Monday, December 26, 2011

optimism update

I forgot one big item on my optimism list. I'm very optimist about the decline of religion (the rise of secularism, the replacement of evangelical flavors of religion with unitarian flavors of religion, the replacement of mythological flavors of religion with secularly spiritual flavors of religion, e.g. Confucianism or Buddhism although I would like to see other such varieties gain popularity as well, etc.)

Interesting on this topic.

Funny on this topic: How many Unitarian Universalists does it take to screw in a lightbulb? Answer: We choose not to make a statement either in favor of or against the need for a light bulb. However, if in your own journey you have found that light bulbs work for you, that is fine. You are invited to write a poem or compose a modern dance about your light bulb. During next Sunday's service, we will explore a number of light bulb traditions, including incandescent, fluorescent, three-way, long-life and tinted; all of which are equally valid paths to luminescence.

(For the record, I love Unitarians. Honestly and unqualifiedly.)

On that note, happy winter holiday of your choosing, so long as you don't try to shame or look down on anyone else for not choosing with you. For those of you looking for a secular occasion to adopt (and why would you abstain from festivities on principle? Make up your own best-of-all-worlds festivities!), ACGPD is an easy way to go, although the lunar eclipse was a good cause as well this year. I don't buy into the winter solstice idea (only the summer solstice - why celebrate the onset of the worst time of year??). I do buy into the back-to-school celebration idea, although that was more pertinent when I actually had classes and actually lived somewhere else for school.

(Edit: as far as decorations go, I'd like to promote the traditional Christmas tree, because they smell nice, topped with a 17-point star, because they're awesome*, and decorated with rainbow lights, because they're... well they're just also awesome. And little plastic balls if you have cats :)

*yes, you have to make it according to Gauss's instructions in order to qualify as awesome. Choose your own skip pattern though.

Sunday, December 25, 2011


Norwegian Wood, by Haruki Murakami - Best book I've read in a long, long, long time. Every time I set it down I was left in a Murakami trance for hours. I can't even pinpoint what was so good about it. The writing style is simple yet beautiful, sparse yet vivid; the plot is mundane yet riveting; the characters are ordinary yet compelling; and most amazingly, a book that exclusively examines relationships doesn't end up in naval-gazing tedium, but rather draws you so deeply into their situation that you can listen to it on your kindle and not even notice the robotic text-to-voice imperfections or that you've been stuck in rush hour traffic for half an hour or that you somehow just ran errands all over town purely on muscle memory, and then you get home and turn it off and sit dazed at your desk for who knows how long trying to remember who you are or what you're supposed to be doing or why you were supposed to care.

Are you there vodka? It's me Chelsea, by Chelsea Handler - It was laying around and the title made me curious. About as mildly amusing as you would expect from a book you only read because it was laying around.

The Undercover Economist, by Tim Harford - Economists will find this boring, but I highly recommend it to anyone who hasn't learned economic theory. It's a very well done qualitative/popsci presentation. Especially those who like to argue about politics without understanding incentives or markets...

Sunday, December 18, 2011

Norway the near-miss-Persian-Gulf-state

Matt Yglesias* has an interesting column.

He shares my concern with a purely open-borders free-trade stance by risk-averse concerns about what happens when a nation puts all its economic eggs in one industry's basket (although he speaks mostly about Dutch Disease, which doesn't strictly have anything to do with that risk particular risk, but with currency valuation and its consequences for domestic industry competitiveness). And it's impossible to disagree with the conclusion that Norway has been very successful at avoiding becoming "yet another Saudi-style oil monoculture".

But, let's review. Dutch disease occurs when a large exporting commodity industry brings a lot of money suddenly into the country which drives up the value of the currency, making foreign goods cheap to export, which reduces the competitiveness of other domestic industries, which leaves the country dependent on this one industry for most of its income. There are a few ways to avoid this outcome. You can artificially reduce the competitiveness of foreign industries through protectionist policies like tariffs and subsidies. You can boost the competitiveness of domestic industries by investing in human capital, research and development, infrastructure, etc. Or you can stop the currency shock by setting aside commodity income as some form of savings.

