Notice that Dave "takes issue with this notion" without offering any supporting evidence of any sort. Neither economic theory nor empirical studies support his wild fantasies about price control's ability to "increase people's faith in market transactions" (when did the substitution of the market by personal connections increase the faith in market transactions? Has he ever heard of the non-market distribution systems, like the Soviet one? That increased the faith in market transactions alright… as did gas rationing in the US in the 70s… as does rent control… as does food-distribution system in modern Africa). Likewise, neither theory nor empirics support his claim that price controls reduce panic demand without increasing price (When did that ever happen?). There is also zero evidence that in emergencies, markets don't clear in rational ways. To the contrary, every emergency area, war zone, and concentration camp has developed robust black markets, which function very rationally.

Dave's post was quite sad, actually. The last paragraph said it all – disliking price gouging is such a morally superior position that anyone who disagrees with Dave’s fantasies on the subject of government regulation is morally depraved.

I second Kate's comments with respect to the lack of evidence for some pretty aggressive empirical claims.

The one that caught my eye was that "the regulations - when publicized appropriately - have the same information forcing effect as higher prices themselves, teaching people that there are supply interruptions and they should change their use patterns until conditions improve."

The same effect? I am highly skeptical that people respond to incentives in higher prices in the same way that they would respond to appropriately publicized public service announcements. I am also am unfamiliar with what it means to use "soft economics" to achieve better results than the market.

Well, I would have liked to see comments to my post appear there (and it would seem fair), but I can respond here too.

Kate: Yes, I've heard of the Soviet system. As a matter of fact, you've mentioned it to me before. I'm pretty sure if you re-read my post, I was making a limited claim that short term threats to punish price gouging have special advantages that long term price controls do not. As for lack of empirical support, I wasn't aware that I had to provide any - especially when the claim I was responding to - that taking large profits in times of civil unrest reduces demand - didn't cite data either. Instead, the argument relies on assumptions about rational market participants that I think the data do not support. As for your last comment, I'd ask you to go read the post I was commenting on, which I think makes the flip of the claim (i.e., that to defend price gouging regulation is itself symptomatic of dubious motives) and decide how seriously to take my comment. I'm sorry you found my post sad. I intended it to be thought-provoking.

Joshua: The claim is that higher prices provide folks with needed information (supply is short). I think it has to be true that when the attorney general goes on the air to say "We're going to enforce our price gouging regulations" citizens are similarly informed about the scarcity of the good. I agree that the question of whether the change in *usage* patterns is the same is an empirical question, but you've misread me if you think I claim that the question is closed.

Let's assume that we have a sudden shortage of a product which is a relative necessity and demand is largely inelastic. If we assume that there are no real substitutes, and that a black market will not develop (due to the short-term nature of the problem and/or high penalties for participating in such a market), is it crystal clear that we want distribution of the good to be determined by price? Here's an example that Mr. Murray gave in his TCS column:

Ah, but what of the hotel owner who has a captive market? Surely they are gouging people when they increase their rates when evacuees show up? Again, the higher price actually helps people. The evacuee is generally willing to pay more for overnight accommodation than the casual traveler because the evacuee has fewer options. Therefore, a higher price actually deters those who don't really need the rooms in favor of those who do. The result is more rooms available for those who really need them.

I think the problem with that solution is obvious: the evacuees may "need" the room more than the casual traveler, but the casual traveler may in fact be able and willing to pay much more than the evacuee. However, the evacuee may be willing to wait a long time to secure the good, while the casual traveler would be deterred by a long wait.

I suppose the proper response is: in times of emergency, the government should step in to help those specifically affected with cash distributions, so that they can overcome the effects of the disaster. But do we necessarily want that money going to merchants who charge higher prices due to scarcity? (Particularly when, as in the hotel example, the costs of producing or securing the good have not changed.)

"There is also zero evidence that in emergencies, markets don't clear in rational ways. To the contrary, every emergency area, war zone, and concentration camp has developed robust black markets, which function very rationally."

I'm dubious, Kate. Markets clear best when information/search costs are low -- i.e., when you can comparison shop. An "emergency" likely features folks unable to spend time searching, unable to risk the chance that this will be the last seller in sight, and therefore subjected to monopoly power. Worse, it may be only perceived monopoly power, because it is unknown whether this seller is the only one around.

