Ragout

The Baker Plan to Eliminate Private Drug Development

2005-10-18T21:45:00.000-04:00

Personally, I'm very sympathetic to the Baker plan to have the government get more involved the later stages of drug development. Progressive economist Dean Baker wants the government to start funding clinical trials and develop drugs that would then be released into the public domain. Although the government spends a lot on pharmaceutical research now, they mainly fund basic research and leave the private sector to take the drugs to market. I'd be happy to throw in a billion dollars of taxpayer money to fund Baker's plan.

I think that eventually, it would probably succeed well enough to justify spending 10 or 20 billion a year. But Baker seems keen to go whole hog, and eliminate private drug research and patents as soon as possible. And that scares me.

Patents have worked tremendously well for hundreds of years. The track record of Baker's system is what exactly? That it would probably succeed just isn't good enough to justify anything more than evolutionary change. The stakes are, after all, life and death. And keep in mind that the promised payoff from the Baker plan is a few hundred bucks a year per person.

And it's easy to think of reasons to be cautious about the Baker plan for eliminating all private drug research. The Christian right is currently fighting the "morning after" pill and even a vaccine that prevents cervical cancer, because they believe both will encourage sexual activity among women. If government were the only institution funding drug development, religious fundamentalists would have a much easier time killing drugs they don't like.

Similarly, if Baker or Marcia Angell were in charge, I'd imagine they'd stop research into the "me too" drugs they're constantly criticizing. Which might leave us with drugs like Vioxx (which causes heart attacks) and not the "me too" alternative Celebrex, which doesn't seem to have the same problems.

So, in the end, I think Baker has a good idea, but he tries to make it sound as radical as possible. Consequently his plan sounds a whole lot worse than it would if he were actually trying to pitch his idea and convince people. I really don't understand what Baker's game is here.

Reforming Drug Research: First Do No Harm

2005-10-17T22:34:00.000-04:00

Presswatching economist Dean Baker has an interesting post over at MaxSpeak, proposing that the government take over financing all drug research. Baker points out that the government already provides most funds for basic research. He thinks the feds could also do at least as good a job as the private pharmaceutical companies that currently fund drug development with the support of profit-raising patents:

Why shouldn't we believe that if we doubled [federal drug research spending], to replace the $25 billion that the drug industry claims to spend on drug research (two-thirds of which goes to research copycat drugs) that we would end up with at least as good progress in developing drugs as what we have at present?

It seems to me the current system works awfully well at producing lots of amazing new drugs. I think we ought to place a high priority on not messing up a good thing. "First, do no harm," as doctors like to say.

So why not start "small" at, say, a billion dollars a year? That would be enough to start developing at least a dozen new drugs, a few of which would eventually prove beneficial. If it looks like we're getting somewhere, we can up the funding in a few years. But it's crazy to start by eliminating patents, which I think is what Baker has in mind. Certainly that's his long-term goal, and he places great emphasis on the evils of patents in his writing.

Really, there's no choice but to start small. To put Baker's idea into action, we'd have to set up a nonprofit or a new government agency, and let them try their hand at picking drugs and running drug trials to test for safety and efficacy. I don't know of any existing institution (other than the drug companies) that can do this.

Baker and lefty Congressman Dennis Kucinich suggest that the NIH and universities could replace the pharmaceutical companies. But drug development is mostly a hard, unglamorous slog. It's not the kind of thing academic researchers want to do. And it involves a large administrativeapparatuss to recruit thousands of patients into drug trials and collect data. Again, the NIH and Universities just aren't set up to do this kind of thing.

So there's no way to start spending $25 billion a year tomorrow, or even in the next decade. We'd have to build entirely new institutions.

So instead of trying to score rhetorical points by proposing to turn the existing system upside down, why try something modest? There will be plenty of political points to be scored when something so obviously reasonable is killed by the right and the drug companies.

The Washington Consensus: "Ideology, not Analysis"

2005-09-13T23:15:00.000-04:00

Brad DeLong discusses a fascinating debate between a Harvard economist and a Yale economist about the sources of India's remarkable growth. A common view is that India's growth acceleration is due to the neoliberal reforms that began in 1991: free trade, lowering tariffs, deregulating business, rolling back the "license raj," and so on.

A typical purveyor of this view is Thomas Friedman, according to Harvard's Dani Rodrik and the IMF's Arvind Subramanian. They quote a Friedman interview with a prominent Indian industrialist, who described "the cumbersome bureaucratic rules and pervasive state ownership that suffocated the Indian private sector," until the 1991 reforms, when "Our Berlin Wall fell."

It was like unleashing a caged tiger. Trade controls were abolished. We were always at 3 percent growth, the so-called Hindu rate of growth--slow, cautious, and conservative. To make [better returns], you had to go to America. Well, three years later [after the 1991 reforms] we were at 7 percent rate of growth.

As Rodrik and Subramanian point out in a compelling paper, the problem with this story is that India's growth in the 1980s wasn't at all slow. In fact, it was just as rapid as it's been since 1991. India's growth miracle began around 1980, long before the neoliberal "Washington consensus" reforms.

The Yale economist, critiquing the Harvard study, doesn't deny that the growth takeoff began long before the 1991 reforms, but argues that India's growth in the 1980s wasn't sustainable because it was driven by an expansion of the public sector. I think the Harvard side rebutted the "growth due to unsustainable deficits" theory in the original paper*, and I won't try to summarize the whole debate.

But I was particularly struck by just how weak some of the "unsustainable growth" arguments really are, despite having apparently gained widespread acceptance, from Friedman's spot on the Times' op-ed page to the hallowed halls of Yale.

The Yale guy says that India's expansion was "fueled by" the expansion of government, including "real wage expansion in the public sector." If you read his footnote, you learn that public sector wages grew a factor of 3.45, while prices grew by a factor of 2.37 during the 1980s. Do some division and take a 10th root, and you find that real government wages grew by 3.8% per year. Is that fast or slow? Well, in the economy as a whole, labor productivity growth was also just about 3.8% per year in the 1980s! (Rodrik and Subramanian Table 1).

So India's bureaucrats were just keeping up with their private sector counterparts! Wage increases just sufficient to keep pace with economy-wide productivity gains are hardly evidence of expanding government. So this Yale argument seems truly lame.**

With academic understatement, Rodrik and Subramanian nicely sum up the problem with their critics:

We suspect that there is a tendency to dismiss the growth of the 1980s because it makes the subsequent reforms less impressive ("since the system was not reformed, any growth that came out should have been unsustainable--i.e., bad growth"). But that would be just ideology, not analysis.

---------------------------

* Rodrik and Subramanian point out that the only way that "Keynesianism run amok" could cause the India's huge surge in productivity during the 1980s is by putting unused resources to work: raising capacity utilization in factors and lowering unemployment. But that didn't happen, they argue, at least not on a large enough scale to explain much of the productivity acceleration.

** This is just econ 101 stuff. The "law of one price" predicts that wages will tend to rise at the rate of economy-wide productivity. If private sector productivity and wages are rising, public sector wages have to rise too, even if government productivity isn't increasing. If public sector wages don't rise, bureaucrats will quit their government jobs and get higher-wage jobs in the private sector, as middle management in a factory, say. So basic economics predicts that public sector wages will tend to keep up with private sector wages, assuming the government doesn't want to see all its best workers leave.

New Orleans Evacuees: "Sort of Scary"

2005-09-07T00:56:00.000-04:00

How to explain the behavior of the authorities?

Why did FEMA turn away the Red Cross and many others who offered to help? Why were the Superdome and Convention Center locked down, and the evacuees forbidden to leave? "Worse than a prison," said one Superdome occupant. "We've virtually made them prisoners," said the Sheriff of neighboring Jefferson Parish, who had ordered his forces to surround a group of evacuees in an open field.

Why were people packed onto busses and planes without being told whether destination was Baton Rouge, Houston, or Utah?

Why were friends and relatives forbidden to enter the city and transport their loved ones to safety? Why were bridges out of town blockaded by armed police? I've seen relatively little about these hard-hearted practices in the mainstream press: the last two links are to horrifying first-hand accounts.

Presumably the authorities placed their first priority on preventing looting in the French Quarter or the wealthy neighborhoods of Jefferson Parish, both a short walk away from the Convention Center and Superdome. One National Guard General described his units' role as conducting "combat operations" against an "insurgency."

Preventing looting is a worthy goal, to be sure. But not a goal that justifies treating tens of thousands of people like animals. And not a goal worth people's lives.

How could the authorities decide to treat everyone in New Orleans as a potential looter, rioter, insurgent? The simplest explanation seems to be that they view poor black people with tremendous contempt and fear. An attitude nicely summed up by Barbara Bush's comment on the evacuees in Houston, "what I'm hearing, which is sort of scary, is they all want to stay in Texas."

Compare and Contrast

2005-08-31T22:13:00.000-04:00

A while back, Jane Galt quoted a passage that caught her eye from a National Review article about Germany:

Indeed, to Schroeder's eye, there is hardly anything worth cutting, right down to the generous dental benefits. "“I do not want to return to an era when you can judge someone'’s wealth by the state of their teeth," ”he observed.

Galt thinks the poor get pretty fine dental care in the U.S., pontificating:

The reason that I comment on this is that one thing you can't tell people's wealth by, in the dog-eat-dog dystopia that is America, is their teeth. Their sports gear, their vacations, their choice of dinner spot, yes, but not their teeth, at least not where I am.

Instapundit Glenn Reynolds found the whole thing pretty funny. Galt's post caught my eye too, because it seemed pretty callous and uniformed. Out of sight, out of mind, say Galt and Reynolds. Last week, the unblinking Malcolm Gladwell supplied the perfect rebuttal in a fascinating New Yorker article about health care in America.

Several years ago, two Harvard researchers, Susan Starr Sered and Rushika Fernandopulle, set out to interview people without health-care coverage for a book they were writing, "“Uninsured in America."” They talked to as many kinds of people as they could find, collecting stories of untreated depression and struggling single mothers and chronically injured laborers--and the most common complaint they heard was about teeth. Gina, a hairdresser in Idaho, whose husband worked as a freight manager at a chain store, had "a peculiar mannerism of keeping her mouth closed even when speaking." It turned out that she hadn't been able to afford dental care for three years, and one of her front teeth was rotting. Daniel, a construction worker, pulled out his bad teeth with pliers. Then, there was Loretta, who worked nights at a university research center in Mississippi, and was missing most of her teeth. "They'll break off after a while, and then you just grab a hold of them, and they work their way out,"” she explained to Sered and Fernandopulle. "“It hurts so bad, because the tooth aches. Then it'’s a relief just to get it out of there. The hole closes up itself anyway. So it'’s so much better."”

....

If your teeth are bad, you'’re not going to get a job as a receptionist, say, or a cashier. You'’re going to be put in the back somewhere, far from the public eye. What Loretta, Gina, and Daniel understand, the two authors tell us, is that bad teeth have come to be seen as a marker of "“poor parenting, low educational achievement and slow or faulty intellectual development."” They are an outward marker of caste. "“Almost every time we asked interviewees what their first priority would be if the president established universal health coverage tomorrow,"” Sered and Fernandopulle write, "“the immediate answer was '‘my teeth.'"”

Like Galt and Insty, I also rarely (but not never) see people with really bad teeth. No doubt like them, the poorest people I regularly come into contact with are behind the cash register or sweeping the floors. I don't ride the early-morning bus with home health care aides or spend much time in slaughterhouses with poultry workers. But unlike Galt and Reynolds, I don't assume because I rarely see it in my day to day life that a problem doesn't exist.

Vioxx: Excess Deaths, Excess Liability?

2005-08-29T22:06:00.000-04:00

In the reporting about Vioxx, I've seen a lot of citations of epidemiological estimates of the excess deaths and heart attacks, which range from 35,000 to 160,000. I've also seen a lot of estimates of Merck's potential liability, which are said to range from a few billion dollars up to $50 billion. But I haven't seen any estimates of the figures you'd actually need to estimate the potential liability. It seems to me that the relevant figures are simply deaths, not "excess" deaths.

"Excess" deaths are something of a statistician's fiction. It's not like there's any clear method to distinguish someone killed by a Vioxx-induced heart attack from a Vioxx user who would have had a heart attack anyway. To see what I mean, consider the estimates of David Graham, the FDA whistleblower-scientist.

Graham finds that about Vioxx users suffered heart attacks (fatal and not) at a rate of about 8 per 1000 person-years. Those using other anti-inflammatory drugs had a rate of 5 per 1000 person-years. The difference, 3 per 1000, is an estimate of excess heart attacks (it's not actually the estimate Graham uses, but it's close).

Since, Vioxx was used for about 10 million person-years, we get:
80,000 heart attacks among Vioxx users
50,000 heart attacks among similar patients using other drugs
which implies 30,000 excess heart attacks.

The final number (30,000) is what the epidemiologists focus on, and what gets reported in the newspapers. But the total number of heart attacks (80,000) is the relevant count of legitimate lawsuits. It's not like the coroner can examine the blood clot clogging somebody's arteries and look for Vioxx molecules. Although randomized trials have shown that Vioxx raises the likelihood of heart attacks, they don't tell us which particular Vioxx user died because of the drug, and which ones would have died anyway. So Merck's potential liability is awfully big (I'm shorting their stock).

From the point of view of economic efficiency, I think Merck is going to end up paying too much, even leaving aside any bogus lawsuits that are filed. Merck ought to be liable for the excess heart attacks and other diseases they caused, not the illness that would have occurred if Vioxx had never been invented. But even though only about a third of the heart attacks suffered by Vioxx users were actually caused by Vioxx, there's no way to tell which are which. So Merck will probably end up paying for all of them.

I don't see any way around this, except for dividing the court awards by three, which might not be such a bad idea.

Vioxx Verdict: AEI Cries "Junk Science"

2005-08-26T00:35:00.000-04:00

Predictably, conservative "tort reformers" are now proclaiming that the Vioxx verdict -- finding the makers of the popular painkiller liable for the death of a patient -- is nothing but "junk science." It's testimony to the power of a conventional narrative that anyone can make this claim with a straight face.

As I blogged a couple days ago, there's pretty overwhelming evidence -- including two large, randomized trials (the VIGOR and APPROVE studies) -- that Vioxx causes heart attacks. Even a layperson can easily verify the medical consensus by reading editorials in the most prestigious medical journals, which express no doubt.

But the American Enterprise Institute's John Calfee claims to know better than the medical journal editors, writing, "junk science now threatens to reign supreme in drug litigation."

The lawyers had to surmount the views of FDA and Canadian expert panels that Vioxx was safe enough to return to the market; evidence that, to the extent that Vioxx was dangerous, it wasn't necessarily any more dangerous than other drugs; and the inconvenient fact that the deceased in the case had died of heart arrhythmia, a cardiac problem not associated with Vioxx.

But none of Calfee's claims are true. Yes, the FDA expert panel did vote 17-15 to return Vioxx to the market, but

"Many of the panel members who were among the narrow majorities approving continued marketing of Bextra and Vioxx did so only with the stipulation that severe restrictions be imposed on their uses," according to one panel member, who added that "he expected that the uses of the drugs would be confined to very limited patient populations."

The transcript of the meeting is online, so it can be easily verified that this is an accurate description of the discussion. Vioxx does have some advantages over other drugs, but the experts say its dangers mean that it should be used at low doses, for limited times, on patients with a low risk of heart disease. In other words, it should be a niche drug, not a multi-billion dollar blockbuster drug.

Yes, the FDA, possibly being overcautious, suggested that alternative drugs like Advil and Aleve might be just as dangerous as Vioxx, but this is just an educated guess, not something based on strong evidence. As Calfee himself points out, "Advil and older prescription and over-the-counter arthritis treatments...[have] not been subjected to long-term clinical trials like the one that seemed to reveal heart problems with Vioxx." If you look at the FDA panel transcript, you'll see that many panel members suggested that Vioxx should only be used after the over-the-counter remedies have failed.

Yes, the doctor who performed the autopsy did find the cause of death to be arrhythmia rather than a heart attack. But that same doctor ended up testifying for the plaintiff that it was a heart attack that triggered the arrhythmia! And University of Michigan pharmacology professor Benedict Lucchesi, "who has studied heart arrhythmias for more than four decades," testified that Vioxx "contributed significantly" to the death.

In other words, Calfee and the "tort reformers" think the views of a pharmacology professor at a top university, the FDA advisory panel, and the editors of major medical journals are "junk science." With the whole medical establishment in the wrong, it's hard to imagine where we can find any "sound science." At the American Tort Reform Association and the AEI?

The Evidence Against Vioxx

2005-08-24T01:41:00.000-04:00

Blogging UCSD economist James Hamilton waded into epidemiology yesterday but appears to have been dragged under by the libertarian Reason magazine. Hamilton devotes most of his post to arguing that "studies suggest but can't prove" that Vioxx causes heart attacks. Hamilton is concerned about a recent jury verdict that found manufacturer Merck liable for the heart attack death of a Vioxx user.

Before going into the specifics, it's worth noting that the editors of the top medical journals don't share Hamilton's doubts about Vioxx's harmfulness. Lancet is the most vociferous:

"The licensing of Vioxx and its continued use in the face of unambiguous evidence of harm have been public-health catastrophes." [Lancet editorial, requires paid registration]

I've put excerpts from the JAMA and NEJM editorials at the end of the post.

Why are these editors so sure that Vioxx causes heart attacks? Because, unlike Hamilton, they don't rely on only one study [the link is to the free, preliminary version. Published version here.]. Since Vioxx was approved in 1999, studies have been piling up demonstrating the harmful effects of Vioxx. The best of these studies (because it was a randomized trial) found that Vioxx nearly doubled the risk of heart attack compared to a placebo.

Hamilton's most misleading statement is that Vioxx causes a "1 in 4,000 risk of dying from a heart attack." The casual reader might fail to notice that Hamilton says "dying from a heart attack," not having a heart attack. In fact, the FDA study Hamilton discusses says that 1 in 75 of those using Vioxx at high doses will suffer a heart attack due to using the drug (and 1 in 397 of those using low-dose Vioxx). These are "excess" heart attacks, in addition to those the patients would normally suffer. Before Vioxx was pulled off the market, it was pretty commonly used at high doses -- 18% of prescriptions, according to the FDA study.

Why the difference? Why does Vioxx seem to cause heart attacks but not deaths from heart attacks? Well, there probably is no difference. It seems much more likely that the issue is simply that many people survive heart attacks, so studies examining deaths rather than attacks will simply have a smaller sample, and so be unable to prove much of anything. In addition, the Vioxx studies have relatively short follow-up periods, mostly two years or less. Longer studies would be needed to observe the fatal second (third, fourth, ...) heart attacks.

Ultimately, Hamilton is upset that a jury is responsible for "evaluating the scientific merits of statistical evidence." While jury trials may not be the perfect system, in this case the 12 members of the Vioxx Jury managed to come to the right conclusion.

------------------------------------------

Editorials in Top Medical Journals

The recent withdrawal of rofecoxib from the market because of adverse cardiovascular events identified in the unpublished Adenomatous Polyp Prevention on Vioxx (APPROVe) study has raised major concerns about the undue control of industry over postmarketing safety data. Topol pointed out that although he and his colleagues published a clear warning about the cardiovascular toxicity of rofecoxib in 2001, the FDA never insisted on a trial to determine the extent of the problem and the manufacturer countered with a "relentless series of publications . . . complemented by numerous papers in peer-reviewed medical literature by Merck employees and their consultants." [Jama Editorial, requires free registration]

The Adenomatous Polyp Prevention on Vioxx (APPROVe) trial, a study of patients with a history of colorectal adenomas, was stopped early because rofecoxib doubled the risk of major cardiovascular events (relative risk, 1.92; 95 percent confidence interval, 1.19 to 3.11). These findings confirmed the increased risk of myocardial infarction previously seen in the Vioxx Gastrointestinal Outcomes Research (VIGOR) trial [NEJM Editorial, requires paid registration].

It's Zucchini Season!

2005-08-08T23:35:00.000-04:00

Zucchini and other summer squash are in season right now and I've been experimenting with Zucchini pancakes. You didn't know you could convert squash into pancakes? Neither did I, but they're quite good.

The goal is to get the zucchini to brown and stay crisp. But the zucchini is full of water, so it wants to steam. One could attack this problem by salting the zucchini, draining it in a colander for half an hour, and finally squeezing out the water. But that's too time-consuming. Also, salting and draining is supposed to make the squash less bitter, which I consider a drawback.

So instead, you grate the zucchini coarsely and mix in some cornmeal (or flour or minute-oats or some other starch). The ratio I've been using is about two tablespoons of cornmeal to 1 medium squash, but it doesn't matter much. The cornmeal absorbs the water and binds the zucchini into a patty. If you saute the patty at this point, you tend to get cornmeal mush with crunchy zucchini embedded in it: the mush seems to protect the vegetable.

This mush is actually pretty good, but I haven't succeeded in browning it. To get a pretty, browned, pancake, you need to stir in a beaten egg too. Oh, and you need more oil than the thin coat you'd use for sauteing. The celebrity chef Emiril has a recipe for Zucchini Bam! cakes along these lines.

And that's how you take a wholesome vegetable, and it a few simple steps turn it into something attractive and delicious, and still maybe somewhat healthy.

Is Zoning Causing a Housing Affordability Crisis?

