I just read an Economist Free Exchange post containing the following chart:
Look at the discontinuity between 49 and 50 firms in the French firm size distribution graph. I would be willing to wager that there is some French industrial policy which kicks in at 50 employees. (ht: petek)
The best sort of technological progress is the sort that does not require new "stuff". It is a little bit magical when someone uses the same inputs and a better process to produce more output. A couple of weeks ago my colleague Alex Fakos gave me an example. He told me about the way telecom companies transfer information through optical fiber cables. The cables are expensive to install, so they are only infrequently replaced. Over time, however, the algorithms for transferring information through the cables have improved, so more and more information can be passed through the same cables. Nothing physical changed--people just thought of better ways to send information.
This morning I had what was, for me, a profound thought--the first input in the supply chain of every production process is a natural resource. Since every important feature of the world was identical a few thousand years ago, everything that is being produced today could have been produced in ancient Rome--if only they had the ideas. The real driver of technological progress is ideas, not electricity, the combustion engine, or computers. If one takes the view that "stuff" is natural resources, then all technological progress is the best kind.
Here is a thought experiment: Suppose that a large research library from today's world was to materialize in Rome during the 1st century AD. Assuming that the Romans don't burn it down, how long would it take them to create working mobile phones? What stops them from doing it in a year? The answer to this question may be related to how technology moves between regions of the world today.
I have been in Washington DC for the last couple of days attending an orientation for an NSF fellowship. The fellowship is for American students working in the sciences. Less than 10% of the group of 200 or so are social scientists. The rest do hard science. I must say it is a little intimidating--they are studying things like nanoparticle electrophoresis and the selection pressures behind insect morphology. I study conspicuous consumption.
Chatting with these hard science people, I realized that they face similar issues to those we talk about in economics. They just call them different names. For instance, hard scientists talk about physical models vs. statistical models. Physical models are based on deep physical theory--conservation of mass, conservation of energy. Statistical models are not tied to any deep theory. They are trained on data to make accurate predictions--a neural network is one example. Compare physical models to structural models in economics and statistical models to reduced form models. Structural models are based on deep assumptions about human decision making, while reduced form models are more concerned with patterns in data--see the microfoundations debate a few weeks ago.
One of the first things we did in this orientation was to dejargonify our project titles. Most of the titles, mine included, went from being totally incomprehensible to someone outside the field to being more or less understandable to an educated non-scientist. I can see that using jargon helps succinctly express complicated ideas to the initiated, but the cost is making your work incomprehensible to lay people. Maybe that is the point.
I have been mulling over schadenfreude for a while now. Introspection leads me to believe that schadenfreude--pleasure resulting from the suffering of others--is something that humans feel. Apparently psychologists have been studying this phenomenon for some time. They even have a theory about why humans feel it: social comparison theory.
What if some people have especially strong schadenfreude, i.e. they really get a lot of pleasure from seeing others justly suffer? Economists like the idea of Pareto efficiency, but people with strong schadenfreude make mutually beneficial trade impossible. If anyone is better off, then the strong schadenfreude people are much worse off. Moreover, these people can greatly improve their own well-being by making others miserable.
Schadenfreude imposes a negative externality on society. A pigovian tax (that fully compensates the victims of schadenfreude) would cause people to totally lose the benefit they received from hurting others, plus these people would be paying a tax which would further reduce their welfare. On the face of it, the best tax would be any prohibitively high one (prison?).
Somewhere in there there is a research topic--I just have to pick through it a little bit more.
This semester I attended a couple of lectures given by Dr. Andrei Vavilov, an expert in the Russian natural gas industry. A major theme has been that due to contracting problems, a classic hold-up scenario, and the inability of politicians to negotiate, a bunch of unnecessary natural gas pipelines are being built. Mainly these pipelines are designed to put as few national borders as possible between Russian natural gas fields and the lucrative Western European market. Natural gas pipelines cost about 8 billion dollars per 1000 km. This is a huge waste of resources.
There are three current projects that fit this bill:
The Nord-Stream project is set to be completed in 2012, and the South Stream project by 2015. The Nabucco Pipeline may or may not be built, but if it goes ahead it is planned to be operational by 2017.
Taking the upper bounds on the prices, we get 70 billion dollars. To put that in perspective, if it was divided evenly among all American, it would be 233 dollars a head. 70 billion dollars is enough to pay for a year of the war in Afghanistan.
The building of these excess pipelines is deadweight loss, and reflects pretty poorly on the governments involved in the negotiations. It is surprising more people haven't heard about this.
There is a certain metaphysical point of view, which I associate with Derek Parfit, which holds that you are not the same person as future you. If people are the same as their future selves, the argument goes, then why do current people discount the future-- they procrastinate and reschedule painful appointments for the more distant future.
Suppose that you agree that a person is not the same as his future self. Then should it be illegal for the current person to behave in ways that benefit themselves at the expense of their future selves? Think cigarette smoking. There is an externality argument to be made for taxing cigarettes. If someone else is smoking in the bar I am at, it is bad for my health, and I should be compensated for the additional risk I am taking on. Some of the legislation seems to go beyond this, however. On NPR today I heard a report about legislation aimed at preventing cigarette companies from marketing to children. If someone, say a young adult, chooses to smoke at home, who are they harming? The answer is their future selves. If you agree with the metaphysical position above, then maybe you think that government is correct in correcting a market failure--the harm current people are doing to future people.
