting revenue stream to carefully add capacity at bottlenecks, and also to ease local land use regulations that have proven a barrier to increased density in high-productivity regions.”
  • Reorient secondary education toward apprenticeships, which means realism about the utility of lofty academics for many, perhaps the majority, of low-income pupils. More shop class will not cut it. Ed Luce in FT (MR link): “Fifteen per cent of taxi drivers in the US have a degree, up from 1 per cent in 1970. Likewise, 25 per cent of sales clerks are graduates, against 5 per cent in 1970. An astonishing 5 per cent of janitors now have a bachelor’s degree…Germany channels roughly half of all high-school students into the vocational education stream from the age of 16….More than 40 per cent of Germans become apprentices. Only 0.3 per cent of the US labour force does so.”
  • Accelerate the regulatory accommodation of of autonomous vehicles, which will will open up lots of part-time employment to disabled and elderly people.
  • Reduce occupational licensing. Only hardcore libertarians (Institute for Justice) have so much as mentioned occupation licensing, which is a literal prohibition on working, and none of the legislatures that turned solid red in the last cycle have lifted a finger to solve the problem. In an under-reported story, it was absurd licensing that kicked off the Tunisian Revolution (link) and continues to be immolatory problem in that country (link), and Egypt’s unrest is largely due to an economy strangled by a military mafia monopoly (link).
  • Stricter requirements for disability insurance. (This American Life on the massive expansion of disability cases.)
  • Reign in the power of local governments to exclude new residents, just as Republicans have reigned in their powers to raise taxes. Many local regulations on building housing are immigration policies under a boring name, and like any immigration policies they keep people from being matched with great jobs. They could say, “Cities are treating working Americans like illegals!”

bespoke = bebroke?

by lewis500 on May 12, 2013, no comments

Arlington, VA is spending about $1 million each for 24 bus stops: link. Each has room to shield about 15 people, although not very well.

$1 million bus stop; Washington Post

The Post says that federal and state money paid for 80% of the cost. It’s an old story: other people’s money.

Transportation funding might not change. So what could lower the ceteris paribus cost of infrastructure? Clearly, local government will always be sloppy with other people’s money, but a imprudence has different results at Wal-Mart than at Tiffany’s.

Over the past century, Americans have gotten way better at making food, electronics, and cars. On a per-unit-of-input basis, we’ve gotten worse at higher education, basic health care and government construction. What stands out here is that everything bespoke has gotten more expensive and gold-plated.

Economists would point to Baumol’s cost disease, but I think we should really look at technological improvement in rent-seeking. Clearly, over many years of working government projects, a contractor will get better at something profitable. Will it learn to make a given quality of bus stop with the least labor and material? Or will it learn to capture more and larger contracts with the same labor and material? It looks like the latter has been more profitable.

I think the only way ahead is to substitute mass-produced projects for bespoke ones. A giant firm selling alike projects to thousands of governments really might profit from cutting costs and prices. For construction, this model is harder than it is for toys. Construction requires “boots on the ground” and hence local firms and hence monopolistic bidding and rent-seeking.

But given the surplus here is tremendous, I think someone will find a way to take a bite of it with lower costs. By surplus I mean not merely the profits in the construction industry but the immense societal gains of building more and better infrastructure…time saved, lives saved, stress reduced, communities beautified. We are already seeing tiny points of light in highway construction, through a field called “accelerated construction.” Here is a New York Times article about a bridge built out of pre-made parts in Massachusetts. Here is an FHWA site about the practice. I sincerely hope mass production of infrastructure proceeds apace.