2005-08-03T00:14:00.000-04:00

Everyday Economist Steve Landsburg wrote one of his better columns in Slate last week, eschewing his usual attempts to push neoclassical economics to the point of absurdity. Instead, he summarizes a recent empirical study Urban Economists Edward Glaeser and Joe Gyourko. Writing in the Cato Institute's Regulation, G&G argue that

If policy advocates are interested in reducing housing costs, they would do well to start with zoning reform. Building small numbers of subsidized housing units is likely to have a trivial impact on average housing prices (given any reasonable demand elasticity), even if well-targeted toward deserving poor households. However, reducing the implied zoning tax on new construction could well have a massive impact on housing prices. [A somewhat more technical paper is available too]

Glaeser & Gyourko's argument is mainly empirical: they claim to be able to split land prices up into two components: the fixed cost of acquiring a parcel of land, and the marginal cost of an additional square foot. Landsburg calls this fixed cost the "mystery component," and summarizes G&G's argument pretty well:

When you buy a house, you're not just paying for the land and construction costs; you're also paying for a building permit and other costs of compliance. You've got to get the permits, pass the zoning and historic preservation boards, ace the environmental impact statement, win over the neighborhood commission, etc. If Glaeser and Gyourko are right, that's the mystery component right there.

I'm fairly sympathetic to this idea. These kind of regulatory costs surely drive up the price of housing in some places, especially in the exurbs, where zoning can restrict minimum lot sizes to 10 acres or more.

Unfortunately, G&G aren't talking about the exurbs or even the suburbs, they're talking specifically about big cities like San Francisco and Dallas. There are certainly restrictions on building in central cities too, but there are also a lot of other possibilities for the "mystery component" of land prices.

The most obvious candidate the "mystery component" of land prices is just the pre-existing housing stock, which is going to impose a lot of constraints on any new construction. One can't simply tack on an extra story to an old building because land prices are high.

If you want to replace a single-family home with three townhouses, you have to tear down the existing house first. Even with a vacant lot, builders have to do a careful survey to avoid cutting underground utility cables and pipes, and to avoid building on land with toxic waste like a leaky underground oil storage tank. And of course, just finding a vacant lot in big built-up cities is costly and time consuming. Not to mention that building one infill house at a time is more expensive than building house as part of a large tract of new homes.

All these drive up the price of housing. And they're also all probably captured in G&G's "mystery component" of land prices, because they're all fixed costs. They're start-up costs that need to be paid before construction can begin, and that don't vary much with the size of a lot. It costs the same to tear down an old house whether or not it has a big yard.

So, there's at least some reason to think that G&G's promise of a "massive" drop in house prices from weaker zoning is exaggerated. But even if they're right, it isn't clear that weakening zoning in big cities is a good idea. Zoning is most needed precisely in big cities where people live close together. Sure zoning has costs but as G&G acknowledge -- at least in the more technical version of their paper, if not the Cato version -- zoning has benefits too.

If There Was a Housing Bubble In Wichita, Would Anybody Notice?

2005-07-19T22:36:00.000-04:00

It's often said that while there may be a housing bubble in some areas, especially big cities on the coasts, there isn't a nationwide bubble. But what if there's a nationwide bubble in land prices, a bubble that's being masked because land just isn't a very big of house prices in many places?

One of many recent statements of the idea that the house price boom is confined to a few areas comes from Greenspan-in-waiting Ben Bernanke, who recently argued,

While speculative behavior appears to be surfacing in some local markets, strong economic fundamentals are contributing importantly to the housing boom.

Econbrowser James Hamilton chimes in:

I would argue that the huge differences across communities in the rate of housing appreciation are much more of an embarrassment for Bernanke's critics than they are for him. Those challenging Bernanke's interpretation are forced to suggest that there are hundreds of separate little bubblets, expanding in different communities at curiously different rates.

One simple point that hasn't received enough attention is that standard statistics on house prices actually measure a combined package that includes both land and the structures built on them. And we'd only expect a bubble in land prices, not in the price of structures.

The price of structures will be constrained by the cost of construction. The price of houses themselves (absent the land) can never rise much over the cost of building a new one. On the other hand, it's pretty hard to make more land, so as long as the demand is there, the price can go sky-high. In the language of econ 101, the supply of structures is quite elastic, while the supply of land is very inelastic.

It may seem a little odd to think of land and houses being sold separately. But if we imagine a city with abundant land, it's pretty easy to see that we can't have much of a boom in house prices. If house prices start to rise, people will just build new homes instead of buying existing ones, and prices will fall back down.

In high-priced areas, land is a bigger portion of house prices, probably a much bigger portion. In Wichita, land may be only 10% of the price of a house (of the house-land package, that is), while in New York City, 90% of the value of a house or apartment may be in the land. If land prices doubled nationwide, people in Wichita would hardly notice: the price of a house would only go up by 10%. But in New York, house prices would go up 90%.

At a casual glance, this is exactly what's been happening, with booming prices in places that already were high-priced to begin with. We can test this a little more rigorously by estimating a regression. The regression tests whether there is a relationship between rents in the initial year and later house price growth. The idea is that cities with high rents are cities where land prices will be a high fraction of the house price package.

Specifically, I regressed the percent change in house prices from the first quarter of 2000 to the first quarter of 2005 (measured by Freddie Mac) on the natural log of median rent in 2000 (from Census 2000), for the 163 metro areas tracked by Freddie Mac. The graph below plots the data. The regression equation is

(House Price Growth) =	-5.74 +	0.97 ln(Median Rent),	N=163 Metro Areas
	(-8.43)	(9.18)

The t-statistics in parenthesis indicate that the effect of initial rent on later house price growth is highly significant. The t-stat of 9.18 is way above the cutoff of 2 that's usually taken to indicate statistical significance. The correlation between the two variables is 0.59, fairly high. So the regression shows that the boom in house prices has been much stronger in areas where rents were high initially. I also tried this regression with initial house prices (also from the census) instead of rents, and got similar but slightly weaker, results (t=7.6).

Now the regression doesn't prove that there's a bubble, that the house price boom has been driven by speculation rather than favorable fundamentals like low interest rates. But it does provide an answer to Hamilton's "embarrassment ... of hundreds of separate little bubblets." It suggests that something is happening nationally, not just in certain cities.

Is there a nationwide bubble in land prices, that's being masked by the fact that we only measure the combined price of land and structures? I'm not sure, but it is consistent with the data.

[hat tip to Bostic, Longhofer, Painter, and Redfearn, whose recent paper emphasizing the importance of land prices inspired this post]

Why Lee Got Spanked at Antietam

2005-07-14T23:40:00.000-04:00

A debate over a statue of Confederate General Robert E. Lee at the Civil War battlefield of Antietam has turned into a debate in the Washington Post's letters to the editor over Lee's role in the battle. I'll leave aside the question of why we need yet another statue commemorating someone who rebelled against his own country in order to protect white Southerners' right to own slaves.

Antietam was the bloodiest single day in the Civil War with about 6000 killed, split fairly evenly on both sides; more casualties than at D-Day in WWII. At the time, it was seen as a great Union victory. After all, Lee had invaded the North, accomplished little, and been forced to retreat back to Virginia. It was the victory Lincoln had been waiting for in order to announce the Emancipation Proclamation. And in Europe it pretty much ended all talk of Britain intervening on the Confederate side. According to historian James McPherson, Antietam was the crucial turning point in the war.

Perhaps aware of this history, one Lee critic writes to the Post's editors,

The monument commemorates a battle that Lee lost -- a battle that many of his officers (those who survived) thought never should have been fought. The Battle of Antietam forced the Confederate general to retreat to Virginia.

Another Post reader responds:

Lee was outnumbered roughly 2 to 1, yet through his usual generalship he managed to fight to a tactical draw. Yes, he withdrew to Virginia, a retreat that is still taught at West Point today.

The problem with this argument is that it was no accident that Lee's army was outnumbered at Antietam. Lee's troops had melted away on the march north. As Lee reported just before setting out, "The army is not properly [equipped?] for an invasion of an enemy's territory. It lacks much of the material of war, is feeble in transportation, the animals being much reduced, and the men are poorly provided with clothes, and in thousands of instances are destitute of shoes."

Yet Lee invaded anyway, hoping that his army's deficiencies would be made good by an uprising of Marylanders in support of the Confederacy. But that never happened.

Lee lost perhaps 10,000 of his 55,000 man army on the road north, according to McPherson. Lee himself put the figure higher, reporting a few days before the Antietam battle, "One great embarrassment is the reduction of our ranks by straggling, which it seems impossible to prevent with our present regimental officers. Our ranks are very much diminished-I fear from a third to one-half of the original number." Although many of those stragglers may have eventually turned up, the fact is a large proportion of Lee's army didn't make it to Antietam, either because they were too weak and underequipped to make the journey, or because they deserted.

Lee thought desertion was common, reporting after the battle that "Some of the stragglers have been gathered in, but many have wandered to a distance, feigning sickness, wounds, &c., deceiving the guards and evading the scouts. many of them will not stop until they reach their distant homes." One of his generals thought that Confederate troops were throwing away their shoes in order to have an excuse not to march on.

So Lee lost the battle long before any of his men ever saw Antietam Creek. He lost it when he sent his underfed, underclothed, demoralized men off on their march northwards.

Economists Say the WTO Hurts the Chinese Poor?

2005-07-13T01:19:00.000-04:00

Titled "Rural Poor Aren't Sharing In Spoils of China's Changes -- Costs of Goods Rise, Standard of Living Falls" Peter Goodman's article in today's Washington Post claims that conditions are dire in rural China.

A recent study conducted by the World Bank found that incomes among rural Chinese -- about three-fourths of the total population -- have declined slightly in the years since China entered the WTO, while urban residents have enjoyed modest gains.

Economists say this trend underscores the downside of globalization: While free trade has proved highly efficient in generating wealth, it has failed to share the spoils, intensifying gaps between rich and poor, urban and rural.

No doubt "economists" do say that globalization can increase inequality, in the sense that the rich may gain faster than the poor gain. But I would be pretty surprised to find any economist claiming that the poor have actually seen declining incomes in China. If it were true, this would be a huge turnaround from the 1990s, which saw tremendous reductions in poverty, according to numerous studies, including another World Bank report:

Between 1990 and 2000, the number of people living on a dollar per day fell by 170 million during a period when total population rose by over 125 million. Over the past two decades, China accounted for 75 percent of poverty reduction in the developing world.

According to Goodman, this new World Bank study of the WTO also contradicts China's goverment statistical agency, which reports continuing declines in poverty since China entered the WHO.

Of course, Goodman is wrong. He's entirely misunderstanding the World Bank's recent report on the impact of the WTO on China.

He seems to be talking about chapter 15 of the report, "Welfare Impacts of China's Accession to the WTO." The key word here is impacts. The World Bank is predicting the effect of the WTO on the distribution of income in China, based on an elaborate model. They're not describing trends; they're predicting that, in the short run, the dramatic gains experienced by China's rural poor will slow down a little. In the first few years, rural poverty will continue to decline, but not by quite as fast as it would have without the WTO.

Specifically, the World Bank predicts that China's poorest city-dwellers will enjoy an increase of 1.5% in income due to the WTO, but the poorest of the rural poor will lose 6%. Until recently, China's had high tarriffs on agricultural imports. As part of joining the WTO, China is reducing these tarriffs. In the short run, this will result in cheaper food for urbanites, but lower farm incomes. The World Bank authors are careful to note that they do not consider the long run. That is, they do not consider that in the long run, cutting farm incomes will spur migration to the cities, where incomes are much higher.

Now a 6% reduction in income is pretty serious for someone living on a dollar a day, but it's important to put that in context. According to the official Chinese statistics I cited above, rural incomes grew 6.8% faster than inflation last year, while urban incomes grew 7.7%. Perhaps the slower growth in rural incomes is due to the WTO. But China's spectacular growth means that the temporary 6% reduction predicted by the World Bank was wiped out in a single year.

Bush Winning on Social Security?

2005-07-06T23:31:00.000-04:00

Karl Rove thinks so, according to a story in today's Washington Post.

"We've been probably to some degree too successful" in selling private -- or personal -- accounts, White House Deputy Chief of Staff Karl Rove said yesterday.

There was a recent poll he had seen that found that about 40 percent of those who disapprove of Bush's performance on this issue actually want private accounts, explained Rove....

"I think their attitude," he said, "is: 'I disapprove of the president's performance on Social Security because he hasn't gotten it done.

Rove's optimism seemed pretty overheated, so I thought I'd look at a few polls. I guess you shouldn't be to quick to dismiss Rove's analysis, since he does seem to have something of a point.

According to a recent Pew Foundation poll, for example, it turns out that although only 29% of the public approve of Bush's handling of Social Security, 47% favor "a proposal that would allow younger workers to invest a portion of their Social Security taxes in private retirement accounts, which might include stocks or mutual funds." So polls rating his handling of Social Security do seem to make Bush's support on the issue look weaker than it really is.

On the other hand, support for private accounts falls dramatically -- from 48% to 27% -- if it is described as "reducing the rate of growth in guaranteed Social Security benefits for future retirees," according to a Washington Post poll. Since the Bush guys have been talking about exactly that (a "clawback" that would reduce traditional Social Security benefits for those who chose to put their contributions into private accounts) it seems like Bush and Rove are going to have a hard time selling their plan. Unless they mislead the public, of course.

Perhaps that's why there's no mention of the clawback on the White House Social Security page. And perhaps that why polls asking which party people "trust more to protect the Social Security system and retirement benefits" show the Democrats coming out on top by 46% to 36%.

Worthless Wine Reviews

2005-06-29T22:56:00.000-04:00

I'm not much of an expert on wine, but I though I'd pass on one of the few good pieces of advice I've ever read: buy the importer, not the wine. That is, you're unlikely to go wrong if you buy a wine distributed by a good importer. There's a list at the end of the Slate article in the link. I can personally vouch for Robert Kacher's $10-$12 wines from the south of France.

Sure you can read wine reviews in a newspaper or on the web, but what good will it do you? You won't remember the name of the wine when you go to the liquor store, and if you write it down, the liquor store won't have it anyway.

I once made a serious effort to track down a great wine I had in a New York restaurant. It turned out to be sold in one store in NYC, and was impossible to find in DC, even if I'd been willing to special-order a case.

Of course, with wines produced in large quantities, reviews are worthwhile, so let me point out that Zardetto Prosecco (a sparkling, dry white wine) is great and widely available. As I learned at DC's second-best pizza place (Matchbox is the best) it pairs really well with pepperoni pizza.

New Blogroll

2005-06-29T22:41:00.000-04:00

I've updated my blogroll, attempting to come up with a list closer to what I'm currently interested in. The most notable changes are the addition of a "Sixth Sense" section (science blogs); the replacement of Pierre Carion, who I suppose has gone back to France, with two French-language economics bloggers; and a bunch of new economics blogs. The most interesting of the last group is Jim Hamilton, who looks like he's as good at econ-blogging as he is at time series econometrics.

Oh, and I've long since lost interest in what Glenn Reynolds has to say.

Traveling to France? Don't Forget the Soap.

2005-06-23T21:41:00.000-04:00

What's up with cheap hotels that don't provide soap? This is a common practice in France and other European countries. Last week I stayed in a bed & breakfast in rural France that was very nice, and a real bargain at 37 euros ($45) a night. It was relatively low-end, I suppose (no TV, no 24 hr front desk, 3 flights of stairs to my room) but they provided a clean, attractive room with a private bath, and it was certainly no youth hostel. The one guest I chatted with seemed to be an old-money, wealthy Brit. But no soap.

As a good economist, I've got to consider the possibility that the hotel is striking an efficient bargain with its guests. In the U.S., lots of half-used hotel soap bars are probably wastefully thrown out at the end of a stay. Suppose that the little soap bars cost a dollar a day, but that European hotel guests can supply their own soap, taking it home at the end of their stay, for only fifty cents a day. Then an economist might speculate that the typical American is happy to pay the extra half-dollar for the convenience, while Europeans prefer to pack a bar of soap and save a few euros. The average European is poorer than the average American, after all.

But as I've said, the guests at this "chambre d'hotes" didn't seem particularly hurting for money. Further, why not offer a choice? For American hotels where the default is to provide little bars of soap, this might be difficult, requiring the maids to be informed of the different preferences of each guest: high "transaction costs" might prevent American hotels from offering fewer services to their guests who'd prefer that. I guess that's why I can't get a discount by offering to re-use unlaundered towels a second time, or even a third.

But soapless European hotels could just sell bars of soap at check-in (and how about little bottles of shampoo as well). I'd certainly buy one, even at an inflated price, rather than have to run out to the convenience store to pick up a bar of my own. Is it possible that cheap European hotels actually do this, but no one's told me?

Lancet Quiz: The Answer

2005-06-23T08:01:00.000-04:00

Someone has finally attempted my quiz! Commenter Kevin, writes:

My guess is that Apfelroth's reasoning is along these lines: in a war, refugees move away from potential flash-points, so the population is not geographically disributed as it would be for a census. So if your sampling method derives from census data, you will be over-sampling violent areas.

No, that's not it. Kevin's critique has previously been raised (very mildly) by Daniel Davies, one of the Lancet study's defenders. But that isn't what Apfelroth is saying, and in my opinion the reliance on old Iraqi census data is probably only a small problem with the Lancet study.

The problem is this. Within cities and villages, the Lancet researchers drew a rectangle around the village on a map, divided the rectangle up into a grid of hundred meter squares, and chose one of the squares at random. Each square had an equal chance of being chosen even though they might have had vastly unequal populations.

Apfelroth, who seems to be knowledgable about surveys, knows that if you draw a random sample of neighborhoods with unequal populations, you have to re-weight the sample to take that into account. Apfelroth is saying that given the available data, this would have been very hard: "it seems quite likely that the grid rectangles created by driving around in a war zone were much smaller than the original census tracts used in the 'cumulative population lists'."

In fact, Apfelroth is being overly generous here, assuming that the Lancet researchers did their best with this crucial step (weighting for neighborhood population). Either he is giving them the benefit of the doubt, or he's failed to notice that the Lancet researchers did not take this vital step at all: they weight all neighborhoods (100m grid squares) equally. Because they weight neighborhoods with different populations equally, their sample is much more likely to choose people from low-population neighborhoods: neighborhoods near parks or rivers, neighborhoods on the fringe of the city.

Anyone who's ever seen a Red/Blue map of Presidential voting in US counties will understand this phenomenon. Bush won the vast majority of the counties in the US (something like 80%) covering the vast majority of the country's land area (maybe 95%, I'd guess). Suppose you took a poll using the Lancet grid method. First you draw a rectangle around the continental US, then you choose grid squares within the rectangle at random, and finally you survey 30 people in each selected grid square. You'd end up with a lot of grid squares with no one living in them at all (e.g., oceans and deserts). And the vast majority of people interviewed would be in the rural, "Red" areas. A survey like this would likely find that 95% of respondents voted for Bush even though only 51% of the country actually did.

Now the Lancet study isn't quite this bad. It was only after they'd chosen a set of cities, towns, and villages to survey, using a reasonable method that gave a larger chance of selection to larger towns, that they began drawing rectangles and grids, and sampling land-area rather than people.

This is a huge problem for the Lancet survey if fighting is heavier on the fringes of cities than in the center, as I think it is. It will cause an overestimate of the death toll (I think a large overestimate). My commenter, Kevin, thinks that fighting is more common in city centers. If he's right, the Lancet study is still biased, although towards finding too small a death toll. It's hard to be sure who's right about where most fighting has occurred, but I'll make my case in a later post.

Deadly Europe

2005-06-13T04:56:00.000-04:00

There seems to be a fad among right-wing authors to write anti-Europe and especially anti-France books. But whenever I travel in Europe I'm always shocked by how dangerous it is, which I attribute to a lack of trial lawyers. In this regard, it's a right-wing paradise!

In Europe, people ride bikes without helmets, signs inform the innocent tourist of beautiful views atop cliffs without guard rails, narrow twisty staircases are unlighted, bread slicing machines are self-serve. Today I saw a Belgian fire escape, which was a ladder bolted to the side of a building. No doubt it was much safer than all the adjoining buildings, which lacked fire escapes entirely.

You may say, big deal, only fools would walk off a cliff, get their finger caught in a bread-slicing machine, or smash their head in a fall from a bike. But during my fairly brief time in Europe, I've see a friend trip and fall flat on his face at an unexpected step in the middle of a long, unlighted hallway, and later ride a bike at high speed into a chain stretched across an unlighted road entering a campsite. Myself, I've smashed into a closed glass door at the bottom of a spiral staircase.

No one was seriously hurt in any of these mishaps, and no doubt many will say that this only proves that I and my companion are fools. Indeed, after his spills, I explained to my friend that he had behaved extemely foolishly. But, in hindsight, I think I would have rather have dangerous areas lit up, safe egress from burning buildings, and a bike helmet protecting my head -- even at the cost of government intervention and tort actions by agressive trial lawyers.

Mississippi Michelin

2005-06-10T18:31:00.000-04:00

So, I'm in Europe, investigating how the lifestyle compares to Mississippi. Specifically, I'm in Leuven, Belgium, a university town of 100,000 or so about 30 minutes from Brussels. Leuven has two Michelin-starred restaurants. I highly doubt that the same, nor anything similar, can be said of anywhere in Mississippi.

Lancet: Extra Credit Assignment

2005-06-07T12:40:00.000-04:00

I've occasionally debated the Lancet study with their blogging defenders for some months now. It's been frustrating, and I've finally figured out why. The Lancet defenders, such as Lambert and Davies, both know something about the kind of statistics that starts "assume you have a random sample," but aren't particularly knowledgeable about the actual mechanics of obtaining a random sample.

The most egregious example is Lambert's casual dismissal of the cogent and well-informed criticisms of Professor Stephen Apfelroth of the Albert Einstein Medical school. Lambert dismisses Apfelroth's criticisms as "just speculations the sampling was not done correctly." In fact, Apfelroth's criticisms are hardly speculations: they're the expert opinion of somebody who obviously understands survey methods. They could be taken straight from a textbook on sampling. Here, Lambert is just sneering from ignorance, and I doubt that he understands Apfelroth's criticisms, which are indeed pretty terse.

