The Econophysics of migration

Sorry to report a second time in the same week about Econophysics (previous one here), but this paper is just too bizarre for me not to mention it. Physicists have the reputation, maybe wrongly, to be very bright, methodical and keen on logic. Hence it surprises me very much when I see this papers were completely senseless arguments are made and empirical evidence is tortured in utmost comical ways. I sincerely hope only the worst physicists make it to Econophysics and my view of the profession is thus biased.

Today's paper is by Anca Gheorghiu and Ion Spanulescu. It starts with a paragraph that has nothing to do with the topic, migration, that should from the onset get a failing grade from any teacher beyond elementary school:

Physics, is the most suitable for modelling the economic phenomena and structures or financial-banking operations, because it takes into consideration the process variables characteristics and permits to use some procedures – including the mathematical one – especially probability theory for minimizing or eliminating such influences depending on human factor and unexpected phenomena also, which cannot be predicted by direct methods.

I think was is meant here is that physics bring statistics to the table when analyzing economic phenomena. Color me shocked. After some generic rambling about what causes migration, the authors come up with an "improved model" (their words) for migration: the net present value of migrating equals the benefits of migrating less the costs of migrating. We are making significant progress here.

But there is more. Gheorghiu and Spanulescu then think of various countries as attractors, and that various forces pull people towards the respective countries. This is of course the well-known gravitational model of migration. Hey, we are using Physics concepts in Economics already! Of course, these economic attractors must be defined, and they come up with gold reserves, computers per capita and the proportion of Internet users. Why is a bit a mystery to me. The empirical evidence is presented in the form of histograms, which are apparently at the forefront of statistical techniques.

For some reason, the authors then compare countries to positive and negative electric charges and try to somehow fit migration into the Coulomb's law of electrostatic interaction. I was hoping the electro-magnetic properties of gold, the use of electrons in computers and the Internet would justify the previous empirical analysis, but no, it is entirely forgotten. Indeed, there is no attempt to somehow link all this to any measurement.

I have already spent too much time on this.

Understanding Chinese household savings

It is not a secret that the Chinese are saving like crazy. The big question is why their savings rate is the highest in the world and it was addressed before on this blog (more below). The explanation that this would have to do with life-cycle considerations as the population ages (with few children) has by now been largely dismissed. So can explain it?

Riccardo Cristadoro and Daniela Marconi make the point that we really need to look at households, as firms or the government have not increased their savings rate sufficiently. Using panel data, they notice that the savings behavior differs markedly across provinces. And one aspect that does as well vary across provinces is the provision of social services and the access to credit. Thus they tie the high savings rate with the need to build up precautionary savings. This corroborates my previous posts about increased idiosyncratic risk and the reform of the public pension system.

Coping with the crisis in transition and developing countries

The recent crisis is a truly global phenomenon, yet we hear only about how the developing economies are struggling. There were some reports about food riots over the surge in primary good prices, and then these economies faded away from the limelights. While the BRIC have not relatively very well, what about the others?

Rasmus Heltberg, Naomi Hossain, Anna Reva and Carolyn Turk report that the fluctuations in the terms of trade and prices in general have been very hard on developing and transition economies. Using households level data from 17 countries, they observe that the crisis was in fact harder than in the developed economies where it originated. Indeed, there is mention of skipped meals, reduced education, increased crime and societal issues. While I would have expected the export sector to be most affected, it appears the informal sector, which mostly caters to domestic needs was hit the hardest. And we can only hope this will be over soon, as the impact seems profound and long-lasting.

The econophysics of religion

Other social scientists do not like when economists venture into their turf and challenge established methodology. We economists, of course, welcome these instances of Economics imperialism because we believe our methods are superior. But other scientists also venture into Economics, and the most brazen are the physicists. They created a new field, Econophysics, that has been met with bewilderment or amusement by economists, including myself (Exhibits A and B). Economics and Econophysics largely ignore each other. Here is now the paper that will be with upmost bewilderment in social sciences, Econophysics applied to religion.

Marcel Ausloos studies a religious sect, the Antoinists, over 80 years. Lacking information about membership, attention is turned towards income and expense reports. Ausloos tries to find patterns and regimes in the data in order to understand the growth and decay of the cult. While there is some discussion of GDP and demographics, the exercise is all about fitting the evolution of expenses with a mixture of geometric growth and a sine wave in three different regimes. Why is not clear. And I also do not think the data has been adjusted for inflation. A really bizarre study, I doubt scholars in religion will learn much from it, and economists certainly nothing except to view this field even more with suspicion.

On the advantages of open-source econometrics

I find it surprisingly difficult to encourage my colleagues or my employer to go for open-source software. It appears to be very difficult to overcome the idea that if it is free, it cannot be good. But open-source software is not only good and free, by its very name, it is possible to look at its entrails. This implies that if there is an error, anybody can find it and fix it. With proprietary software, errors are much more difficult to detect, and then one is at the mercy of the publisher to do something about and disseminate a patch. You may think this is not a problem for you, well let's see whether this example hits closer to home.

