October 2014

I wanted to note this disagreement between P.E. Gobry and Noah Smith because it allows me to pull out my favorite underappreciated David Hume quote.

Gobry:

Science is the process through which we derive reliable predictive rules through controlled experimentation. That’s the science that gives us airplanes and flu vaccines and the Internet. But what almost everyone means when he or she says “science” is something different. … Since most people think math and lab coats equal science, people call economics a science, even though almost nothing in economics is actually derived from controlled experiments. Then people get angry at economists when they don’t predict impending financial crises, as if having tenure at a university endowed you with magical powers.

Smith:

One way of systematically understanding the world is just to watch it and write down what happens. “Today I saw this bird eat this fish.” “This year the harvest was destroyed by frost.” “The Mongols conquered the Sung Dynasty.” And so on. All you really need for this is the ability to write things down. This may sound like a weak, inadequate way of understanding the world, but actually it’s incredibly important and powerful, since it allows you to establish precedents. … A second way of systematically understanding the world is repeated observation. This is where you try to make a large number of observations that are in some way similar or the same, and then use statistics to identify relationships between them. … The first big limitation of empirics is omitted variable bias. You can never be sure you haven’t left out something important. The second is the fact that you’re always measuring correlation, but without a natural experiment, you can’t isolate causation. Still, correlation is an incredibly powerful and important thing to know. … Experiments are just like empirics, except you try to control the observational environment in order to eliminate omitted variables and isolate causality. You don’t always succeed, of course. And even when you do succeed, you may lose external validity – in other words, your experiment might find a causal mechanism that always works in the lab, but is just not that important in the real world.

Hume:

Mankind are so much the same, in all times and places, that history informs us of nothing new or strange in this particular. Its chief use is only to discover the constant and universal principles of human nature, by showing men in all varieties of circumstances and situations, and furnishing us with materials from which we may form our observations and become acquainted with the regular springs of human action and behaviour. These records of wars, intrigues, factions, and revolutions, are so many collections of experiments, by which the politician or moral philosopher fixes the principles of his science, in the same manner as the physician or natural philosopher becomes acquainted with the nature of plants, minerals, and other external objects, by the experiments which he forms concerning them.

There’s got to be a better way to prep for class. First I read the assigned text, taking notes while I’m reading either in the back of the book or, when space runs out, in a little pocket notebook that I carry. Then I read through those notes, highlighting specific passages or commentary that might be relevant for lecture and discussion. Then I re-type some (hopefully more coherent) version of those highlighted notes in a Word file, organizing them in some kind of thematic fashion or outline. (Sometimes I divide that step up into two: first, I retype all the highlighted notes in a Word file; then I organize those notes into outline form in a new Word file.) Once I have some basic sense of the themes I’ll be talking about and the passages I want to focus on, I prepare my lecture (whether it’s a grad seminar or an undergrad class, I always do some interwoven combination of lecture and discussion). All the while I’m trying to do some secondary reading to help me figure out what the hell is going on in or around the text. There’s got to be a better way to prep for class.

A standard piece of advice to researchers in math-oriented fields aiming to publish a popular book is that every equation reduces the readership by a factor of x (x can range from 2 to 10, depending on who is giving the advice). Thomas Piketty’s Capital has only one equation (or more precisely, inequality), at least only one that anyone notices, but it’s a very important one. Piketty claims that the share of capital owners in national income will tend to rise when the rate of interest r exceeds the rate of growth g. He suggests that this is the normal state, and that the situation prevailing for much of the 20th century, when r was less than g, was an aberration.

I’ve seen lots of discussion of this, much of it confused and/or confusing. So, I want to offer a very simple explanation of Piketty’s point. I’m aware that this may seem glaringly obvious to some readers, and remain opaque to others, but I hope there is a group in between who will benefit.