One of Apfelroth's more obscure, but more important criticisms is this:

When a town or village was selected from the "cumulative population lists for the Governorate", the survey team then "drove to the edges of the area and stored the site coordinates"....it seems quite likely that the grid rectangles created by driving around in a war zone were much smaller than the original census tracts used in the "cumulative population lists".

If you understand survey methods, you'll understand why it is a huge problem that the grid rectangles were much smaller than the original census tracts. And if you think about this issue further, you'll realize that the Lancet method is highly likely to oversample rural areas and the fringes of cities. It is precisely these low-density areas where the fighting was most intense.

So here's the extra credit question. If you get it right, I promise to take your criticism or praise of the Lancet study very seriously. Why does the discrepancy between the grid rectangles and the census tracts imply that the Lancet study oversamples low-density areas?

UPDATE: So, as of June 10, no one's attempted my quiz, even though I know plenty of people have come over here from the Lancet discussion on Tim Lambert's blog. Come on people! I'll settle for an explanation of what Apfelroth is talking about!

Lancet's 100,000 Deaths in Iraq Vindicated? Nope.

2005-06-06T01:12:00.000-04:00

Before the presidential election, the medical journal Lancet released a study finding 100,000 excess deaths from the Iraq war. Since then, the UN and the Iraqi government have released a new study (the ILCS survey) with a much larger sample size, finding about 24,000 deaths from military action. This has lead to much uninformed crowing from the right, charging that the Lancet study has been refuted.

The fact is, the two numbers are not comparable. The Lancet figure is for all excess deaths, and is based on higher post-war rates of violent crime, disease, infant mortality, and so on. The ILCS figures ask specifically about deaths due to combat.

But now Tim Lambert, an Australian computer science professor who delights in denouncing bad science and has written over 40 blog entries defending the Lancet study, has done some new calculations based on the Lancet data. And he's crowing that the Lancet study has been "vindicated."

Lambert calculates that 33,000 deaths occured as a direct result of fighting in Iraq. Since the Lancet study is for a slightly longer period (18 months vs. about 14 months), Lambert concludes that the numbers suggest about the same death rate. But Lambert's claim of "vindication" is just as flawed as the earlier right-wing debunking.

First, the ILCS counts include Iraqi soldiers killed during the war, while the Lancet counts include only civilians: so say the authors of the two studies. More importantly, Lambert's figures exclude Falluja, where most of the deaths in the Lancet sample occured. If Falluja were included, the Lancet figure would be 189,000, almost an order of magnitude higher than the ILCS numbers.

Now, I don't deny that there are good reasons to exclude the Falluja cluster. But if you do so, you can't claim to have an estimate of war-related deaths in Iraq. You have an estimate of war-related deaths in areas without intense combat. In a letter to the editor of a British newspaper, the Lancet authors give a good description of the issue.

Our study found that violence was widespread and up 58-fold after the invasion; that from 32 of the neighbourhoods [i.e. excluding Falluja] we visited we estimated 98,000 excess deaths; and that from the sample of the most war-torn communities represented by 30 households in Fallujah more people had probably died than in all of the rest of the country combined.

Fallujah is the only insight into those cities experiencing extreme violence (ie Ramadi, Tallafar, Fallujah, Najaf); all the others were passed over in our sample by random chance. If the Fallujah duster is representative, there were about 200,000 excess deaths above the 98,000.

Perhaps Fallujah is so unique that it represents only Fallujah, implying that it represents only 50-70,000 additional deaths. There is a tiny chance that the neighborhood we visited in Fallujah was worse than the average experience, and only corresponds with a couple of tens of thousands of deaths. We also explain why, given study limitations, our estimate is likely to be low.

I hope I'm not belaboring this obvious point: excluding the part of the sample where the most war-related deaths occured means that the Lancet's 100,000 and Lambert's 33,000 are likely to be underestimates. Indeed, in other contexts the Lancet's defenders on the web have emphasized this point: the Lancet figures are conservative, probably very conservative. Lambert himself has written that "excluding Falluja biases the results downwards."

Now Lambert appears to have changed his mind. I've raised this point with him on his blog, and as near as I can figure out, his reply is that Falluja was uniquely violent, or that the period covered by the Lancet survey after the ILCS survey had ended (April or May 2004 to September 2004) was uniquely violent. So he claims that the figures from the two surveys are comparable even when the Falluja data is dropped from the Lancet survey.

In fact, both surveys cover time periods with similar amounts of fighting. Both surveys exclude the second round of fighting in Falluja during November 2004. Both surveys include the intense fighting during the conventional war itself (March-April 2003). In general, there has been a lot of fighting in Iraq, at many times and places.

According to the Iraq Body Count, 7,981 civilians were killed from the start of the war until March 2004, a time period covered by both surveys. During intense fighting in Baghdad, Falluja, and other parts of Iraq in April 2004, another 1,165 civilians were killed. These deaths are covered in the Lancet survey, but some were probably missed by the ILCS, which was in the field from March 22, 2004 to May 2004. Finally, 1,696 civilians were killed in fighting from May 2004 to September 2004, a time covered only by the Lancet study. The Iraq Body Count figures are based on newspaper reports, and so probably miss many deaths. But there's no reason to doubt that they get the time pattern of deaths about right.

Some calculation shows that the ILCS survey covers about 3/4 as long a time period as the Lancet study, during which time about 3/4 of the deaths occured. Although there was a lot of fighting in April 2004, the next 3 or 4 months were relatively quiet. So Lambert's claim that "the intense fighting was mostly after the ILCS was conducted" is simply false.

And that leaves us back where we started. Lambert's 33,000 figure of war-related deaths in a narrow population (excluding areas of intense combat and the deaths of soldiers) just isn't comparable to the ILCS figure of 24,000 for the whole population. If anything, the similarity of these numbers suggests that something went very wrong with the Lancet study. It certainly doesn't "vindicate" it.

[The Iraq Body Count figures can be found here (for the first 50 days of the war) and here (for the period since then)].

Okrent Swings and Misses

2005-05-31T19:45:00.000-04:00

In his final column as NY Times Public Editor, Daniel Okrent charged that "Paul Krugman has the disturbing habit of shaping, slicing and selectively citing numbers," but didn't bother to give a single example to support his smear. Krugman, not surprisingly, took offense. Okrent responded with some specific charges, which Krugman then devastatingly rebutted, as has Brad DeLong.

Krugman and DeLong don't bother to answer some of Okrent's sillier charges, so let me. Okrent criticizes:

[Krugman's] 1/27/04 assertion that the cost of unemployment insurance "automatically" adds to the federal deficit. This two-fer misrepresents a pair of facts: that unemployment insurance is largely borne by the states, and that major federal contributions to the states come about only because of an act of Congress, which is hardly automatic.

Obviously this quibble over Krugman's use of the word "automatic" has no relevance at all to Okrent's original charge: that Krugman misrepresents the numbers. Presumably this is why Krugman and DeLong ignore it.

But it's worth pointing out just how ignorant Okrent's complaint is. Krugman is just using the absolutely standard language economists use to describe this particular aspect of fiscal policy, called "automatic stabilizers." For example, in his undergraduate textbook, Greg Mankiw (recently chair of the Bush Council of Economic Advisors) writes

"The system of unemployment insurance automatically raises transfer payments when the economy moves into a recession, because unemployment rises." [Mankiw, Macroeconomics, 1992, pp. 324-5]

which is something you'll find in pretty much any macro textbook, whether written by a conservative or a liberal. And which is almost exactly what Krugman originally wrote. Let me join with Jonathan Chait is saying that Okrent ought to be ashamed.

Laser-Guided Infrared Thermometers

2005-05-27T21:02:00.000-04:00

A surprising number of people come to this blog searching for "Alton Brown infrared thermometer," since I once mentioned that Alton Brown recommends them. Apparently, he isn't enthusiastic enough to include infrared thermometers on his list of essential equipment. So, here's my take.

Infrared thermometers are in widespread use by mechanics for measuring engine temperatures and such, but can be also measure the temperature of a pan. You just point it at a metal object a few feet away, using the built-in laser beam to target it, and press the trigger to display the temperature.

I've never actually used one, but I recently gave one as a gift. The Bonjour thermometer is sold on lots of cooking sites. If you want to save a few bucks, I'm pretty sure that the Bonjour is just a repackaged version of the Mastercool infrared thermometer marketed to auto mechanics. They both look the same, and have the same temperature range. The recipient of my gift, who's quite a good cook, says he hasn't used it to test a pan yet, but that it seems to works very well on his boat engine.

Personally, I'm pretty happy with the usual methods of judging the temperature of a hot pan: watching how easily the oil flows; waiting for the butter to foam up, the olive oil to give off its characteristic aroma, or the oil to begin smoking; or throwing on a test piece and listening to the sizzle. Hmm, maybe it really isn't all that simple. I guess it would be nice to have an infrared thermometer. And they sure are cool.

Electronic Vote suppression

2005-05-24T23:31:00.000-04:00

Perhaps the most important finding in Card and Moretti study I blogged about last week is that electronic voting seems to suppress turnout. The Berkeley economists found that voter turnout fell by seven tenths of a percentage point in counties that switched to touch screen voting, compared to what would be expected based on past voting patterns and county demographics. Card and Moretti suggest that some voters may distrust or be intimidated by the machines, and so been deterred from voting.

It seems to me that there's a much more obvious interpretation: electronic voting machines don't work all that well. There's certainly tons of anecdotal evidence. In the last election, watchdog groups recorded over 2,000 reports of trouble with voting machines, three quarters specifying electronic voting machines.

Here's a report from one precinct in Houston.

"Yes, there are six machines down. I got there at 8 o'clock and they have been down since," said voter Melita Warren. "The tech does not know how to fix it. She is reading the manual, so therefore I should have been at work a long time ago."

A private election watchdog group documented the problems.

"Long lines, an hour and half wait. People (were) coming in at 7:30 and not leaving until 9," said Mary Huffine.

No one knows how many voters left before casting their ballots and there is no way of knowing who will be back
...

All 11 E-slate machines are working now, but it took three technicians to come out to the site and fix the problem. Officials told Eyewitness News it should have just taken one technician to do the repairs.

There is still no word on what exactly was wrong with the equipment. Several other precincts were having problems with the E-slate machines as well.

For more, see VotersUnite.org or the National Academy of Sciences Electronic voting Committee.

Shit Storm in Iraq

2005-05-23T22:36:00.000-04:00

Via Tim Lambert's blog Deltoid, I learn about the Iraq Living Conditions Survey, a new report by the UN and the Iraqi government. Serious problems in Iraq include not only electricity and violence, but also a shortage of water. Before the war, only 5 percent of urban households lacked safe supplies of drinking water. It's now 40 percent. Typically, the problem is weekly breakdowns of the water supply. Shockingly, 38 percent of urban households, and more in rural areas, report that the "cost of drinking water restricts consumption." I bet they're not taking too many baths either.

Iraqi pipes aren't so good at bringing in the water, or at carrying it away. According to the survey, 37 percent of Iraqis now lack "improved sanitation." Before the war, it was 7 percent. "Improved sanitation," by the way, doesn't mean a flush toilet -- a private outhouse counts. The third of Iraqis without improved sanitation typically use a public outhouse, or just a hole in the ground. No doubt this is why interviewers saw raw sewage around 40 percent of urban households, or less politely, human shit in the streets.

No Large-Scale Vote Fraud Say Berkeley Economists

2005-05-18T23:40:00.000-04:00

According to Berkeley Economists David Card and Enrico Moretti, it doesn't look like there was systematic fraud associated with electronic voting. An earlier analysis (by a team of Berkeley sociologists) had found that in Florida, Bush improved on his 2000 vote share more in counties that used touch screen voting. [sociologists' study here]

Card and Moretti used similar techniques to examine voting in all U.S. counties. Holding constant past election outcomes, they also found that Bush improved his vote share in counties with electronic voting, by about a fifth of a percentage point on average. But this effect is mainly found in states where you wouldn't expect to find fraud. Electronic voting seems to have helped Bush in states with Democratic elected officials; it doesn't seem to have helped much in swing states. [free copy of the paper here]

As Card and Moretti emphasize, this doesn't prove that Bush didn't steal a few counties here and there, say in Ohio and Florida, but it does seem to rule out widespread fraud based on electronic voting. Personally, I'm inclined to believe it because Card and Moretti are such prominent researchers (Card is a winner of the prestigious Clark Medal), because they're hardly Bush supporters, and because I think they're honest.

What Would Jesus Eat?

2005-05-13T00:20:00.000-04:00

Dieting guides and religious books are consistent best sellers. So what could be more natural than combining the two?

Stephen Arterburn, the host of a Christian radio show and author of Lose it For Life, says: "If you want the world to notice Jesus, it helps to look and live like Jesus. We don't do this so we can look in the mirror and be more attractive. We do it so people can look at us and see Jesus."

Apparently, there are now several "Jesus wants you to be thin" diet books out. Even one called What Would Jesus Eat by Don Colbert. Colbert's books don't show up on conventional bestsellers lists, probably because many are sold at Christian bookstores and such, but he claims to have sold over 4 million copies of his many books.

Prize Money, Informants, and the Noose

2005-05-11T22:55:00.000-04:00

Corporate Law prof Stephen Bainbridge has a fascinating post about the economics of Patrick O'Brian, or, more accurately the economics of the British Navy in the 17th-19th centuries. You didn't know the British Navy had economics? Well, when Naval vessels captured an enemy ship, they got to keep it, or more accurately, the captured ship and its contents were auctioned off or purchased by the crown, with the victorious captain getting a quarter of the proceeds. "Head money" was also paid for enemies killed or captured.

As readers of O'Brian's terrific novels know, this system didn't give quite the right incentives. The system of prize money and head money did encourage captains to fight rather than flee, but not necessarily to fight the right battles. Captains could make more money, much more safely, by attacking enemy merchant vessels rather than warships. And with frigate captains often on the other side of the world from their superiors, monitoring them wasn't easy.

According to The British Navy Rules: Monitoring and Incompatible Incentives in the Age of Fighting Sail, the academic paper by Douglas Allen that Bainbridge is discussing, the British Admiralty offset these incentives in several ways. They frequently executed captains for avoiding combat and treated captains who fought but lost leniently. The French navy of the time did precisely the opposite.

Also, the admiralty tried to set things up so that lieutenants would report captains who shirked their duty. Lieutenants were required to keep logs and couldn't be fired by the captain. Perhaps most importantly, it was very difficult to get promoted to captain, at least for those without connections. A lieutenant who ratted out his captain stood a chance of getting his job.

It was a complicated system with lots of flaws, but it worked well enough that the British navy ruled the oceans for over a century.

Anyway both the post and the paper are well worth reading. They give a lot of insight into the events of O'Brian's scrupulously researched novels (the conflict between lieutenants and captains was news to me). They're also fascinating examples of what economists call the "principal-agent problem." That is, the problem faced by managers who want their employees to work hard, but aren't able to easily measure the quality of their work.

To take one example, the problems faced by the 19th century British Admiralty sound a lot like the dilemmas faced by police chiefs today, as described by George Kelling of "broken windows" fame. How to make sure that beat officers are doing their job instead of snoozing in their patrol cars? How to allow cops the freedom to take initiative without giving too much opportunity for corruption? I'm not sure, although our experiments with prize money for the police haven't turned out that well.

Fighting Obesity

2005-04-29T00:43:00.000-04:00

Policy Prof Mark Kleiman has an interesting post discussing policy responses to the obesity epidemic, motivated by the debate over a new report that merely being overweight doesn't increase mortality (although being obese does).

Kleiman basically suggests that we should change land-use patterns (building more walkable neighborhoods) and regulate convenience foods in some way. Although these measures seem commonsensical, and maybe they're worthwhile things to do, I think they're largely beside the point.

Kleiman's common sense proposals go both too far and not far enough. They go too far because they're very broad-based interventions affecting everybody, when the real problem is relatively concentrated. Obesity mainly kills people with BMIs of 35 or more, about 13% of the population. In other words, for a 5'8" person, obesity starts to have serious consequences when you're around 65 pounds or more overweight.(*)

They don't go far enough because they don't do that much. Personally, I doubt that making more walkable cities will do anything at all (that's a subject for another post, but see Cutler et al for a convincing argument on this point). But even if better land use planning does help, it's not going to help anytime soon. It's a project for decades. What kind of response is that to a problem that's killing 100,000 or more per year?

I'd guess that regulating fast food and such is more likely to be helpful (again, see Cutler et al), but is this really likely to happen? I can imagine the government cracking down on candy and soda in schools, but not much else. Are there really any junk food regulations that have any chance of gaining political support?

Finally, even if commuting by car and the availability of junk food are what's making Americans so fat, that reversing these trends will make us thin. Once you gain weight, it's very hard to lose it: your body adapts a new set point, and fights to keep you at your new weight. So, these proposals might help the next generation, but not necessarily help people who are already overweight.

So, what' s a better response? First more research, into things like appetite control drugs, and breast feeding. Yes, breast feeding. It looks pretty likely that influences in the womb and the first months of life have a large effect on obesity later in life.

Also, there's really only one thing that' s been proven to result in significant weight loss, and it's not dieting, excerise, or drugs. It's gastric bypass surgery ("stomach-stapling"), which about 140,000 people have every year, in spite of the dangers. You want a policy proposal? How about insisting that Medicaid and private insurance cover this procedure?

-------------------------------------

(*) The Flegal et al study discussed in the Times reports that there are about 30,000 extra deaths per year among the obese (BMI 30-35) and 82,000 among the extremely obese (BMI 35+), so lower levels of obesity aren't necessarily risk-free. Or maybe they are safe: the study looks at several decades of data, and BMIs of 30-35 increase mortality only in the 1970s, not the 1980s or 1990s. But in any event, I think it's fair to say that obesity is only a serious health problem for about 13% of the population.

Bubbly

2005-04-27T01:00:00.000-04:00

In the Washington Post last week, Daniela Deane reported that "some economists warn of a housing bubble" in the DC area, with prices up 89% over the last five years. On the whole, it's an excellent article, with Deane suggesting that there is a bubble in the DC area.

Among the symptoms that some say point to a bubble: a widening gap between rental and ownership costs, a spike in the number of investors rather than occupants buying, and a ever-tighter affordability squeeze. Much of the boom in recent years has been sustained by low interest rates...But the consensus among economists is that interest rates will rise at least a little this year.

One economist who disagrees is David A. Lereah of the National Association of Realtors, who argues:

"Right now, most local areas have a lean supply of homes. And the Washington area is creating tens of thousands of jobs rather than losing any."

Continuing, Deane writes that

the Census Bureau reported last week that the Washington area added 75,000 residents last year, making it the fastest-growing metropolitan region outside the Sun Belt.

Today, the National Association of Home Builders writes in to the Post, making the same point about DC's rapid population growth.

One problem with this explanation is that the city of DC and the inner ring suburbs of Arlington and Alexandria aren't growing rapidly. They're actually losing population. And yet they have some of the steepest increases in house prices.

In fact, within the DC area, there's almost an inverse relationship between population growth and house price growth. Except for the outer-ring Loudon and Prince William counties, the fastest growing counties have the slowest rates of price growth.

Now, one could certainly reconcile the facts with the realtor's theory, by thinking about supply as well as demand. Maybe job growth is attracting people to the DC area but they're relatively indifferent betweeen living in the city or the suburbs. If there are supply constraints in the city, say because there isn't much land left to build on in the city, increased demand could drive up prices in both the city and the suburbs, even without the city's population growing.

But I'm not all that convinced. The suburbs with the most rapid population growth are fairly far out. It's hard to believe that many of Loudon's new arrivals would move to DC if only there were more housing available. And supply constraints in the city would have to be awfully severe for the city to actually be shrinking. So my money's on a bubble.

	% Change from 2000-2004
County	House Prices	Population
Loudoun	83.8	41.0
Prince William	95.5	19.9
Calvert	35.8	16.0
Charles	46.3	12.7
Frederick	46.2	11.5
Howard	48.1	7.6
Montgomery	68.4	5.5
Prince George's	48.0	5.2
Anne Arundel	59.1	3.9
Fairfax	80.4	3.4
Alexandria	91.3	-0.1
Arlington	82.0	-1.8
District of Columbia	85.9	-3.2

Source: Washington Post (house prices seem to appear only in the print version); Census Bureau

Fafblog Takes Down Juan Cole

2005-04-20T23:33:00.000-04:00

Over at Fafblog, Giblets digs up yet another outrageous claim by Middle East pundit Juan Cole.

True, the quote in Fafblog doesn't compare to such expert Cole pronouncements as "[I]f Sharon and AIPAC decide that they need the US government to take military action against Iran...it is likely that the US government will do so." Or "Colin Powell was pushed out as secretary of state because he sought to rein in Israeli Prime Minister Ariel Sharon" (linking to an article which says no such thing).

And Giblets apparently can't manufacture anything as bizarre as blaming Israel for 9/11 (Cole believes that Israel "helped drag the United States into a hot war with terrorists"). Or as dishonest as writing a long screed about Israel's failure to meet with the the Palestinians after 9/11, without mentioning that high-level talks began within two weeks.

But as usual Fafblog is right on target, quoting Cole as saying "dynamite is explosive...If you feed it to your pig, your pig will explode." Hilarious! Just as the real Cole makes absurd pronouncements about Israel's behind-the-scenes manipulation of U.S. foreign policy, so the Fafblog Cole makes ridiculous statements about the properties of dynamite.

