Open science in commercial firms

Universities engage in research and put result in the public domain because it fosters the public good. In recent years, though, they have put more focus on patenting research results in order to obtain more revenue in the face of dwindling income from public sources. For-profit firms, though, seem to follow the opposite evolution. They hire more and more researchers to let them publish their results in scientific journals instead of patenting them. This even happens to economists who get hired, for example, by Google, Microsoft, Yahoo, AT&T, and commercial banks to conduct research. For the economists, I kind of understand it as a way to secure top talent when needing advice in complex markets. For hard sciences and engineering, my prior is that these firms have realized that patenting has become very inefficient as seeking exclusivity is now more of a lawyer's than a scientist's job.

Markus Simeth and Julio Raffo have another interpretation of what is happening in for-profit firms, and it looks like what is happening for economists. Using a dataset of R&D performing firms in France that they match with academic publications, they find that the old way of just collaborating with academics is not sufficient to acquire knowledge from the technology frontier, you need to hire them full-time. Adopting the academic discourse and disclosure allow to also benefit from it. And like firms participating in the open source movement, I suppose participating in the open dissemination of science also buys you some academic credibility that can attract top talent.

Efficient raffling of public goods

How to provide for a public good is a classic problem, because of the free-rider problem. If you cannot compel users, say, by forced taxation, it is very difficult to recoup the cost of a public good, unless the cost is very low and the benefit to each user very large. That rarely happens. So how can you solve this conundrum when philanthropy is not available?

Jörg Franke and Wolfgang Leininger propose an algorithm based on raffling the cost of the public good. This ideas is due to John Morgan: you create a lottery for the benefit of a socially beneficial public good. Then, people are willing to pay more for the public good. But this is not yet sufficient for the efficient amount of public good. To get there, Franke and Leininger propose an unfair raffle. The raffle tickets are sold at prices that look like Lindahl pricing, but unlike in the Lindahl model, there is no coercion. While there are already many charity raffles that would fit the Morgan model, I cannot think of an unfair public-good raffle in practice, though.

Dealing with congestion: fast lane or toll booth?

In densely populated areas, road congestion is a fact of life. As time lost in traffic has an economic value, it is then quite obvious that road pricing with flexible tolls that depend on the current level of demand should provide for the most efficient allocation of cars through time and space. But there are some logistic problems with this, and depending on the actual implementation, some privacy issues as well.

Mogens Fosgerau claims one can get a Pareto improvement with a toll lane that is reserved for particular users if demand is inelastic. The lane set-up is similar to airport check-in, where privileged people get their own lane and others can use it if it is available. I question, though, that demand is inelastic. That may well be the case in the short-run, but people do adapt to road prices on a longer horizon by changing schedules and locations. Also, I find the comparison to tolls unfair in the sense that the benchmark is a coarse toll: there is a single toll price that is only activate when there is congestion. Obviously, it creates peak loads right before and after it is in place. That can easily be improved with more flexible pricing.

The evil of patents

The litigation saga between Apple and Samsung over minute and obvious details of their respective phones makes good entertainment but has little economic value. In fact, it highlights all that is wrong with the current state of the patent system. Sadly, textbooks still teach how temporary monopolies are beneficial for innovation, yet evidence is now overwhelming that they hurt the production of innovation, the use of innovation and well-being in general. I have blogged a few times about some cases and discussed some research in this regard.

Michele Boldrin and David Levine have written a nice summary on where we stand on the usefulness of patents. They have written a very good book about the topic a few years ago (buy it or download it for free, the authors are consistent with their message). The new working paper provides a shorter breviary with the main arguments for and (mostly) against patents, with a few updates from the literature and case studies. The major point is that despite a huge increase in the number of newly granted patents, there is no evidence of increases in R&D expenditures or total factor productivity. In fact, there is a growing consensus that technological progress is slowing down and we should get used to slower growth in the economy. Patent law and practice is at least partly responsible for this.

How to reduce public procurement waste

The biggest waste of public resources probably comes from irregularities in procurement practices. When the bidding process is rigged, or there is none, not only does it cost more, but the job may be not be handed to the best person. The problem is that the checks and balances on procurement processes are costly and it is difficult to get complete enforcement. However, one can threaten with audits and possible punishment, and hope that will induce officials to be more careful. Does this work?

