With a Russian frigate closing in and a half-dozen U.S. warships within shouting distance, the pirates holding a tanker off Somalia's coast might appear to have no other choice than to wave the white flag.

But that's not how it works in Somalia...

[...]

How can a bunch of criminals from one of the poorest and most wretched countries on Earth face off with some of the world's richest and well-armed superpowers?

"They have enough guns to fight for another 20 years," Ted Dagne, a Somalia analyst in Washington, told The Associated Press. "And there is no way to win a battle when the other side is in a suicidal mind set."

From Gordon Tullock's "The Edge of the Jungle" in Explorations in the Theory of Anarchy (1972, edited by Tullock as well):

Suppose that there is a quantity of meat and two lions, one larger than the other, who want it. If they fight, the larger has a good chance of winning and the smaller, therefore, will not only get nothing to eat but will probably be quite severly injured. Nevertheless, we do observe occasional fights...

[...]

The smaller lion is rationally designed to engage in irrational behavior. The smaller lion imposes upon the larger lion the prospects of physical injury, even though the prospects of physical injury for the smaller lion are much greater. ... This requires that the smaller lion sometimes fight and fight hard.

Here is one survey from last night that casts doubt on the Median Voter Model, although I admit there may be others with better wording or better samples (please, post them in the comments if you know of any). And while the views of the public may be less libertarian than economists, the survey also indicates that the public has libertarian sympathies. Main findings of the survey:

1. 45% opposed and 30% supported this specific modified rescue plan (rather than a hypothetical modified plan as in earlier surveys).

2. Crucially, 47% were worried that the federal government would do too much in reacting to the current crisis (as opposed to 36% saying "too little").

3. In fairness (and confusion, considering the above), the same survey indicates that 44% of Americans think the rescue plan will help the economy, compared with 23% thinking it will hurt the economy (though if you add in the "no impact" group of 19% the two sides are roughly equal).

Yesterday I blogged about a recent Rasmussen survey on the financial market bailouts, in which only 7% of voters supported using tax dollars to keep firms like Lehman Brothers solvent, with 65% of voters favoring bankruptcy. That's an over 9:1 ratio against bailouts. That survey was from early last week.

Of course, there are many different surveys on this issue. Pew has a poll from this week, which apparently the White House is touting, which shows 57% of Americans favor the government "investing billions to try and keep financial institutions and markets secure," with 30% opposed. Or, an almost 2:1 support of bailouts.

An L.A. Times/Bloomberg poll from this week shows 55% of Americans oppose the government using "taxpayers' dollars to rescue ailing private financial firms" and 31% supporting, or slightly less than 2:1 opposing bailouts. The L.A. Times article emphasizes that the wording of the question means the surveys are measuring different things, and this is certainly true. Framing is very important for survey research. But the Mueller Test rests on constraining the mental options for funding a given proposal.

"Investing billions" can be done in any number of ways, such as by raising taxes, shifting funding from other projects, or by deficit finance. "Using taxpayers' dollars" constrains the choice set slightly, but even here we see a big effect. My prediction is that if the choice set is contrained further, e.g., by asking "if your tax bill went up by $2,500 per person in your family," the effect would be larger.

Which level of constraint in question-asking is appropriate? I have attempted to address this issue before, but in the end it is a normative question. My normative answer is that if we really want an accurate reading of public opinion, the actual tradeoffs (as best we can predict them) should be accurately portrayed and constrained in the survey questions.

Only seven percent (7%) of voters think the federal government should use taxpayer funds to keep a large financial institution solvent. Sixty-five percent (65%) say let the company file for bankruptcy.

This paper reports estimates of consumers' preferences for plans to improve food safety. The plans are distinguished based on whether they address the ex ante risk of food borne illness or the ex post effects of the illness. They are also distinguished based on whether they focus on a public good -- reducing risk of illness for all consumers or allowing individual households to reduce their private risks of contracting a food borne pathogen. Based on a National Survey conducted in 2007 using the Knowledge Network internet panel our findings indicate consumers favor ex ante risk reductions and are willing to pay approximately $250 annually to reduce the risk of food borne illness. Moreover, they prefer private to public approaches and would not support efforts to reduce the severity of cases of illness over risk reductions.