Of course, dynamite is perfectly edible. In small doses, it's even a medicine, being just stabilized nitroglycerine. In large doses, dynamite is poisonous, but a dynamite-eating pig still isn't going to blow up. The whole point of inventing dynamite was to create a stable explosive, requiring a blasting cap to set it off.

Indeed this Fafblog entry is subtle, and many may have misinterpreted it as a defense of Cole. So subtle is the humor here, that I suspect it was written not by Giblets, but by the Medium Lobster.

And Fafblog's intricate joke isn't limited to merely putting a silly claim into Cole's mouth. Giblets continues, hysterically deriding Cole's expertise in "pig history and pig theory," fields of study which obviously provide little insight into the properties of dynamite. Here, Giblets (if it is Giblets) deftly skewers Prof. Cole for spouting nonsense when making pronouncements outside his area of expertise, which is middle eastern religion. As Fafblog implies, Cole's insights into Shiites, Bahais, and Sunnis obviously don't prevent him from making strange conspiratorial claims about Israel's secret orders to President Bush, demanding that he fire his subordinates and invade various countries.

The Squid Strategy

2005-04-18T23:04:00.000-04:00

When you're wrong, and someone points it out, the squid strategy is a good one. Fill the water with black ink to confuse the issue, and hope that observers will through up their hands and decide that it's all too complicated, and too much trouble to judge who's right.

This kind of thing is pretty common in politics, but via Brad DeLong, I learn how Harvard Economist Caroline Minter Hoxby is doing the same thing. Hoxby wrote a well-known paper arguing that competition, in the form of numerous school districts, improves school productivity and student outcomes.

In fact, relatively simple statistical methods find no such relationship, but Hoxby argues that there might be an omitted variable or reverse causation problem. For example, maybe good school districts get bigger, reducing competition. This doesn't seem like such a serious statistical problem to me, but Hoxby famously proposed to solve it using rivers as a "natural experiment." Her idea is that metropolitan areas with more rivers will have more school districts, because in the olden days, it was hard to cross rivers to go to school. To simplify a little, Hoxby shows that metro areas with more rivers have higher student achievement, and so concludes that competition is a good thing.

Recently, the tenured Hoxby has been criticized by the untenured Jesse Rothstein of Princeton, who says he found numerous errors in her study.

Hoxby responds with the squid defense, rebutting Rothstein's weaker and more technical points, while not providing a convincing answer to his main criticism. An economist at the Lowest Deep blog is taken in, writing: "Rothstein's criticisms are lengthy, technical, and enumerated in excruciating detail." So is Brad DeLong, writing "I don't have a dog in this fight. I'm not qualified to judge this fight."

Oh! It's all so complicated! How can a layperson judge! Or even an economist in another subfield! Well let me explain these deeply technical and complicated matters.

The whole thing hinges on the definition of rivers. Hoxby says that it's small rivers that determine how many school districts are formed in a metropolitan area, not big rivers. Rothstein pretty convincingly shows that Hoxby's provides no clear definition of small rivers, and that her decisions about what to count are extremely subjective. When you replace Hoxby's measure of "small rivers" with a simpler measure of the total number of rivers in a metro area, Hoxby's results disappear.

As far as I can tell, Hoxby doesn't dispute this point. She has a vague and unconvincing argument that her count of small rivers is the right one, but mostly tries to shift attention to more technical matters.

Rothstein's main point is summarized pretty well in his abstract

In an influential paper, Hoxby (2000) studies the relationship between the degree of so-called "Tiebout choice" among local school districts within a metropolitan area and average test scores. She argues that choice is endogenous to school quality, and instruments with the number of larger and smaller streams. She finds a large positive effect of choice on test scores, which she interprets as evidence that school choice induces greater school productivity. This paper revisits Hoxby's analysis. I document several important errors in Hoxby's data and code. I also demonstrate that the estimated choice effect is extremely sensitive to the way that "larger streams" are coded. When Hoxby's hand count of larger streams is replaced with any of several alternative, easily replicable measures, there is no significant difference between IV and OLS, each of which indicates a choice effect near zero. There is thus little evidence that schools respond to Tiebout competition by raising productivity.

See, not really all that complicated. It's all a matter of which streams count as large and which as small. The depressing thing is that Hoxby is a better writer than Rothstein, so I bet she's going to win the debate.

President Bush: MP3 Pirate?

2005-04-11T22:59:00.000-04:00

According to the NY Times, Bush has an iPod.

So, is he listening to legally purchased music, or is he saving a few bucks by engaging in illegal file sharing? Well some of his tunes are legal:

Bush...does not take the time to download the music himself; that task falls to his personal aide, Blake Gottesman, who buys individual songs and albums...from the iTunes music store.

But, wait; those legally purchased songs aren't the only music on Bush's iPod:

The president also has an eclectic mix of songs downloaded into his iPod from Mark McKinnon, a biking buddy and his chief media strategist during the 2004 campaign

Surely Bush isn't listening to pirated music! After all, just the other day, his administration asked the Supreme Court to ban the file sharing service Grokster. Bush promised to "restore honor and dignity" to the White House, and surely there's no honor in thievery. Still, it's hard to see what "songs downloaded into his iPod from Mark McKinnon" can mean, except that Bush is listening to illegally copied music, presumably songs from McKinnon's CDs.

I suppose McKinnon could have borrowed Bush's CDs and ripped selected songs for him. And if you believe that, you'll also believe that Bush wants to save social security.

Wooden Carpet

2005-01-28T01:05:00.000-05:00

Wouldn't it be great to have a wooden carpet?

Perhaps as a throw-rug under a coffee table. Or it could be a fancy replacement for the plastic sheet sometimes put down over carpet to make rolling a desk chair easier. I picture something like a polished wooden serving tray (except bigger). For some reason, nobody seems to sell this, perhaps because a big slab of wood would be hideously expensive and easy to break.

About the closest thing available are thin slats of wood attached to a cloth backing. The problem with this one is that it seems fairly low quality, judging from the picture, and you can probably only buy it in the US if you're willing to import a shipping container's worth.

Not quite the same, but nicer and actually available in the US, are bamboo rugs. New-Agey hand-painted wooden rugs are also available.

Silence

2005-01-28T00:48:00.000-05:00

I haven't been blogging for a while. Mainly, it's because I'm depressed and befuddled over the election.

Depressed because there's really no point in discussing economic policy with Bush in the White House and the Republicans controlling Congress. They're going to run the country into the ground no matter what anybody does, so why bother? Befuddled because it's just hard to switch gears after so many months of thinking about Kerry vs. Bush.

If I were a better person, I'd could be thinking about ways to achieve change at the local level. Or I could be criticizing Bush's plan to privatize Social Security and turn this country into Argentina by running up the debt, like everybody else. Or I'd be proposing ways for the Democrats to win with new and improved rhetoric. Or I would at least be reading the newspaper.

Instead, I've been shopping and rearranging my personal finances. So that's what I'm going to blog about until I get bored of it, or get interested in politics and policy again.

Why Drug Spending is Soaring

2004-12-03T01:32:00.000-05:00

A friend directs me to an excellent critique of Marcia Angell's "The Truth About the Drug Companies," by Malcolm Gladwell in the New Yorker. There's lots of good stuff in the article, but one point that was new to me was the fact that spending on prescription drugs is rising rapidly not so much because of price increases for existing drugs, but because drugs are being more widely used, and because new drugs keep being discovered.

The second misconception about prices has to do with their importance in driving up over-all drug costs. In one three-year period in the mid-nineteen-nineties, for example, the amount of money spent in the United States on asthma medication increased by almost a hundred per cent. But none of that was due to an increase in the price of asthma drugs. It was largely the result of an increase in the prevalence of usage—that is, in the number of people who were given a diagnosis of the disease and who then bought drugs to treat it. Part of that hundred-per-cent increase was also the result of a change in what’s known as the intensity of drug use: in the mid-nineties, doctors were becoming far more aggressive in their attempts to prevent asthma attacks, and in those three years people with asthma went from filling about nine prescriptions a year to filling fourteen prescriptions a year. Last year, asthma costs jumped again, by twenty-six per cent, and price inflation played a role. But, once again, the big factor was prevalence. And this time around there was also a change in what’s called the therapeutic mix; in an attempt to fight the disease more effectively, physicians are switching many of their patients to newer, better, and more expensive drugs, like Merck’s Singulair.

Asthma is not an isolated case...All told, prescription-drug spending in the United States rose 9.1 per cent last year. Only three of those percentage points were due to price increases, however, which means that inflation was about the same in the drug sector as it was in the over-all economy. Angell’s book and almost every other account of the prescription-drug crisis take it for granted that cost increases are evidence of how we’ve been cheated by the industry. In fact, drug expenditures are rising rapidly in the United States not so much because we’re being charged more for prescription drugs but because more people are taking more medications in more expensive combinations. It’s not price that matters; it’s volume.

The source for much of this is probably IMS Health, a massive health industry consulting firm, although I couldn't find any recent figures from them (I did find a popular article by one of their executives with some nice charts). But figures from the BLS and CMS confirm the IMS story, at least in broad outline.

Over the last 10 years, per capita drug expenditures have more than tripled, while drug prices have gone up 47% and overall inflation has been 28%. The drug price index is for existing drugs, so it isn't affected when a cheap old drug is replaced by an expensive new drug.

In the end, I think Gladwell is mostly right, but not entirely. To the extent that drug spending is soaring because more people are getting access to needed drugs or because old drugs are being replaced with better ones, we probably shouldn't worry. But the part of the increase in the drug spending, according the Angell, is expensive new drugs that aren't any better than the cheap old ones they replace.

----------------------------------------------------------------------------------

Social Security Shell Game

2004-12-01T01:32:00.000-05:00

NYU Economist Nouriel Roubini has a detailed post explaining why the Bush administration's plan to partially privatize Social Security is a "con" and a "scam." Bush wants to allow workers to put part of their Social Security taxes into a private account. Of course this would create "transition costs," since the government would need to come up with extra money to pay current retirees. Details are sketchy, but it appears that Bush plans to finance the transition costs through borrowing.

The claim is that this scheme will somehow resolve Social Security's long-term shortfall. CNN quotes an administration flack as saying "It's important to view the financing of the transition as a down payment...If we fail to act, we face a $10 trillion gap."

The problem is that this plan doesn't actually do anything. It doesn't solve the funding gap, doesn't increase national savings, and doesn't increase national income.

That's because today's borrowing to fund the transition costs has to be paid back eventually through new taxes. In other words, the private accounts are illusory: they returns they generate will just go to pay the taxes for the borrowing used to fund them. Equivalently, the Bush plan is just a new tax, one that the government returns to workers in the form of private accounts.

What Bush gives with one hand, the private accounts, he (or some future, more responsible President) takes with the other in the form of higher taxes. The "ownership society," he calls it.

More Depressing News From Iraq

2004-11-30T01:42:00.000-05:00

A couple of weeks ago, I posted a comment on Crooked Timber about the Lancet study's estimate of violent death in Iraq:

The numbers I find fairly plausible are the figures for the post-war violent death rate. I think the authors’ definition of a household is less problematic over shorter time periods.

They don’t actually seem to report the violent death rate specifically, but some calculations show it to be 1.8 per 1000 (6 per 1000 with Falluja). Which translates to 44,000 per year (152,000 with Falluja). So, they’re pretty big numbers.

In the US, very high murder rates are something like 0.7 per 1000 (in places like Detroit in the early 1990s). So living in Iraq today seems to be something like living in the worst neighborhoods of Detroit, DC, or New Orleans during a high crime year.

So, even though I don’t believe the study’s pre-war figures, I think these are pretty depressing and believable numbers.

Today, Crooked Timber links to another depressing story from Iraq, about the soaring murder rate there:

“Our morgue was designed to cope with between five and ten bodies a day,” explained Kais Hassan, the harrassed statistician whose job it is to record the capital’s suspicious deaths. He gestured into the open door of a refrigeration unit at the stomach-turning sight of tangled corpses inside, male and female, shaded with the brown and green hues of death. “Now we’re getting 20 to 30 in here a day. It’s a disaster.”

This is just the latest in an endless series of these stories. The NY Times ran a good story a few months ago, and there have been many others. It seems like every reporter in Iraq pays a visit to the Baghdad morgue (but only a single visit).

The Baghdad figures, which count "suspicious deaths" that the morgue investigates as possible murders, provide a nice check of the Lancet numbers. The morgue numbers suggest a post-war suspicious death rate in Baghdad of 1.5 per 1000, double or triple the pre-war rate of 0.5 or 0.7 per 1000. (Click on "###" at the end for details about the calculations).

The post-war rate is right in line with the Lancet figure of 1.8 violent deaths per 1000 in Iraq excluding Falluja. But the Lancet estimate of 0.1 violent deaths per 1000 before the war is very different from the morgue figure. Admittedly, these are pretty back-of-the-envelope calculations, but I think they provide some evidence that the Lancet study got the post-war death rate fairly right, but botched the pre-war rate.

The London Times quotes a morgue doctor who offers a pretty good summary of the whole mess.

The mortuary staff cannot agree whether the present situation could be described as better or worse than that which existed under Saddam Hussein...The staff also remember when hundreds of victims of mass execution were dumped by the Baathist authorities at the mortuary and relatives were too frightened to collect them.

"Better or worse is irrelevant — they’re both bad," Dr Hassan said. "And it could have been so easy for the Americans. Why did they disband the army and police last year and allow those weapons and munitions to pour into the hands of criminals in our streets? Why did they leave us for a year with no national army and police? I don’t know. Now we all suffer — them and us. Am I depressed? All the time."

Figures from the NY and London Times suggest that there will be about 7500 "suspicious deaths" investigated as possible murders by the Baghdad Morgue by the end of this year. In 2002, before the war, there were 3,500 such deaths according to the Boston Globe, while the NY Times suggests that pre-war rates of suspcious deaths were about a third present rates.

Baghdad is a city of about 5 million people, which gives us a "suspicious death" rate of 1.5 per 1000 after the war, and 0.7 per 1000 before the war (or maybe 0.5 if the NY Times is right). So the post-war violent death rate is right in line with the Lancet's estimate of 1.8 per 1000 (with the caveat that suspicious deaths in Baghdad aren't the same thing as violent deaths in Iraq). But the pre-war violent death rate in the Lancet was 0.1 per 1000, very different from the rate suggested by the figures from the Baghdad morgue.

Some Notes on Medical Innovation (with links)

2004-11-23T01:42:00.000-05:00

It's too late at night to write anything coherent about the vast problem of how to fund medical research, so I'm going to try brain-dumping some notes that I hope to develop in future posts. When reading this blog entry, just pretend that each paragraph is a Powerpoint slide.

Dean Baker has proposed establishing an NIH agency to commercialize drugs that would then be made available for free to the public. Dennis Kucinich has introduced a (sketchy) bill.

What is this supposed to accomplish? Lower drug prices. Medical innovation determined by social priorities rather than profit.

It's true there's a conflict between social priorities and profit. An example is that new uses for off-patent drugs may not be profitable to research. Also unprofitable are drugs for poor people such as anti-malarial drugs for the third world. Hollis provides other examples. Kucinich and Baker also think socially unworthy "lifestyle" drugs are researched instead of cures for deadly diseases. This last is the kind of ascetic position I'd expect from vegetarian like Kucinich.

But the proposed agency will also face a conflict: whether to fund research that leads to lower drug prices or more innovative research. An example is "me-too" drugs which lower prices but don't offer much benefit. Baker endorses this kind of research. "New uses for old drugs" could be very valuable medically but won't lower prices, since the drugs are already off patent. Baker and Kucinich don't seem interested in this.

How is the agency supposed to develop new drugs? Presumably by funding clinical and applied research instead of academic research. That's what Baker says and it's probably an accurate-enough description. But this is an oversimplication. See this debate between an industry blogger and an academic blogger about the role played by the government, academia, and industry in developing new drugs.

The present regime (Bayh-Dole Act of 1986) developed precisely because academic research wasn't being commercialized. You don't get tenure for saving lives! (Or for developing drugs or transferring knowledge to industry). So the law gave patent rights to government-funded research to the researchers (often academics). Most universities now actively try to patent and license their research. See GAO, MIT study, Sampat survey.

"Principal-agent problem"--NIH doesn't want to do it. "Phase III clinical trials" -- the final and most expensive stage of developing a new drug -- are boring. High-powered researchers don't want to do it, since it's mainly a management task (recruiting patients and collecting data). Academics and the NIH probably don't want to do it: they want to discover new proteins and such -- make exciting new breakthroughs. That's why industry needs to be bribed to do it. That's why Bayh-Dole bribes universities and academics to do it. If we give tens of billions of new dollars to the NIH, why should we expect it to be spent on clinical/applied research? It will just be spent on the kind of academic research that scientists enjoy and value. Baker and Kucinich have given no thought to this problem.

The current system of medical research is vast and complicated, developed over 100s of years, and enshrined in the Constitution (patents). The old-fashioned conservative position -- respect for the wisdom embodied in traditional practices -- has a lot to be said for it. A few hundred million or a billion for a pilot program is a good idea, but we ought to be cautious before tinkering with a complicated system that's working pretty well.

Cheaper Drugs

2004-11-22T02:17:00.000-05:00

My weekly post is up on Angry Bear, arguing that we ought to decouple proposals for alternative ways to finance research and development of new drugs from the goal of making drugs cheaper.

Flu Snafu: FDA Knew

2004-11-18T23:44:00.000-05:00

The Washington Post's David Brown reported yesterday that the FDA knew about contamination problems at the Chiron flu vaccine plant a year ago, but did almost nothing -- even though the company asked for help.

FDA managers overruled its inspection team [in 2003] and made its fixes voluntary rather than mandatory.
...
The one FDA action he said was a mistake was the agency's failure to send its full report on the June 2003 inspection to Chiron in September 2003. That was when the FDA decided to make its improvements voluntary, not mandatory. Chiron requested and received the full report in June 2004.
...
The company had asked for a meeting with FDA officials after the 2003 inspection, but the agency never granted one, apparently because it did not view it as necessary.
...
[With minor exceptions], the FDA dealt with Chiron entirely from a distance, through letters, e-mails and telephone calls.

Comparing these latest revelations to press reports from before the election, the news is not so much that the FDA knew about the plant's problems a year ago, but how little action they took. After the election, it's front page news with the headline "U.S. Knew Last Year of Flu Vaccine Plant's Woes." Last month, it was in the business section, with the headline "FDA denies knowing scope of plant's mess."

Don't Re-Frame, Organize

2004-11-18T01:57:00.000-05:00

Via Brad DeLong, I learn of a fascinating article written by someone who spent seven weeks knocking on the doors of undecided voters in Wisconsin.

Time after time, undecided voters would agree vociferously with every single critique I offered of Bush's Iraq policy, but conclude that it really didn't matter who was elected, since neither candidate would have any chance of making things better.

So reports Christopher Hayes of TNR. He concluded that many voters have a

fundamental lack of understanding of what constituted the broad category of the "political." The undecideds I spoke to didn't seem to have any intuitive grasp of what kinds of grievances qualify as political grievances. Often, once I would engage undecided voters, they would list concerns, such as the rising cost of health care; but when I would tell them that Kerry had a plan to lower health-care premiums, they would respond in disbelief--not in disbelief that he had a plan, but that the cost of health care was a political issue. It was as if you were telling them that Kerry was promising to extend summer into December.

I was struck by how similar his analysis is to the views of cynical political scientist Larry Bartels, who has written extensively about the widespread failure to reason about political and policy choices, for example, about the estate tax:

Millions of ordinary Americans say that the federal government should spend more on a wide variety of programs, that the rich are asked to pay too little in taxes, and that growing economic inequality is a bad thing -- yet simultaneously support a policy (estate-tax repeal) whose main effect would be to reduce the tax burden of the very wealthy, constrain funding for government programs, and further widen the growing gap in economic fortunes between the rich and the rest of American society.

Elsewhere [pdf], Bartels points out that this failure isn't confined to the stupid and poorly educated, citing the support of many intellectuals for Stalin in the 1950s, long after it should have become clear that he was a monster.

How ought candidates to respond to this? Many say the answer is a change in rhetoric. Hayes argues that the Democrats need to "rebuild a popular, accessible political vocabulary." Over at the DLC, the New Donkey says "Democrats should focus on developing a broad, national message for change on all the challenges facing the country, since our "targeted" messages [about particular issues such as Social Security], some of which violently oppose "change," don't seem to be succeeding very well." Others speak of adopting the right Lakoffian "frames" for the Democrat's message.

I don't doubt that what candidates say, in speeches and TV commercials, is important. But it doesn't seem like fine-tuning the message is the obvious way to persuade voters who see high health costs as a fact of life, rather than the outcome of political choices. And I don't have high hopes that framing the issues correctly will win votes from those who don't believe that politicians tell the truth, and don't believe that politicians have the ability to make life better, even if they wanted to. I guess improving the message is better than nothing, since people have to vote for one of the candidates.

So what will work? It seems to me that more important than developing new language is building institutions that can speak to voters from a position of trust. Labor unions and the mass media are obvious examples. Ralph Nader, for all his faults, is pretty good at coming up with innovative ways to fund citizen's groups. If I were George Soros, I'd buy People magazine, and order them to run embarrassing pictures of Bush (say, the shot of him giving the finger to the camera) and pictures of maimed Iraqi children.

Which isn't to say that organizing is the only answer, in the absence of the right poetry and prose, message and issues. Just that all three things (and probably more) are needed.

