Sunday, November 4, 2012

Is the argument for price gouging autistic?

Not in the literal, diagnostic sense of course - but in the high-on-analytics/low-on-social-context sense. I don't know (I'm probably too deep into economics to be impartial on calling this), but I think commenter Lord makes important points:

"They obviously don't need price to steer goods anywhere; they are more than willing to direct them to where they are wanted, they operate under monopolistic competition after all. Not nice [this is in response to my post stating that even if gouging serves a social function, it certainly can't be considered a nice thing to do] is important because it demonstrates a social cost, one that isn't represented in the market, or perhaps is and is very high which is why raising prices is an anathema to most everyone. Raising prices is an autistic solution to a problem that doesn't really exist."

Joseph Fetz counters: "Given that there are only a particular amount of goods available at any given time, what incentive do suppliers have to steer a greater amount of goods to one particular region vs that of another region? Further, what spurs greater production of the good in a world of scarce resources?"

Lord responds: "As opposed to shipping gas to stations without electricity or shipping chainsaws to Canada for next years use? What great industries are waiting for disaster to do what they never would otherwise? What industries are so profitless that being deprived of price gouging they cease to exist? Autistic is written all over this."

I certainly agree that (1.) the importance of the signal is easy to overstate, and (2.) WTP and ATP are probably not going to be binding constraints here - who shows up first is. As painful as it is to pay $10 or whatever it would get up to for gas, the number of people that would actually be priced out of that market is small. My suspicion is that what is really sorting people is not the price anyway - it's the access and the patience, and gouging wouldn't change that.

As I said in the last post - if you are actually incurring costs to get the gas in there, obviously no one should begrudge you charging a higher price.


  1. And, as *I* said in the other post, not charging the higher price is indeed a cost.

  2. More directly, there is one and only one thing in short supply, places with electricity (and probably natural gas). So should we charge those with it double to make up for those without it? It is actually in excess supply since there is little effective demand, so perhaps it should fall instead. I am sure in economists fantasies utility problems would be solved overnight if we did the former but it just isn't realistic. Yet we are asked to believe that propagating the problem to all other goods will improve economic efficiency. It is quite unlikely it would as it is lack of access to alternative sites, not any shortage of gasoline or hoarding by consumers. Long lines are probably exquisitely suited to deter hoarding, more so than any price could be.

    1. I don't advocate anything you're talking about, nor do I agree with it. The only reason that a good should be exchanged at all is because the owner of that good, and the person on the other side of the transaction, both agree to the terms of the exchange. What those terms are is not my business. It goes without saying that I have no idea what the price of that particular good should be, only those who are parties to the exchange know this, because they are the ones who ultimately determine it though their individual valuations of the goods involved in the exchange.

    2. That is probably the greatest weakness of establishing price gouging as a principle, it actually encourages hoarding as speculators anticipate ever rising prices forming a bubble as they collectively try to corner the market, each believing they can exit before the crowd and make their killing in the process. The crash occurs, but not without generating a lot of damage in the process.

  3. My suspicion is that what is really sorting people is not the price anyway - it's the access and the patience, and gouging wouldn't change that

    Well, yeah: non-market allocation tends to be the result of price ceilings (placed above the market clearing price). So instead of anyone willing to pay getting gasoline, the first people there do, as you note, or the most patient or willing to pay in other non-monetary, non-saleable, illiquid ways. And a background of price-gouging laws is effectively a price ceiling, albeit a more ill-defined one, which is worse.

    It's very naive on Lord's part to say, "[herp derp], they already know to bring gas there, I guess prices must be irrelevant" -- since prices affect the margin of production. Sure, those able to make normal profits redirecting their gas will, but marginal cases won't.

    1. Maybe I'm not reading this right, but did you just call me "herp derp"?

    2. You aren't getting it: without any restraints on the market, I'm guessing non-market rationing would still be the deciding factor.

    3. @Daniel_Kuehn: I'm not getting it because of your guess?

      @Joseph_Fetz: No, I was just inserting Lord's implicit admission of his own naivete.

    4. I am not missing the marginal case but rather that it operates in precisely the opposite direction as you think it does. This is a difference of normal markets and crises. In normal markets higher prices are a signal to increase supply; in crisis, they are a signal to reduce it for fear of even more severe shortages in the future. It reinforces hoarding rather than opposing it.

