Information, Markets and the Subprime Mess

Though not specific to Oregon, the mortgage market mess is certainly quite relevant given that it affects Oregonians just a s much as anyone else. What strikes me most about the sub-prime mortgage crisis is how it paints, in stark relief, just how important information is for efficient market outcomes. This is a general theme, not specific to mortgage markets.

In a nutshell, the story of the sub-prime mortgage mess goes like this: Lenders, eager to generate new mortgages that they were finding willing buyers for, kept dipping deeper and deeper into the barrel of qualified borrowers. After a while they were scraping the bottom and lending to very risky buyers, compensating the risk by offering very expensive loans. But the expense of the loans was usually delayed by a small period (one, two, three years) - enough time, in other words to off-load the paper and a lot more to follow before the expense kicked in and ignited a wave of defaults. Here is where the information problem kicked in. The wholesalers who were buying the paper and bundling it with a lot of other paper and selling bulk mortgages to investment banks didn't really know what they were buying. Only the people who wrote the loans really know who it was they were lending to. Credit rating agencies were relied upon to provide evaluations of the loans, but they were not effective in assessing the increasing risk of these sub-prime loans.

Then it happened, the expensive parts of the loans kicked in, the default rates started skyrocketing and all of a sudden investment banks, taking huge losses, were not so interested in buying up new paper, especially sub-prime paper but even mortgages from borrowers of moderate risk. This meant that the amount of money available to lend to home buyers dried up precipitously. All because of a lack of information.

Information is essential for efficient markets - it is their life-blood. One of the key assumptions in the familiar Adam Smith 'invisible hand' story is complete information - essentially that everyone who participates in a market knows everything about everything. In this case when there is so little information, speculation ensues and all of a sudden a credit market that looked so efficient, collapses. What is striking to me is that when serious market failures happen (like the lack of credible information in mortgage markets) regulation is an appropriate thing to discuss. What type and how much is debatable, but hoping the market will correct itself in the face of such a lacuna of information is not good governance.

Update: an excellent article in the Wall Street Journal today on Bernanke and the Fed and their response to the sub-prime mess.

Economist's Notebook: Correlation is not Causation

Yesterday (Sunday) the Oregonian ran an op-ed piece by two OSU economists who have studied land values in areas with and without stringent land-use restrictions and found that land use restrictions have not depressed land values. From this one might be led to conclude that land use restrictions do not depress land values (as the headline writer seemed to do). But the evidence they produced doesn't actually tell us much about that.

NB: The authors are very careful to refrain from any causal statement - I am commenting on how this research will likely be presented, not on any error on their part.

To find a correlation (or lack of one) between land-use restrictions and property values does not tell us that land-use restrictions have had no affect on property values. In fact, I can imagine that places that are experiencing or are anticipating high growth rates are more prone to enact land-use restrictions. These restrictions might serve to limit growth and thus, after-the-fact, the growth rates in property values might look the same in low-growth no restriction areas. But the contra-positive (which is unobservable because it never happens) could very well be that without restrictions land values would appreciate a great deal faster than 'average.'

It is interesting that the authors point out themselves the problem with drawing causal conclusions from correlations, but in the opposite direction (they fail to point out the reverse argument that would reduce the value of their data to proponents of M49):

Because of the complex ways that land markets work, it is easy to confuse the effects of land-use regulations with "after-the-fact" opportunities created by those same regulations. For example, regulations that limit the supply of developable lands may give the impression that lucrative development opportunities abound for property owners. However, as we have demonstrated elsewhere, this perceived increase in property value is not the same thing as a reduction in value because of regulation.

To give an example supporting the quote above: suppose you have a property just outside an urban growth boundary and within the boundary property values are skyrocketing. What can you say about the affect of this land-use restriction on the value of your property? You cannot say that without the land-use restriction on your property it would have appreciated as fast as those within the boundary because in the absence of the land-use restriction, those properties might not have appreciated at nearly the rate that they did.

Still the message is the same, these are correlations - any conclusion about the affects of land-use restrictions on property values drawn from such data (anecdotal or survey data) would be inaccurate - including any claim about how land-use restrictions lower the values of private property. So, in the end, any real attempt to measure the true causal link is probably a fool's errand.

The reason I bring this up is not because I do or don't like land-use restrictions. Simply that I think the more sophisticated people become when confronted with statistics, the better the public policy that will result because you reduce the effectiveness of populist appeals.