Stefan Litschig and Yves Zamboni discuss an interesting experiment from Brazil. There, local authorities take part in an audit lottery that gives them a 4-6% yearly probability of being drawn for an audit. That does not seem like a high probability of getting caught. However, for a one-time experiment, 120 counties were selected and told that 30 of them would be audited a year later. Did the 20 point increase in the audit probability have an impact? Oh yes, the probability of irregular procurement practices went down by 17-20%. The irregular provision of health services was not affected, either because it is more difficult to measure, or because potential punishment does not include jail. Conclusion: for better government practices, audit more frequently, randomly, and threaten jail.

On democracy and tuition hikes

Some students in Quebec have now been on strike for over three months over a law that would increase the tuition in all universities (they are all public) by a total of about US$1500 over five years. This seems a rather trivial amount for a US student, but in countries where tuition is free or almost free, this is not trivial. The apparent violence of the protests, which have gone all the way to sabotaging the subway system, and the daily protest marches show there is some deep issue at play. Let me add my grain of salt on two points.

The first is about democracy. I am all for popular uprisings, demonstrations and marches when there is a failure in the democratic process that leads the government to take decisions that are against the public good. Frankly, I do not see where the failure of democracy is in this case. The law was adopted by a democratically elected government. While Quebec is a province with severe corruption issues (for Western standards), the electoral process seems clean. Polls appear to show wide support for the government's policies. Even the striking students are a minority in the student population. The street should not hold the democratic process and sound policy making hostages.

Which brings me to the second point. Apparent popular support in the polls may be a reaction to the violence and radicalization of the student movement. It may not be about sound policy. But it should. Indeed, the main argument for low tuition is that it makes university access affordable to everyone. That is right, but it is also a gigantic gift to the rich, who send their children much more frequently and much longer to university. If you add the costs and the taxes, giving free tuition is equivalent to a very regressive taxation. I do not think that this is the goal. The goal is to get everyone to pay their fair share in education, for which the future personal benefits in present value are very large. Tuition should be subsidized because of the positive externalities of education, but those that benefit the most from it should also pay the most for it. If students cannot afford studies right now, then grants and loans can overcome that. But the fact that some students cannot afford to study should not lead to a policy where higher education is free, or almost free, for everyone.

The Quebec government is right on this one, and the street is wrong.

Are multipliers larger than we thought?

In the last years, much of the debate on fiscal stimulus vs. austerity was centered on the measurement of government spending multipliers. And to a large extend this was a debate between those how used dynamic stochastic general equilibrium (DSGE) models, finding small multipliers, and those using reduced form models, finding large multipliers. Both strategies have pitfalls, the structural one in that the model may be miss-specified as it is always an abstraction of a complex reality, the reduced-form one because of the Lucas Critique.

Patrick Fève, Julien Matheron and Jean-Guillaume Sahuc make the point that there could be a source of downward bias in the estimation of the elasticity in structural models. It arises from ignoring the endogeneity of government expenses combined with complementarity between public and private consumption. With exogenous expenses, the elasticity is 0.97 for the United States, with endogenous ones, it is 1.31. No small potatoes.

The cost of US fiscal imbalance

It is obvious that the US government is currently spend more than normal. It is also obvious that the current level of deficits is not sustainable. But just how bad would it be if the current fiscal imbalance would be maintained? We have models that can calculate welfare costs of such policies, but because the current policy is not sustainable, it is impossible to solve them and get a number. The only solution we can have is to look at what it would cost to delay going back to a sustainable path.

Bertrand Gruss and José Torres do with a heterogeneous agent DSGE model. According to their results, postponing the return to a sustainable path (defined by a pre-crisis debt/GDP ratio) by two decades leads to a permanent loss of output of 17% and a consumption equivalence loss of 7%. That is gigantic. The losses come mostly from the long run, thus it is all about whether the short-term gain is worth the long-term cost. To understand this result, it is thus important to understand the model.

This is a model of occupational choice (entrepreneur/worker), with idiosyncratic employment risk, no borrowing at the individual level, distortionary taxation and lump sum government transfers. There is another public good, wasteful government spending. A major role of the government is thus to insure agents against shocks, which they can also self-insure with accumulating Treasury bonds. Everybody knows exactly the path of policy. There is no aggregate shock except for fiscal policy. To summarize, the only beneficial thing the government does is insure people against shocks, to some extend. Otherwise, the government is harmful: it takes away goods to destroy them and distorts household choices in adverse ways to finance this. It is then no surprise that almost any government spending is harmful, no matter what the circumstances. In particular, the fact that there could be, for example, a temporarily very high unemployment rate that justifies larger public expenses is ignored here.