How to Save Social Security

2004-11-17T02:14:00.000-05:00

Now that Larry Lindsey is no longer around to reassure Bush that the numbers add up on his plan to partially privatize Social Security, Bush faces the unwelcome prospect of joining the reality-based community. Lindsey's idea was the the stock market would rise so much that excess gains in the new personal accounts could be taxed to pay the transition costs, keeping the benefits flowing to current retirees.

Fortunately, there's still one economist who believes in a free lunch: Kevin "Dow 36,000" Hassett. The rumor mill says he has the inside line to head Bush's Council of Economic Advisors. Last I heard, he was still predicting that the Dow would quadruple. If anyone can make the Bush Social Security plan work, at least on paper, it's Hassett.

And I bet he could use the work, now that his book's out of print.

[hat tip: bakho]

Vandals for Ignorance and Disease

2004-11-17T01:23:00.000-05:00

Ever wonder why you can't sleep if the room is too hot or two cold? Some people would rather you didn't know.

IOWA CITY, Iowa, November 16 -- Vandals dumped chemicals, damaged computers and freed research animals at the University of Iowa.

Ever wonder why winter feels harsher when it begins than it does after a few months? Some people would rather you remained ignorant.

The vandalism included laboratories where research animals were housed. An undetermined number of mice and rats were missing. More than 30 computers were damaged, university officials said.

Ever wonder why all mammals keep their body temperature between 97 and 100 degrees Fahrenheit, and humans at 98.6? Some people would rather we never learned the answer.

Researchers could not specify the extent of damages to the laboratory or how much research would be set back because they have not been allowed to enter the building since the incident, which was discovered Sunday morning. Some researchers said months or even years of research could be lost if computer hard drives sustained significant damage.

Perhaps you want to take a class in cognitive psychology? Some people would rather you didn't.

Because of uncertainty about the extent of the damage, including the deliberate dumping of chemicals, university police evacuated the building Sunday and don't expect to reopen it until after the Thanksgiving break.

One of the victims wonders just what kind of strange impulses motivated the "Animal Liberation Front."

"What they did to the animals was worse that what they could accuse us of doing," said Mark Blumberg, a cognitive and behavioral neuroscience professor who conducted animal research at Spence Laboratories. "There were animals that drowned because of this. It was horrible. How they think that they're doing something that is for the benefit of animal rights is beyond me."

Blumberg is the scientist whose research into sleep and temperature regulation I've been describing. I don't have the scientific background to appreciate his research, but his peers seem impressed, awarding him the "Distinguished Early Career Award." He's written a popular book too, "Body Heat: Temperature and Life on Earth." I've ordered a copy in solidarity, and because it sounds pretty interesting.

Culture Warrior Juliet Schor

2004-11-15T01:44:00.000-05:00

My latest post on Angry Bear discusses the work of Juliet Schor, the left's counterpart to "family values" advocate James Dobson (and I mean that in the nicest possible way).

Falluja and the Propaganda War

2004-11-15T01:32:00.000-05:00

A commenter asks if I really believe that Juan Cole is wrong when he says "the razing of Fallujah is precisely the sort of action that may provoke an al-Qaeda response and will in any case aid in al-Qaeda's ability to recruit angry young Muslims."

Well, I can hardly claim to be sure, but I really think there's a good chance that Cole is wrong. As far as the fight against terrorism goes, it seems to me that razing Falluja is actually killing some Al-Qaeda fighters or their equivalents (something we haven't done a lot of in Iraq), and if we win, our victory will probably demoralize Al-Qaeda. That's the upside.

Balanced against that is the possibility that we lose, and Falluja falls back under the control of the insurgents in a few months. But that wouldn't be a big change: Iraq would remain a quagmire, just as it is already. So I don't think it would be a big boost to Al-Qaeda's morale.

Do civilian casualties outweigh these factors? I doubt it, because civilian casualties in Iraq are nothing new. I doubt that killing a few more civilians, or even a few thousand more, is going to anger anybody who isn't already plenty outraged by the previous tens of thousands we've killed.

Why do They Hate Us? Does Juan Cole Have the Answer?

2004-11-11T15:33:00.000-05:00

As I've written before, John Kerry probably won the Arab American vote by 2-1, and did better still among Muslim Americans (see Zogby's final exit poll too). This is in sharp contrast to Juan Cole's claim that John Edwards' support for Israel in the VP debate would cost him the support of these voters. Prof. Cole, an expert on the Islamic world, wrote: "With just a slight change in rhetoric, Kerry and Edwards could probably avoid alienating most of these Arab Americans and Muslim Americans."

Cole's mistake was to attribute much more importance to the Israeli-Palestinian conflict than it actually has among Arab and Muslim Americans. These voters were very upset about Ashcroft's infringements on their civil liberties, and also put a high priority on the same issues as most other voters: the economy, the war on terror, social issues, and so on. On a list of eight issues, "Israel-Palestine" was consistently rated as least important by Arab Americans.

My main point here isn't to mock Cole's ineptness when he writes about matters outside of his narrow specialty (that's just a secondary point). My point is that if Cole can misinterpret public opinion so badly in this case, how reliable are his assessments of public opinion in the Arab world?

A typical example of Cole's assessment is: "the razing of Fallujah is precisely the sort of action that may provoke an al-Qaeda response and will in any case aid in al-Qaeda's ability to recruit angry young Muslims." And of course, many analysts make similar claims.

Let me enumerate why it's a lot harder to discern the opinions of Arabs than of Arab Americans.

First, Arab and Muslim Americans, like Cole, are Americans. We can expect Cole to have special insight into the psyche of his fellow citizens, and also into the details of issues that motivate them, such as the economy.

Second, vote choices are concrete and relatively easy to measure, unlike support or membership in al-Qaeda.

Third, there is a fair amount of reliable polling data about the attitudes of Arab and Muslim Americans, while polling data from the Islamic world is sparse, and in important cases nonexistent (e.g., Saudi Arabia).

Fourth, America has a free press and freedom of speech, so we can read what Arab and Muslim Americans have to say about the presidential election, and who their organizations support. This isn't generally the case in the Arab world.

Fifth, Arab and Muslim Americans are a fairly diverse group (Muslim, Orthodox, Catholic, or secular in religious matters; originally from various countries; Democrats and Republicans) but surely nowhere near as diverse as the billion or so in the many countries of the Islamic world.

Sixth, Cole isn't just discussing rhetorical support for Al-Qaeda, which is a view shared by millions or maybe hundreds of millions of Muslims (I don't know). He's discussing Al-Qaeda's recruiting, which is something that involves tiny groups of people: hundred or thousands. And out of hundreds of millions of people, there is a lot of room for small groups to have idiosyncratic views, even views entirely divorced from reality. If you look at the web sites of right-wing hate groups in the US, for example, it's not at all clear that they pay much attention to current events (here's one devoted to "racial holy war").

So I think that anyone who claims to be able to identity with any certainty the effect of the Iraq War on support for terrorism -- much less one particular battle -- is full of it.

New Uses for Old Drugs

2004-11-10T01:24:00.000-05:00

Political Animal Kevin Drum reports that

A new heart drug that targets African Americans reduced deaths from advanced heart failure by 43% and reduced hospitalizations by a third, setting the stage for it to become the first drug approved for only one racial group. [from an LA Times article by Thomas Maugh]

An old study, when re-analyzed, found that a combination of two generic drugs was superior to the usual treatment for Blacks but not for Whites. A few years later, clinical trials enrolling only African Americans found that the two drugs, in combination with a third, were better than the third alone. It's quite possible that the 3-drug combination is effective for people of all races, but this hasn't been tested. One reason for running the clinical trials with Blacks only was U.S. patent law:

[Nitromed] holds a patent on using the combination to treat all races, but that patent expires in 2007. NitroMed has a patent on using it to treat blacks, and that patent is good until 2020, preventing anyone from bringing out a generic version of the combination pill before then. Physicians could, however, prescribe the two drugs individually to either blacks or whites. A dose of the generic drugs costs about 44 cents.

Drum focusses on the racial angle ("Racist Drugs?" he asks) but also says:

Isn't that lovely? Maybe it works for whites too, but there's no money in it so no one's going to bother finding out. For blacks, there is enough money in it to make testing worthwhile, but in the end it turns out that BiDil [the generic combination] is mostly a marketing gimmick that provides a way for NitroMed to charge them more for drugs that already exist. It's almost like the worst of both worlds, isn't it?

I don't think this is fair to Nitromed, and I think this criticism is contradictory. Nitromed says that will probably spend $39 million [p. 45 of the pdf] on the clinical trial. This estimate doesn't seem to include adminstrative overhead, or the opportunity cost of the funds. The treatment probably won't make it to market until 6 years after the start of the trial. Nitromed's money could have been earning interest during those years instead of being spent on study expenses. At a guess, the total expense of the trial could easily be double the direct costs, say $80 million. And, of course, the drug combo might not have panned out.

So Nitromed better be able to "charge them more for drugs that already exist" if we want them to have an incentive to take the risk and invest in the research in the first place. Kevin would like the research to be done to see if the treatment works for other races. If we want the private sector to do it, we'd better insure that there's some money to be made. (The alternative is a bigger public sector role, which I really ought to blog about one of these days).

Really, I think this is a pretty happy outcome, not the "worst of both worlds." There's probably not enough research done to find new uses for old drugs. So I'm glad that you can get a patent for finding a new use for an old drug, or even just packaging two drugs into one pill. It sounds to me like this -- frequently criticized -- aspect of the patent system has worked well in this case.

Revealed: How the Democrats can Recover

2004-11-04T22:38:00.000-05:00

Aren't people making a little too much of the fact that 22% of voters said that "moral values" was the most important issue to them? This statement, from Thomas Friedman's column, is pretty typical:

"The Democrats have ceded to Republicans a monopoly on the moral and spiritual sources of American politics," noted the Harvard University political theorist Michael J. Sandel. "They will not recover as a party until they again have candidates who can speak to those moral and spiritual yearnings - but turn them to progressive purposes in domestic policy and foreign affairs."

Well talk is cheap, and if Democrats can win with a little God talk about social justice or whatever, that's fine with me, although I'm skeptical about the odds of this working. But I think Sandel and Friedman are a little too quick to prescribe a cure, without taking time to make an adequate diagnosis.

I happen to know exactly what the Democrats could have done to win Tuesday's election, and what they have to do to "recover as a party" -- they need to convince 1.5 - 2% of voters to choose them instead of the Republicans. If they persuade religious conservatives to switch that's fine. If they persuade some of the rest of the Republican coalition to switch -- the 60% or so that don't think "moral values" is the most important issue -- that's fine too, and seems a lot more likely.

Maybe nominating a Southern Baptist is the way to win next time, or maybe it's protectionism, as Aussie Economist John Quiggen suggests, or maybe it's something entirely different.

Finally, it's worth keeping in mind that social issues are likely to play out pretty differently next time. The Supreme Court, with three or four Bush appointees, seems likely to overturn Roe v. Wade sometime soon. If that happens, pro-choice Bush voters (about a third of his total), may realize that they're made a terrible mistake.

Think Small!

2004-11-03T22:26:00.000-05:00

Maxspeak has the best commentary at the moment on what the next four years will be like and how progressives can win. I disagree with him, though, when he says, in discussing a future progressive program: "think big. We want vision, not laundry lists."

I say: think small! This is the internet era. Think narrow-casting, not broadcasting.

Here's one example. Kerry probably peeled off the about 200,000 Arab-American votes in key battleground states, according to Zogby. He also did very well among Muslim Americans (not necessarily the same group: the large majority of Arab Americans are Christian). Kerry won 93% of the Muslim vote, according to one survey (although the survey doesn't sound particularly reliable). A huge issue for these voters, who've been targeted by Ashcroft, was civil liberties, a topic which probably didn't crack the top ten list of issues discussed in the campaign.

Thinking small, Grover Norquist wants to defund Democratic constituency groups like unions, trial lawyers, social service organizations, and foundations. What can Democrats do (at the state level, presumably) to help them out?

A shift of about 1.8 million voters would have won Kerry the popular vote, and a shift of much less could have won the electoral vote, if it came in the right places. So, what are Estonian Americans thinking? What was Kerry's stance on Bear Baiting (a hot topic in Maine)? On ethanol? (Undoubtably Kerry was pro-ethanol, a big issue for Iowa corn farmers). Nuclear waste? (Nevada). Admittedly, ethanol and nuclear waste didn't put Kerry over the top. But what if he had proposed something for the Iowa potato growers too?

Election Post-Mortem

2004-11-03T22:08:00.000-05:00

I don't know why Kerry lost (although I like the theory that it was an uphill struggle to begin with) or what the Democrats can do to win next time, but I do have a couple thoughts about the exit polls.

The early exit polls (finished at 6pm, I think) had Kerry up 51-48. The final results seem to be the opposite, with Bush winning 51-48. The exit poll sample was about 10,000 at that point, I recall, which means a shift of 2 points in Bush/Kerry lead is outside the margin of error. The actual shift of 6 points is way outside the margin of error.

Micky Kaus, of all people, has the best explanation for the exit poll's failure I've seen:

Why were the exit polls so off? My nominee is Reason 3 from Mystery Pollster's list--"Voting patterns may be different early in the day." Specifically, angrier voters vote earlier. This year, Kerry voters were angrier, so angry that they lined up at the polls as soon as they could in the morning and got disproportionately counted by the NEP survey-takers. Unfortunately, they could only vote once, and their vote was cancelled by the less angry Republicans who sauntered in later in the day.

This rings true to me. I was anxious to cast my vote against Bush, so I voted early in the day, which I rarely have done in the past.

One other thing the exit polls show is that there's been a shift in Party ID of about 3-4 points. The final exit poll, weighted to reflect actual turnout, found 37% Dem., 37% Rep., 26% Ind. I've been arguing all along that this was happening, since it showed up in so many polls, while others preferred to blame the polls for drawing bad samples. Too bad my final prediction -- a big victory for Kerry -- wasn't vindicated too.

A Nailbiter

2004-11-03T00:47:00.000-05:00

I've been over at a friends house all night, watching TV elections coverage, so I have very little idea what's going on with the elections.

Judging from Mark Kleiman and Josh Marshall's reports, Ohio will be the new Florida. The Republicans seem ready to try and to steal the election by keeping the 300,000 provisional ballots from being counted. That's assuming they haven't stolen it already. According to Marshall: "a lawsuit strategy from Republicans is causing delays and shutdowns in precincts that remained open to allow people who were already in line to vote. Lawsuits create delays; folks leave."

One thing to keep in mind: Kerry has run an excellent campaign. One of Kerry's pollsters, writing on Sunday, argued convincingly that the fundamentals favored Bush. Wars usually help incumbents. The economy, while much worse than it should be, is good enough that election forecasting models based on economic statistics have been predicting a big Bush victory.

My point is, if we wake up tomorrow to a court battle over who was really elected President, we ought to rally around Kerry. He shouldn't have to face criticism for not doing better, or the kind of sniping that was all too frequently directed at Gore.

Early Exit Poll Results!

2004-11-01T21:55:00.000-05:00

Monday, November 1, 9:55 PM. Early exit poll results announced by Fox News a few minutes ago show George Bush heading to a landslide victory with at least 324 electoral votes!

For the second election in a row, Fox scoops all the other networks and is the first to call the election. This year, their call came only 24 hours in advance, but still in plenty of time to beat their sluggish rivals.

Congratulations Fox News! Congratulations President Bush!

[hat tip: Gary]

Another Hilarious HTML Joke

2004-11-01T21:45:00.000-05:00

Seen on a bumper sticker: </BUSH>.

Conservatives & Moderates: Vote Kerry! Vote Gridlock!

2004-11-01T02:55:00.000-05:00

My weekly post is up at Angry Bear, discussing how pleased moderates can expect to be with a gridlocked Kerry Presidency.

UPDATE: 100,000 Killed in Iraq War?

2004-10-31T13:06:00.000-05:00

I see that Fred Kaplan at Slate has scooped me (see my previous post), speaking with Beth Osborne Daponte, who also doesn't believe the Lancet estimates of 100,000 dead since the beginning of the Iraq War:

Daponte (who has studied Iraqi population figures for many years) questions the finding that prewar mortality was 5 deaths per 1,000. According to quite comprehensive data collected by the United Nations, Iraq's mortality rate from 1980-85 was 8.1 per 1,000. From 1985-90, the years leading up to the 1991 Gulf War, the rate declined to 6.8 per 1,000. After '91, the numbers are murkier, but clearly they went up. Whatever they were in 2002, they were almost certainly higher than 5 per 1,000. In other words, the wartime mortality rate—if it is 7.9 per 1,000—probably does not exceed the peacetime rate by as much as the Johns Hopkins team assumes.

Kaplan buries the Daponte critique, and emphasizes -- to my mind, overemphasizes -- questions about the Lancet study's large confidence interval. It is worth noting that Daponte's take on this study is extremely credible.

Daponte, a respected demographer at Carnegie Mellon, had her 15 minutes of fame in 1992 when she publicly challenged the first Bush administration's estimate of civilian casualties. Daponte used demographic methods much like those in the Lancet study, and found that 158,000 Iraqis (military and civilian) died during the Gulf War and its aftermath. For her efforts, Daponte was fired by the Census Bureau, though her job was eventually saved after she initiated legal action and drew strong support from her colleagues and the academic community. She has sinced published her findings about Iraqi mortality in prestigious academic journals.

In other words, Daponte is an expert on the issue and would certainly not be afraid to challenge the Bush administration and endorse the Lancet study if she thought it were correct. I'd guess she'd probably be eager to support the Lancet findings. But she doesn't think those findings are correct.

100,000 Killed by Iraq War?

2004-10-29T23:28:00.000-04:00

According to a new study by Roberts et al published in the prestigious medical journal The Lancet, the Iraq War has resulted in about 100,000 "excess deaths," mainly due to violence.

If the Iraq War has truly resulted in an additional 100,000 deaths over a year and a half, that would greatly undermine what I had thought was an important justification for the war. Specifically, I had hoped that the war would allow us to end the sanctions that were killing 50,000 Iraqi children a year, according to an earlier study in The Lancet (by Ali & Shah).

The Roberts et al study finds that the death rate increased from a pre-war 5.0 to 7.9 per thousand after the invasion (excluding Falluja). Roberts et al exclude Falluja from most of their results because the extremely high death rate there is estimated very imprecisely with their methods. Given Iraq's population of 24 million, this increase in the death rate amounts to an additional 70,000 deaths per year, or 105,000 over the 18 months between the invasion and the study.

Roberts et al Estimates of Mortality in Iraq (Lancet 2004)



                 Pre-        Post-

                 Conflict    Conflict

Death Rate

(per 1000)          5.0         12.3



Death Rate

Excluding Falluja                7.9



Infant Mortality

(per 1000)           29           57

I doubt these estimates are correct.

The problem is that the pre-conflict figures are not consistent with other, better, studies. Ali & Shah's 2000 study of infant mortality in Iraq interviewed a sample 40 times larger. The World Health Organization and US Census Bureau have also published estimates of pre-war mortality in Iraq. These other studies show much higher pre-conflict death rates.

Roberts et al's a post-war death rate of 7.9 per 1000 is only slightly higher than the WHO's pre-war estimate of 7.6 or the Census Bureau's estimate of 6.4. These differences are within the margin of error, suggesting that any change has been too small to measure with the 1000-household Roberts et al survey.

The closest that Roberts et al come to discussing other pre-war estimates is when they state:

The preconflict infant mortality rate (29 deaths per 1000 livebirths) we recorded is similar to estimates from neighbouring countries. [20]

This is true enough. An infant mortality rate of 29 per 1000 is in line with the levels found in places like Egypt, Jordan, and Iran. But it's hardly relevant, since pre-war Iraq was in much worse shape than these other countries. In fact, if you go to the web page cited in the Roberts et al footnote, you find Iraq's pre-war infant mortality rate listed as 107! They make no mention of this.

The Roberts et al post-war infant mortality rate of 57 is in fact much lower than the pre-war rate of 108 estimated by Ali & Shah (although it's not much different from the rate of 62 estimated by the WHO and Census Bureau). If we believe the Ali & Shah figure, which is the basis for the claim that sanctions were killing 50,000 Iraqi children a year, it turns out that most of these deaths have been eliminated. The war, by ending the sanctions, has resulting in about 40,000 fewer Iraqi children dieing each year.

Alternate Estimates of Pre-Conflict Mortality in Iraq



                   Ali & Shah     WHO     US Census

                 1994-99        2000    2000

Death Rate

(per 1000)             --         7.6      6.4



Infant Mortality

(per 1000)            108          62      62



Infant Mortality

in Autonomous Areas

(Northern Iraq)        59          --



Source: Ali & Shaw, Lancet, 2000.  WHO.  U.S. Census Bureau.

Have Voter Registration Drives Paid Off the The Democrats?

2004-10-28T18:17:00.000-04:00

Political consultant James Carville is famously skeptical of appealing to new voters: "You know what they call a candidate who's counting on a lot of new voters? A loser." He's basically right.

Polling figures suggest that this year's massive voter registration drive will likely increase increased Kerry's margin by a percentage point or two. Nothing to sniff at, but still relatively modest. Let's go through the calculations.

Voter registration rates have increased by 5.3 percentage points over the rate four years ago, comparing Pew surveys from October 2000 to October of this year. A total of 82.4 percent of American adults are now registered. If the newly registered vote at the same rate as others, 6.5% of all voters will be new registrants (5.3/82.4).

The Pew figures may be low. At the beginning of the month, an ABC poll reported bigger registration gains than Pew. More recently, Gallup reported an 11 percentage point gain in registration, compared to four years ago. Both surveys have much smaller sample sizes than the Pew figures cited above though.