What I would love to see discussed with M49 is the social cost / private right division. Like in my previous post (yes, 'I like to hear myself talk' is the obvious conclusion).

Beeronomics: Hops Shortages and the Price System

The current shortage of hops that is affecting the craft brewing industry seems to be a nice teaching moment about the role of prices in the market economy. First some background, courtesy of Beervana: there is a severe hop shortage thanks to some bad weather and dwindling acreage and since supplies take a while to produce, they can't be ramped up overnight. So the bad news in increasing prices - but wait - are higher prices bad news? Well, no.

Sure high prices are bad in the fact that beer prices will follow and that will make it a bit harder for all of us to enjoy beer. But it is these very same high prices that will fix the problem. Without prices going through the roof, no new acreage will be planted, perhaps even more acreage will go to some other, more lucrative, crop. Without high hop prices, in other words, there would be continual hop shortages! Astute observers will complain that if it is mostly a weather story then my contention is a bit thin, but the other way of looking at it is with high enough hop prices, more production will occur and that production will be spread and thus the total exposure to one adverse weather event will be reduced - hopefully leading to lower supply and price fluctuations in the future.

So, let's hear it for high prices!

This leads to a reminder of what the price system does in general. It transmits data about relative demand and supply to all the important economic actors, buyers and producers, and assures that the 'right' amount of stuff is produced. What is remarkable is that prices do this all on their own - they serve as a humongous coordinating device - and they do it incredibly well. Attempts to replace prices with something better have always failed: they have led to shortages, black markets and inflation. This is one of the true wonders of a market based economic system.

Is Voting on Taxes for Specific Services a Good Idea?

In addition to voting on Measure 49 and Measure 50, I am also being asked to vote on a measure, 02-64, that would levy $0.90 per assessed $1,000 to support services in Benton County, most notably the County Sheriff's department and justice system. Last year, Corvallis residents were asked to support a levy to support local schools. Measure 50 is, of course, a special tax on cigarettes to support children's health care access. And so, time and again, we see voters being asked to support a tax for a specific public service or infrastructure project.

Some celebrate this type of measure-driven public policy arguing that it gives voters more control over what they want government to do. Economics tell us, however, that this is a poor way to provide public goods - goods that have at least partial non-excludability and non-diminishability. As an example of public goods consider police protection: the fact that I enjoy police protection does not exclude any other person from enjoying it and the fact that I may need police assistance from time to time does not in any substantial way diminish what is available to others. This is true of other services (fire, ambulance, street cleaning, etc.) and infrastructure (roads, bridges, libraries, etc.). The reason that voting for the funding of specific government functions is a poor way to do things can be easily seen in this simple example I have been teaching undergrads for 10 years. The story goes like this: suppose a town beside a river is considering building a bridge over the river. There are only five residents in this town and they each would value having that bridge differently. Suppose their own private valuations for having the bridge are: $800, $600, $300, $200, $100. In other words one person would be willing to pay up to $800 to have the bridge built, the next $600, etc. Note that the total benefit to the town of a bridge is $2000. Suppose that the bridge would cost $1600 to construct. Socially then constructing the bridge would lead to $400 in surplus. But will it be built if an equal tax were to be voted on? No. The proposed tax would be $320 and only the top two valuation residerns would vote yes. The tax would fail a popular vote 3 to 2. What about asking for voluntary contributions? It is likely that each individual would contribute little, hoping that others will contribute more. So, in the end the community is likely to loose the $400 in surplus a bridge would provide. What if the government had a general fund from which it could draw the $1600? Given the $400 surplus, this would be an easy choice to make - build the bridge, everyone is better off.

Now there are ways to make the tax progressive that might help. (But note that I have said nothing about income or wealth - just valuation: so the $800 person might be the poorest but work directly across the river and without the bridge has to travel 50 miles each day) Sill there is a good chance that socially beneficial projects or services will not be provided. A bigger problem arises in the from of free-riders. Since I can enjoy a public good if you pay for it, what is best for me is to enjoy it without contributing (like OPB for example). It is these same motivations that provides the rationale for government provided services in the first place. Allowing voters to decide on them al-a-carte style makes no sense: it makes it likely that public goods will be underprovided. This is why a goverment staffed with skilled beaureaucrats is the model we use - it allows us to escape many of the problems of free market provision of public goods.