Suppose that you are a debtor, facing an interest rate r, and that your income grows at a rate g. Initially, think about the case when r=g. For concreteness, suppose you initially owe $400, your annual income is $100 and r=g is 5 per cent. So, your debt to income ratio is 4. Now suppose that your consumption expenditure (that is, expenditure excluding interest and principal repayments) is exactly equal to your income, so you don’t repay any principal and the debt compounds. Then, at the end of the year, you owe $420 (the initial debt + interest) and your income has risen to $105. The debt/income ratio is still 4. It’s easy to see that this will work regardless of the numerical values, provided r=g. To sum it up in words: when the growth rate and the interest rate are equal, and income equals consumption expenditure, the ratio of debt to income will remain stable.

On the other hand, if r>g, the ratio of debt to income can only be kept stable if you consume less than you earn. And conversely if r < g (for example in a situation of unanticipated inflation or booming growth), the debt-income ratio falls automatically provided you don’t consume in excess of your income.

Now think of an economy divided into two groups: capital owners and everyone else (both wage-earners and governments). The debt owed by everyone else is the wealth of the capital owners. If r>g, and if capital owners provide the net savings to allow everyone else to balance income and consumption, then the ratio of the capital stock to (non-capital) income must rise. My reading of Piketty is that, as we shift from the C20 situation of r ≤ g to one in which r>g the ratio of capital to stock to non-capital income is likely to rise form 4 (the value that used to be considered as one of the constants of 20th century economics) to 6 (the value he estimates for the 19th century)

This in turn means that the ratio of capital income to non-capital income must rise, both because the capital stock is getting bigger in relative terms and because the rate of return, r, has increased as we move from r=g to r>g. For example if the capital-income ratio goes from 4 to 6 and r goes from 2 to 5, then capital incomes goes from 8 per cent of non-capital income to 30 per cent[^1]. This can only stop if the stock of physical capital becomes so large as to bring r and g back into line (there’s a big dispute about whether and how this will happen, which I’ll leave for another time), or if non-capital owners begin to consume below their income.

There’s a lot more to Piketty than this, and a lot more to argue about, but I hope this is helpful to at least some readers.

[^1]: Around 20 per cent of GDP is depreciation, indirect taxes and other things that don’t figure in a labor-capital split, so this translates into a fall in the labor share of all income from a bit over 70 per cent to around 50 per cent, which looks like happening.

My travelogue continues … By the way, check out friend-o-the-blog Sam Bikinoraion‘s blog – he is also going round the world this year, and seems to be visiting a load of my favourite places, which I didn’t fancy taking the kids to. This episode takes me through Greece, and is posted a bit in arrears, as I headed off to the desert after the events described herein …

[click to continue…]

Not surprisingly, the US Supreme Court’s non-decision on equal marriage has caused plenty of debate, including John H’s smackdown of NR’s Matthew Franck.

The discussion got me thinking about the broader problem of legal reasoning, at least in its originalist and textualist forms, and also in precedent-based applications of common law. The assumption in all of these approaches is that by examining (according to some system of rules) what was legislated or decided in the past, lawyers and judges can determine the law as it applies to the case at hand. There are all sorts of well-known difficulties here, such as how words written a century ago should apply to technologies and social structures that did not exist at the time. And it often happens that these approaches produce results that seem unacceptable to most people but for which a legislative or constitutional fix is impossible for some reason.

It’s always seemed to me, though, that there is a much bigger problem with this approach, namely the implicit assumption that “the law” actually exists. That is, it is assumed that, if the appropriate procedure is used to interpret the inherited text, and applied to the problem at hand, it will produce a determinate answer. But why should this be true? The same law might contain contradictory clauses, supported by contradictory arguments, voted in by different majorities, and understood at the time of its passage in contradictory ways. Most notably, the same constitution might grant universal freedoms in one place, while recognising slavery in another.

At a minimum, such contradictions mean that there is no determinate law on the particular points of difference. But the problem is worse than this. The law rarely prescribes an exact answer in a specific case. The standard view of legal reasoning is the principles can be extracted from case law, then applied to new cases. But contradictory laws and contradictory cases produce contradictory principles. The ultimate stopping point is the paradox of entailment: a contradiction implies anything and everything.

I don’t have a fully worked out answer to this problem but I think it underlies a lot of the disquiet so many people feel about legal reasoning (apart from the ordinary disappointment when the answer it produces isn’t the one we want).

