Month: September 2009
If traders are located 100 miles away from an exchange, they face a
delay of one millisecond whenever they seek to trade a price via their
computer screen. Few serious investors can afford to be that late to
prices that flash so quickly. The blink of a human eye takes 300
milliseconds; many traders now operate in the smaller realm of
Here is the longer story. I liked this bit:
Mr Greifeld said there might have to be measures to ensure speeds
within the co-location facilities were the same. “We might have to give
everybody the same length cable, believe it or not,” he said.
For the pointer I thank Christian Meyer.
In an impressive new paper, Karthik Muralidharan and Venkatesh Sundararaman provide evidence on the power of teacher incentives to increase learning. The paper is impressive for three reasons:
1) Evidence comes from a very large sample, 500 schools covering approximately 55,000 students, and treatment regimes and controls are randomly assigned to schools in a careful, stratified design.
2) An individual-incentive plan and a group-incentive plan are compared to a control group and to two types of unconditional extra-spending treatments (a block grant and hiring an extra teacher). Thus the authors can test not only whether an incentive plan works relative to no plan but also whether an incentive plan works relative to spending a similar amount of money on "improving schools."
3) The authors understand incentive design and they test for whether their incentive plan reduces learning on non-performance pay margins.
The results are as follows:
We find that the teacher performance pay program was highly effective in improving student
learning. At the end of two years of the program, students in incentive schools performed
significantly better than those in comparison schools by 0.28 and 0.16 standard deviations (SD)
in math and language tests respectively….
We find no evidence of any adverse consequences as a result of the incentive programs.
Incentive schools do significantly better on both mechanical components of the test (designed to
reflect rote learning) and conceptual components of the test (designed to capture deeper
understanding of the material),suggesting that the gains in test scores represent an actual
increase in learning outcomes. Students in incentive schools do significantly better not only in
math and language (for which there were incentives), but also in science and social studies (for
which there were no incentives), suggesting positive spillover effects….
School-level group incentives and teacher-level individual incentives perform equally well in
the first year of the program, but the individual incentive schools significantly outperformed the
group incentive schools in the second year….
We find that performance-based bonus payments to teachers were a significantly more cost
effective way of increasing student test scores compared to spending a similar amount of money
unconditionally on additional schooling inputs.
Surprisingly, since absent teachers are a big problem in India, reduced teacher absenteeism per se does not appear to be the primary mechanism by which incentives improve learning. Instead the primary mechanism appears to be more intensive teaching, including more homework and classwork and better attention to weaker students, this greatly increases the relevance of these results to teaching in the developed world.
Addendum: See also Karthik's comments on the comments at 26.
Banerjee and Mullainathan offer a different
explanation, one which shows how temptation might interact in a unique
way with poverty. Suppose, they argue, that there are certain goods
that people are tempted by, such as candy, or coffee, or cigarettes.
Then suppose that as people get richer, they spend a decreasing
proportion of their income on these goods; not a smaller absolute amount, but a smaller proportion.
(There’s only so much money you’re likely to spend on cigarettes, no
matter how rich you get.) Finally, suppose you’re realistic enough to
know that you’ll be just as tempted in the future by these items as you
In sum, your “long term self” knows that you will
spend money on temptation goods in the future, but places no value on
that spending. (Your long term self doesn’t like the fact
that you’ll spend money on cigarettes, even though your today self
wants it.) Knowing that you will spend this money amounts to a
“temptation tax” on future wealth. This is a disincentive to save for
the future. Why save today? After all, your future self will just
squander the money on cigarettes!
But as you get wealthier, the effective “tax rate”
is lower, because temptation goods are a smaller proportion of your
income. With a lower tax rate, your disincentive to save shrinks.
Perversely, if you expect to be wealthier in the future, you have a
greater incentive to save and invest! Banerjee and Mullainathan show
that this can create a poverty trap. When you expect to be poor in the
future, you are less likely to save and invest, which keeps you in
poverty. When you expect to be wealthy in the future, you are more
likely to save and invest, which makes you wealthier still.
In his new book he writes:
In my view, the adverse repercussions of Lehman' failure could have been substantially reduced if the federal regulators had made clear that they would protect all holders of Lehman's commercial paper with a maturity of less than 60 days and guaranteed the completion of all trades with Lehman for that period.
As I interpret that recommendation, it is to guarantee the obligations which are vulnerable to run-like behavior, but not to guarantee debt obligations more generally.
Addendum: James Kwak comments.
2. Health care expenditures, in graphic form.
4. "These factors combine to make our era the most consistently and consequentially deluded and unadaptive of any era ever.": Robin Hanson is on a roll. Or how about this:
Our dreamtime will be a time of legend, a favorite setting for grand
fiction, when low-delusion heroes and the strange rich clowns around
them could most plausibly have changed the course of history. Perhaps
most dramatic will be tragedies about dreamtime advocates who could
foresee and were horrified by the coming slow stable adaptive eons, and
tried passionately, but unsuccessfully, to prevent them.
I'm interested in understanding why MR has such a high-quality comments section. I'd like you to consider this passage, from today's Guardian (not today's Onion), and try to write high-quality comments on it.
The statement, read out by Archbishop Silvano Tomasi, the Vatican's
permanent observer to the UN, defended its record by claiming that
"available research" showed that only 1.5%-5% of Catholic clergy were
involved in child sex abuse.
Let's see how you do. If you can indeed produce high-quality comments, it means you're better than the other blog commentators. If you can't, maybe it means that Alex and I are in some way better with regard to what we post and how we present it. In that case, once our splendid framing is off-scene, you revert to your usual, rotten selves. I want you to end up with most of the credit.