Election Special: Measure 49

This post presents an almost impossible task. What I aspire to do in general here is to try and give readers some sense of what economics has to say about an issue or specific policy so that they can use this information (or not) when they form opinions. As Measure 49 is an amendment to Measure 37, the issues here are pretty much the same. It is a complicated, messy situation and economics will not really help much in the details. So I am going to go the other direction to talk about land use planning in general and what economics has to say about that.

I once read a letter to the editor in The Denver Post in which a person was frustrated that, in building his new home, he had to comply with a multitude of building codes. Why, he asked, should the government interfere with what I do on my private property? If I build a house that is unstable and it collapses on my head, I suffer the cost myself. His conclusion: government should stay out. The absurdities of this statement are fairly self-evident, for example I immediately thought of things like the faulty wiring that starts a fire that eventually spreads to neighbors and the cost of the fire fighters that have to come and put it out. These are direct costs to society. There are also others' rights to consider, like the interest of the state to protect the health and well-being of this person's children in the face of unstable construction, etc. There are also the indirect costs of, for example, this person's health care costs in the face of inadequate insurance when the house collapses. All in all building codes are pretty easy to defend government intrusions into private citizens rights.

But how far is too far? Does a city code that stipulates that houses cannot be painted pink make sense, for example? Pink houses create a cost for neighbors if it lowers their own property values. But so might a slightly brown lawn or not impeccably manicured shrubs. This is also not just an economic question but a legal question: at what point do we restrict private rights for the public good? What economics is clear about is the justifications for governmental infringement. Markets with externalities and public goods markets can have outcomes that are quite inefficient. Land use planning is no different: values of surrounding properties, traffic and congestion, pollution, stresses on common-property resources (e.g. groundwater) and protection of natural areas are all very legitimate economic reasons for governments to engage in land-use planning. However, deterring private investment, creating inflation in property values by limiting the available land, adversely affecting individuals property values by the imposition of new land use regulations are all legitimate economic reasons for limitations on government intrusion into private property rights as well.

Beyond this, very little is objective. I do believe that over time these government intrusions are going to have to get deeper and our society is going to have to accept the larger coordinating role governments will have. This is a simple demographic and resource scarcity argument. More people and increasingly scarce resources call for more attention paid to the externality and public goods problems we face because the inefficiencies become more severe. Consider a simple story. What if there is one road that connects a rural area to the urban core. Before any development only a few cars per hour travel on it. Now suppose first 100, then 200, then 300 houses are built that use this road as the only access. The road become increasingly congested but still flows. Then perhaps 400 houses starts to create delays, 500 houses creates 30 minute delays and 600 houses creates 2 hour delays, etc. In other words, the next house contributes even more to congestion than the house before. Thus the stakes of land use planning become increasingly higher as time passes. So if I owned a piece of land in this area and I built the very first house, I would not have contributed anything to congestion. If I built house 601, I would have contributed much more to congestion. Is it fair then that I might have had the right to build before any houses were built, but now that 600 have been I no longer have that right. Perhaps: economic conditions have changed and the social cost of your house is now no longer tolerable. On the other hand, that 601st house created construction jobs, bought materials and helped keep house prices lower. What I assert is that over time the social costs continue to increase rapidly, but the social benefits do not increase at nearly the same pace.

What the passage of Measure 37 symbolized for me then, was a rejection of this idea that governments have an ever increasing role to play in land use planning. That private rights trump social considerations. It appears that this is where minds are at in Oregon these days. However, proponents of 49 believe that Oregonians did not appreciate the role government plays in land use planning at that 37 has shown Oregonians, well, just what I mean when I say externality and public goods problems. I agree things like 37 and recessions seem to focus attention on what government does and voters seem to take government less for granted in the wake. I will be very curious to see how 49 fares, because I believe this is one of those moments when we will get a sense of how people feel about the appropriate public interest/private right division.

As an aside, I imagine people will often talk about 'fairness' in response to M37 and M49. But I am honest when I say I don't understand what they mean. Do changing land use restrictions during a single ownership represent unfair government intrusion? Is it OK at the time of transfer of ownership? Is is 'fair' for others to so congest my local road that I have to leave 30 minutes earlier to work than I did 10 years ago? Since we are talking of private v. social interests, it is impossible for me to speak of fairness in the equation, because then it makes me have to impose some kind of moral judgement on the weight of different welfare effects (do 100 affected neighbors count the same as one private land owner?). But then, I am an economist, so I must be kinda dense...