In a post today, Andrew Gelman quotes himself on the silliness of microfoundations in economics. The replies in the comments to this post, and his original post, are very nicely written, and represent diverse opinions.
Ok, now let's gather around the campfire. The economics lore on microfoundations begins with the Phillips curve. Until the 1970's, mainstream economists believed that there was a permanent well-defined trade-off between inflation and unemployment. A central bank could lower unemployment, but only at the expense of increasing inflation. Then in the 1970's there was both high unemployment and high inflation--so-called stagflation. On the back of this episode, Bob Lucas made a famous critique of treating historical macro-level relationships like the Phillips curve as fixed when doing counterfactual policy experiments. Unless analysis begins at the level of (unchanging) technological possibility and consumer preferences, altering government policy may affect historical or aggregate relationships which earlier economists took at fixed.
This argument still seems convincing to me, at least. The Gelman post above was written in response to Paul Krugman's post on a similar topic from a couple of days ago. Krugman's points are reasonable, but I don't think that he is really complaining about using microfoundations. Boiling what he wrote down, Krugman has two main points. The first is that reduced form macro relationships are good for intuition, and help us to rapidly interpret the world. The second is that the microfoundations used in economics are patently wrong, and not even a good approximation to the way people behave. Both points are well taken, especially the second. I agree that reduced form relationships aid intuition, but ultimately I would like to be able to derive them from first principles if need be.
Krugman uses a bunch of metaphors from the natural sciences. Physicists often use Newtonian laws as approximations of low energy interactions, even though they know that they are incorrect. If push came to shove, however, the exact calculations could be made--physicists know which events are appropriate to be approximated with Newtonian laws and which are not. The problem is that economists don't know when it is ok to use reduced form relationships, because economists (debatably) don't have a good approximation to how people actually behave. Krugman's second point is right on the money. To really understand the way macro-level economics works, microfoundations are really important. However, that doesn't mean that any old assumptions about the way people act are preferable to just using macro-level relationships. Microfoundations have to at least reasonably approximate the way people actually behave. As far as predicting policy affects goes, garbage in garbage out.
Scientists and Canadians are perennially heckling us Americans about the English measurement system. This is nothing more than base 10 snobbery. We Americans prefer base 2, the most parsimonious number base and the language of computers. Behold the beautiful base 2 English system, and the ugly metric system. I am proud to use measurements that aren't based on the number of toes I happen to possess. (source: wikipedia)
The title is a fair description of the position taken in a provocative new paper by Alberto Giubilini and Francesca Minerva. The reasoning goes something like this: newborns do not yet want to live (they just want to avoid pain), thus a newborn is not harmed by being killed painlessly. In the case that parents and society are harmed by the existence of the newborn, euthanizing the baby is ethically correct.
The paper is not very tightly argued. The authors speak a lot about rights. They make the questionable claim that "a necessary condition for a subject to have a right to X is that she is harmed by a decision to deprive her of X." They also claim that "those who are only capable of experiencing pain and pleasure...have a right not to be inﬂicted pain." It is not clear to me where these rights are coming from. Certainly common-sense morality dictates that it is wrong to cause a baby pain, and also that it is wrong to kill a baby.
Even so, it strikes me that whatever the merit of the paper's argument in the baby case, a similar argument could be used to justify eating meat. If animals don't want to live, then the authors might argue that there is nothing wrong with depriving them of life. Of course, animals clearly dislike pain, so the authors might assign animals the right "not to be inflicted pain." This would be enough to make being veganism ethically correct in today's world.
I think the main structural problem with the argument is that it is not clear what "want to live" means as I have been using it above. I am paraphrasing the paper here, so to be fair let me put "want to live" in the authors' own words:
"Both a fetus and a newborn certainly are human beings and
potential persons, but neither is a ‘person’ in the sense of
‘subject of a moral right to life’. We take ‘person’ to mean an
individual who is capable of attributing to her own existence
some (at least) basic value such that being deprived of this
existence represents a loss to her"
What does it mean to want to exist? Babies seem to have consciousness--why would you suppose that they do not attribute some basic value to their own existence? The same goes for animals.
The paper is certainly provocative, and it focus the debate on a valid point--why should the legal status of a human being change on either end of the birth canal?
Alternative energy is closely associated with the green movement in my mind. Greens stand for non-violence, ecology, democracy, and social justice. At the same time, the alternative energy industry has an interest in a high oil price. One thing that drives high oil prices is instability in the middle east. I wonder how wind power executives feel when they hear Iranian officials threaten Israel or civil war erupts in Syria. I plotted crude oil prices vs. the price of GAAEX, an alternative energy index fund. Although recently the two have diverged, the day-to-day jumps track each other pretty closely .
When I mentioned this idea to my officemate Felix, he said that it reminded him of the green paradox of German economist Hans-Werner Sinn. According to Sinn, the expectation of greener energy policy in the future makes the owners of "dirty" energy resources want to extract and burn their reserves more quickly.