These measures are not equally good. The last one allows the country to smooth the commodity income over time, reducing the impact of booms and busts in that particular commodity market on national income, and allows the benefits of the commodity income to be spread out over generations in whatever way desired. Investing in education/R&D/infrastructure is great for obvious reasons. Artificially changing the competitive balance via subsidies and tariffs, though, distorts incentives in a cherry-picked set of industries chosen via political whim, which can lead to major inefficiencies as soon as the situation changes at all (such as, for example, the current comical butter shortage that motivated Yglesias's post.) Beyond that, it can actually worsen the currency situation, since tariffs on imported goods reduces demand for foreign currency, driving the value of the domestic currency even higher.

With this in mind, Yglesias is first of all clearly much too quick to lump together Norway with those Saudi states. Ceteris not at all paribus. The culture and institutions and fiscal choices are radically different in many ways; the protectionist policies of the Norwegians likely explain a tiny fraction of the difference in overall outcomes. In particular, I'm guessing (by which I mean I'm pretty sure but don't want to compile proof just for a blog post...) they have a much stronger and longer-term commitment to investment, both for its own sake and as the 2nd Dutch Disease mitigation method I mentioned above, than most Saudi-style oil states.

Second of all, he glosses over the differences between those mitigation methods. Since the combination employed by Norway has been successful overall, he claims that the particular protectionist tactic is vindicated. That's obviously not a valid argument. (In fact, he mentions that the national savings fund is a much more important aspect of their efforts to avoid Dutch Disease, but then glosses over the differences between the methods again.)

All in all, though, it's nice to hear support for protectionism being motivated by rational long-sighted economic logic, rather than the short-sighted knee-jerk yet ubiquitous "stop the mean foreigners from stealing our jobs" rhetoric. I'm so happy about that I almost don't want to bother nitpicking with the details. In fact, I'd call this more of a clarification than a counterargument, since his ultimate point is just that avoiding Dutch Disease is a good goal and the Norwegians should be admired for trying, even if it leads to silly butter shortages.

*It's nice to see him writing longer form stuff now that he's at Slate. (At least, I think it's a higher percentage of his posts, at quick glance...)

Friday, December 16, 2011

self-fulfilling beliefs

A much more humorous examination :)

(I don't know why he felt the need to invent another word for it though, especially with as awkward a definition as that. Oh well.)

Monday, December 12, 2011

college as signaling/education/life-coaching

I wouldn't go as far as Bryan, but I agree with his general point that college is often more of a signaling device (I went to college, therefore I am the type of person who gets into and finishes college, therefore I'm a good hire) than an actual educational system in which you acquire the skills you need to excel at your career.

I do, though, think that part of the reason is that college is designed to be good at one kind of education, and that kind of education isn't very general, and out of some combination of inertia and a concern about maintaining credibility and the fact that the pure signaling mechanism already works pretty well at motivating enrollment, that hasn't changed very much over time as college becomes more universal. For example, a traditional college could emulate music conservatories more closely, for music majors, or a trade school education, for people intending to go into those careers. Or they could simply offer more classes that could replace/accelerate/enhance on-the-job training in a variety of careers.

But, despite the fact that I admit that part of the reason college is oversold for a large fraction of students is an avoidable consequence of its current design, and that a major, or even primary, function of it is to be a signaling mechanism, I also wouldn't go nearly as far as Seth Roberts:
At Berkeley (where Bryan went and I taught) and universities generally, the highest praise is brilliant. Professor X is brilliant. Or: Brilliant piece of work. People can do great things in dozens of ways, but somehow student work is almost never judged by how beautiful, courageous, practical, good-tasting, astonishing, vivid, funny, moving, comfortable, and so on it is. Because that’s not what professors are good at.
Um, what? Academics are trained to be brilliant and judged by their brilliance. That's what academia is, a brilliance factory. When you enter academia as a student, of course this is also the standard you will be judged by.

Even if college adopted a more flexible educational paradigm that actually prepared people for careers in the real world (or the arts, I guess, since the standards Seth is describing seem to be relevant mostly to those), it's ridiculous that college would be the place for learning those things. Certain skills are better learned, or must be learned, in the course of life, not in a structured learning environment. Inspiration, passion, courage, etc. cannot be taught because they come from within, emerging organically and unforcibly through the course of life's varied experiences.