Oregon Econ Blog Comments & M50

This blog will be exactly one month old tomorrow. I was nervous about starting it - I spent six years as a professor in a state, Colorado, where the political climate was so toxic that even attempting a neutral blog could invite a barrage of trouble. So it was with much hesitation that I began this one. Part of my goal was expository, especially as a way to get my students thinking beyond the classroom. Part of it was to get people discussing policy in Oregon. I make no claim to have some monopoly on the truth and I am acutely aware of the limits of economic thinking. Nevertheless, I want to push economic arguments and have people dissect and push back with other economic arguments or arguments from other perspectives. However, I didn't really think the reality would live up to my aspirations.

I was wrong.

Though I have only had a few commenters on this blog so far, the comments have been better than good - they have been excellent. Every single comment so far has made me stop and think hard about the comment itself and my original argument. For that I thank all of you. I now have renewed energy and faith in this endeavor I am excited to get more students involved and to have more commenters from around the state sharing their perspectives with me, challenging me and the economic way of thinking, and discussing policy.

Two recent comments about my post on M50 really struck me as so good that I decided to reply in this post rather than in the comments.

First, 'keith' writes:

I don't understand, exactly, how smokers cost the state more than non-smokers -- certainly not the $11. I understand that smoking is associated with illness and a shorter life-span. By contrast, non-smoking is associated with a longer life-span; ending, of course, in illness at some point. I don't mean to be crude, but everybody eventually gets ill and dies; and being sick is expensive, and some of those costs are borne by the state. If smokers' illnesses are more expensive than others, I would think that at least some of these costs would be defrayed by the fact that they are likely to pull for a shorter time on social security, retirement insurance, and in general the social expenses of living old.

This is such an excellent point, I had to mull it over for a long time and I still do not have a good reply. My first response was, 'well yes, but smoking has direct costs associated with it that are not reflected in the price, living well does not.' But then I played devils advocate: if eating whole foods, leafy greens and exercising make you live longer, and if living longer is costly to society (and I agree it is in terms of health care for the aged) then why don't we impose Pigouvian taxes on whole grain bread, lettuce and health club memberships? From a pure economics perspective the answer is that the two are equivalent. A second answer was one of welfare economics: it is socially optimal to have people live as long as possible for this allows the highest 'utility' possible for a society (apologies for the econ-speak, but I have to indoctrinate you all in little doses...). Smoking lowers societal utility by shortening life-spans and is costly on top of that. This argument has some merit but largely fails by the counter argument that smokers are engaging in an activity of which personal costs are well known and yet since they do it, they must be engaging in a utility maximizing behavior including the lost utility from the extra years not lived (N.B. see addiction not below). So finally, I have only one good response (but no good data to back it up): it is very likely that living a healthy and active lifestyle saves as much in societal costs during the prime years as it is costly during the extended old years. Playing the devil's advocate again, maybe you could make an argument that nicotine makes people so extra-productive in their working years, that it compensates the extra costs from health complications in later years - but I personally doubt this very much (but maybe I have stumbles on an interesting research question: do countries that have a high incidence of smokers have higher overall productivity than low smoking incidence countries?).

Second, 'Chris Lowe' asks a more philosophical question about economic thinking:

Your discussion of externalities is interesting but the application to smoking reflects one problem I have with marginal utility economics at times. To me the greatest negative externality about smoking is the premature loss of human lives, not reduced to a balance sheet of income, productivity and costs, but persons and all the human ties that envelop them, cutting of which harms others. I've felt such smoking costs myself, observed it in others, and fear it for certain friends.

I have an interesting (and I'm going to guess very unsatisfactory) answer to this, again from an economist's perspective: I agree that there are deep emotional and I guess I'll say 'human' costs associated with smoking, especially for those who are close to someone who has smoked and then is stricken with a debilitating illness. But economists would call this an additional social 'cost' associated with smoking and as such should be included in the Pigouvian tax calculation. Of course it is impossible to measure (the thought experiment would be for an individual: how much would I be willing to pay to not have this person I care about stricken...). But the thing is, these costs are, I believe, internalized by smokers themselves, meaning a tax is unnecessary. To put this in personal terms, my father has smoked all my life and I worry constantly about what this will mean in the future in terms of my own pain and suffering (as well as his, of course). If I could put that into monetary terms (I can't) and if that could be reflected in the price of cigarettes through a tax, then his decision to smoke would be a decision based on how much enjoyment he gets from smoking versus the cost of the cigarettes plus the cost of my suffering. But of course he (I hope) already internalizes how much his loved ones will suffer from the increased likelihood of serious illness when he decides to buy that next pack of cigarettes - so the tax is, in a sense, there.