  4. "As painful as it is to pay $10 or whatever it would get up to for gas, the number of people that would actually be priced out of that market is small. My suspicion is that what is really sorting people is not the price anyway - it's the access and the patience, and gouging wouldn't change that."


    If everyone demanded fixed and equal quantities of gasoline, your argument would make sense. Because in that world, the only choice consumers make is whether to buy that fixed amount of gasoline or to not buy any gasoline at all. Thus, the only way prices would ration gasoline consumption would be to price some consumers "out of the market". So in that world, consumers would be divided into those who get their desired amount of gasoline and those who do no get any. And the welfare consequences don't seem to be that different if we ration gas using prices or using waiting lines.

    However, this is not a good approximation of reality. In the real world, we can not only decide whether to buy gasoline or not, but we can also decide HOW MUCH gasoline to buy. If the price of gasoline goes up to $10, most drivers probably would still buy gas as Daniel predicts, but they would probably buy much LESS of it. And that is very important if we want resources to go to their most valued uses.

    Think of it this way. Let's say I value my first gallon of gas at $10 and the second at $3. If gas is $3 per gallon, I will buy 2 gallons right? And normally that's fine. But what if there are only 2 gallons left at the gas station and the guy behind me in line would also be willing to pay $10 for his first gallon of gas? Then we are losing $7 of value by giving that second gallon to me instead of the other guy.


    1. Daniel,

      Also, there is no reason to guess too much. We have actual estimates for how responsive people are to changes in the price of gasoline that can give us at least some idea how much people will cut back on gas consumption if prices go up. i

      The price of gas in your neck of the woods should be around $3 per gallon (

      If prices went up to $10 per gallon, that would represent a 233% increase in price. Since the short-term price elasticity of demand for gasoline is around -0.3, this means raising the price of gasoline to $10 would reduce the quantity of gasoline demanded by 70% (233*-0.3).

      Not exactly small potatoes. :P

    2. You really can't use non-disaster gas price elasticities to talk about demand for gas in a disaster zone.

      Yes, when I see a high price I drive a littler farther to get it.

      When I don't know if the next place has gas and when gas isn't just in my car - it's in my generator that's giving me heat - my demand is going to be far more inelastic.

      I haven't forgotten price theory at all.

      Don't you forget your search theory.

      Search frictions in a disaster zone are non-trivial. Search frictions in Falls Church, Virginia are (and in that case we can rely more exclusively on price theory and traditional gas elasticities).

    3. Daniel,

      I will reply to your point on the elasticity of demand in one second. First, I just want to say that it makes me sad that you ignored the theoretical meat of my post and instead choose to jump on the elasticity footnote.

      My main point was that even if you believed that who showed up to the gas station first was critically important for who could buy gasoline after the hurricane (which is what it seems you meant by saying "patience" and "access" were important) prices would still play a very important role in rationing. This is because people don't demand fixed quantities of gasoline (e.g. 10 gallons or none at all). Instead, they can adjust HOW MUCH gasoline they buy based on the price they have to pay.

      This is important. Because it implies that to direct resources to their most valuable uses we don't need to price anyone "out of the market" per se. Instead we want to encourage people to economize on the amount of gas they consume. Prices will do that because it will influences choices at the margin. Queuing won't.

      THAT is what I meant by you shouldn't forget your price theory.

    4. Now, on to your point about elasticity estimate. First, I want to point out that I never said that the elasticity of demand would remained unchanged. What I said was that I was using the published estimate to give us AN IDEA of how sensitive consumers are to price changes.

      It is hard to have a fruitful discussion about quantities without putting numbers to it. And now that we have a concrete starting-point we can actually start saying something useful. You think the elasticity of demand will change after the hurricane. Great. I agree.

      So will demand be more or less price elastic after the hurricane?

      You tell one story for why it might be less elastic. But I could also see demand for gasoline being MORE elastic. For example, after the hurricane many businesses will be closed. That means that many (most?) of the people that are employed by that business wont have to go to work. And if consumers anticipate that many businesses will be closed, they have less reason to go out themselves. So on the whole, consumers might be willing to sit on a half-tank rather than fill up for $10 per gallon (especially if they thought regular service would resume shortly).