I don't think that suggesting the fallibility of markets is nearly so ridiculous that it is fair to call it "sad," a "wild fantasy," or whetever other hyperbole comes to the mind of someone with greater faith in unrestricted free markets. Indeed, a great many free-marketeers might view "emergencies" a narrow exception to their stances against regulation.

Scott:

1. Low information costs are neither necessary nor sufficient for markets to clear.

2. The absence of a monopoly (or perceived monopoly) is neither necessary nor sufficient for markets to clear.

3. I never said that free markets are infallible. I just said they are better than non-market-based distribution systems.

4. There is not a shred of empirical evidence showing that markets fail to clear in times of emergencies. Dave still hasn't produced such evidence, even after being challenged by both me and Josh. Robust black markets that emerge in every emergency zone is the evidence to the contrary.

5. I am not responsible for what other free-marketeers believe. We are not an invitation-only club.

Matt:

Handing out money/vouchers to disaster victims increases the supply of needed goods. E.g., if housing vouchers were easily cashed at the bank, many people might be renting rooms in their homes to NOLA refugees. Many landlords might be willing to fill in apartments that currently stay empty. Many developers might be putting up emergency shelters well-equipped with electricity and running water.

Likewise, if food-and-other-necessities vouchers were easily cashed, people would rush truckloads of food, diapers, toothbrushes, and clothing to disaster areas. Alternative: waiting weeks for the government to deliver the same stuff.

We can say it stinks that someone profits off other people's misery, but I would take the world where food, diapers, and housing are increased in supply and available for all disaster victims over the world where necessities are theoretically cheap, but not available to people without connections. And I don't even need to cite the famous:

"It is not from the benevolence of the butcher, the brewer, or the baker that we expect our dinner, but from their regard to their own interest."
Adam Smith

Kate: One minor follow-up is that I don't know that it's fair to repeatedly (i.e., in both of your posts) hit Dave for not producing empirical evidence.

You'd have to concede that blog postings necessarily can't survey all relevant empirical evidence. For example, in your initial post, you assert: "There is also zero evidence that in emergencies, markets don't clear in rational ways. To the contrary, every emergency area, war zone, and concentration camp has developed robust black markets, which function very rationally." In your follow-up, you assert, "Robust black markets that emerge in every emergency zone is the evidence to the contrary."

I don't know that there's been meaningful study of every single emergency zone. Many emergence zones, by being such dire emergencies that civil society breaks down, don't allow for meaningful study. Have you evidence of not only black markets, but "robust" black markets that "function very rationally"? I think you're going beyond the empirical evidence yourself.

But my main point is that we can't fault Dave for not giving us a boatload of citations or for not undertaking an empirical study. If he did these things, it'd be a law review article, not a blog post. If we criticize blog posts for not producing supporting empirical evidence... well, that's a shot that can be taken at any blog post (except for some of Volokh's, some of which are virtually law review articles!), which is to say it doesn't prove much.

I think it is important here to distinguish between seller monopoly power and a change in supply conditions.

Scott, your comments seem to concerned with the former rather than the latter (or perhaps both?). You are surely correct that markets may be more competitive when search costs are low. But I dont think that is a problem here. And as Kate points out, markets clear under monopoly conditions, but there is loss of social welfare due the seller's ability to create artificial scarcity.

The scarcity here is VERY real. The increase in price is not a function of the elimination of competition such that there is now a seller of necessary goods with monopoly power. The increase is due to a change in supply conditions faced by all competing sellers. Let me know if I misunderstand what you are saying here.

I think Matt nails the question when he says "But do we necessarily want that money going to merchants who charge higher prices due to scarcity?" From an economic perspective, I think the answer is clearly yes. We want higher prices to allocate the scarce resource to its most valued use, and we want higher prices to send valuable information to the market to attract more supply.

Surely, a byproduct of this is that today's sellers get a windfall because supply conditions have changed. I dont have a problem with this. Many do, but my question to them would be to show me a non-market allocation mechanism that is superior on these dimensions. As I stated before, I dont think there is any empirical evidence that one exists.