As a final note: there is a whole economic literature on addiction and 'time-inconsistent preferences' that argues that people start smoking making rational marginal benefit versus marginal cost calculations but fail to understand or fully internalize the nature of addiction (do any of us really know what it is like until you are there?). And so ex ante rational choices become quite irrational ex post. If this is true of many smokers than Pigouvian taxes do an even better job of leading to a socially optimal level of smoking.

Election Special: Measure 50

The November special election is coming up in Oregon and in it Oregonians are being asked to vote on two measures, 49 and 50. 49 amends measure 37 and is something I will address in the coming days. Today I am going to give an economist’s perspective on Measure 50. Measure 50, if passed, would increase the state tax on tobacco and dedicates the new revenue to provide affordable health care for uninsured children in the state. Measure 50 brings up a host of economic questions, questions to which the answers are not at all clear. As a result, and I’ll say this at the outset, I remain conflicted about this bill.

Let’s begin by outlining the economic issues that I see as pertinent to this measure. First is the nature and size of the externality that comes with smoking. Second is the price elasticity of demand for cigarettes. Finally, and complicatedly, is the issue of the uninsured in society and what should be done. Now I’ll address these in order.

Externalities are the costs and benefits of an activity that are not born or enjoyed by the person (group, agent, firm, etc.) engaging in the activity. When externalities are costly we call these negative externalities and when they are beneficial, we call them positive externalities. To give an example of both, I can relate a story from my graduate student years. When I was in the first year of my Ph.D. studies at Cornell (the first year in economics programs are notoriously intense) I lived in an apartment house directly above a couple who were physics Ph.D. students. (In other words, wicked smart and slightly deranged) Anyway, the woman was an accomplished violinist and many evenings as I was studying furiously, the wonderful sounds of classical music from her violin could be heard clearly in my apartment. It was beautiful and soothing and if there had been such a market, I would have gladly paid her some small sum to keep it up longer. Her music, therefore, was a positive externality something that imparted benefits to non-participants and for which there was no remuneration. In the absence of remuneration she played less music than was socially optimal. In stark contrast her highly erratic boyfriend would sometimes, inexplicably, late at night turn his stereo up to eleven and blast the most god awful goth rock so loud you would have though it was coming from my own stereo. His was most definitely a negative externality and one which he did not have to pay the cost of my discomfort (save for the fact that I screamed at him once – and that was the last time it happened). Thus, since he did not have to pay the price of others discomfort, he played his music too loud and too often relative to the socially optimal amount. I make this long digression to point out that externalities are not just factories belching toxic smoke, but are often individuals undertaking some activity that imposes a cost on others. (By the way, as I write this, an OSU groundskeeper is blowing leaves around under my window with the loudest leaf blower ever conceived)

So, finally to the point: Cigarette smoking produces two main negative externalities. The first is the smoke that is potentially harmful to non-smokers in the vicinity. The second is the smoking related illnesses whose costs are often partially (or fully) covered by the state and federal government. These are both costs that do not accrue fully to the persons partaking in the activity. So, for this reason I have absolutely no sympathy for ‘smokers rights’ arguments, just as I have no sympathy for those that would argue liberty in support of no helmet laws, for example. Prove that you are fully self-insured and I’m happy to have people do all of the life threatening things they want. But if your going to do something that potentially imposes a cost on me and I (or my agent, the state) have a right to intercede. Given that these externalities exist, there is a very good economic argument for state intervention. The state regulates tobacco and taxes it. In fact there is even a theory of optimal taxation given externalities, it is called Pigouvian Taxes. The idea is pretty simple. Suppose we knew that each pack of cigarette imposed a sum total of $1 in external costs. By taxing cigarettes at $1 a pack, we would align the private costs of cigarette consumption with the social costs of cigarette consumption and the result of the market would be efficiency. In addition, the revenue raised from this tax can be used to defray the costs of the harmful effects of the activity.