In other words, the model ignores the potential benefits of current deficits. Not very useful.

Why are the poor more generous?

Poor people are more generous than rich people when giving to charities. Their donations to churches are proportionally also higher. That seems rather puzzling, as the marginal utility of consumption is higher for a poor person, to she should be less willing to give something away. Of course, you could also argue that this person is poor because she is giving away too much. But there should be a better explanation.

Julio Rotemberg offers one. He bases his model on the fact that people like making donations to causes they can identify with: they are OK with making anonymous donations in such a case because it helps like-minded people, and they like it when others agree with them, and are thus willing to help them. The model yields multiple equilibria, some in which the rich give the bulk of donations, others where the poor do. This can help rationalize why the proportion donors varies so much from one country to another, and so independently from the size of the welfare state. Some parameter values lead to people donating in the hope it will encourage others to do so. These equilibria should be dismissed, according to results I recently discussed. The model also seems to build on the fact that there is little crowding out of private charity from public funds, which I doubt according to recent evidence I discussed here. But the model predicts nicely that there is a strong incentive for charities to differentiate themselves to capture donations, even if this not increase total donations.

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.

Egoistic giving

Why do people give to charity? It could be because they care about others. It could be for tax reasons. It could be to offload some guilt. It could be to improve one's standing in society. It could be to encourage others to do the same. Or it could be a combination of all the above. In a more fundamental way, the question is whether philanthropy is egoistic or altruistic.

To address this, Dean Karlan and Margaret McConnell examines donors to Yale University. During a telephone campaign, some potential donors were told about the opportunity of being recognized as donor in the college's newsletter. They responded by giving more frequently (+3%) and more (+14%) than the control group. Variations in the amount necessary to be listed did not yield significant effects, though. One can conclude from this that people donate at least in part to elevate their status and/or to encourage others to give.

To disentangle to two, Karlan and McConnell use a better controlled environment, the laboratory. They work with undergraduates to see whether the latter are willing to part with US$5 to help fight AIDS, tuberculosis and malaria. In some rounds, donors are announced, but no difference in giving is noticed by varying the timing of the announcements. One can thus conclude that giving to encourage others to give is not an important factor, even with such small amounts.

A market for IP addresses

IP addresses face exhaustion, at the least those under the standard IPv4 format, and by some reports they should have been all used up already. What has helped delay the inevitable is probably the fact that there is now a market for IP addresses, yet it is not clear that the market is working efficiently. The reason is that IP addresses are allocated in blocks, and fragmenting the big IP allocation table makes it more difficult to manage it. For technical reasons, each allocation needs to be a square in the table. Thus, if a square is partially unused, it can only be split in multiple squares, increasing their number. Routers need to keep each possible square in memory, and their multiplication slows routing. And as IP addresses are privately owned and managed, there is no way to control this negative externality.

Benjamin Edelman and Michael Schwarz propose a market mechanism that should make the allocation of IP addresses more efficient. They suggest a "spartan rule:" in each bilateral trade, one of the two traders is designated as "extinguished," i.e., as prohibited from trading with other extinguished ones. As one can be extinguished only once, this implies that the number of cuts N in the IP table is limited to the number of initial holders of IP blocks. The analysis is static and under certainty, implying that the implicit rental price of an IP is zero as long as there is still a free one. But with the proposed rule, I do not see how one could necessarily reach exhaustion after the N cuts. It all depends on the initial allocation: one can end up with free IP addresses and no possible moves. In addition, once we add uncertainty and dynamics, there is going to be strategic behavior as being extinguished is a potentially costly absorbing state. I am thus not convinced of the arguments in this paper.

Of course, the easiest would be for everyone to switch to IPv6, which would give a sufficient number of IP addresses to last for a long time. But IPv6 devices cannot communicate with IPv4 devices (large scale IPv4 to IPv6 translation is cumbersome), which gives little incentive to switch until there is substantial critical mass. In other words, another situation like Y2K is approaching, and nobody has an incentive to do something about it. The more efficient market allocation will delay this, but also will make it even more urgent when it happens, because more addresses will need to switch, and they will have less time for it.