In the negative part of Professor Hayek's thesis there is a great deal of truth. It cannot be said too often -- at any rate, it is not being said often enough -- that collectivism is not inherently democratic, but, on the contrary, gives to a tyrannical minority such powers as the Spanish Inquisitors never dreamed of.

Orwell went on to criticize Hayek's favored system, capitalism:

But he does not see, or will not admit, that a return to "free" competition means for the great mass of people a tyranny probably worse, because more irresponsible, than that of the State. The trouble with competitions is that somebody wins them.

The last sentence indicates that Orwell has missed the most important insight about the market: mutually beneficial exchange. In most economic competitions, there are not winners and losers. The game is not zero sum. Hayek was not much help on this point referring to "the game of catallaxy" as "the best way to understand" the benefits of the market system. He tries to reassure us that the game is not zero-sum, but even a great mind like Orwell was trained to think otherwise when the words "game" and "competition" are used.

(Thanks CLR for the pointer)

Actually we are borrowing money from Japan to buy oil from Canada. Doesn't make for as scary a soundbite.

True, China is the number 2 holder of U.S. Government debt, but non-scary countries the U.K., Brazil, and Luxembourg also rank high on the list. Also true, Saudi Arabia is the number 2 exporter of oil to the U.S. (Mexico is a very close third), but we import more oil from Canada than from all Persian Gulf countries combined. Canada, Mexico, and the U.S. Virgin Islands alone account for around 32% of oil imports.

Dr. Thomas DiLorenzo: "Gerry O'Driscoll also endorses the Fannie/Freddie bailout/taxpayer rippoff/moral hazard generator."

Dr. Don Boudreaux: "Gerry O'Driscoll wisely warns against bailing out Fannie Mae and Freddie Mac."

My reading is that Dr. DiLorenzo is being slightly uncharitable, but I will leave that up to you.

Maybe you've guessed, based on my coyness about mentioning the party names, that it is the Democrats who are pushing this pro-market change, while Republicans are in favor of anti-property rights zoning laws. Maybe Jason is on to something.

As the reporter correctly notes (probably thanks to Dr. Horwitz), about $1 of this is due to inflation, but I don't think she quite understands the full force of this fact. That means that a gallon of milk, over the past 10 years, has gone up 31 cents (3 cents per year) more per gallon than if it had increased at the same rate as the average consumer product. Of course, 3 cents here, 3 cents there, it can really add up across the consumer's full budget. But we also need to look at the other side of the budget, income.

The latest personal income data is for the 1st quarter of 2008, at $34,546 per capita. A decade ago, the same figure was $22,753. That is an increase of 51.8%. The "milk index" presented to us in the story went up 46.0% over the same time period. So income is increasing faster than milk prices. These means that the real price of milk has fallen over the past decade; not by a lot, but it certainly hasn't gone up. What is this food crisis everyone is talking about?

I know that conveying economic logic, as well as throwing lots of numbers at people, doesn't always come over well in 5 second blurbs on TV. But it also doesn't seem like it would have been that hard to put up the income data underneath the milk price data, and compute the percent changes.

I hate to sound all cheery and Panglossian about these things. The economic effects of Fed-generated inflation and the somewhat related economic slowdown are certainly real factors for concern. But the world is not falling to pieces.

The letter is signed not only by the CEO of Northwest, but also 11 other major airlines. The letter can be read here. CNN is reporting the story, so I assume it is legit, but it is just so illogical (at least, without a public choice angle). Here is the main piece of evidence offered:

Twenty years ago, 21 percent of oil contracts were purchased by speculators who trade oil on paper with no intention of ever taking delivery. Today, oil speculators purchase 66 percent of all oil futures contracts, and that reflects just the transactions that are known.