Polling data on registration rates have one major advantage over the numbers reported by state election boards that have been widely reported: they avoid double-counting. Many state election boards are slow to purge non-voters, so many remain on the rolls long after they've died or moved. I'm pretty sure that I'm still registered at every address I ever lived at in New York state. And reported figures make no distinction between new registrations and people who are just re-registering after a move.

The poll figures don't provide a partisan breakdown, but some recent surveys suggest that new voters are strongly pro-Kerry, by a margin of 59-40 (Gallup) or 60-35 (Ipsos). Combining the Pew figures on new registrants with the new voter polls suggests a net gain of 1.3 points for Kerry (5.3% of the 25 point Ipsos lead), or perhaps 1.0 points (Gallup). If the Gallup registration gain of 11 ponts is right, these figures would double.

So, Democratic gains from voter registration drives are relatively modest. For one thing, much of the gain is presumably just offsetting new voters signed up by the Republicans. Still, every little bit counts: the election could easily be decided by a percentage point or two.

Click on "####" for the table.

End Date of Poll    Adults    RV    %RV

----------------    ------  -----   ----

28-Jun-00           2,174   1,673   77.0

23-Jul-00           1,204     918   76.2

10-Sep-00           2,799   1,999   71.4

8-Oct-00            1,331   1,009   75.8

22-Oct-00           1,263     997   78.9

29-Oct-00           1,963   1,508   76.8

5-Nov-00            2,254   1,829   81.1

TOTAL               12,988  9,933   76.5

TOTAL October       4,557   3,514   77.1

  

13-Jun-04           1,806   1,426   79.0

10-Aug-04           1,512   1,166   77.1

14-Sep-04           2,494   1,972   79.1

21-Sep-04           1,200     989   82.4

26-Sep-04           1,200     948   79.0

3-Oct-04            1,233   1,002   81.3

TOTAL              11,013   8,810   80.0

TOTAL October       2,801   2,309   82.4



Source: Pew, "About the Survey".

French Wine Boycott Fails

2004-10-28T01:30:00.000-04:00

Curious about why Australian wine seems to be permanently on sale at the delis I frequent, I found an article by Frank Vannerson called "Wine, Francophobia and Boycotts," on the Liquid Assets web site. (Liquid Assets is a newsletter for investors in high-end win run by Princeton labor economist and oenophile Orley Ashenfelter).

In January, 2003, just before the Iraq War, anti-French sentiment heated up. As you may recall, french fries were briefly renamed freedom fries. French toast became liberty toast. The makers of French's Mustard felt compelled to point out that they weren't really French at all. French wine sales fell, leading to some early reports of success for the boycott.

I clipped the figure below out of the Vannerson paper. The graph shows the share of the US wine market captured by French, Italian, Australian, and other producers. The data is from supermarkets, so it's mostly limited to low-end wines, under $10 a bottle. The figure shows that sales of French wine have been plummeting while sales of wine produced by our ally Australia have soared. At first blush, this seems to suggest that the boycott was successful.

However, the figure makes clear that French wine sales had been falling steadily for over a year before the boycott, and the post-boycott fall was merely a continuation of the trend. Vannerson's article demonstrates this statistically. At the same time, Australian wine sales have been increasing, by about 80% in less than two years (again, in line with pre-boycott trends).

According to the Worlds of Wine (sort of a wine blog, written by a wine historian) wine imports, at least at the low end, rise and fall with exchange rates.

With the problems in the American economy (deficits, massive trade imbalances), the U.S. dollar has lost value dramatically against the Euro, the currency now common to the biggest European wine-producing states, France, Italy and Spain...As a result, French and Italian wine imports to the U.S. fell 15 per cent in 2003....No surprise that it's the Australians who've hopped in to fill the gap created by falling European imports.

So where the boycotters failed to hurt French winemakers, Bush succeeded -- by running up the deficit and weakening the US dollar.

Source: Adapted from Vannerson, Liquid Assets.

National Sales Tax Hits the Big Time

2004-10-26T23:47:00.000-04:00

I'm heartily sick of the Presidential election. So let's talk about the South Carolina Senate race instead, where a major issue has been the Republican candidate's support for a national sales tax.

Democratic attack ads against GOP House candidates who support a 23 percent national sales tax are causing a stir in several states, with Republicans demanding that TV stations drop them.

The ads, running in seven House districts, target Republicans who support HR 25. The bill would eliminate the federal income tax, estate tax and payroll taxes and replace them with a 23 percent sales tax. The issue has been a mainstay in the Senate race in South Carolina, but the Democratic Congressional Campaign Committee's ads have expanded it to three House districts in Texas and one each in Georgia, Indiana, Kansas and North Carolina.

Indeed, if you look at Inez Tenenbaum's web page (she's the Democratic candidate for Senate in SC), most of the ads criticize her opponent for supporting a national sales tax. Here's a typical excerpt:

My opponent, Jim DeMint, has a big idea. A new 23 percent federal sales tax on just about everything we buy. Like milk, bread and groceries. Clothing, new tires, going to the movies. Even prescription drugs. What we really ought to do is cut taxes on middle class families.

DeMint, along with Tom DeLay and over 50 Republican members of Congress, has endorsed a national sales tax that would replace almost all federal taxes. Tenenbaum's ads are a little misleading, because they don't mention that DeMint envisages eliminating other federal taxes. (she has a press release that's more accurate). On the other hand, she doesn't challenge DeMint's 23% figure, even though most experts say a rate of 50-60% would be needed for a "revenue-neutral" replacement tax.

Also, her implication that the sales tax really amounts to a tax increase is basically correct. According to the Institute on Taxation and Economic Policy, a revenue-neutral sales tax would amount to a 16% average tax increase [pdf] for South Carolinians, an additional of $1,731 per year. The ITEP is a liberal-leaning think tank, but this conclusion is hard to dispute. The sales tax is regressive, and South Carolina is a relatively poor state. So even if taxes stayed the same on average in the country as a whole, South Carolina and other low-income states would be pay a bigger share.

The regressivity of the Republican consumption tax proposal is truly breathtaking. According to the ITEP, the poorest 20% of families would see their taxes rise from $455 to $4,432, almost a 10-fold increase. The wealthiest 1% would see their tax bill fall by almost two thirds (from $213,000 to $77,000). The increase for the poorest 20% is almost literally breathtaking -- taking your breath away is ultimately suffocating, after all -- since the poorest 20% in South Carolina currently get by on an average of $8,245 after federal taxes. Under the Republican's sales tax plan, they'd have to squeeze by with $4,268.

One More Progressive Criticizes Cole

2004-10-26T22:56:00.000-04:00

Juan Cole has called on his readers to defend Professor Joseph Massad of Columbia University, under attack from "a concerted campaign..by the American Likud, aimed at getting him fired." It's not so clear what makes Massad's critics -- the NY Daily News, NY Sun, and Congressman Tony Weiner -- "American Likud," but we'll let that slide.

To his credit, Cole doesn't defend Massad's ideas, only his academic freedom. I agree that Massad shouldn't be denied tenure because of his political views -- although he's someone who's political views hardly seem distinguishable from his scholarship, which is what tenure reviews are supposed to be about. Looking at some of Massad's writings on the web, my favorite quote is this (from a newspaper column):

"Such practices [as the torture of Abner Louima by NYC police] clearly demonstrate that white American male sexuality exhibits certain sadistic attributes in the presence of non-white men and women over whom white Americans (and Brits) have government- sanctioned racialised power."

Progressive blogger Stuart at the New Appeal to Reason has more, particularly noting the irony that Massad's leading defender (who runs the web site Cole links to) is someone who banned Israelis from serving on the editorial board of her academic journals.

Flu Snafu: Rampant Price Gouging

2004-10-26T01:29:00.000-04:00

Many conservatives and libertarians charge that price gouging laws have contributed to the flu vaccine shortage. It turns out, though, that price-gouging laws aren't very effective. In fact, according the the Washington Post, there are actually firms that specialize in price gouging:

The larger [distributers] are also free to sell to smaller distributors, who then sell to even smaller suppliers that are willing to place orders for flu vaccine early in the season at high prices on the chance that they will be able to resell it for even more when the flu season hits hard. That bet pays off particularly well when dramatic shortages occur.

A salesman for Stat, in a Sept. 13 e-mail sent to pharmacy buyers, predicted this year's turn of events, saying that it was "my favorite time of the year! Hospital beds are filling up" and "all the little darlings are back in school coughing up their diseases . . . " Chiron was having production problems, he wrote, and if that continued, wholesalers "can pretty much ask and get what ever the old market will bare," a copy of the e-mail shows.

[CEO] Alley said his salesman was writing tongue-in-cheek, and that "no one wants to see kids hacking." But the pricing comments, he said, are "the truth. It's happened for the last seven flu seasons."

A recent survey found that 80% of hospitals had been contacted by a company offering to sell the vaccine at $30 or more a dose (four times the price charged by the manufacturer); 20% were offered vaccine at $80 or more a dose (10 times the manufacturer's price). Only about a quarter of hospitals responded to the survey, so these figures are probably overestimates (if hospitals receiving inflated price quotes were more likely to respond to the survey).

The price gougers may actually do some good, if their purchases of vaccine early in the season encourage the manufacturers to make more: they're providing a buffer stock.

But whether price gouging laws are good or bad, I take these two articles as evidence that they aren't that effective. There's an active secondary "speculative" market in flu vaccine, and most hospitals have access to to high-priced vaccine, if they want it. It's like ticket scalping: illegal many places (though not everywhere) but still common as concerts.

The Ownership Society

2004-10-25T02:35:00.000-04:00

My weekly post, with my continuing attempt to understand why Bush designs such horrible policies is up at the Angry Bear.

Libertarians on the Flu Snafu

2004-10-22T00:53:00.000-04:00

I've come across a lot of libertarians blogging about the flu. The best libertarian flu post, by the way, is on Cafe Hayek, mocking the multiple explanations offered for the shortage. Marginal Revolution has a good post too, discussing why drugs offer manufacturers a better opportunity to price discriminate than vaccines.

In general, though, there are two basic problems with the libertarian posts. First, they want to claim that there are price controls or something similar on flu vaccine, though this isn't true (see libertarian one, two, three, four) . Second, although many of them discuss vaccines in general, none of them make the obvious libertarian argument against mandatory vaccination.

Let's take mandatory vaccination first. The government requires vaccination for many childhood diseases (though not the flu), which is enforced by keeping unvaccinated children out of school. The obvious reason for this is that these diseases can be transmitted from child to child, so an unvaccinated child may catch the disease and spread it to others. Hence, vaccinations have positive externalities and ought to be subsidized, which the government approximates by requiring them and providing cheap vaccination shots to children.

One implication is that, for most vaccines, the government is strongly intervening to increase demand, presumably to the benefit of vaccine manufacturers. And yet we've had several recent shortages of childhood vaccines, just as we've had shortages of flu vaccine.

From a libertarian point of view, one would think that mandatory vaccinations are a blow to liberty. Even if libertarians don't think young children should be able to make their own decisions about vaccines, you'd think they would object to parents being forced to vaccinate their children. So why aren't the libertarians screaming about "forced vaccinations" or "vaccination slavery" or at least decrying this reduction in our freedom to go unvaccinated? I'm just wondering.

Second, libertarians and others want to claim that there are price controls on flu vaccines. The strongest counterargument is that the price of flu vaccine has gone up by almost a factor of five since 1996. Many conflate price gouging laws with price controls. But price gouging laws apply to all products, not just flu vaccine. So if these laws are so important, why do they cause shortages only of vaccines, and not other products?

Others conflate large government purchases with price controls. The government does make large purchases of many vaccines, often buying half the supply, which it does by putting contracts out for bid. However, it does not do this for the flu vaccine, despite many claims to the contrary. (Alex Tabarrok initially implied that the government is a major purchase of flu vaccine, but to his credit, has since corrected it). According to the GAO:

"Most influenza vaccine distribution and administration are accomplished within the private sector, with relatively small amounts of vaccine purchased and distributed by CDC or by state and local health departments."

However, even for the childhood vaccines, where the government is a large purchaser, it is not at all obvious that government action, taken as a whole, reduces the demand for vaccines. To repeat the obvious: the government makes childhood vaccination mandatory. Surely this offsets at least some of the price reduction caused by the government's large purchases.

Black Box Voting

2004-10-20T01:47:00.000-04:00

My prediction for the Presidential election is: Kerry wins in a landslide, and then right-wingers spend the next four years charging ballot fraud.

That's what seems to be happening in Venezuela. Today, a friend emailed me an essay about the latest round in the battle over Venezuela's Presidential recall vote of two months ago. Venezuela's election was conducted using a mix of paper ballots and electronic voting. The electronic voting machines produced a paper trail, a print-out of the voter's choice that was deposited in a ballot box. The election was observed by the Carter Center, which, among other things, conducted an audit comparing a large random sample of the paper print-outs to the votes recorded in the machines.

In other words, the election seems to have been run a whole lot better than most US elections. And the result was a landslide, meaning that minor, Florida-style glitches wouldn't matter. And yet the Venezuelan opposition, and US right-wingers, have been screaming about electronic ballot fraud.

There's even been a battle of the econometricians (incumbent/opposition). The opposition first charged that there were a suspicious number of polling places where multiple machines showed the same number of "yes" votes. That is, in one polling place two machines might have both recorded 283 "yes" votes; in another, two machines might have both recorded 391 "yes" votes; this was alleged to have happened too many times. The charge has since been refuted, with both sides' statisticians agreeing that the ties were coincidence, not evidence of fraud.

My impression is that the claim about suspicious ties was absurd on its face, and yet it generated several months of heated charges and counter-charges. And of course, now there's a new charge, with the opposition and their statisticians claiming that the Carter Center's audit wasn't truly random. I don't know whether the new charge is true or not, though I'm inclined to believe the Carter Center, which says the elections were fair.

But either way, I wouldn't be surprised if Venezuela's ongoing fight over ballot fraud (or lack thereof) is in our future. After all, US right-wingers have already been practicing up for the battle.

Chat with Cook

2004-10-20T01:25:00.000-04:00

I've been meaning to post something about last week's Washington Post online chat with astute political observer Charles Cook. Here's a few excerpts that relate to recent blog posts of mine, but Cook has sharp comments on almost every poll topic that's come up this year.

Fooling Ourselves

"My advice to people is to not pay too much attention to any one poll, there is a temptation to cherry pick, to focus on the one or two polls that tell you what you want to see happen the most, and ignore all others as methodologically flawed. I would look at the averages of polls that are published in various places, an average of many polls is most likely to give you a truer picture than any one."

The Best National Polls Trump the State Polls

"Given that a quarter of [state-level] polls are complete garbage and another quarter fairly suspect, I think that [counting electoral votes] is very problematic. Unless someone happens to be privy to the much more sophisticated (and expensive) polling that is being conducted for the two parties, the chances of anyone accurately calling all of the 11 states that we are calling toss ups (Colorado, Florida, Iowa, Minnesota, Nevada, New Hampshire, New Mexico, Ohio, Pennsylvania, West Virginia and Wisconsin) are pretty slim. If the margin in this race is more than one percentage point, the Electoral College vote won't matter, if it is inside of one percent, then there are too many states that are too close and the state level polling, even the good ones, won't be of much use, much less these three-dollar state polls that are flying over the internet....

"...stick to polls that are done over the telephone (NOT internet) and conducted by real live people, not "push #1 for Bush, #2 for Kerry...) like Rasmussen or Survey USA. They have no idea of they are interviewing nine years old or not."

I think Cook's argument for ignoring the state polls is very compelling: if the national margin is over one percent, the electoral votes will follow. If the national margin is closer than that, the state polls will also be too close to tell you anything.

Still, I can't restrain myself from checking electoral-vote.com every day.

Flu Vaccine: "A Very Attractive Business"

2004-10-19T01:17:00.000-04:00

Via Political Animal and Denise Gellene of the LA Times, I learn that two new companies hope to enter the US flu vaccine market. Apparently they see great profit opportunities in the US.

GlaxoSmithKline, the largest vaccine maker in the world, and ID Biomedical, a small Canadian company, have announced plans to sell flu shots in the U.S. ID Biomedical could enter the market as soon as next year.

That would reduce the nation's reliance on two vaccine makers — and the odds of another massive shortage.

The competitive interest in making flu vaccines could dispel the notion that there is no money to be made in the business. In fact, over the last five to six years, the wholesale price of a flu shot has jumped to more than $8 from less than $2, far outpacing increases in production costs. What's more, the market is growing. Demand has risen to about 80 million doses annually from half that in the mid-1990s. U.S. public health authorities aim to vaccinate 150 million Americans annually by 2010.

"It is a very attractive business," said Anthony Holler, ID Biomedical's chief executive.

A very attractive business? No doubt that will come as a big surprise to readers of the mainstream media or conservative blogs, which have subjected us to an endless stream of nonsense like the following:

[Wyeth's] exit is part of a long, slow industry-wide flight away from flu vaccine, which has simply become more trouble than it's worth...Even under the best circumstances, vaccines have never been very attractive investments. [David Brown, "How U.S. Got Down to Two Makers Of Flu Vaccine."]

Readers of Ragout, of course, have known all along just how dubious these claims about unprofitability really were. Two weeks ago I cited flu vaccine maker Chiron's pre-contamination optimism:

[The newpaper article's] claim that it's hard to make money selling vaccines is also disputed by a surprising source: Chiron, which says that flu vaccine prices have risen enough to justify substantial investment in the U.S. market. Chiron's president testified before Congress earlier this year, "Pricing of influenza vaccines has reached a level that allows manufacturers to invest in maintaining facilities to meet FDA standards and in expanding manufacturing capacity in order to meet the increased demand." Prices have risen in the past few years because 3 of the 5 former manufacturers have left the U.S. market.

Perhaps the mainstream reporters can be forgiven for not uncovering these facts. After all, I relied on hard-to-find, almost secret sources of information.

Survey USA and the Virtues of Transparency

2004-10-19T00:23:00.000-04:00

A correspondent takes me to task for being too hard on SUSA, one of the five firms that produce most of the state-level polls. I didn't mean to be. In fact, I think that they're probably the best of the five. (Zogby, another of the five, has a very good reputation, but their state polls are based on internet panels, not telephone polling).

I assume that SUSA is the best of the group because they release by far the most information about their polls, which I think is a very strong indicator of quality. For one thing, it shows they have confidence in their work. For another, and I think this is more important, releasing this kind of information probably helps SUSA understand how well their methods are working so they can catch errors and improve their methods.

This works in other fields too: "We've generally found over the years that it's a good sign when a label gives you a great deal of information about a wine," report some experts (p. 191).

Vaccines, Vitamins, Vegetables, and Price-Fixing

2004-10-17T23:24:00.000-04:00

What does the flu vaccine shortage have to do with rotten vegetables, overpriced vitamins, and price-fixing? Perhaps more than you'd think. Find out more by reading my weekly post at the Angry Bear.

Fooling Ourselves

2004-10-17T16:33:00.000-04:00

According to an ABC News poll, 88 percent of Bush supporters expect him to win. Most Kerry supporters have the opposite forecast: 67 percent expect Kerry to win.

Why the disconnect? I think it's because people tend to pay attention only to information that confirms their prior beliefs, and ignore news that conflicts with their beliefs. For example, if you look at conservative blogger Instapundit's front page today, the only horse race poll mentioned shows Bush with a big eight-point lead. The liberal Atrios, meanwhile, cites two polls, one showing Kerry up by three, another showing Bush up by only two.

This practice of trumpeting good news and ignoring or downplaying bad news is also found, to a lesser extent, on sites more focused on the nuts and bolts of the campaign. If these sites mention bad news, they quickly dismiss it. Bad news polls always seem to be taken on the wrong day, have the wrong mix of Republicans and Democrats, have the wrong likely voter model, or something. The Republican Redstate.org tends to dismiss polls showing Kerry gaining; the liberal Donkey Rising and DailyKos tend to dismiss polls showing good news for Bush.

Wishful Thinking

2004-10-17T16:00:00.000-04:00

This week's prize for wishful thinking goes to the conservative ex-warblogger Steve Den Beste, who has an elaborate analysis of how the polls have been manipulated by the liberal media to show a Bush lead. No, that's not a typo: Den Beste believes that the polls were manipulated by a dozen or so separate "liberal" media organizations in September to show a Bush lead:

The [poll] data for September, however, is clearly an anomaly. The data is much too consistent. Compare the amount of jitter present before September to the data during that month. There's no period before that of comparable length where the data was so stable....
In September, I think there was a deliberate attempt to depress Kerry's numbers, so as to set up an "October comeback". Of course, the goal was to engineer a bandwagon....

No collusion was needed because everyone knew "the script" for September ("temporary Republican convention bounce") and for October ("Kerry comeback because of the debate").

Den Beste's case hinges on the fact that the polls were less noisy in September, which he thinks proves that they were manipulated in some way. But perhaps there's some alternative explanation, that doesn't involve charging every polling organization with fraud? Den Beste is plotting a moving average of polls, and as we get closer to the election, more polls are conducted, more frequently. So an alternate theory is that as we get closer to November, the moving average is based on more underlying data, and so tends to be smoother.

Nah, couldn't be. If the polls are accurate, Kerry's gaining. And all good conservatives know that's impossible.

UPDATE: Tim Lambert, John Lott's nemesis from Down Under, is also skeptical. He shows that the variation in the September polls is about what you'd expect from averages of half a dozen 1000-person polls. One more strike against Den Beste's poll rigging charge.

The Best National Polls Trump the State Polls

2004-10-14T23:17:00.000-04:00

The web is swimming in state-level polls this year. The spiffyest is electoral-vote.com, but they're everywhere. Slate has a state poll map, as does Real Clear Politics, MyDD, the Wall Street Journal, Pollkatz, and no doubt many others. Sam Wang at Princeton does some complicated calculations combining all the polls (he currently gives Kerry a 27% chance of winning).