I suppose I was imagining a situation where demand and supply were both fairly inelastic, like the hotel example offered by Mr. Murray in his column. I agree that if high prices could actually increase the supply in response to scarcity, the pricing mechanism is doing its job. And in most cases, supply will be elastic enough to accommodate an expansion in response to scarcity. But if you have a finite supply which is greatly exceeded by demand, you have to allocate, and allocating based on who can pay the most may be unfair in some circumstances. (They who can pay the most may not be those who *need* it most.) However, these circumstances may not exist all that often. I don't think gas is like the hypothetical hotel room -- it may be a necessity, but there are substitutes, conservation measures could be taken, and high prices may encourage greater production, etc.

Austan Goolsbee has an interesting post (with empirical evidence!) up on Slate about price gouging occuring not during the initial scarcity, but instead hanging on after the crisis has passed. As he notes, "gouging, if it occurs, typically begins when the stations' costs start to come down again. The stations in the study took about twice as long to cut prices when their costs decreased as they had to raise them on the way up. It was after a crisis ended that their profit margins shot up." I'd be interested to hear other folks' thoughts on this. It sounds like these results would contradict standard economic theory, which would say that at least one station would cut its prices as soon as it could, leading to a chain reaction where all of them would eventually come down to the proper market price.
Here's the link: http://www.slate.com/id/2125814/

Scott: I am with you on not demanding full citations to every well-known claim that a person makes in a blog. Such demands only retard the conversation.

But when someone makes a claim that (a) contradicts modern econ theory (that markets clear prices even when info is very poor and monopolists are present); (b) contradicts well-known evidence on the behavior of markets (that market-based distribution systems do not cause panic and overconsumption, but personal-connection (or lottery, or whatever else) based distribution systems cause panic and overconsumption – and this is true for long-term and short-term substitution of markets by alternative distribution schemes); (c) contradicts lots of case studies and anecdotal evidence about the emergence of black markets in the least regulated and most “panicked” environments – then, such contrarian would have to show the basis for his claims. Notice that I didn’t ask for “peer-reviewed, high-quality econometric studies.” Any empirical evidence coming from a reputable source would do. Or, for that matter, any decent theoretical model.

Otherwise, counter-intuitive, counter-theoretical, and counter-common-knowledge claims are appropriately called “fantasies” (and many people would have used a much harsher word here).

Let me note that I'm actually not in favor of anti-"gouging" price controls on retail gas stations outside New Orleans. That market strikes me as sufficiently competitive that, in Milwaukee, I don't feel tat the local station has much power to "gouge" me.

What I would support are more localized price controls in/near the emergency areas. Even if markets "clear," that's a low threshold; the "new equilibrium" may be sub-optimal due to information problems (i.e., high search costs), monopoly power, and behavioral quirks such as retailers' risk aversion (which may lead them to jack up the price more than necessary to cover costs of restocking), buyers' risk aversion (stock up because you never know if this is the last supply available), and both sides' exaggerated perception of the permanence of the emergency condition (i.e., "we'll never see this product again"). Disclaimer: This catalogue of behavioral quirks is off the top of my head (and no, I don't have survey data of New Orleans refugees to back it up!).

On the other hand, I'm cognizant of the fact that price controls will dis-incentivize others to bring more supply to the emergency area.

But to follow up on the plea for empirical data: it is an unanswered empirical question whether, in the peculiar circumstances of any particular emergency, (a) the dis-incentive effect of price controls outweighs (b) the positives of limiting inefficiently high prices. (As to (b): prices would come down eventually, without controls, if they truly were "inefficiently high" -- but "eventually" may not be good enough in an emergency situation.) In short, I'm dubious that we can rely solely on first principles to answer the question, "should there be price controls in this particular local emergency situation?"

And to clarify a point Josh correctly notes that I left ambiguous: by "monopoly power" I mean the power of a seller to charge a high price because the buyer may fear that this is the last place s/he can buy the commodity, due to the emergency condition. I don't just mean "supply scarcity creates monopoly power," which you rightly note is an incorrect proposition cited by some who support price controls.

What is "inefficiently high price"? I've heard of inefficiently restricted supply, but how can a market-clearing price be inefficient?

A price could be inefficiently high if it is based on parties' misconceptions about scarcity.