Well that solves it, speculation has gone up from 21 percent to 66 percent of oil contracts. Speculators must be the culprit!

Just a few quick questions: twenty years ago, who was buying the other 79 percent of contracts? Why aren't they classified as "speculators"? Are they simply saying that in 1988 spot contracts were the majority of transactions, and now futures contracts are? It's unclear from the letter, and it is also unclear that this trend would be unique to oil markets. I suspect similar trends could be observed for commodities where the real price has fallen in the last 20 years.

The only other recent ACLU post I can find with more than 10 comments is one on mandatory ultrasound laws, which has 24 comments. This makes sense when you realize that post is from June 26, the day of the Heller decision, and the thread was hijacked to discuss the case. That's right, the ACLU did not even blog about what a presidential candidate that has a working relationship with the ACLU called "one of the Supreme Court’s most important rulings on behalf of liberty" until 5 days after the decision (but they did have 5 blog posts on that date).

I would also be very interested to here the ACLU's definition of a "collective right ," and whether gun rights are the only negative liberty falling into this category. 

But I would like to ask a more fundamental question: How do we even know there was an "outbreak"? It's not as if there had been no cases of salmonella last year or the year before. The CDC tells us that every year there are 40,000 reported cases, but those are only the ones that actually get reported and "the actual number of infections may be thirty or more times greater." So there are possibly at least 1.2 million cases per year, or 100,000 per month. This "outbreak" has involved 869 people in about 2 and a half months. This is an outbreak?

Of course, they will tell you that this is a rare strain of salmonella, which only 3 people were reported to have in 2007. And there were apparently common factors found among those that were ill: 1.) "consumption of raw tomatoes" and 2.) "persons who ate at restaurants." Now seriously, how many of you (at least, before the FDA/CDC generated scare) had ate a raw tomato and ate at a restaurant within the past week? I would guess this group of people is something like half of the U.S. population. And a few hundred people got sick.

My guess is that the effect is actually random, but government agencies and the media like to see patterns where none exist. Fooled by randomness. And once the "outbreak" is announced, I'm sure that more people get tested, thus more reported cases. Some of those around 1 million unreported cases are now getting reported. And the false outbreak scare has cost the food industry at least $100 million (I've seen estimates more than double that).

Now Lou Dobbs is one-upping Krugman and calling for Bush's impeachment over this. Seriously? Oh yes, I'm sure if you doubled the FDA and CDC budgets, they could get to the bottom of this, plus prevent all those 1.2 million cases from happening in 2009. Yeah, right. We live in a risky world, much less risky than the past, but still risky. It would be better if we the people, the government, and the media just acknowledged this fact.

First, the "lost tribe" of the Amazon that wasn't so lost (via LRC). They do exist, but the back story was not exactly true:

Far from being unknown, the tribe's existence has been noted since 1910 and the mission to photograph them was undertaken in order to prove that 'uncontacted' tribes still existed in an area endangered by the menace of the logging industry.

The disclosures have been made by the man behind the pictures, José Carlos Meirelles.

Meirelles admitted that the tribe was first known about almost a century ago and that the apparently chance encounter that produced the now famous images was no accident.

Second, the "pregnancy pact" at a Massachusetts high school that might not have been. It appears there is basically no evidence that it existed:

"There was definitely no pact," 17-year-old Lindsey Oliver [one of the pregnant girls] told "Good Morning America" on Tuesday.

Mayor Carolyn Kirk also denied a pact existed after a closed-door meeting with city, school and health leaders on Monday.

Kirk said she and Superintendent Christopher Farmer have been in touch with [Principal] Sullivan [the source of the rumor], and he was "foggy in his memory" about how he came to believe there was a pact.

I would be willing to wager that the retractions got less attention (other than by bloggers) than the original, "shocking" stories.

I have previously speculated that the metric system may be inferior to other measurement systems, because of the possibility of large errors.