The Internet makes you happy

We have previous established that the Internet, contrarily to conventional wisdom, makes people more social. Does this also mean that people with Internet access are happier? Of course, one should take into account that those without Internet, at least nowadays, are likely to face hardships like low income and education.

Thierry Pénard, Raphaël Suire and Nicolas Poussing do such an analysis for Luxembourg and find indeed that Internet users are happier, especially among those with lower incomes. This is also true when taking into account the intensity of Internet use. This implies that making the Internet accessible to lower socio-economic classes can improve welfare, possibly significantly. Of course, one has to take with a grain of salt studies of happiness based on surveys that ask for subjective self-evaluations. That grain of salt may be bigger when one considers who small Luxembourg is. The approach then becomes similar to the randomized experiments in the development literature where results for a small set of villages are difficult to apply to other contexts. Yet, Luxembourg is surprisingly diverse, so maybe these results are generalizable. Readers, you can now safely that you are now happier from being on the Internet and reading this.

Public consumption and the business cycle

One aspect of government purchases the current crisis has highlighted is how volatile they can be. Quite obviously, they are influenced by politics, to the point of complete reversal between massive spending and severe belt-tightening within months as in the US and the UK. But there could also be a more systematic component that is linked to the business cycle. After all, the government may be trying to improve the welfare of its constituents and for example substitute public consumption for lacking private consumption, or the same for investment.

Ruediger Bachmann and Jinhui Bai look at this using an augmented real business cycle model. They claim that 25-40% of the variance of public consumption can be accounted for by shocks to total factor productivity once implementation lags and costs of public consumption, as well as taste shocks to public vs. private consumption. I am no particular fan of taste shocks, as they are the symptoms of a modeler who is giving up on trying to explain something and simply equates the error term in the Euler equation to a shock. Then much is driven by how this shock is calibrated, in this case to match a four year electoral cycle and some data moments. When I think about shocks in this context, I think indeed about who is in power to decide on public expenditures. But that is not completely exogenous. Indeed, the state of the economy has an impact on who gets elected or reelected. And this can be calibrated without trying to match the data moments one is trying to explain.

Public pensions are not sustainable, even in Norway

By now, everyone must be aware that populations are getting older and that this puts some serious strain on pension systems. Unless one plans far ahead or is blessed with substantial sustained growth, some problems in financing retirement will appear. But there must be some place that is going to do fine, say a country with a forward-thinking government, a recently reformed pension system, a well managed endowment of natural resources and a small and smart population, like Norway. Right?

Wrong, say Christian Hagist, Bernd Raffelhüschen, Alf Ering Risa and Erling Vårdal. To come to this conclusion, they use generational accounting, which measures the fiscal sustainability of the public sector and in particular the publicly funded retirement pensions. The latter went this year through a significant reform, which includes pension indexation below wage growth, benefits adjusted to be actuarially fair if life expectancy increases further, and work incentives for elderly. It turns out the pension reform has helped substantially for the sustainability, about as much as the presence of the endowment of oil and natural gas. But that is not going to be enough, even with higher oil prices and an exceptionally well managed petroleum wealth. And for those hoping that future growth of the economy or higher fertility would help, well at least in the case of Norway this would barely help. To close the gap, a 17% increase in taxes would be needed, and they are already very high in this country. So, if Norway cannot make it, how could countries with inactive governments and little or poorly managed endowments make it?

The Internet did not raise a generation of loners

The image of the basement-dwelling World-of-Warcraft-playing loner is often shown as an example of the adverse impact of the Internet on social capital and in particular social interactions. Whether this is true is not so obvious, as the Internet also makes possible social interactions that could not exist before, as this blog shows in a limited way.

Stefan Bauernschuster, Oliver Falck and Ludger Woessmann study the impact of broadband Internet on social capital using a natural experiment in Eastern Germany. There, some choice by the telecommunications provider resulted in 11% of East German households to be on OPAL lines instead of DSL, which better supports high speeds. Using the German Socio-Economic Panel, they measure social capital with the frequency of going out, visiting friends and performing volunteer work. They find that Internet access has no visible impact on social capital. To the contrary, for children it seems to enhance social capital, possibly because it makes them aware of new opportunities to interact in real life. This is in stark contrast with television use, which has many times been shown to be detrimental to social capital, likely because it is a one-way communication, while the Internet can build two-way communication.

