Month: June 2004

Retail gas pricing

“Zone pricing,” is what the gasoline industry calls price discrimination – wholesalers charge less to stations in zones with many stations and more to stations in zones with few stations. Legislators take one look at “zone pricing” and assume that they can lower prices by requiring wholesalers to sell to everyone at the same, “non-discriminatory,” price.

Let’s assume that the legislators are succesful in lowering prices in high-price zones. Do you think that the retailers in these zones will pass the price reductions on to their customers? Of course, not. The reason prices are high in zones with few stations is that stations in these zones have greater market power. It’s this fundamental fact that makes prices higher in these zones – all price discrimination at the wholesale level does is change who gets the profits. With price discrimination the wholesalers get the profits, with uniform pricing the retailers get the profits.

So consumers in high-price zones don’t benefit from ending zone pricing but what about consumers in the low-price areas? If forced to charge a single price do you think that wholesalers will charge the lowest of their zone prices? Of course not – they will charge an average of their zone prices. As a result, consumers in highly competitive zones will face higher prices under uniform pricing.

Thus uniform pricing makes retailers better off at the expense of wholesalers and consumers.

Not sure if the analysis is right? Our colleague, Bart Wilson, is the author (with Cary Deck) of an excellent new paper on retail gasoline pricing. Wilson and Deck setup an experimental market with retailers, refiners and gasoline customers (the latter are computer agents) and find exactly these results. The Wilson and Deck paper is powerful evidence because an experimental market is a real market – it just happens to be a real market under the careful control of an experimenter.

Wilson and Deck also analyze divorcement (requiring wholesalers to divest themselves of retail stations – very bad for consumers because of the double monopoly problem) and the puzzling phenomena called “rockets and feathers” – the tendency of prices to rise faster with costs than to fall with costs. The full paper is here but they have also published a very good “executive summary” in Regulation, one of my favourite journals.

The largest known prime number

We just found a new one, and it has seven million digits. Here is the bottom line:

Mersenne primes are an especially rare type that take the form 2^p-1, where p is also a prime number. They are named after a 17th Century French monk who first came up with an important conjecture about which values of p would yield a prime. The new number can be represented as 2^(24,036,583)-1. It is the 41st Mersenne prime to have been found.

Here is the full story.

Note also that the number was found by a consortium of private computers, designed to exchange spare computing power:

GIMPS volunteers download a piece of software that runs in the background on their computer. A central server distributes different prime number candidates to each machine, which use spare processing power to test whether it is a genuine prime or not.

Here is more information, plus how to volunteer. Or perhaps you would prefer to search for alien lifeforms.

Blogging from Poland

My time in Paris is over and I’ve arrived safely in Poland. I’ve found exactly what I’d expected. It is a modestly bustling economy, but not setting the world on fire. The slowdown of the last two years appears to be over. Corruption is low for a transition economy, social capital appears to be high, and there is an emerging middle class. The government has serious fiscal problems, but then again they probably should be running deficits, though not at 51 percent of gdp as they are doing. Overall it is hard to see them not turning the corner. If you would like an illustrative lesson as to how the world is a better place than it was twenty years ago, Poland is exhibit B after China as exhibit A.

I might add that Warsaw has excellent food, arguably the best in Eastern Europe after Budapest.

I’m blogging just across from the famous statue of Copernicus. Did you know that he offered one of the earliest statements of the Quantity Theory of Money?

Media Bias

Discussions of media bias are a sort of political Rorschach test – what you see depends on what you already believe. Despite this, I think the evidence is consistent on a few points.

1. Journalists tend to be more liberal than the average American. A recent Pew poll confirms previous studies of journalists. In this study, journalists did not describe themselves as liberal (which previous studies did find to be very common) but their attitudes tended to be very liberal. For example, 51% of Americans thought homosexuality should be accepted, while 88% of journalists in the national media thought so.

2. Very few journalists describe themselves as conservative. In the same Pew poll, 33% of Americans described themselves as conservative while 7% of journalists said they were conservative. Previous studies have found that journalists overwhelmingly vote for Democrats, except during the 1950’s – probably because Eisenhower was the most liberal of recent GOP presidents, when compared to the competition.

3. Analysis of American news consistently shows biases. In the 1990’s, for example, numerous studies found that it was relatively rare for ex-GOP leader Newt Gingrich to get positive press coverage. I found it interesting that one study found Bob Dole’s coverage was about 50% positive/50% negative. I wonder what it was about the future Viagra spokesman that made him lovable by the media?

Those who see the media as dominated by conservatives probably focus on Rush Limbaugh, the Fox network and some other high profile conservatives. They could also justly point out that media is big business, and owners probably favor legislation that benefits them (a la Rupert Murdoch), even though some media entrepreneurs such as Ted Turner are quite liberal.

So here’s my analysis of media bias: the journalism profession, on the average, is quite liberal and they also believe in the ethic of objective reporting. This creates a situation where much reporting is probably informed by liberal values, but presented in a “manner of fact” way. As a result, there are a lot of angry conservative readers and viewers who are flustered by this type of reporting.