The bias also relates to uncertainty regarding coordination of behavior. If a meeting is scheduled to begin at 10:00AM, my experience is that the meeting will not begin at 10. Some people will show up early, thinking the meeting starts at 10 and not wanting to be late. Some will show right at 10. Some will show up late, thinking the meeting starts "around 10."

Speed limits are another coordination problem. If the speed limit is 55, can I be ticketed for going 56? 60? 65? There seem to be vague rules of thumb, I assume occasionally enforced by courts, but it all seems quite arbitrary. Stated simply, the bias is "when numbers are stated in multiples of 5 or 10, the number loses some of its focal point quality."

I have read Nudge, and have lots to say but really do not know where to begin (or where the "review" would eventually go). Overall I think the book is very confused and confusing, above all since the authors are often ambiguous as to whom the advice is directed: individuals, firms, or government (as Mario Rizzo stressed mulitiple times in a discussion with Thaler).  But I do have some "nudges" for the "rounding bias." And just to be clear, these are suggestions for private meeting organizers and owners of private roads (I expect local governments love the revenue generated from the uncertainty).

For starting meetings (or parties), I have occasionally used the trick of starting at "odd" times, such as 3:03PM. Starting 3 minutes past the hour becomes focal (perhaps simply because it is novel), and the repetition of 3 sticks in your mind. This seems to have had some positive effects, but maybe this is just confirmation bias. For speed limits, posting non-round numbers would be a similar nudge, such as this 29 MPH speed limit at the private Valdez Marine Terminal in Alaska (according to this page the Marine Terminal also has speed limits of 11, 16, 21, and 26).

Reading the article itself should raise some red flags. First is the statement that "The research proves that destructive ability of earthquakes on Earth increases alarmingly fast and that this trend is set to continue, unless the problem of 'global warming' is comprehensively and urgently addressed." Note the use of the word "prove." Is there some direct connection, even a theoretical one, between global warming and earthquakes? The article never says, but claims that the author has "proved" this fact and "proved" that "the global annual energy of earthquakes on Earth began increasing very fast since 1990." The article also points out that the increase in seismic activity "is not theoretical but that it is an Observable Fact." Yes, Observable Fact, capitalized. (In the "published" article it is labeled "Observed Reality.")

The work is all done by Dr. Tom Chalko, an "Australian scientist." Although the article doesn't say so, a quick Google search shows that the article was published in the NU Journal of Discovery. "NU" stands for Natural University, which, their website tells us, is "nowhere in particular and yet everywhere on Earth." So basically, it is a website, which publishes this "journal" on a very occasional basis (this article is the fifth published since 2001). Dr. Chalko is also on the editorial board. Apparently, Chalko has a PhD in Laser Holography ("The Science of Light") and also writes books on new age healing. None of this is mentioned in the article, and no other scientists or anyone else appears to have been consulted for the CBS piece.

So obviously, this guy is not a geologist or climatologist. But so what? Hasn't he, as he states, compiled all the earthquake data from USGS between 1973 and 2007? And didn't he find a strong change in trend 20 years ago? What of that Observable Fact that he "proved"? Okay, so there is no causative relationship proved. So what?

Take a look at the graphic accompanying the CBS story. Although there are no labels, I assume time is on the x-axis and some measure of earthquake activity on the y-axis. Notice that big jump about one-third in? It seems a bit big for a one-year change, with two distinct time trends before and after. Any idea what could have caused that? Well, let's ask USGS. First they tell us:

Although it may seem that we are having more earthquakes, earthquakes of magnitude 7.0 or greater have remained fairly constant throughout this century and, according to our records, have actually seemed to decrease in recent years.

Hmm, that's interesting. Anything that might have caused a change in the data in the last 20 years, as CBS and Dr. Chalko report?