I for one would be damn pissed off if I was paying for an education and got a life coach instead.

Saturday, December 10, 2011

lunar eclipse

I finally saw (most of) a total lunar eclipse last night. Got up at 4:30am after a two-hour nap on an office couch, grumbled a lot about how if my parents hadn't sent me to bed before totality during the eclipse in junior high or OSSM had let us stay outside for the one in high school and if it hadn't been cloudy during the last three I tried to see I wouldn't have to settle for this one starting in the middle of the freaking night instead of comfortably in the evening hours, got the scope and binoculars set up on the roof of a building in Alameda, and was blown away for the next two hours.

About halfway through the partial eclipse, through the eyepiece of my telescope

Partial eclipse over San Francisco and the Bay Bridge

Peak Totality

totality over bay bridge; starting to get light out

And then it started to get light out. It was never unhazy, and the marine layer really started coming in towards the end, so we lost the moon entirely around 6:45. Totality ended at 6:56 and the sunrise was at 7:18, so we didn't see it emerge from totality (or the selenelion). The haze was unfortunately too thick to get a clearish picture through the telescope during totality, and you can see that even these are pretty fuzzy.

I was amazed how dark it got. Typical sequences of images of lunar eclipses vary the exposure length so that the overall brightness isn't too drastically different from the beginning to the middle of the sequence. This is of course necessary - any exposure long enough to capture the moon at peak totality would be horrendously overexposed if used to image the moon normally. Despite knowing this, I was surprised how dim the eclipsed moon was in reality.

Thoroughly awesome. As was going to sleep at 7:30 afterwards...

Thursday, December 8, 2011

social norms, morality, and psychology

Social norms are fundamentally multiple-equilibrium outcomes. In a given context, some outcome (how much to give to charity, what the housework/schoolwork balance for kids should be, what age defines adulthood, how to treat animals, rules for dating, rules for hygiene, rules for hospitality, the list is endless) somehow is chosen as the equilibrium norm, through some process that is not well understood. Then through some other process that is only becoming well-understood currently, this norm is enforced and perpetuated organically.

The best candidate for that process is, essentially, social- and self- image. We obey norms because we want to be seen, and see ourselves, as good people who obey norms, or to avoid the punishment that comes from breaking them.

But, inevitably, the line between "good" and "conformist" and between "caring about being good" and "caring about being seen as good" is blurred. This is partially because there are certain situations in which we think there is something like an absolute morality. In these contexts (random cold-blooded murder, for example), every society has the same equilibrium norm. Then there are other situations where there's near consensus, like monogamy, or rape, or violence towards women, etc. And other situations where there's a little less consensus, like marital fidelity or child labor. And this continuum of consensus continues all the way to the opposite end of the spectrum where norms of formal dress or small talk lie.

Within this smooth continuum, it's hard to tell exactly how much of a social norm is an arbitrary equilibrium choice and how much is a result of absolute morality. Even on the far end of the spectrum, we can plausibly say that it's morally wrong to ignore arbitrary norms of dress, because of the destructive disruption it can cause, or some such explanation.

In addition to this fundamental ambiguity that leads people to forget about the arbitrary aspect of norms, the moral motivation for adhering to any norm is often reinforced since the easiest way to keep the masses in line is to keep them from thinking too much about the rightness or wrongness of an action and simply prescribe a code of behavior as morally absolute. Sure, religion and other ethical codes don't usually dictate behavior in the realm of workplace attire, but they reinforce the moral interpretation of norms overall. Soon, the arbitrariness is forgotten (or never realized to begin with, after enough the moral rhetoric prevails for enough generations.)

This may keep society functioning in an orderly manner, but the sword cuts both ways. The more we motivate our actions by telling ourselves that it's the right thing to do, the more we interpret others actions likewise. Thus, a friend's gaffe is interpreted (usually not illegitimately - I'm describing a rational expectations equilibrium here) as a lack of concern about being a good person, rather than simply lack of concern for social signaling, or a reasoned objection to the governing norm, or a reasoned situational deviation from that norm.