But the rub is, of course, that we do not know exactly what these costs are. The “Economic Fairness Coalition of Our Oregon” (an advocacy group the supports the measure) claims that, all tolled, the true ‘cost’ of a pack of cigarettes is $11.16. Suppose this is true, then the proposed tax is wildly inadequate. What if it is a lot less? Is the tax potentially too high? I think common sense dictates that the $11.16 figure is probably pretty extreme, but that (given the rising costs of health care) $2.025 that is being proposed is probably not enough to make it truly in line with actual costs. But true evidence is so hard to come by that any attempt at a true Pigouvian tax is really a shot in the dark. Still, on this score I would have to score one for the measure, I think it doubtful it is too high and it is a step in the right direction. Oh and one little side note that observant readers will notice: the real cost of an individuals smoking depends on many things, so the cost we are looking for is an average and thus some will pay ‘too much’ and some ‘too little’ if a Pigouvian tax were enacted.

The next issue is how much revenue will be raised by an increase in taxes. Yes, the per-pack tax will rise, but the number of packs consumed will thus fall. Consumers will stop smoking or smoke less. They will switch to substitutes (nicotine gum?). They may start purchasing more cigarettes when on vacation or in duty-free shops. Whatever the cause, to estimate the reduction in sales of packs of cigs in Oregon, we need to know the price elasticity of demand for cigarettes. This is simply, in percentage terms, by how much the quantity of cigarette packs will fall, given a 1% rise in price. I looked the economics literature and, surprise, surprise, estimates vary widely. I would say that about -0.50 overall is about average (or even a bit overstated) and it is a bit higher for youngsters. So this means we should see about a 1% decline in cigarette consumption for a 2% raise in price. Since Oregon’s tax increase would raise prices roughly 20%, we should see a 10% drop in sales. So revenues should increase by roughly 10% (though equilibrium price should settle even higher) of which the states share has increased 71% so tax revenues could be expected to increase by about 61%. This looks like about an extra $162 million. So my extremely crude back-of-the-envelope calculations are pretty close to what I have seen is the state’s projections (though I am not sure why this is supposed to rise dramatically after the first biennium as they predict).

Finally, the last issue is a population without adequate health insurance. This is not really related to smoking and is probably the reason I have so much trouble with this bill. Perhaps I am hopelessly naïve, but this bill is pure populism: let’s tax smokers (boo) and give kids health insurance (yeay). If smoking and its affects are the point of concern than create policy that deals with this. Use tax revenues to support health care for uninsured sufferers of emphysema and lung cancer. If uninsured children are a concern than create a new plan that is funded from general revenues. This creating and ear marking of specific taxes is troubling to me. I don’t like the ends-justify-the-means arguments and I don’t like it in policy either. ‘Sin’ taxes are a political expedient, but are not necessarily good policy (see the beer tax debate from the last legislative session).

But in the end I’ll probably vote for it. What the hell? I don’t smoke!

The Nobel in Economics

Often the best place to find a lay-person's description of the nature of the work that has led to the prize is the Nobel organization itself. Here is their description of the work of Hurwicz, Maskin and Meyerson:

Optimizing Social Institutions

In the mid 20th century, economists found themselves in need of a new theoretical framework with which to tackle the comparison of fundamentally different types of economic organization, such as capitalist and socialist institutions. Discussions between the likes of Oskar Lange and Friedrich von Hayek led to the development of the idea that economic institutions could be viewed as communication mechanisms, and set the stage for Leonid Hurwicz to formulate a general mathematical framework for analyzing institutions implementing collective decision making. His 'Mechanism design theory', first introduced in 1960, has developed into a powerful and widely-applied tool.

Whether one considers auctions, elections or the taxes we pay, our lives are governed by mechanisms which make collective decisions while attempting to take account of individual preferences. Such mechanisms are designed to deliver the greatest social good despite that fact that individual participants may act for their own gain, rather than for the general well-being of society. Studying such mechanisms is the goal of mechanism design theory, and mechanism design can be described as the art of producing institutions that align individual incentives with overall social goals.

Mechanism design theory is a branch of game theory (which psychologists refer to as the theory of social situations), and extends the application of game theory to ask about the consequence of applying different types of rules to a given problem. As a method of demonstrating which mechanism, out of all conceivable allocation mechanisms, gives the optimal result, mechanism design theory can be applied to problems as diverse as the auctioning of radio frequencies to mobile phone companies to the building of social welfare systems. Mechanism design theory lies at the heart of many organizations whose operation we now take for granted.