A partial explanation [for the perception of more earthquakes] may lie in the fact that in the last twenty years, we have definitely had an increase in the number of earthquakes we have been able to locate each year. This is because of the tremendous increase in the number of seismograph stations in the world and the many improvements in global communications. [My emphasis]

Oh, well it all makes sense now. But how many people that read the headline on CBS, or MSNBC, will do any fact checking? Of course, none of this means that climate change is not happening, and that there will not be some negative consequences from it. But journalists' willingness to publish the results from any "study" without doing the most basic fact checking about the data or whether there was actually causation demonstrated is definitely a worrying trend. It also seems to be correlated with increasing CO₂ levels.

Update (6/19 at 5:02pm): The story was still the lead CBS's SciTech page mid-morning, but now it appears to be gone -- down the memory hole with no explanation. But don't worry, the link to MSNBC story still works and it is also on Yahoo (and the CBS graphic which was prominently displayed on their SciTech page hasn't been erased yet). The MSNBC and Yahoo stories are actually press releases with the source listed as "Dr Tom Chalko." The CBS story was the press release cut-and-pasted to appear like a real story, but the link to Chalko's article and his byline were removed. Good for CBS if they removed pseudo-science (or rather, non-science), but shouldn't they run a correction as the lead story for the next 12 hours or so?

Update (6/20): The CBS story is now in Google cache. There is apparently disagreement between CBS and the AP over the origin of the story (rather than just the press release). Climate Audit is also following the story.

The following statement by Krugman got me thinking:

One amazing decision [by the USDA] came in 2004, when a Kansas producer asked for permission to test its own cows [for mad cow disease], so that it could resume exports to Japan. You might have expected the Bush administration to applaud this example of self-regulation. But permission was denied, because other beef producers feared consumer demands that they follow suit. 

Several things are curious about this piece of evidence cited by Krugman. Above all, it seems quite odd to use this as a counter to Milton Friedman's claim that "private companies would avoid taking risks with public health to safeguard their reputations and to avoid damaging class-action lawsuits." It seems they were trying to self-regulate. What stopped them? Not the free market, but, as Krugman accurately notes, the USDA itself, which was headed by "a former food-industry lobbyist." This is market failure? Is that the best you got? If this is evidence that the Bush administration is more about corporatism than capitalism, I'm all for it, but Krugman doesn't quite pitch it that way.

There are interesting historical parallels a century before, also related to testing of cattle. Salmonella poisoning in tomatoes on my cheeseburger was not the concern, but rather a much more deadly disease: tuberculosis. The year is 1908, and Chicago (a major cattle-processing center) has just enacted an ordinance to limit the spread of tuberculosis, requiring testing of cattle and pasteurization of milk. As reported by Olmstead and Rhode, competitors in the nearby

Elgin area used their influence to capture the Illinois legislature. Their main ally was long-time Speaker of the House, Edward Shurtleff. ... After helping these interests capture the committee investigating milk health measures, Shurtleff spearheaded the passage of the 1911 state law prohibiting cities from requiring tuberculin testing. [My emphasis]

But again, oddly, Olmstead and Rhode use this as a piece of evidence in support of market failure. Their argument is actually more sophisticated than Krugman's: differing regulatory regimes across localities and states increased information asymmetry problems, allowing unscrupulous cattle owners to profit while endangering consumer and cattle health. The asymmetry problems would seem to be even larger with more localized regulation, that is, at the level of the individual cattle owner. But this still seems like a better example of regulatory capture than market failure. How this information problem was eventually solved is quite interesting, but I will save the full story for another day. Here is a brief preview.

Olmstead and Rhode provide an account in a companion article to the one linked above, and they emphasize a cooperative federal-state program to test cattle for bovine tuberculosis. It is interesting to note though, the absence of any mandatory federal regulations, either for testing of cattle or pasteurization of milk. The testing program they describe was voluntary, and, believe it or not, there was never a federal regulation mandating pasteurization of milk, even to this day (yes, really, I mean it). The U.S. Public Health Service has long issued model legislation for states, but it was never mandated (recall the claim above that state and local regulations could not work). I have a radically different explanation for the historically important process of eliminating tuberculosis, but am still in the process of collecting all the relevant facts, so you will just have to wait.