The state polls are a lot of fun. But I think the national polls are the ones to watch.

In theory, state-level polls should be better. Even though each one typically has about half the sample size of the national polls, their combined sample size is big. And, after all, it's state electoral votes that determine the President (at least usually). But in polling, having a truly representative sample usually outweighs sample size. Overall, the state polls have weaker methodologies than the best of the national polls, and I suspect their samples are less representative.

Five polling companies put out most of the state polls: Rasmussen, Survey USA, American Research Group, Zogby Interactive, and Strategic Vision. All have flaws. Strategic Vision and ARG put out very little information about their methodology, which is a big strike against them in my book. Strategic Vision is a Republican consulting firm, which also raises warning flags.

Zogby interactive, the state poll reported in the WSJ, is an internet poll of a panel selected by Zogby to be representative. (Zogby puts out a national telephone poll too). Zogby claims that 90% of likely voters are on-line and that their weighting allows them to produce results similar to traditional telephone polls, but still this is still a new, relatively untested methodology.

Rasmussen and SUSA are automated "robo-polls," that use a recorded voices and accept touch-tone responses. Here's SUSA's description:

SurveyUSA surveys are conducted in the voice of a professional announcer. SurveyUSA is the first research company to appreciate that opinion research can be made more affordable, more consistent and in some ways more accurate by eliminating the single largest cost of conducting research, and a possible source of bias: the human interviewer.

It goes on, and is fairly convincing, as far as it goes. Rasmussen says pretty much the same, arguing that their recorded voice is actually better than old-fashioned humans.

What neither mention is how they sample within households. The best polling firms take pains to choose someone at random, for example by asking to speak to the adult with the most recent birthday. Neither Rasmussen nor SUSA claim to do this, and I highly doubt that they do. They just "speak" with whoever answers the phone.

Second, SUSA gets a response rate of about 10%, probably because people are more likely to hang up on a robo-voice than a human being. Rasmussen doesn't say, but their response rate is probably similar. For comparison, the best of the national polls get response rates of 25-30%.

Finally, to make up for their less representative samples, all three firms have to weight more intensively than many national polls. In particular, Zogby and Rasmussen weight by party ID. SUSA doesn't say, but my guess is that they do too. Although this methodology has some supporters, the weight of opinion (including mine) is against it.

Since I think the big national polls -- ABC/WaPo (but not their tracking poll), NBC/WSJ (but not their state polls), Democracy Corps (despite their partisanship), CBS/NYT, Gallup/CNN/USA Today, Fox, Pew, and the LA Times -- have the most representative samples, I'll take them over the state polls any day.

Inequality Soars Under Bush's Strong Leadership

2004-10-14T01:39:00.000-04:00

At last! The debates finally featured some discussion of poverty and inequality. Kerry went out of his way to bring up the issue towards the end. Looking over the transcript, he didn't seem to be responding to any question in particular, but he made a strong statement:

The American middle class family isn't making it right now, Bob. And what the president said about the tax cuts has been wiped out by the increase in health care, the increase in gasoline, the increase in tuitions, the increase in prescription drugs.

The fact is, the take home pay of a typical American family as a share of national income is lower than it's been since 1929. And the take home pay of the richest .1 percent of Americans is the highest it's been since 1928.

These figures are very likely from the recent comprehensive study by Piketty and Saez, which would make them figures for 2000. I don't mean to criticize: this study is currently the gold standard of inequality research and I'm impressed that Kerry (or one of his advisors) is talking about it. In contrast, I rarely have any idea where Bush's figures could possibly come from.

The most recent figures available [big pdf] are from the Census Bureau's Current Population Survey (table A-3). They don't cover the truly rich, like the Piketty and Saez study, but they're informative about the trend in inequality. Since Bush took office, the situation has continued to deteriorate.

The figure below shows two common measures of inequality: the ratios of the 90th and 95th percentiles of household income to the 10th percentile. After increasing for 15-20 years, inequality began to level off in the late 1980s. But since Bush took office, it's been on the upswing.

In 1995, a few years into the Clinton recovery, the 10 percent of households lived on $10,501 or less (in 2003 dollars), while the 95th percentile household made 12.9 times more: $135,448. By 2003, the 10th percentile household had advanced by only $35, while those at the 95th percentile enjoyed an extra $18,672 per year, now making 14.6 times more than the lower-income households.

Final Presidential Debate: Instant Analysis

2004-10-13T22:38:00.000-04:00

Kerry wins narrowly. Bush's worst moments were the three times he tried to dodge the question and change the subject to education.

* Jobs being outsourced? Education.
* Higher minimum wage? Education.
* Affirmative action still needed? Education.

It's worth mentioning that the federal government plays a relatively minor role in funding education, covering only about 10% of the costs. The rest is paid for by state, local and private sources. So Bush's 49% increase in federal funding for education, even if true, only amounts to about a 5% increase in the country's spending on education. That's hardly enough to be the solution to all the problems Bush claims will be solved.

Oh, and I noticed during the hugs at the end that Bush is wearing his [back brace/posture support/battery pack/evil alien overlord] or whatever it is, again. Salon has a picture of Bush wearing it while driving on his dude ranch, which leads me to believe that it's a back brace or something similar (it leads Salon to conclude that it's a radio transmitter).

How to Lie With Budgets

2004-10-12T01:45:00.000-04:00

One trick politicians use is to shift money from one budget line to another, and then boast about how much one line-item has been cut or how much the other has increased, as the case may be. It's like an alcoholic boasting about how much less beer he's drinking, without mentioning his newfound taste for whiskey.

Bush is a master at this kind of budget switcheroo. For example, he likes to boast, "we will double the number of people served by our principal job training program." But he doesn't mention that at the same time he's cutting every other job training program by even more.

In my weekly post on Angry Bear, I discuss how Bush seems to be doing pretty much the same with homeland security.

Bush Admits Tort Reform is Unimportant

2004-10-09T01:40:00.000-04:00

In the 2nd Presidential Debate tonight, Kerry said that tort reform (to keep down the cost of lawsuits) was worth doing, but that Bush tries to make it seem more important than it is. "It's less than 1 percent of the total cost of health care," Kerry said.

Bush, making what he seemed to think was a devastating rebuttal, replied

He says that medical liability costs only cause a 1 percent increase. That shows a lack of understanding. Doctors practice defensive medicine because of all the frivolous lawsuits that cost our government $28 billion a year.

The total cost of health care is about $1.6 trillion. A little math shows that Bush's reply is really just a quibble. While Kerry says lawsuits raise health costs by 1 percent, Bush, with his self-proclaimed greater understanding, knows that it's really 1.8 percent!

Republican Social Engineering

2004-10-07T22:17:00.000-04:00

Last week Atrios posted some thoughts about how the tax system encourages people to hold their wealth in extremely illiquid assets.

...The tax system encourages people to hold their money in real estate through the home mortgage interest deduction (and other various programs which encourage people to buy). It also encourages people to save in various investment accounts which can't be tapped without serious penalty until retirement (or, for a couple of other purposes)....The net effect of all of this is, among other things, that even people with relatively decent incomes and who are actually managing to accumulate some wealth, still feel they are living "paycheck to paycheck" (once they've sent their mortgage payment and thrown a few bucks into their retirement account) because most of that wealth can't be converted into cash as needed.

I've long thought that the point of encouraging homeownership is to make people more conservative. The canonical example being that homeowners band together to keep out undesirables -- Blacks, the poor, and so on -- who might lower their property values. Renters don't do this kind of thing. But overall, my thoughts are pretty half-baked.

Conservatives, however, have been doing some deep thinking about how to keep Republicans in power through social engineering. George Will wrote yesterday,

Bush's "ownership society" is another step in the plan to reduce the supply of government by reducing the demand for it...Hence Bush's menu of incentives for private retirement, health, education and savings accounts.
Conservatives hope such measures will encourage aptitudes that will make the welfare state compatible with traditional American individualism and self-reliance. And conservatives hope such aptitudes will result in Republican attitudes, especially among the elderly and other people with portfolios of equities.

Will praises a recent exposition of conservatives' long-run plan to "reduce the demand" for government and create a more Republican electorate by Jonathan Rauch (National Journal, July 26, 2003). I've only skimmed it, but I was struck by this quote from Republican activist and bane of the Greatest Generation, Grover Norquist:

"Twenty years from now...who's demanding extra government if I have a 401(k) medical savings account, I've pre-saved for my old age, I have control over where I send my kids to school? Investing in smaller demand for state power down the road is a rational position."

Uh, maybe the demand for more government will come from people whose investments didn't pan out and face spending their retirement in poverty? So, I'm not sure if Norquist's schemes will pay off, but I am sure that it's important to understand them.

Juan Cole, Conspiracy Theorist

2004-10-07T22:10:00.000-04:00

Middle East expert Juan Cole claimed today that the U.S. is violating international treaties, and common decency, by manufacturing biological weapons.

Juan Cole writes that Saddam Hussein "had to be at least a little afraid of US retaliation, and it actually does have nuclear and biological weapons." As far as I can tell, "it" refers to the U.S. This is a remarkably serious charge to toss off so casually, without any elaboration, citation, or evidence or any kind. Cole is fond of this kind of wild accusation. Last August, he suggested the Israel has stocks of biological or chemical weapons (it wasn't clear which).

Of course, I don't know what kind of secret weapons production facilities the U.S. or Israel might have. But neither does Cole. He's making incredible claims about secret programs, without demonstrating that he has access to the secrets. That's the very definition of a conspiracy theorist.

Flu Vaccine Shortfall, Yet Again

2004-10-06T21:37:00.000-04:00

Yesterday, Chiron corporation, which supplies about half of all U.S. flu vaccine, announced that it was suspending production, due to contamination problems uncovered by British regulators. Although this year's shortage will be particularly severe -- the CDC is calling for those at low risk of flu to voluntarily forgo flu shots -- it isn't unusual. There was also a shortage of flu vaccine last year and in 2000. Other vaccines are also regularly in short supply.

James Bernstein, writing in Newsday, attributes the shortages to the fact that "only two major companies in the world manufacture flu vaccines...Experts say making vaccines is not a good business." Actually, however, there are only two companies licensed to sell flu shots in the U.S. Britons, for example, obtain their flu vaccine from six different companies, and so aren't expecting a shortfall.

Bernstein's claim that it's hard to make money selling vaccines is also disputed by a surprising source: Chiron, which says that flu vaccine prices have risen enough to justify substantial investment in the U.S. market. Chiron's president testified before Congress earlier this year, "Pricing of influenza vaccines has reached a level that allows manufacturers to invest in maintaining facilities to meet FDA standards and in expanding manufacturing capacity in order to meet the increased demand." Prices have risen in the past few years because 3 of the 5 former manufacturers have left the U.S. market.

Last time around, Conservatives blamed this situation on price controls and legal liability issues, and no doubt they will again. Both of these explanations are truly a triumph of ideology over facts. First, there are no price controls on vaccines in the U.S. The government does buy about half of some vaccines, at discounted prices, and distribute them cheaply, but this isn't a price control. Manufacturers aren't forced to sell, for one thing. For another, the government purchases only small amounts of the flu vaccine (see this GAO testimony).

Liability issues are also a red herring. Vaccine manufacturers managed, almost two decades ago, to persuade Congress to exempt them from liability laws (see the GAO report). People who are harmed by vaccines can be compensated through an arbitration process, but can't win sky-high damage awards. Manufacturers continue to complain that lawyers are able to find loopholes in the exemption, and bring their cases to court. But this is minor, compared to liability issues with pharmaceuticals or other products.

UPDATE: I learn from Mark Kleiman that flu vaccine isn't included in the Vaccine Injury Compensation Program that removes injury cases from the courts for some vaccines. But this is largely because there's been relatively little litigation, and vaccine manufacturers aren't that interested in being included.

I see three possible causes of the persistent shortages, one near and dear to conservatives hearts, two more popular with progressives:

* Misregulation, with the FDA making it too hard to become licensed in the U.S., and closing plants with unpredictably shifting regulations. (See the GAO report).

* Monopoly power. With so few suppliers, it's natural to expect that they would try to raise prices by restricting supply. Remember the California energy crisis of a few years ago, or the world-wide vitamin price fixing case.

* Positive externalities. When deciding whether to get immunized, people don't consider the social benefits: flu-free people can't spread the disease to others. Economists usually call for subsidies to overcome this problem.

I have no idea which of these three problems are the most important.

White House Becomes Bush Campaign Headquarters

2004-10-05T23:41:00.000-04:00

Labor blogger Nathan Newman complains that Bush has turned the OMB web page into a campaign commercial. "It looks just like a campaign site. Democrats should demand that Bush compensate the public for the staff work spent on this design," Newman writes.

Of course he's right. I've spent a lot of time poring over Kerry's campaign web site. But I quickly realized that WhiteHouse.gov is a much better place to look for campaign literature than the Bush's official campaign page.

Check out this page on the WhiteHouse.gov, "President George W. Bush: Record of Achievement." It's not particularly different in tone than that equivalent page on his campaign web site. The White House page is actually a more informative campaign document than the GeorgeWBush.com version, and doesn't have the annoying background music

And then there's the federal budget document, which is now filled with full-color pictures of Bush signing bills, Bush with happy children, Bush saluting the flag, etc.

And then there's the $300 tax rebate in 2001, which came with government funded direct mail letter praising Bush,

"We are pleased to inform you that the United States Congress passed and President George W. Bush signed into law the Economic Growth and Tax Relief Reconciliation Act of 2001, which provides long-term tax relief for all Americans who pay income taxes."

And then there's Bush's $13 million taxpayer funded advertising campaign touting the Medicare changes, that reminded many observers of a campaign ad.

This can't be legal, can it?

Cheney Bashes Small Business

2004-10-05T23:00:00.000-04:00

Overall, it seemed to me that Cheney lied a lot less than he usually does, probably expecting close scrutiny of his remarks. But one lie did leap out at me.

Cheney claimed that rolling back Bush's tax cuts for the rich would also raise taxes for 900,000 small businesses, forcing them to lay off employees. This claim has already been thoroughly debunked by Factcheck.org and the Brookings-Urban Tax Policy Center. The real figure is about half of what Cheney claimed.

Bush frequently makes the same claim on the campaign trail, praising small businesses as "job creators," and charging that Kerry's plan will raise their taxes: "Ninety percent of small businesses pay tax at the individual income tax rate, because they're either subchapter-S corporation or a sole proprietorship."

So it was more than a little hypocritical when, later in the debate, Cheney charged that Edwards had used a "special tax loophole" to avoid taxes during his days as a lawyer. The loophole? Incorporating under subchapter-S, of course!

In their stump speeches, subchapter-S corporations are virtuous job creators, but when their opponent starts a perfectly typical corporation of this type, he's a tax dodger. What a cheap shot.

NOTE: I've re-written this post since the first version was too long-winded.

UPDATE: It looks like I missed some lies. Who'd of thought Cheney would lie about meeting Edwards? Still, I don't think it was more than 4 or 5 lies. So I stand by my original claim: a relatively honest night for Cheney.

230,000 New Registrations in Cleveland: Is That a Lot?

2004-10-04T22:04:00.000-04:00

With voter registration deadlines starting to bite, there's been a spate of articles about swelling voter rolls. Most of these articles cite figures that aren't that easy to interpret, such as Cleveland's total this year of "230,000 new registrations, more than double the number in 2000." We don't know how this compares to the total number of voters in Cleveland, how many people registered twice, how many people were re-registering after a move, or how many people left the rolls as they died or moved out of the state.

Buried at the end of their report, ABC reports some more useful figures.

What's ultimately essential is who votes, and current polling suggest higher-than-usual turnout. Interest is high, and registration drives across the country may be having an effect. Compared to an ABC News/Washington Post poll at this time in 2000, Americans are six points more likely to say they're registered to vote; and registered voters are six points more likely to say they're certain to vote, and 18 points more likely to be following the election very closely.

ABC's poll had a sample of 1800 adults, so we have a gain of 6 points (+/- 2.6). Pew always reports figures for the percent registered, so their data is available for a larger sample (although just what period to look at is a judgment call). Pew finds that voter registration rose from 77 to 80 percent in the last four years (+/- 1.6).

In other words, between 4% (Pew) and 7% (ABC) of voters this year will be new additions to the rolls, resulting from the intense registration efforts of the parties and the 527s and the high interest in this election. This assumes, of course, that the newly registered vote as the same rate as the rest of the electorate. The ranks of the newly registered are presumably higher in the swing states, where the registration drives are concentrated and voter interest is strongest.

Polls of registered voters will presumably capture these new additions. But there's a good chance that polls of likely voters are screening them out. With the race looking like it will be tight this year, 4-7% is a large number of voters, plenty big enough to swing an election.

Click on "####" for the table.

         

End Date of Poll    Adults    RV    %RV

----------------    ------  -----   ----

23-Jul-00           1,204     918   76.2

28-Jun-00           2,174   1,673   77.0

10-Sep-00           2,799   1,999   71.4

8-Oct-00            1,331   1,009   75.8

22-Oct-00           1,263     997   78.9

29-Oct-00           1,963   1,508   76.8

5-Nov-00            2,254   1,829   81.1

TOTAL               12,988  9,933   76.5

TOTAL October       4,557   3,514   77.1

          

13-Jun-04           1,806   1,426   79.0

10-Aug-04           1,512   1,166   77.1

14-Sep-04           2,494   1,972   79.1

21-Sep-04           1,200     989   82.4

26-Sep-04           1,200     948   79.0

3-Oct-04            1,233   1,002   81.3

TOTAL               9,445   7,503   79.4

TOTAL Sept/Oct.     6,127   4,911   80.2



Source: Pew, "About the Survey".

Conservatives' Turn to Bash the Pollsters

2004-10-04T21:05:00.000-04:00

A couple weeks ago, Kerry was behind in the polls, and liberal poll analysts rushed to attack the pollsters, charging that the Republicans' gain in Party ID proved that the polls' methodology was defective. Even MoveOn joined the chorus, taking out an ad to attack Gallup. Since this shift showed up in half a dozen of the most respected polls, there seem to be an awful lot of incompetent pollsters out there.

Now Kerry is surging in the polls, and it's the conservatives' turn to denounce the pollsters. Wishful thinker John Fund, writing in the Wall Street Journal, spouts exactly the same argument as the liberals did a few weeks ago.

If you buy Newsweek's methodology, one out of nine voters has changed their party affiliation in the last month. That might explain why the poll found that one out of eight voters changed their mind on how they will vote too. But how many people do you know who have switched parties and/or changed their mind on the presidential race in the last month? Frankly, the Newsweek poll is more volatile than any electorate could be.

Fund's article comes highly endorsed by Donald Luskin, so it's no surprise that the reasoning is "poor and stupid," with the claim about the volatility of the electorate backed with no evidence at all. Fund just knows better than the data, using the Jimmy Breslin "common sense" technique, I suppose.

In fact, researchers who have actually looked at the data have found that to have 1 in 9 voters change their answer to a Party ID question over the course of a month isn't at all unusual. Pew researchers on several occasions have re-interviewed poll respondents soon after a presidential election. It turns out that when you ask the same people the same question about party affiliation, between 1 in 4 and 1 in 6 give a different answer than they had a month or two prior to the election.

Bush: Election Year Moderate

2004-10-04T03:34:00.000-04:00

My weekly post on Angry Bear discusses Bush's attempts to cast himself as a moderate by recycling his broken promises on health care from 2000.

Regular posting will resume this week on Ragout. I hope to evaluate Bush's newfound love of community colleges and job training programs, figure out exactly what those "community health centers" that Bush keeps promising are, give the lowdown on soaring voter registration, and maybe even discuss my position on the Iraq war, as one commentator has requested.

The Race is Still Kerry's to Lose

2004-09-29T17:58:00.000-04:00

The latest Democracy Corps poll has Bush ahead 49-46 in a 3-way matchup. The survey is conducted by Democratic partisans James Carville and Stanley Greenberg, but it's pretty much in line with most other polls (except Gallup). The Democracy Corps survey also includes a generic question about Presidential preferences: "Now let me ask overall, do you think the country should continue in the direction Bush is headed or go in a significantly different direction?" Bush doesn't do as well on the generic question: the split is 47-51, against Bush's direction.

Since the same group of people were asked both questions, this means that 2% of voters are currently supporting Bush but think he needs to take the country in a different direction. And 5% of voters think we're heading in the wrong direction, but aren't supporting Kerry. So there's a majority that has doubts about Bush, but haven't been sold on Kerry yet. The debates will be Kerry's best opportunity to convince the swing voters that he's an acceptable alternative to Bush.

"The Uninsured are Sicker and Die Sooner"

2004-09-28T19:27:00.000-04:00

Many people believe that the uninsured are able to get access to health care: at the emergency room, at the county public hospital, surely somehow. A 1999 poll found that 57% of Americans believed that the uninsured are "able to get the care they need from doctors and hospitals." Although it's true that the uninsured get some care, they get a lot less than those with insurance, and suffer worse outcomes. In short, a lack of health insurance is deadly.

The Institute of Medicine has issued a series of reports about the health outcomes of the uninsured. The IOM is part of the National Academy of Sciences, so these are consensus reports of a committee of experts, and about as authoritative as it gets. The overall conclusion is:

Uninsured adults have a 25 percent greater mortality risk than adults with coverage. About 18,000 excess deaths among people younger than 65 are attributed to lack of coverage every year.