Matthew Franck advances a number of arguments for thinking the Supreme Court taking a pass on gay marriage is similar to the Dred Scott decision. I think he missed one that is at least as good as several of the ones he offers. [click to continue…]

When a militarily powerful country tries to govern the affairs of millions of people on the other side of the planet, we shouldn’t be surprised that chaos results …

That’s of the grab from my latest piece in Inside Story, commenting on the utter incoherence of US (and therefore Australian) policy in the Middle East. An extended version:

How could it be otherwise? A rich and militarily powerful country has taken it upon itself to govern the affairs of millions of people on the other side of the planet, of whom it knows nothing. Its emissaries routinely elevate particular individuals, ethnic groups, religious sects and political parties as favourites, then just as quickly dump them in favour of new friends. Its tools vary randomly from overwhelming force to plaintive exhortation, with no clear or consistent rationale.

The key observation is that, with the exception of slavish obedience to the whims of the Netanyahu government, the US has switched sides on almost every conflict in the Middle East in the space of a couple of years.

My policy recommendation to the US is

an announcement that, from now on, the people of the Middle East would be left to sort out their problems for themselves. In particular, it would be useful to state that the United States has no strategic concern with Middle Eastern oil, and that energy policy is a matter for individual countries to determine according to their own priorities.

Inside Story doesn’t appear to take comments so read there (lots of other interesting stuff) and comment here.

… so begins this comment on a recent thread. I don’t have to read any further to know that the subsequent comment will be both hostile and silly.[^1]

My surname is mis-spelt fairly often, reasonably enough in the case of people who’ve only heard it and have to guess at the unstressed vowel. But it happens surprisingly often when all that is needed is to transcribe the text in front of them.

Likewise, I occasionally get people addressing me as “Mr” because they feel the need for a title and choose the default.

Neither, by themselves guarantees hostility and stupidity. But in ten years of blogging, I’ve never seen an exception to the rule that together, they imply exactly that.

Is this just me? Do other bloggers and commenters find that particular forms of address predict the content of comments? And, if so, which ones?

[click to continue…]

A couple weeks back the estate of Jack Kirby reached a settlement with Marvel, the day before the Supremes were set to take the case. This was a surprise, as the Kirbys were 0-2 in the courts, to that point. I hadn’t paid attention but I figured their case was weak, although they had my sympathy. I had read stuff like this from early NY Times articles:

WHEN the Walt Disney Company agreed in August to pay $4 billion to acquire Marvel Entertainment, the comic book publisher and movie studio, it snared a company with a library that includes some of the world’s best-known superheroes, including Spider-Man, the X-Men, the Incredible Hulk and the Fantastic Four.

The heirs of Jack Kirby, the legendary artist who co-created numerous Marvel mainstays, were also intrigued by the deal. Mr. Kirby’s children had long harbored resentments about Marvel, believing they had been denied a share of the lush profits rolling out of the company’s superheroes franchises.

Marvel made out like bandits, treating Kirby badly. But that’s not a legal argument.

What made Marvel blink at the last minute? The terms of the settlement can’t be revealed, so it’s hard to say.

[click to continue…]

Since I blogged about Arendt and Eichmann on Erev Rosh Hashanah, I figured I’d do the same on Erev Yom Kippur.

Actually, there’s a reason I’ve been thinking about the Arendt/Eichmann controversy of late: it’s heating up again. This time, prompted by the publication in English of Bettina Stangneth’s Eichmann Before Jerusalem. I’ve been reading the book, which offers a full-scale reconsideration not only of Eichmann but of how Eichmann presented himself at court in Jerusalem. In the background, inevitably, is Arendt’s Eichmann in Jerusalem.

Stagneth’s is an uneven book, which starts out with great promise (its opening set piece is almost worthy of Arendt), but performs the nearly magical feat of being both tendentious, maniacally repeating its argument over and over again, and wayward; it’s both polemical and dilatory.