      So it seems to me it could go either way. But honestly the demand for gasoline could become much more inelastic and the quantity of gasoline would still drop significantly. Why? Because the price increase you imagined (from $3 to $10) is JUST SO FREAKING BIG (233%).

      For example, say that the elasticity of demand was cut in half so it was -0.15 instead of -0.3. Now a rise in the price of gas to $10 will lead to a ~35% decrease in quantity of gasoline demanded. That is still a pretty big drop.

      Now, maybe you think the elasticity of demand will drop even further. Maybe close to zero? If that's the case, I'd like to know why. There is no theoretical reason I can see for thinking it will drop so low. And I don't know of any empirical analysis that would give us guidance.

      But maybe that isn't what you're saying? The fact that it isn't clear what your position shows that there are dangers to discussing quantitative topics in vague and non-quantitative terms.

    5. I don't see how this is unrelated to the elasticity point. I thought I was commenting on this point.

      When I say that people have highly inelastic demand for gas, I am disagreeing on a practical level with your assertion that "people don't demand fixed quantities of gasoline". In a disaster, we expect they do and that it would be highly insensitive to price.

      Arguing the elasticity point is addressing that issue, and it's not abandoning price theory.

      Let me put it this way: search theory obviously doesn't overthrow price theory, we agree, right? It just complicates it.

      My claim is that search frictions are big in a disaster, and that when you add search frictions in with supply disruptions and high demand inelasticities, you (for all intents and purposes) have a vertical supply curve and a vertical demand curve. That's not abandoning price theory - that's saying that price theory probably doesn't have a happy ending in this. What does price theory tell you happens in the case of a short run with two parallel lines? Am I misjudging this?

      Now, let's say they're not perfectly vertical. Regardless, the price at which they do meet is going to be quite high, and we start getting into humanitarian concerns where ATP is more decisive than WTP. When humanitarian concerns kick in, queuing starts to look more appealing.

      I'm not trying to come out as a proponent of anti-gouging laws or anything. I'm still not on board with those laws. Because for every really bad disaster area there are marginal areas too where demand is less inelastic. But if we think of the really hard hit areas I think you have to realize that search frictions matter and that we're probably not dealing with gently sloping supply and demand curves.

      I disagree with your second to last sentence. If you know you only get so much gas because of a quantity regulation you absolutely will ration your use of it. Why wouldn't it?

    6. Daniel,

      As I mentioned in my previous post, there are reasons to think that demand will be less elastic after the Hurricane. But there are also reasons to suspect that it will be more elastic. So there is no theoretical reason I can see for leaning one way or the other. However, if (for whatever reason) we do think demand will be less elastic, then it could be MUCH more in elastic than it is in normal circumstances and we would still expect to see relatively large quantity changes because increasing the price of gas to $10 represents such a HUGE percentage change in price. I don't think you would see this unless you played around with the numbers. That's why doing the math is so important.

      Now, obviously, if the elasticity of demand fell to near zero, then you might have a point. In that case, it might require REALLY large increases in price to clear the market and ATP concerns might creep up. But I don't see any theoretical reason to believe that demand will become THAT inelastic after a hurricane. You say the following....

      (1) "When I say that people have highly inelastic demand for gas, I am disagreeing on a practical level with your assertion that "people don't demand fixed quantities of gasoline". In a disaster, we expect they do and that it would be highly insensitive to price."

      What do you mean "we", kimosabe? Why should "we" expect demand for gasoline to become fixed at specific quantities after the hurricane? There is certainly nothing in standard price theory that would make me think that. Nor is there anything in my personal experience that would suggest that.

      In 2005, my hometown was hit by a flood and became a natural disaster area. The next day or the day my family took my brother's car and my mother's car to fill them up with gas just in case there was a shortage. I think the price was $5 per gallon or so and they filled up both cars. If the price was $10 per gallon, maybe they would have only filled up one car. If the price was $2 per gallon, maybe they would have filled up both cars and bought gas cans to fill those up as well.

      This is why I contend that prices will encourage people to economize on gas while waiting lines won't. However, you seem to disagree with this claim as well....