Since the whole journalism profession is pretty much liberal, it’s probably hard for more conservative owners to impose their will on the newsroom – although they try quite hard sometimes. The end result – an untapped market of conservative viewers that can be catered to by the likes of Limbaugh and the Fox network. The Al Frankens of the world probably focus on the high-profile conservatives (like Rush) and ignore the average news reporter, while the Limbaugh’s obsess over the New York Times – a model for much of the journalism profession – and ignore successful niche players such as Fox.

How to choose a restaurant

Where oh where should we eat?

Perhaps the common social choice problem that any of us face in practice is when we find ourselves in a group that must choose one restaurant at which all of us will eat. We propose a method where, similar to the I-choose-you-cut rule for dividing a cake, individuals in the group take turns restricting the set of choices for the group. Specifically, under our method the first person restricts the set of restaurants to a certain number the second person restricts the set to a smaller number and so on until the last person in the group selects one restaurant. We derive a formula for choosing these numbers such that – under a natural assumption about individual preferences.the probability that the group will choose any individual’s favorite restaurant is equal for each individual.

That’s from an interesting recent working paper by Tim Groseclose and Jeff Milyo.

My recipe is simpler: “Go Where I Say,” though I am less likely to use it in repeated game settings. And I am willing to defer to Randall Kroszner and John Nye, among others.

I do wonder why collective choices are not made in more efficient ways. Overall there is not enough deference to expertise and too much interest in finding a “lowest common denominator” of taste within the group. The real problem is to allow those who know to exert their influence, but without appearing like bullies. Yes I know the group cares more about harmony but that is precisely the problem: outlier tastes in the group end up frustrated. Maybe we should sever food decisions from all others, if this is possible. Then people could cede to the food expert, but without fear of future bullying in all other areas. As is sometimes the case in politics, the question is not “what is the proper social welfare function?” but rather “how do we get the right thing done?”

Who has the highest tax burden?

Number one is France, Belgium and Sweden are right behind. I am shocked to see China come in fourth, and to see the U.S. (New York State at least) beat out Canada.

The relevant metrics do not measure tax burden fully. Instead the rankings measure a weighted average of various marginal rates; average tax rates do not seem to enter into the calculation. Nor do they seem to consider how much of the tax is actually paid, or what you get for your taxes.

Since 2000, the tax burden, as measured in this fashion, has dropped in 22 countries, risen in 13, and held steady in 15.

The least taxed country is now United Arab Emirates, pushing long-time winner Hong Kong into second place. Next in line for low marginal rates are Singapore, India, South Africa, and Indonesia, an odd mix of countries. Then comes Ireland.

Here are extensive links to the data.

Note that economic theory says that marginal rates should be of primary importance. But often in the data it is the average tax rate that matters. Why is this? Could it be that liquidity effects are paramount? High average rates then suck away cash and end up affecting decisions at the margin. Or consider another channel. The tax system is very complex and full of loopholes and deductions. The average rate might in fact be the best measure of the true marginal rate we have.

Postscript: I know this is a market-oriented economics blog, but hey, I am also a contrarian at heart. I cannot help asking: which countries would you rather live in? Visit?

One view is “France is great, but it would be even better with lower taxes.” Another is: “France is great, and for reasons which bring both its wonders and its high taxes.” I am perpetually torn between these two views, often co-blogger Alex tries to push me more toward the former.

Punitive damages

California Governor Arnold Schwarzenegger is proposing to tax punitive awards at a 75% rate. The idea has some merit, punitive awards are intended to punish the defendant but to serve that purpose it is not necessary that they should flow to the plaintiff. Moreover, punitive awards often have more to do with the defendant’s deep pockets or other arbitrary factors like whether the plaintiff is headquartered in or out-of-state than with proof of malice. Giving punitive awards to the plaintiff, therefore, may over incentivize plaintiff’s and their lawyers (under the plan contingency fees would be calculated on the compensatory portion of the award only).

Schwarzenegger, however, is putting forward the proposal as a revenue source and calculating revenues under the assumption that punitive damages will not decline after the tax. On the contrary, as with any tax we can expect fewer punitive damages after the tax than before. In this case, that’s a benefit of the tax.

But tort reformers shouldn’t be too happy, however, because there are ways around the tax. In particular, the tax will encourage settlements. Suppose the plaintiff and defendant estimate that there is a good chance of a 10 million dollar punitive award only 2.5 million of which will go the plaintiff. See the gains from trade? The plaintiff and defendant can agree to settle at say $5 million, making both of them better off. In this way, punitive awards decline but “compensatory” awards increase. Increasing the incentive to settle is not necessarily a bad idea but it doesn’t help revenues.

Of ocourse, the Governor is being disingenuous when he claims this is tax policy not tort reform. Just like bad ideas, good ideas usually don’t get passed on the merits but need some otherwise irrelevant impetus (like a big budget deficit).

Addendum: Glen Whitman says we should burn the money – which is not as crazy as you might think!