One potentially fascinating angle of the book, which I haven’t seen Stangneth develop, at least not yet, is why Arendt wasn’t more interested in Eichmann’s performance at Jerusalem as a performance. Arendt, after all, had an especially theatrical conception of politics, understanding all that we do in the public sphere as a kind of performance, a mask we wear, a role we inhabit. And no one reading those opening pages of Eichmann in Jerusalem could fail to see just how theatrical is her sense of the “show trial” in Jerusalem. And yet Arendt refuses to apply those insights to Eichmann himself. Rather than see him as performing a part (Stangneth does a good job of showing that that is exactly what Eichmann was doing at Jerusalem), Arendt sees Eichmann as being subsumed by, or subsuming himself in, his role. That is, in part, his blankness, his banality, for Arendt. It’s understandable that Arendt would resist seeing Eichmann in Jerusalem as a performance: that is, after all, the point of her book. Even so, it’s a fascinating wrinkle in the story, one that I hope Stangneth will pursue at some point in the book.

Back to the Arendt/Eichmann wars. They seem to flare up every decade or so. What’s truly astonishing is that the wars continue today, more than a half-century after the publication of Eichmann in Jerusalem. With the exception of Darwin’s On the Origin of the Species, what book has managed, 50 years later, to stir up so much wrath and rage? With books like the Bible or Capital, it’s more understandable: they, after all, are immediately linked to a political or religious movement. But Eichmann in Jerusalem is not.

Or perhaps it is… [click to continue…]

This, from Ars Technica, is pretty extraordinary:

bq. In the early 2000s, William “Trip” Hawkins—founder of video game publisher Electronic Arts—was living the good life. … Hawkins had a peculiar way of keeping his cash flow up; he wasn’t paying all the taxes connected to the proceeds of some of his stock sales. Instead, he participated in a tax sheltering setup designed to produce on-paper “monetary losses” to offset the gains. The scheme was all done through accounting firm KPMG, which used convoluted Swiss and Cayman Islands deals that eventually raised the eyebrows of Internal Revenue Service (IRS) tax auditors. The IRS and the California Franchise Tax Board eventually cried foul. In 2002, the IRS notified Hawkins’ lawyers that the tax shelters, accounting for about $60 million in claimed losses, wouldn’t be allowed for the tax years 1997 to 2000. This meant that Hawkins would be on the hook for millions in back taxes on all those EA stock profits. Still, Hawkins continued living a jet setter’s life until around the time he filed for bankruptcy protection in 2006. For instance, a government legal filing said that Hawkins’ private jet had cost $11.8 million in 2000 and had an “operating” cost of $1 million annually.

[click to continue…]

When I was sixteen and seventeen I did my 5th Year of secondary school twice. Amidst grinds, tears and two to three hours of Honours Maths homework each night, I just could not make it past Christmas and still understand what was going on. (The obvious and practical response; take Ordinary Level Maths instead and accept that a career in Medicine was out, just didn’t seem to present itself.) For two years I hungrily repeated the exercises in the small part of the curriculum I understood, and threw myself with increasing desperation and diminishing returns at the rest. The last chapter I remember mastering was called something like ‘Sequences, Series and the Binomial Theorem’.

Happily, understanding – at least a little – the concept of geometric progressions has turned out to be one of the most useful and widely applicable bits of Maths I could have picked up. It crops up everywhere; understanding the spread and gravity of DDOS attacks, why mouse infestations need to be hit early, why skimming stones on water is so hard, and how a young woman settling for less money than a man at the beginning of her career may still be paying for it when she’s middle-aged.

The definition of a geometric series or progression is ‘whenever a term of a sequence is a constant multiple of the preceding term’. When that multiple is greater than one, the numbers will get very big, very fast. If, for example, the multiple is two, you’ve got ‘exponential growth’, a mathematical term of art that’s often used inaccurately elsewhere to describe rapid but not geometric increases. Real exponential growth tends to sound pretty grim when the term is correctly applied in epidemiology.

At dinner the other night, I learnt that the rate of increase of cases of Ebola in certain African countries has been modeled as a geometric progression for weeks, if not months.* Since at least August, the number of new Ebola infections has started to double every month. Common sense dictates that the more people infected, the more people who will be infected. Mathematics predicts chillingly just how bad it will be. The battle to stop the spread of this disease reaching the threshold where it is now running like wildfire has already been lost. [click to continue…]