      (2) "I disagree with your second to last sentence. If you know you only get so much gas because of a quantity regulation you absolutely will ration your use of it. Why wouldn't it?"

      Well let's wait a second here. What do you mean by "quantity regulation"? I'm talking about using queuing or waiting lines to ration goods instead of prices. That is to say, the price of gas will be fixed at pre-Hurricane levels (say $3 per gallon) and will be sold on a first-come first-serve basis until it is gone.

      In that case, there is no reason to think that waiting lines will make people economize on gasoline consumption. This is because waiting lines represent a fixed cost. They won't change the **marginal** cost of gasoline. Once it is my turn in line to buy gas, I will buy gas until the marginal value per gallon equals $3 just like before the hurricane. The only thing that has changed is that I have a smaller consumer surplus due to the up-front cost of waiting in line.

      Why does this sound like a "fair" way to allocate gasoline (if we want to get into humanitarian concerns)? We are basically saying "if you can get to a pump in time, buy as much gasoline as you would have before the hurricane if not more". Since there is less gasoline to go around, I would think the "fair" thing to do would be to tell consumers "only buy as much gasoline as you 'need' because we have much less of the stuff to go around."

  5. You know, classical economists believed supply determined price and demand determined quantity. This is because as a rule, firms face constant or decreasing returns to scale.

    If you think about it, generally a store will open more branches if demand increases. They will only increase price if it is a positional good or one with a fixed quantity.

    Raising prices does undermine a sense of fairness in most situations. If people know costs have increased, they're less likely to begrudge. But raising prices is surefire way to create outcry (and has done many times in the past).

    I think the whole 'demand increased, it's OK to increase price' is an autistic approach. No concern for human cost, just a weird concern that everything obeys so called 'laws' of economics (which of course, are not laws at all, but that's for another day).

    /ramble over

  6. All the market fundamentalists here seem to be ignoring the basic fact that the supply curve of gas is vertical and has shifted significantly to the left. All this discussion of the demand side--and the miraculous ability of the market to respond--totally ignore this.

    Also, to the idea that ATP=WTP. Tell that to the Irish in the 1840s or the Bengalis in the 1870s. Modern famines were precisely caused because ATP<>WTP.

    This gas shortage is obviously a much less serious problem, but the difference is one of degree, not principle.

    Also, the idea that people should just ask their "tribe" for money when everyone is equally bad off or that people should be selling their baseball card collection or their condo in the middle of a hurricane to raise funds is really the most autistic argument that's been concocted here.

    1. Whoops, Indians in the 1870s, not Bengalis.

    2. re: "All the market fundamentalists here seem to be ignoring the basic fact that the supply curve of gas is vertical and has shifted significantly to the left."


      See my reply to Wayne above. I am still an anti-anti-price gouger. That will help in the long run and it will help in the areas that were hit more marginally. But the idea that this is going to provide some happy solution to the problem is not very serious.

      We probably need to let prices do their thing but have serious quantity regulations.

    3. Andrew,

      Since I was called out in reply to this post, I hope I am not being labeled a "market fundamentalist".

      I am well aware of how markets can fail. And I am not necessarily opposed to market interventions after a disaster. Though not for any of the reasons you point out (you seem to think WTP can exceed ATP, which is not possible).

      I am mostly hammering on the poor quality of Daniel's arguments in the past several posts. Specifically, he seems reluctant to lay out his central assumptions unless pressed. For example, he apparently believes that after a disaster, consumers no longer make decisions on how much gasoline to buy at the margin. Instead, they have fixed-quantity demands. That is a HUGE assumption and a HUGE deviation from the way economists traditionally model consumer behavior. Yet it went unmentioned in any of his previous posts.

      That is either a problem with his writing. Or it could mean it is a problem with his thinking on the subject. Maybe he didn't realize the implicit assumptions he was making until they were pointed out (it happens, no judgement).

      Either way, it seems like he is a long way from having the grounds to make sweeping statements about what ideas are "serious" or not (as above). And the fact that doesn't stop him from making such sweeping statements irks me.

    4. - For example, he apparently believes that after a disaster, consumers no longer make decisions on how much gasoline to buy at the margin.