Update: Dr. Tabarrok passes along a post of his from 2004 regarding the mad cow self-regulation ban that Krugman referenced. He had the regulatory capture angle nailed as the story broke, and correctly notes that USDA has had the "power to decide minimum and maximum testing" since 1913. I should further note that my above summary of his position as "Tabarrok disagrees [that we are 'back in The Jungle']" is not precisely correct. He stated that there is "no evidence whatsoever" for making such a claim, a less sweeping claim than Krugman's (and the correct one, in my judgment).

Earlier this week I saw a similar poll on CNN, but it was actually presented as a tradeoff (see the story and the actual questions). The results are similar to the survey Caplan points to: people are more worried about long lines than high prices. But, as the Gallup poll shows, when you don't present it as a tradeoff, you get huge majorities in favor of dumb policies. We happened to be discussing Caplan's work in class on Tuesday, and I mentioned this CNN poll to the students as evidence that the public is not really that irrational, just subject to some framing effects.

Of course, if the second question about rationing had not been asked, it's possible that some politician could trumpet this survey as a reason for imposing price controls. Maybe someone will anyway, but at least in this case it probably won't get put into effect. But in many cases, the majority's response to non-tradeoff questions are often enacted as policy (and used as support for the policy, and then later social scientists claim that democracies "give the people what they want"). 

I also note that this is already being called "the Mueller Effect" in the comments at Caplan's blog (well, by one commenter so far). If Stigler's Law holds though, this may be known to history at "the Horpedahl Effect" (ugh).

Frightening historical note: support for price controls is much higher today than immediately pre-Nixon price controls (at least when questions are asked in a non-tradeoff manner). I need to head to the library and check the Gallup archives, but this JEH article claims that in June 1971 (two months before price controls), 50% of Americans favored wage-and-price freezes, compared with 53% today! Caveat: those were for general controls, not just gasoline prices. It is also fascinating to witness the dramatic spike in support among Republican elites following Nixon's announcement of the policy. What were their "real" preferences?

This summer I am teaching Public Choice, and we have spent a little time working through the basic calculus of voting model. As, I hope, my students can now articulate, in order for your vote to matter in an instrumental sense it has to break a tie (or force a tie). We also spent plenty of time discussing non-instrumental reasons for voting, but as far as influencing an election, unless the margin is one vote your can not influence the outcome.

Okay, so there is a Republican Congressional primary in Virginia tomorrow, and the candidates are Amit Singh and Mark Ellmore [edit: in the 8th District]. Some mud has already been thrown in public and through private emails, but it's now really heating up, as they say. Singh is somewhat of a newcomer/outsider to the party, and it appears that there is some fear that he may present a challenge and be very important. In fact, the Ellmore campaign thinks that Singh is a very important guy, so important that he personally is allowed to cast 4,655 ballots.

In an email I received from they Ellmore campaign today, they claim that "thanks to Amit Singh, Democrats control the Senate and no judges are confirmed." You see, in 2006 Republican George Allen lost his Senate race by 4,655 votes. Amit Singh did not vote for George Allen. But as far as I know, Singh only possesses one vote. So how could his vote have tipped the election? It's unclear. (Note: Singh was not a public figure in 2006, so announcing who he was voting for could not have influenced anyone to vote for Allen.)

Is this just a single typo in the email? No. The Ellmore campaign goes on to ask "Why wouldn't Amit Singh support George Allen to keep the Senate in Republican control?" Again, Singh's support wouldn't have changed one thing about that election. And further: "Amit Singh helped to sabotage George Allen's Senate race in 2006 that gave control of the Senate to the Hillary Clinton, Barack Obama, and John Kerry Democrats." In conclusion, Republicans don't understand how elections work, in a very fundamental sense.

The Ellmore campaign email makes other accusations against Singh, including linking him to Ron Paul. Oh, the horror! But, oops, seems like Ellmore has kind things to say about Paul as well.

Yah for democracy.