Talha and Yasemin Yalta have looked at a series of econometrics software bundles and in particular tested for their accuracy. It turns out that fixes for gretl, the open-source one, were rapidly applied. The commercial products, though, took up to five years to correct them. And I know of at least one case where the software publisher refused to correct an error, because "published results would then not replicate." And imagine all the errors we do not see because the code is secret...

PS: another interesting point the authors raise is that commercial publishers do not make old versions of their software available for replication purposes. That can be a serious problem, for example if someone were accused of falsifying results, or just being sloppy.

I do not understand US policy

The current situation in the US can be summarized in the following way: the economy is growing again, although not yet as fast as steady state, and unemployment is still high. Most sectors are doing kind of OK except for the construction sector, which operates roughly at half capacity. Budgets deficits are very high, but not at record level once you factor in cyclical effects. Inflation is moderate and on target around 2%, including its expectations. Everyone is sitting on a lot of cash and nominal interest rates are minimal. The crisis has been going on since 1997, 4.5 years.

What is the Federal Reserve doing? Keep in mind that it cannot address structural issues in the economy and cannot only boost the economy in the short term, if at all. Bernanke and co. are keeping interest rates rock bottom until at least 2014. That would be seven years of emergency measures. If you want to instill some confidence in the economy, you do not go about telling everyone the economy needs to stay in the emergency room three more years.

And by the way, if inflation is targeted at 2% and nominal interest rates are at zero for the foreseeable future, that means real interest rates are negative. Take any model and throw in negative real interest rates. The result are pretty ugly, especially if this is more than temporary. Yet this is what the Fed seems to be advocating.

And why would you need such low interest rates? A Taylor rule would call for low interest rates if inflation (or inflation expectations) are too high (correction: too low). They are not. Or if output is below potential. That may well be the case, but I do not think it is to the extend that the Fed believes it is. The rhetoric coming from the Board indicates that the potential output is simply a continuation of the trend line from before the crisis. I do not think it is adequate. The construction sector is gone, and is going to be gone for a while because of an oversupply of residences and structures. Most households suffered a significant wealth shock that is not going to recuperate soon, and consumption has shifted down. Long time unemployed are simply not employable because of sectoral reallocation. Potential GDP has suffered a permanent shock, and while it will get back to the usual trend growth, it will not get back of the previous path.

For the Fed, this means that it is currently chasing a target it will never be able to reach. Its estimate of potential GDP is much too excessive. And the difference is due to structural issues the Fed does not have the means to address.

Which brings us to fiscal policy. Temporary tax cuts of various sorts are good to address temporary weaknesses in consumption. They work best when targeted towards people with high propensities to consume. These are the poor, and more generally the liquidity constrained. These are not the rich, and certainly not everyone. And if the loss in consumption is permanent, then this cannot be address with such temporary measures. You just have to accept it and live with it. Thus: taxes need to go back up, especially for the rich, also because they got an unnecessary free ride in the past years.

And then there is the sectoral problem. Construction is in the dumps. If you want to do something about it, you target public spending towards construction. Not general tax cuts, not across the board stimulus. And the US could actually need some serious infrastructure replacement and improvements. Actually, it is the perfect time to catch up on lost time here: costs are low, interest rates are low.

Now there is of course the stimulus package. But it has had no impact because it was not sufficiently targeted towards infrastructure, It went to state governments, which just used this manna to avoid getting into debt. It went into all sort of pet projects, very few of which had anything to do with construction. And if there was some construction, it went into initiatives that have such long horizons that they are not going to help anytime soon.

So what needs to be done? Washington needs to acknowledge that the targeted path is too ambitious. Bring interest rates back up, This will entice people and businesses to do something with all the cash they are sitting on. The US Treasury needs to focus on infrastructure spending and restructure tax rates. And do not tell me we first have to wait until the presidential election is over.

Shopping hours competition

Firms not only compete with prices, but also with product characteristics. In the retail market, an important characteristic is the opening times. In some areas, for example in much of Europe, shopping times are regulated, the motivation being to give retail workers somewhat "normal" working hours. In some countries, for example the United States, there is much less regulation, and shops have extensive if not around-the-clock hours to satisfy King Customer.

Miguel Flores, who must have won the award for the shortest paper title, studies whether regulation is welfare enhancing when incumbent retailers can prevent entry of competitors by strategically choosing opening times. This essentially comes down to a model of competition through product differentiation. The standard result that regulation is bad when there is little diversity (regulation cannot promote differentiation) and good when there is a lot of it still holds here. The subtlety of the paper is to consider a situation where the incumbent chooses hours of operation, the competitor chooses to enter and its hours, and then they compete on hours. It is thus a two-dimensional space with entry deterrence on one.