In a society where we are already so overly preoccupied with self-analysis and psychology and we overthink  every action and reaction in our friendships and relationships, we don't need to feed the naval-gazing by making it even harder to remember not to jump to conclusions of malice. Remember that norms are flexible and definitely not always agreed on. Clarify expectations when that potential disagreement might lead to misunderstanding and hurt. Clarify intentions after the fact. Give the benefit of a doubt.

Well, that was a long way around to that point :)

Saturday, December 3, 2011


The Wasp Factory, by Iain Banks - Interesting, weird, darkly humorous, engrossing at the time but ultimately not my favorite kind of book. And the ending really didn't tie up enough loose ends in my opinion.

The Crying of Lot 49, by Thomas Pynchon - Unsettlingly engrossing; take a break periodically to reassure yourself that only the character, not you, is toying with sanity. That caution, however, was not meant to contradict or caveat an enthusiastic recommendation. Pynchon disparages this book which makes me want to read his others even more.

The Things They Carried, by Tim O'Brien - I apparently don't like war stories very much. They were ok in a small dose, but I really don't care for having the ficton of a story shoved in my face constantly. Why would you do that? Tell a story, sell a story, insist on its truth, even if it's obviously not.

Thursday, December 1, 2011

repackaged loans

A tiny attempt to clarify one common source of confusion / ignorance: what are repackaged loans (e.g. mortgage-backed securities) and why were they created?

Say there are two types of people in the world, those with high and low credit worthiness. If you loan a high type money, he will always pay you back. If you loan a low type money, he will pay you back only 90% the time.

Banks might be willing to give a high type a $100 loan with a 5% interest rate, knowing that they will definitely get back $105. In order to be willing to loan money to a low type, the interest rate has to be higher to compensate for the risk: loaning $100 to a low type with an interest rate of 17% will on average get back the same $105 as loaning it to a high type (10% of the time you get nothing back, 90% of the time you get $117 back).

This $105 quantity, in either case, is the expected value of the loan. That is, this is the average amount of money you can expect to get back in payments for the loan.

Despite the fact that the expected value of each loan is the same, banks would obviously rather make the loan to the high type, since there is no risk involved. So, if banks for some reason can't charge a high enough interest rate to compensate for the risk of lending to low types (for legal reasons, or because no one wants to take the loan on those terms, or whatever other reason), there are people or businesses out there who are on average good loan prospects who can't get loans. This is a bad thing because if someone can turn $100 cash into $105 of expected value, society in sum is better off if they can get the financing to do that.

This is where securitization (repackaging the loans) comes in. If a hundred banks make a hundred loans to low types, and then split up the repayments equally, no single bank will be burdened by a defaulted loan and each will have the same amount of money on the line as if they each made a single loan as usual. The risk is disappears (in the limit), and any loan with a positive expected profit can be financed.

A mortgage-backed security is a slice of this big pie of loans. It gives you the right to keep a certain fraction of the payments made by a whole bunch of people who took out mortgages. You get back, with a very small amount of risk compared to the riskiness of the underlying loans, the expected value of the loan.

This is a great idea. Risk sharing allows society to undertake worthwhile things that are too risky for individuals to finance on their own, be it home ownership or startup tech firms or pharmaceutical research. And since there is money to be made once the risk is taken care of, this can simply happen through the market. Wins all around.

The problem, as with so many things, isn't with the thing itself, it's with the stupid use of it. Seatbelts and antibiotics are clearly wonderful inventions. But people who drive recklessly or contribute to the evolution of resistant superbugs, thereby hurting themselves and others, need to be held accountable. Likewise, people who blindly purchased mortgage backed securities, thinking they were riskless investments, and then lost a lot of money when it turned out that the expected profit was actually negative, can't blame the securitization process itself. They can only blame themselves for not doing due diligence (which, yes, is very hard when you're dealing with huge pools of loans, and with contracts that are, in actuality, extremely complicated).

(It goes without saying that deliberate misinformation is another matter. But still not something you can get upset at the loan packaging mechanism about.)