State-owned banks in the US?

Many countries have state operated banks that support local development or other objectives that deviate somewhat from those of usual for-profit banks. No such institution exists in the US except for the Bank of North Dakota.

Yolanda Kodrzycki and Tal Elmatad study the Bank of North Dakota in the perspective of the feasibility of a similar bank in Massachusetts. They find that the BND is not a typical bank. While it favors local development, it rarely does so directly, but rather by helping local banks. It thus encourages a network of small and local banks, something that does not quite seem efficient to me. The BND was, however, not particularly useful in periods of crisis, like the agricultural crisis of the 1980s, because it also had financing difficulties. All in all, the bank of North Dakota is very different from state banks abroad, which offer all customer services like private banks and thus help regulate through competition some the excesses of private banking. The BND looks much more like existing development corporation that exist in most if not all US states. If Massachusetts just wants to em ulate North Dakota, it does not seem worth the large cost of the initial bond issue, especially in the current economics context.

Property rights and natural resources

It is a firmly established conventional wisdom that natural resources are best preserved when there are well established property rights. It is the quintessential example of the tragedy of the commons that if everyone is allowed, say, to take water, water will be over-exploited. This wisdom takes, however, a crucial assumption: that once the resources is taken, property rights are well established and uncontestable. What would happen if not?

Louis Hotte, Randy McFerrin and Douglas Wills show that reverting this assumption can have a dramatic impact. Suppose that you took a freely available resource, but that now anyone can contest your ownership of that resource. Depending on the consequences, you may not want to extract in the first place. It thus matters in which way the state is weak. If it is weak in that it gives away rights to natural resources, then there will be over-exploitation. If it is weak in that it cannot enforce property rights in general, and in particular when it comes to bring product to the market, then it is the Wild West and under-exploitation may ensue. Theft is a powerful mechanism to kill markets.

Are military expenses good for growth?

It is obvious that federal fiscal deficits will have to be addressed sooner or later in the US, and seeing how difficult it is to raise taxes, one has to think about how to trim expenses. Of course, the biggest line item is defense, and one can ask what the consequences of cutting these military expenses could be. Critics of those cuts will point to WWII, where the military build-up has pulled the US out of the Great Depression. While I do not quite agree with this interpretation of this anecdote, it is worthwhile to study more generally the impact of military expenses.

Giorgio d’Agostino, Paul Dunne and Luca Pieroni do a literature review and note that out should not just look at the direct impact of expenses. Indeed, a military build-up is also more likely to generate conflicts, and after all a conflict is overall a waste of resources as much effort is spent blowing physical and human capital to pieces. The multiplier argument is also rather vacuous, as these funds could be used for other purposes as well with higher multipliers, in particular when you compare wars in foreign lands versus infrastructure at home. The same applies to the argument that military research has some positive impact on civilian technology (why not simply focus research on the latter?).

This clearly makes it difficult to make a case that military expenses are good for growth. Empirical work is really difficult, like so often with cross-country growth regressions, but d'Agostino, Dunne and Pieroni conclude that the evidence tends towards a negative impact. The only ones that obtain positive impacts are those that include supply-side effects, and those are of course rigged to provide a positive impact.

Starve the beast?

The Republican strategy in the US has been since Reagan to starve the government to prevent it from growing. The master at this has been Bush Jr., who to significantly increase expenses while cutting taxes. And the new crop of parliamentarians has vowed to make these tax cuts permanent, thus forcing even further government expense cuts in the near future. While there can be much disagreement about where to cut, one could first ask whether it is a good strategy in the first place to starve the beast like this.

Michael Kumhof, Douglas Laxton and Daniel Leigh use an elaborate model to come to the conclusion that this is a good strategy if government expenses are useless. But if they provide a public good, then it is not. That seems like a very trivial result, and one that can explain the disagreement between the left and the right in the United States. But there is more to the paper. It shows that for this strategy to be welfare enhancing, cuts need to be done very quickly, and cut services must have little impact, and the taxes that are reduced must be very distortionary. It is very unlikely that all three conditions can be satisfied.