Often, when asked if the charge that modern economics, with so much emphasis on mathematical reasoning and theoretical rigor, is out of touch with reality and no longer of much use, I will counter with a huge list of examples where modern economic research is being used in real life (because this charge is completely baseless and totally absurd). Mechanism design is a very good place to start. For example, this work laid the foundation for auction theory that has been used by the FCC in its spectrum auctions - work in which all of us benefit as the government maximized revenues. And by the way, Milgrom and Wilson, the pioneers of auction theory are pretty good Nobel bets for the near future.

More Beeronomics

Over on the Beervana Blog's 'Honest Pint Project' some commenters have pointed out that the under-pour (or "profit-pour" in a lexicon that is new to me) is just as insidious as the cheater pint if not more so. To this I replied that the under-pour is a full-information market phenomenon to which an immediate reaction can be made (leave no tip*, do not buy another, do not return, tell your friends, etc.). Market efficiency should result. Why I have such a distaste for the cheater pint is that most customers do not know they are being served substantially less than a pint. Market efficiency will not result.

But anyway, in the comments section John Foyston of The Oregonian mentions that as hops and malt prices increase (as they have been doing and are expected to do so in the near future) the practice of under-pouring may become epidemic. Which got my little head thinking - why would a business under-pour and not simply raise prices? One answer relies on the unsuspecting customer who doesn't know they will be given a not-full glass. The problem with this argument is that I believe that for most pubs the customer who shows up once on a whim and never returns again is a relatively small portion of their overall business. Most customers of a pub will have more than one beer in a visit, will visit repeatedly and/or will go based on a recommendation from a friend. In the economics parlance (as I know you are dying to know) a pub beer is an experience good - one which you don't know the quality before you try it, but after you are likely to be a repeat customer if it is a good one. So preying on uninformed customers is probably not a good strategy.

Perhaps the answer lies in the concept of "sticky prices." First, some background. Economists have for a long time understood two things: One, Keynes was theoretically wrong, but his theories do a remarkably good job making sense of the real world macro-economy. Two, that modern macro theory seems completely right, but does a terrible job explaining what we see in the real world. (And here insert your own favorite economist joke about how economists don't really know anything) So economists have searched for the modification to macro theory that will reconcile it with reality. (physicists have superstring theory - economists have sticky prices) To give but one simple example, why do we have unemployment? When macroeconomic conditions deteriorate, why don't wages instantaneously lower to restore equilibrium between supply and demand for labor? Well, many or most wages are not allowed to change instantaneously because they are set by employment contracts, turnover is costly to firms so they hesitate to lower wages, etc. So wages are a sticky price (wages being the 'price' of labor) and with this incorporated into modern macro theory - it works! There are lots of this type of example, prices set in contracts between suppliers and firms, prices that are posted and printed in may forms that are hard to change (think about bus fares as one example) and - yes - even menus themselves.



This last example feels a bit stale in this era of personal computers and laser printers, but there is still a cost to restaurants that comes from changing prices that makes them reluctant to do so too often. (It could also be that they are afraid of annoying frequent customers) So what does a restaurant do in the face of increasing cost of ingredients? One strategy is to start cutting down on portions. Thus, with hop and malt prices on the rise these days, the best advice when one enters a brewpub is caveat emptor!




______________________________
*This is going to be my strategy, because from my restaurant experience I know that if servers start complaining to the bartenders - problem solved. Bartenders hate nothing more than to have servers barking at them.

On Economics and Beer

My friend and I had lunch at the Lucky Labrador brew pub in Portland recently and enjoyed pints of their fresh hop concoction "The Mutt." Yum - think Beaujolais Nouveau in beer, perhaps a little green tasting but very interesting. But that's not the point of this screed. What is the point is that we ordered our pints and were served, wonder of wonders, a real pint! Three cheers for Lucky Lab! In case you are unaware (as most are), the typical 'pint' served in the typical pub in Oregon is no such thing. They are served in what are known as shaker glasses, the very popular, thick, stackable straight sided glasses like the one in the picture. Thing is, these glasses hold about 13oz. Don't believe it? Well, if you have ever bought a souvenir glass from a pub and taken it home and poured yourself a nice bottle of beer in it, you'll no doubt notice that the beer with a bit of head fits perfectly. This is a subject the aforementioned friend has blogged about himself. This has become a pet peeve of mine. Not that I mind pubs using the shaker glasses - I quite like them - but these pubs have no business calling what they serve a 'pint' of beer. (Word of caution, there are shaker style glasses that are a full 16oz, but they are very rare) Update: Join the new Honest Pint Project!!