      No, not true.

      - Instead, they have fixed-quantity demands.

      Highly inelastic demands. Don't make a pretty reasonable assumption about demand elasticities sound pre-marginal.

      You also keep dodging this point that in this environment search frictions mean that a simple supply and demand model can't necessarily bear the load you're trying to burden it with.

      You want me to take you more seriously?

      Stop saying I'm making basic supply and demand mistakes and start engaging the relevant points here I'm making about high demand inelasticity, finite short-run supply, and search frictions. None of these are outlandish or anti-marginal points.

  7. Daniel,

    I don't need you to take me more seriously. And I don't think you are making basic supply and demand mistakes. I think you're modeling consumer behavior in a bizarre way.

    Or at least maybe you are. Now it is hard to tell. Before it sounded like you said that after a hurricane, consumer demand for gasoline becomes fixed at some specific quantity. If the price for gasoline is below the value for that quantity, they buy gas (if it is available). Otherwise they don't. Here's a quote that led me to believe that.

    "When I say that people have highly inelastic demand for gas, I am disagreeing on a practical level with your assertion that "people don't demand fixed quantities of gasoline". In a disaster, we expect they do and that it would be highly insensitive to price."

    If you are only saying that demand for gasoline becomes highly inelastic after a disaster, then that sounds less bizarre. But it is also unfounded. As I said in 2 previous posts now, there are reasons to think it might become less elastic (like the search frictions you mention) and reasons to think it might become more elastic.

    But even if we knew the direction of how elasticity would change, it doesn't tell us much unless we know the magnitude. As I pointed out before, gasoline demand could become MUCH more inelastic than it is now (say -0.15 instead of -0.3) and "affordable" price increases (say from $3 to $10 per gallon) could lead to significant reductions in quantity of gasoline demanded. That should suggest that markets could clear with prices that are high, but not high enough to make ATP a major concern.

    We would only anticipate that clearing the gasoline market would require REALLY large large price increases (and thus raise ATP concerns) if the elasticity of demand fell close to zero. You assert that this is the case. Yet you have offered absolutely zero reason to think that the elasticity of demand would fall THAT far after the hurricane.

    Although that doesn't seem to have stopped you from making some pretty sweeping policy recommendations (quantity regulations?).

    1. PS* when I say "ATP (Ability to Pay) concerns" above, I am really just talking about distribution concerns (working poor not being able to afford gas to get to work type stuff).

      I honestly thought this was what everyone meant in this thread when they talking about ATP concerns. But Andrew's post makes me wonder.

    2. I was talking about adenosine triphosphate.

      Now I really feel embarrassed.

    3. That must be what you meant. Since Andrew said that the key issue is whether WTP exceeds ATP and you agreed.

      I know you couldn't have been agreeing with the notion that Willingness to Pay might exceed Ability to Pay, since as an Economics graduate student you know that WTP incorporates the budget constraint (and thus can never exceed ability to pay).

      Though based on the class schedules you've posted you've never mentioned a price theory or micro course as being part of your core curriculum at AU. So maybe I presume too much?

    4. I am not sure where he said the key issue is whether WTP exceeds ATP (certainly not what I agreed to above). The point is whether the constraint on WTP is ATP. Presumably in a lot of cases we care about, it does. I haven't been convinced that's a big issue here but we certainly bump up against it more. WTP in the economic sense and "willingness" in the sense that we usually use the word are quite different things. That's all I've ever known Wayne to say.

      I'm sick of this taunting shit. If you keep it up and you're going to start getting your comments deleted. The difference between what we think of the situation and what you do is pretty clear. I don't know how insecure you have to be to continually ignore those real points of difference and keep taunting like that.

    5. Daniel,

      I didn't ignore the real points of difference. I directly addressed them in several lengthy posts (maybe too lengthy). For example, you asserted the elasticity of demand for gasoline goes to near-zero after a hurricane. I raised (what I thought were) several reasons to be skeptical of that assertion--in posts that were civil and time consuming to write.

      And in return I either got ignored or a joke about ATP. *shrug* If I put effort into raising the bar of conversation and get lampooned, that makes me moody.


All anonymous comments will be deleted. Consistent pseudonyms are fine.