Now, insurance coverage is obviously correlated with age, income, race, and other factors that affect health. The studies discussed by IOM did try to control for these factors, but probably didn't succeed perfectly. So I was most impressed by the findings that among people with the same illness, the uninsured got less care and were more likely to suffer bad outcomes. The IOM found this over and over: diabetes sufferers got less regular exams, AIDS patients got less effective drugs, even uninsured car crash victims receive less treatment:

To see how uninsured patients fare in a hospital setting, the committee focused on two conditions for which most people are treated regardless of whether they are insured: traumatic injuries and heart attacks. It found that uninsured persons with traumatic injuries are less likely to be admitted to the hospital, receive fewer services if they are, and are more likely to die than insured victims. One statewide study of car crash victims discovered that uninsured victims had a 37 percent higher mortality rate. Another statewide study found that although uninsured trauma patients were just as likely to be placed in intensive care, they were less likely to be operated on or to receive physical therapy.

Where's the Chef?

2004-09-28T18:50:00.000-04:00

I'm spending a week on the west coast, combining a vacation with a little work. I hope to make it home for the first debate on Thursday. I hadn't really planned to blog much this week, but I'm an addict.

More Tolls, Less Gridlock

2004-09-27T04:15:00.000-04:00

The gang at Angry Bear has invited me to co-blog on a regular basis with them, and I've gladly accepted. I'll be regularly posing there once a week on Sunday nights/Monday mornings. I've kicked off the new gig with a post about the wonder and glory of congestion pricing.

Has the Campaign Begun Yet?

2004-09-25T01:32:00.000-04:00

Lots of people (including me) are pretty angst-ridden about the polls showing Bush 5 points ahead or so. But I think the truth is that the swing voters are just starting to pay attention. At this point, somewhere between 86 and 98% of voters are willing to say that they favor or lean towards one candidate or another. It depends a lot on the poll and how the question is asked: 2% undecided in Gallup (with leaners), 14% in Battleground (without leaners), and the rest in between, since mid-September.

But a lot of voters still say they might change their minds. In the Pew poll a few weeks ago, only 80% of voters say there's no chance they could switch. Specifically, 38% of non-Bush supporters say they're "definitely" not voting for Bush, and 42% of non-Kerry supporters say the same.

In the LA Times' 1996 exit poll, the last time we had an election with an incumbent, 29% of the voters said they made up their minds sometime after the debates.

LA Time 1996 Exit Poll
"How long ago did you finally decide how you would vote for president today?"


                          Percent   Cumulative

Today                        6          6

Yesterday                    3          9

Over the weekend             2         11

During last week             4         15

After the presidential

debates                      14        29

After the conventions        9         38

During the primaries         13        51

Before then                  49       100

One Reason Why Houses are So Expensive

2004-09-24T01:34:00.000-04:00

I'm spending a few days in Canada and as usual, I'm hearing about the dispute over U.S. tariffs on Canadian softwood lumber imports. The issue doesn't get much coverage in the U.S.: a Google news search shows that U.S. papers just reprint the wire services story, if they run anything at all. But there's tons of coverage in Canada, where it wins us no friends.

The facts are pretty simple: the U.S. imports about a third of its lumber from Canada and imposes a 27% tariff, arguing that Canada subsidizes its lumber industry by selling trees from government owned land too cheaply. The tariffs were imposed by the Bush administration in 2001, hoping to win some votes from the lumberjacks and sawmill owners, I suppose. Before that, the Clinton administration imposed quotas on Canadian lumber. The US has lost a series of NAFTA and WTO cases, most recently last month, but continues to appeal, dragging out the issue and keeping the tariffs in place.

While this isn't the most important issue facing us, it's not trivial either. There's about $5000-7000 of lumber in a typical house, so homebuilding industry estimates that these tariffs increase the cost of a new house by $1000 seem pretty reasonable.

Bush Takes Cops Off the Beat

2004-09-23T02:19:00.000-04:00

In the first issue of the Economists' Voice, Yale Crime expert John Donohue evaluates ex-President Clinton's claim to a share of the credit for the fall in crime during his adminstration. Donohue writes,

With an astonishing 40 percent drop in the murder rate and a 33 percent drop in the violent crime rate during his two terms in office, Clinton certainly had much to crow about. In sharp contrast, there has been virtually no drop in crime in the last four years.

Donohue gives Clinton credit for reducing violent crime by about 6 to 8 percent, mainly by providing federal funds to local authorities to hire police, which put about 80,000 new cops on the street. In contrast, Bush has pretty much eliminated this funding, perhaps because a lot of it went to big cities where he gets few votes. And local communities have been reducing the size of the police force, after having gone on a hiring spree during the Clinton years (see the graph below). John Kerry has promised to restore funding to hire 100,000 new police officers.

Even leaving aside the increased crime Bush's cuts have probably caused, this is also one way in which Bush's policies have raised state and local taxes. Local governments haven't laid off all 80,000 new cops. They've kept many of them on the beat, replacing the federal money with local money.

I'm pretty impressed by the first article I've read in this new web publication. Kudos to editors Stiglitz, DeLong, and Edlin.

Police per 100,000, from Donohue (Economists' Voice 2004)
.

The Skies Look Good For Kerry?

2004-09-21T01:06:00.000-04:00

A few weeks ago, I discussed the research of Princeton political Scientists Achen and Bartels, who found that voters often punish incumbents in November for bad weather from May to October. According to their data, bad weather cost Gore half a percentage point in 2000, compared to average weather, or 2.5 percentage points compared to perfect weather. The reasons for this are pretty mysterious: perhaps rain or drought puts the voters in a bad mood. My earlier posts speculate about the possible reasons why the weather should matter [post one, post two, older version of Achens & Bartels paper]

So, how's the weather been this year? I've updated Achens & Bartels' "weather index" using data from May through August of this year. It turns out that the weather's been a little worse than average, especially in swing states. The weather's been especially bad in Colorado, Nevada, and Arizona (drought) and also Ohio, Pennsylvania, West Virginia, and Wisconsin (rain and flooding). Florida has had good weather, according to the Achen & Bartels measure, but presumably the hurricanes have changed that.

The table below lists the swing states and their "weather index." Higher numbers mean more drought or more rain/flooding. A weather index higher than 1.8 is worse than average. The next column multiplies the weather index by the factor estimated by Achen & Bartels for the 2000 election. The factor translates the weather into the number of percentage points the weather costs the incumbent.

You can also see the National Climatic Data Center's discussion of the weather here, and a cool map here. The map shows areas with a long term moisture deficit or surplus, which is the measure Achens & Bartels used. Basically, white areas count as good weather, and anything colored counts as worse than average. An awful lot of the map is colored in, so things look pretty good for Kerry.

I also plotted the Bush's gains this year against the weather, to see if Achens & Bartels' measure could "predict out of sample." The vertical axis shows Bush's standing in the polls as of September 13, less his percentage in the same state in 2000. For example, Bush is polling 3 points ahead of his 2000 showing in Arizona, and about the same as 2000 in Florida. The horizontal axis shows the predicted points lost due to the weather (again, compared to perfect weather).

If you fit a line through the scatterplot, it's upward sloping and statistically significant. But that's the wrong slope! So far, Bush has been doing better (exceeding his 2000 results by more) in states with worse weather! Of course, the race could change a lot in the next six weeks and probably will, but so far the weather doesn't seem to be having the impact predicted by the political scientists.




                                  Percentage Points Lost 

                       Weather    by Bush (compared      

                       Index*     to perfect weather)    

                       -------    ---------------------- 

 Missouri               1.2       1.6                  

 Arkansas               1.2       1.6                  

 Florida                1.2       1.7                  

 Maine                  1.4       1.9                  

 North Carolina         1.5       2.0                  

 New Hampshire          1.6       2.1                  

 Minnesota              1.6       2.3                  

20th Century Average    1.8       2.5                  

 Iowa                   1.9       2.6                  

 Oregon                 1.9       2.7                  

 Washington             2.0       2.8                  

 New Mexico             2.1       2.9                  

 Tennessee              2.3       3.2                  

 Michigan               2.4       3.3                  

 Virginia               2.6       3.6                  

 Wisconsin              3.0       4.1                  

 Colorado               3.1       4.2                  

 Nevada                 3.2       4.4                  

 Ohio                   3.3       4.5                  

 Pennsylvania           3.5       4.8                  

 Arizona                3.8       5.3                  

 West Virginia          4.2       5.8                  

  

Note: Weather Index calculated based on PHDI for May-August,

2004 as in Achen & Bartels (2004).



Source: Authors calculations, data from NOAA.

Posted by Hello

Novak: We've Lost the War

2004-09-20T08:43:00.000-04:00

Prince of Darkness Bob Novak announced in his column today (not yet posted on the web) that Bush plans to withdraw U.S. troops from Iraq, even if withdrawal will leave Iraq to civil war. Novak doesn't mention it, but a U.S. withdrawal would leave neighboring Iran in a strong position to dominate the country.

The military believes "there are insufficient U.S. forces in Iraq to wage effective war," writes the conservative columnist. According to Novak, whose administration contacts are stellar, "well-placed sources in the administration are confident Bush's decision will be to get out" after the U.S. elections. "

If Novak is right, it would be nice to see Bush facing reality for once. But Novak's other prediction is scary and all too plausible given Bush's steadfast refusal to fire anyone despite the most egregious failures. Novak says Paul Wolfowitz and Condoleezza Rice will be promoted as a reward for the Iraq disaster, becoming the Secretaries of Defense and State.

NBC's Helpful Advice to Kerry

2004-09-19T18:47:00.000-04:00

Elizabeth Wilner, political director for NBC News, explains that "today's crop of Democrats are right to worry that they lack the laser focus and ruthless efficiency of the current GOP."

Her proof? The Bush Campaign's "hard-nosed," "ruthlessly efficient" response to the CBS memos about Bush's National Guard string-pulling:

"GOP partisans outside the Bush campaign quickly questioned the authenticity of the documents that CBS claimed to have unearthed, while the Bush campaign itself stayed out of the fray."

Compare that to the "sluggish" Kerry response to the Swift Boat lies:

"Kerry advisers insisted that taking the high road amid attacks on Kerry's Vietnam service by the Swift Boat Veterans for Truth was the way to go."

Now some might suggest that the key difference is that the Republicans have a media empire -- Fox News, Clear Channel, talk radio -- while the Democrats don't. The pro-Republican media was quite willing to relentlessly repeat the Swift Boat charges, despite the overwhelming evidence that they were lies.

But for NBC's political director, such a theory doesn't even warrant any consideration. To her, it was obviously foolish and incompetent for the Kerry camaign to take the "high road," but brilliant for the Bush campaign to stay "out of the fray."

Cell Phones Killing Polls?

2004-09-18T01:20:00.000-04:00

Via electoral-vote.com, I learn that Jimmy Breslin, a great reporter and schmoozer, but not such a great statistician, has proclaimed the death of polls:

Anybody who believes these national political polls are giving you facts is a gullible fool.

Any editors of newspapers or television news shows who use poll results as a story are beyond gullible. On behalf of the public they profess to serve, they are indolent salesmen of falsehoods.

This is because these political polls are done by telephone. Land-line telephones, as your house phone is called.

The telephone polls do not include cellular phones. There are almost 169 million cell phones being used in America today... There is no way to poll cell phone users, so it isn't done.

He's wrong about the last point: the reason why pollsters don't call cell phones is because it's illegal, just as it's illegal for telemarketers to call cell phones.

Breslin is right that cell phones present a problem, but they don't present a very big problem, or even a very new problem. As of February of this year, about 6% of households have dropped land lines in favor of cell phones, according to the FCC [pdf; page 2, footnote 2]. That means they can't be reached by telephone polls. But this isn't anything new: 20 years ago 8% of households didn't have telephones at all. Even today, 6% still don't have either a land line or a wireless phone.*

More generally, cell phones are just a new reason for low response rates. In any poll, pollsters never contact most of the people they set out to, even though these people have land lines. People use caller ID or answering machines to screen their calls, they hang up on pollsters, or they're just never home when the pollsters call.

Response rates have been dropping over time. Today, most telephone polls have response rates of less than 30% [pdf]. Response rates of 10% [pdf] are common. This sounds like an awful problem, but there are numerous studies that show that response rates don't actually affect the results that much. Samples with low response rates remain pretty representative demographically (in terms of race, age, sex, and education) . (See this brief review of the literature by ABC's director of polling, and this brief review by two high-powered academics, presumably consulting for the British polling firm YouGov).

Even if, say, young people are less likely to be found by pollsters, because they only have cell phones or because they're always out partying, or whatever, that's not enough to conclude that the polls are biased. Weighting the poll data to compensate for any unrepresentive demographics can always adjust for it. To have problems, the young people with cell phones instead of land lines also have to have different voting patterns than other young people. This is possible, but I don't see any obvious reason why this should be true.

Casual evidence also suggests that low response rates don't matter much. Overnight polls get pretty much the same results as polls taken over longer periods (which get higher response rates because they call people back multiple times). And, as I pointed out in a post yesterday, election eve polls are typically spot on, at least within a couple percentage points of the final result.

So, adding cell phones to the mix could turn out to be the final burden that breaks the camel's back. But they're really just one of many things that make pollsters lives difficult, and all the other problems haven't caused polls to become worthless so far. It's definitely too early to throw out the polls and go with the Breslin plan of just making stuff up. "Common sense would say that the majority of the 18 to 25 who do vote would vote for the Democrat," he writes. Although common sense would certainly be a much cheaper way of gauging opinions, I'm pretty sure that polls remain a lot more reliable.

* FCC, [pdf] table 1, page 6. My guess is that some cell phone only households are answering the survey question wrong and the true percentage of households without any phones at all is about 5%.

Addendum: You can also read the response of pollster John Zogby, who Breslin misquoted, on Rox Populi.

Which Pollster is the Best?

2004-09-17T01:10:00.000-04:00

Answer: all the major firms did pretty well in 2000, except for Rasmussen and Battleground:

* Gore/Bush Tied: Harris phone, Harris interactive (email).

* Gore +1/+2: Zogby, CBS
* Bush +1/+2: Gallup/CNN/USA Today, Pew, IBD/CSM/TIPP, ICR

* Bush + 3: NBC/WSJ, ABC/WashPost

* Bush + 5: Battleground

* Gore + 9: Rassmussen

Who Loses When Businesses Drop Health Insurance?

2004-09-17T01:01:00.000-04:00

An anonymous commentator asks, "I don't quite understand the argument that workers will be left out in the cold [when business stop providing health insurance]. Isn't that equivalent to saying that they will experience a decrease in their effective wages, and wouldn't employers reduce wages now if they could?" That's a good question that I didn't really give much thought to in my previous post.

Kaiser estimates that 2.6 million workers will lose health insurance under the Bush plan. The idea is that at some firms, most workers would prefer to buy health insurance in the private market, garner Bush's new tax breaks, and get higher wages instead of employer-based health insurance.

The catch is that some workers get large benefits from employer-based health insurance -- the old and the sick, those with sick children or spouses. They won't be able to buy matching coverage in the private market except at a much higher price than their raises, if they can buy it at all.

By the same token, some workers don't get much benefit from health insurance -- the young and healthy. They may choose to save their money and forego health insurance. But some of them will inevitably become sick or injured, and become a burden on the taxpayers or their family.

On average, everybody loses, because private insurance has serious adverse selection problems. Private insurers charge a premium on the assumption that anyone who wants insurance must be sicker than average. So private insurance will allow for a lot less risk-sharing.

So, I rest my claim that some workers will lose out on the observed fact that some workers cost more to insure than others, but they don't seem to "pay" extra for their insurance (in the form of lower wages).

So I think it's true in practice, but big deal: is it true in theory? In a frictionless neoclassical world, I suppose that it might not make any difference whether the health insurance was employment-based or private. Employers would charge the sickly more for their health insurance, just as insurance companies do. Obviously, employers don't do this, presumably because it would be costly to screen new hires this way, not to mention probably being illegal. So these frictions are useful: they're what allow us to escape the usual adverse selection problems of insurance markets and pool risks among employees at a particular firm.

Also, I think many employers make an implicit bargain with their workers to cover the increased insurance costs if the worker develops a costly illness. So even if all workers are offered the same expected compensation when they're first hired, over time some will turn out to need more health care than others. If the employer decides to break the deal and drop health insurance, some workers are going to gain and some are going to lose big. It's really a mistake for the government to use tax breaks to encourage firms to break these implicit contract with their workers.

Tomorrow's Front Page Today!

2004-09-15T23:55:00.000-04:00

Ragout, 3 hours after Bush's speech at the Republican Convention: "Typical. Bush proposed about $3 trillion in additional spending and tax cuts in his convention speech, and then had the gall to criticize Kerry for proposing to spend $2 trillion."

Washington Post, 11 days after Bush's speech at the Republican Convention (Mike Allen): "The expansive agenda President Bush laid out at the Republican National Convention was missing a price tag, but administration figures show the total is likely to be well in excess of $3 trillion over a decade."

Not only was Ragout first, but our analysis came with a generous helping of sarcasm sorely lacking in the Post's article.

But never let it be said that the Washington Post wasted 10 days and 21 hours without adding any value to the story. Here at Ragout, we conservatively estimated the transition costs of partially privatizing Social Security as "at least $1 trillion." The Post reports that it's closer to $2 trillion. Ragout estimated the cost of Bush's tax cuts as $1.0 to 1.6 trillion, the Post puts them at closer to $2 trillion. So Bush's agenda of tax cuts for the rich and decreased security for seniors is going to cost about $4 trillion, without even considering smaller items like his new tax giveaways for the oil industry.

AEI's "Independent" Analysis of Bush Health Plan

2004-09-15T23:27:00.000-04:00

The American Enterprise Institute's Joseph Antos has come out with a new analysis comparing the Kerry and Bush Health plans. One claim that caught my eye was that the Bush plan would newly insure 6.7 million people at a 10-year cost of $39.4 billion. That's $5,881 per person over 10 years, meaning less than $588 in the first year, since costs presumably rise over time. That seems pretty implausible. It's also quite a bit different than the Kaiser foundation's estimate, prepared by MIT health economics guru Jonathan Gruber, that 1.3 million people will gain coverage under the Bush plan.

In other words, Gruber estimates that Bush's plan amounts to a drop in the bucket compared to the 45 million uninsured, and the AEI's estimate of 6.7 million isn't that big either. As even the AEI admits, Kerry would cover 27 million new people.

The authors of the AEI study don't offer any analysis justifying its (implausible) figure, but in an earlier study, Antos cites the Bush administration's Council of Economic Advisors as the source. The CEA's discussion pretty much amounts to: let's just assume 15% of the uninsured will gain coverage.

Several different studies have examined the likely effects of the health insurance credit on insurance purchases. Pauly, et al. (2001) find that a $1000 refundable tax credit would likely increase the participation rate among the uninsured by 21 to 85 percent. Gruber (2000) finds smaller effects, closer to 10 percent, but analyzes plans with premiums that are much more expensive than those described above. Other studies focus on average premiums, not the best offers available. Even with the most conservative assumptions, the health insurance credit would substantially increase participation in health insurance markets. If even 15 percent of those uninsured for a full year (or 10 percent of those uninsured for part of a year or more) take advantage of the health insurance credit, 6 million people would be newly covered.

That's the CEA's entire discussion of the calculations.

The difference between the CEA/AEI figures and the Kaiser/Gruber ones turns out to rest not so much on differing assumptions, as on the fact that the CEA/AEI calculations are incomplete. The CEA and AEI look only at the benefits of the Bush plan (uninsured people gaining insurance) and ignore the costs (employers dropping health insurance coverage).

[Click to Continue]

The Bush plan has two components. The first is aimed at low-income households: a tax credit of up to $1000 per person or $3000 per family, phased out as income rises. The second is mainly valuable to high-income people: making premiums for privately purchased health insurance tax deductible. The catch is that the private health insurance has to be a high deductible plan combined with a health savings account (HSA). An HSA is essentially a super-IRA, with still more tax breaks that are most valuable to the rich.

The CEA analysis addresses only the tax credit for low-income households. Gruber estimates that the credit will be used by 3.1 million previously uninsured people, but that this will be offset by 1.3 million people whose employers drop health coverage. The CEA acknowledges that Guber may be right, that some businesses might drop coverage:

Employers may choose not to offer health insurance at all if many of their employees can take advantage of the credit and purchase insurance individually, and receive taxable wages in lieu of employer health insurance contributions.

But the CEA argues that the credit only affects low-paid workers, many of whom don't have health coverage to begin with, adding, "Furthermore, employers make the decision to offer health insurance based on all of their employees, so they are unlikely to stop offering insurance simply because a minority of their employees become eligible for the health insurance credit."

So the CEA argues that no employers will drop coverage (that's what they assume in their calculations). But the CEA's argument only makes sense because they're limiting their analysis to the low-income tax credit! They ignore the other piece of Bush's plan, the tax deduction, which is aimed at highly-paid workers!

So employers can drop health coverage, offering their workers higher wages instead, knowing that the Bush plan will give tax breaks to both low and high paid workers. Some workers, those in the middle perhaps, will inevitably be left out in the cold when businesses stop offering insurance. Gruber looks at both pieces of the Bush proposal, not just the one benefiting low-income people. He finds that 3.9 million uninsured people will buy private coverage with the tax breaks, but 2.6 million will lose employer based coverage, for a net gain of 1.3 million newly insured.

If anything, my guess is that Gruber is overly optimistic (as I've argued before). That's wintry Mark Scmitt's guess too. Bush's plan radically changes a system that is currently strongly biased towards employer-based health coverage. It's radical enough that the effects are hard to predict and could be big; Bush's plan could totally disrupt the health insurance market. At the very least, It's pretty clear that the number of people losing insurance because of the Bush plan ain't going to be zero.