Which finally brings me to the economics point: a classic market failure for which government intervention is appropriate is what is known as 'asymmetric information.' One of the most classic economics papers of all time is a fairly simple story of used cars and shows that since owners of used cars know much more about the true quality of the car than prospective buyers, market inefficiencies can result. Thus governments can step in an enact truth in advertising laws, anti-fraud laws, etc. In my current example, most customers do not know they they are being served considerably less than a pint and when informed of this, often react in outrage. But how are they to know, do we expect customers to bring in their own measures? Of course not, we rely on the government to enact and enforce laws that protect us - the free market will simply not work properly (as it would if we could all tell just by looking precisely how much we are being served). True this is 'small beer' in terms of import, but switch the commodity in question to gasoline and you can see how quickly a small problem can be made big. Note that what is required here is a non-market intervention, an extra-market enforcement of laws that assures market efficiency.

Economics and Beer cont...

What is perhaps most surprising about the beer case is that this is not a new problem brought about by a popular glass. The British have been requiring a government stamp on pint glasses used in pubs since 1699 to assure the 'punters' that they are getting a right measure of the best bitter. (Something by the way that is changing due to the vagaries of the EU) There you have not only an economically correct government regulation, but one that works beautifully - all parties are immediately satisfied that a full imperial pint is being served and the government has little to do, but licence glass manufacturers and verify on occasion that they are meeting the 20oz requirement. So why not in the US?

Payday Loans

Here is a topic which is a good example of the occasional enormous chasm that divides economists from non-economists and makes 'normal' people think that we just don't 'get it.' The argument for the regulation of the payday loan industry in Oregon (which, among other things, limits the amount of interest that can be charged at 36%) seems overwhelming: these businesses are praying on the poor and vulnerable and indenturing them into debt servitude, furthering their misery and contributing to their poverty. Yet, I don't buy it and I worry that the poor and vulnerable will be made worse off with such regulation, not better.

Let's begin the analysis with the root problem: access to credit for the poor and/or those with poor or no credit histories. This is a considerable hardship because credit provides flexibility when dealing with limited and often transitory income. Access to credit can also be a key to escaping poverty by allowing investments in productive assets (like education) that can increase future incomes. This is what I would call the disease.

Here is a symptom: Because credit at (for lack of a better term) mainstream financial institutions is inaccessible, a host of businesses have cropped up to provide credit to this population. They have been criticised for having exorbitantly high interest rates and short repayment periods. The subtext to this critique is that they are abnormally profiting from other's misfortune. Given as evidence of the scale of the problem are the vast numbers of payday loan shops. But these critiques strike me as completely misguided. The fact that there are many payday loan shops suggests to me that the industry is highly competitive, and therefore that the interest rates they charge are reflective mostly of the costs of doing business in small-scale loans and high delinquency rates. So the proposed cure will cause firms to exit the industry - worsening the disease by further limiting access to credit to the poor. For those that remain and are limited to 36% interest, they will impose more stringent requirements to limit their credit to only the least risky of their clients - again limiting access to credit for the poor.

So I find this a totally misguided policy. I think that payday loans are a problem, but that the problem is with access to credit for this population. What government could do that would be more appropriate perhaps is to mandate that banks, credit unions and thrifts extend credit to this population. To do so these institutions would end up charging more for credit for all and thus this would be a type of transfer from the relatively well to do to the relatively less fortunate. But it would be aimed at the disease and not the symptom.

Finally, one argument that really bothers me (and I am perhaps typical of economists) is that these payday lenders are predatory because borrowers are naive and don't understand what they are getting into when they borrow money. I find this incredibly patronizing - basically "the poor are dumb." While it is true that education and socio-economic status are highly correlated, intelligence is not. And even if people (in general) are not too savvy about understanding the implications of these loans, they tend to be very small and short and one experience is likely enough for borrowers to learn (unlike, say, sub-prime mortgages). It would be one thing if these lenders were accused of fraud (like, say, some sub-prime lenders) and I would be in favor of any sanction against such practices, but the argument is not fraud, but the failure to comprehend. But you don't have to just take my word for it, the issue of whether payday loans are really predatory has been studied very carefully by economists at the fed who find that, in fact, the population of payday loan customers looks very similar to customers of mainstream financial institutions in their delinquency rates and that the payday loan industry appears to be quite competitive.

So, in my view, here is a policy that just gets it wrong - it attacks a symptom, not a disease and is likely to hurt the very people it intends to help.