I found this at the excellent Twitter site of Michael Nielsen, recommended by an MR reader. He also refers us to this interesting article on neurology and athletic performance and this piece on the surprises of mathematics.
Here is his blog and here is his blog post on the future of scientific journals. Here is Michael on the future of science.
Most of all, I like his six rules for rewriting.
Hail Michael Nielsen, who justifies Twitter all on his own.
Doing the math, you find that on a per-mile basis, a drunk walker is eight times more likely to get killed than a drunk driver.
The subtitle of the book is Global Cooling, Patriotic Prostitutes, and why Suicide Bombers Should Buy Life Insurance and you can pre-order it here. The authors are…come on guys…need I tell you?
The Harper and Collins press blurb offers this summary:
"SuperFreakonomics challenges the way we
think all over again, exploring the hidden side of everything with such
- How is a street prostitute like a department-store
- Why are doctors so bad at washing their
- How much good do car seats do?
- What's the best way to catch a
- Did TV cause a rise in crime?
- What do hurricanes, heart attacks, and highway
deaths have in common?
- Are people hard-wired for altruism or
- Can eating kangaroo save the planet?
- Which adds more value: a pimp or a
I would stress different angles. My favorite part of the book was the presentation of the List-Levitt critique of experimental economics. In particular the authors discuss whether the subject participants are more cooperative to begin with and also whether they are primed to please the experimenter. The biographical information on John List is fascinating. There is a very good revisionist account of the Kitty Genovese story; the neighbors didn't perform as miserably as many people think. Terrorists are especially likely to rent rather than buy, especially unlikely to take out life insurance (which doesn't pay off in cases of suicide), and likely to have a large number of cash withdrawals relative to other transactions.
Geo-engineering, as a response to global warming, receives more pages than any other single topic.
This book is recognizably in the style of Freakonomics, a book I suspect you already have made up your mind about. I will say only that SuperFreakonomics is a more than worthy sequel, a super sequel you might say. If you're a fan of Freakonomics, you'll like this too. This really is the fall season of big, big books.
I am surprised how many people still think Twitter is a fad or a waste of time. I view Twitter — or some modified future version thereof — as everlasting. Most of all, the search function helps you tap into a real time conversation on just about any topic you want, including the lecture you just gave. Google is wonderful but it's hard to sort through the mess and figure out where the conversation is now. For sampling opinion on either movies or music, Twitter is essential, or even for researching a forthcoming blog post. Think of it as Google focused on one time-slice and giving the weight of crowd opinion no more than linear force. If an opinion is more common it will receive more tweets but otherwise your search brings up the splat, ordered by chronology, and thus it is more idiosyncratic than the first Google search page and often in a good way.
At least now, the people on Twitter are smarter on average than the people whose choices feed into Google. I am not sure that particular benefit will last forever,
If you can find some people worth following, so much the better. But the value of the medium doesn't much depend on what they had for breakfast.
Many people use Twitter to ask for advice; I have yet to learn how to do this well.
Please let me know, comments are open. If you don't have a mental model of my intellectual tastes by now, let the heavens fall. And if you can, please give their Twitter handle.
Adam Smith, a loyal MR reader (yes that is his name), writes to me:
I had a very MResque thought today I wanted to share with you. Why are the typical lengths of albums across different music genres so different? In particular, I was thinking most of my rap albums are at least over the hour mark and many run all the way up to the 80-minute maximum. They're usually packed with intros, skits, and lots of 5 minute tracks that have extended intro and outro instrumental beat only sequences. My metal albums, on the other hand, have an average run length of no more than 40 mins. Most albums are between 8 and 10 tracks with little in the way of tangential material. These different run-times show up in other places too. For example, my older jazz albums (i.e. Kind of Blue, Time Out, Blue Train) typically run around 45 mins with a half dozen or so tracks yet my newer jazz albums like MMW's The Dropper run almost the whole 80 mins. Also, prog. rock bands tend to produce much longer albums than garage rock bands. Even adjusting for the fact that prog bands emphasize longer musical passages, they could compensate by just having fewer songs or garage rock bands could just have twice as many (like the White Stripes did on their first album).
Is there a relative price argument for these differences? Or even signaling? Perhaps there is a rat race among rappers to signal they're capable of coming up with enough material to fill out the maximum length, even if it includes lots of filler. Perhaps the recording costs are lower as instrumentation relies so heavily on sampling. Maybe metal runs into diminishing returns after 30 mins or so where the listener becomes numb to the intensity.
I'll offer a few points:
1. The average career of a rapper is short. A long CD increases the chance that something will "stick" and the rapper won't get too many other chances to try.
2. Some metal bands develop great loyalty among their followers and achieve durable franchises. That gives them a lower discount rate and they are more inclined to save up material for the future. Plus they are marketing an overall sound — rather than clever particular innovations — and if the first forty (five?) minutes don't convince you nothing will. Rap songs probably have a higher individual variance.
3. Many older albums are short for technological reasons, plus the albums were due in relatively rapid succession for contractual reasons. In the 1960s there was lots of technological advance in music, so if you sat on the sidelines for a few years you could become obsolete.
4. It is relatively easy for a contemporary jazz artist to tack on additional improvisations and he can choose standard compositions if necessary. Other forms of popular music cannot expand quantity so easily without hitting a wall in terms of quality. One prediction here is that "compositional jazz" albums should be shorter in average length than albums of jazz improvisation, contemporary that is.
5. If you wanted a somewhat strained explanation, you could argue that the longer CD is a more bundled product and it will make economic sense as a form of price discrimination, the more varied the valuations of the audience. This would require that rap CD buyers have a higher variance of marginal valuation.