Let's say local gas stations and I each think there will be gas rationing next week. They'll charge $4 per gallon and I pay it -- but I also start decreasing my gas consumption, e.g., (a) I start bicycling to work, which is an inferior substitute because it wastes time and raises my odds of physical injury because I'm out of shape, or (b) I buy my food from the expensive nearby corner store rather than from the supermarket or Sam's Club that's several miles away.

In that situation, due to shared misperceptions of scarcity, sellers are selling less gas, and buyers are shifting to inferior substitutes. You are correct that markets still clear at $4, but that just means we don't have a shortage or a surplus.

I think Joshua put it well talking about monopoly: "as Kate points out, markets clear under monopoly conditions, but there is loss of social welfare due the seller's ability to create artificial scarcity." Analogously, outside the context of monopoly, shared misperceptions can result in a market-clearing price that is sub-optimal in terms of social welfare.

The (accurate) response to my point could be, "well, if the market price was based on misperceptions, those misperceptions will clear up once it becomes clear gas isn't that scarce, so the situation will correct itself." That is true, but in a local emergency, it may be troubling to "wait it out"; the short-run welfare losses may be a substantial concern if they mean people aren't getting critical goods.

A point of Kate's with which I agree is that often, government payments/vouchers are a superior substitute for price controls. I always wonder why some of my fellow liberals think gov't should jawbone drug companies into giving away drugs to the poor (whetehr at home or abroad); if there's a compelling policy justification for getting these drugs to those who can't afford them (and I think there is), then the government should hand out the pills after buying them on the open market (i.e., paying the drug companies the market price, to preserve drug companies' incentive to produce).

But in emergencies, I don't see how that works. When New Orleans is suffering Noah's flood, and there is neither electricity nor communications nor police presence, how do we distribute vouchers to hotels/boardinghouses/customers? How do we determine eligibility for vouchers?

While anarchy also makes it complicated to enforce price-gouging laws, a simple ban on XYZ price is easier to administer than a voucher distribution system, especially one that must be deployed quickly by the same governments that can't get their acts together to keep people alive....

Scott, you comment that "outside the context of monopoly, shared misperceptions can result in a market-clearing price that is sub-optimal in terms of social welfare."

But why do the misperceptions exist? The parties' best perceptions about market conditions are reflected in the current price. Your claim is that inefficiency will result from market prices where these perceptions about market conditions may be incorrect. I think there is a problem with this argument that can be addressed by asking where the misperceptions come from.

One likely reason for mistaken perceptions is that information is costly. Market participants economize on these costs when making decisions. I do not think it is right to assume, if these misperceptions exist, that they are exogenous. If they exist because of costs, which is likely, it is certainly efficiency for agents to incorporate these costs into their choices.

The same argument would apply to the concept of mistaken perceptions of "monopoly power." (Not to mention, a great way to correct misperceptions about market power is to allow the high price to attract entry).

I do not want to speak for Kate, but I think this explanation is one reason that the concept of "inefficient market-clearing price" is troublesome.

Right, the "inaccurate perceptions" I suggest do result from infromation costs, or imperfect info-processing ability. But I'm makign the behavioral-ec suggestion that the info-processing errors are likely not random, but biased in one direction -- in an emergency, biased in favor of exaggerating the extent of the emergency.

Yes, in the mid- to long-run, the sitation would correct itself due to both entry (as yuo suggest) and the simple fact that correct info would eventually become more obvious (e.g., "hmm, it's been 2 weeks and no rationing yet..."). I'm just making the limited suggestion that in certain emergency situations, the short-term costs of the market working itself out (i.e., the costs of the higher prices resulting from erroneous info) may be deemed too costly.

A couple of clarification questions on your definition of inefficient consumption and prices.

(1) Suppose Manolo Blahnik shoes are clearly outside my price range. I, however, just bought a pair of their stilettos because I saw a photo of Sarah Jessica Parker wearing shoes just like these. It later turned out that Sarah Jessica Parker in fact owns a similar-looking pair of Jimmy Choo’s. Does it mean my purchase of Manolos was inefficient?

(2) Suppose that, with no fault of Manolo’s, some number of its customers held the same erroneous belief re: Sarah Jessica Parker and the shoes. Does it mean Manolo’s prices were inefficient?