08
Oct
2013
Obama to Nominate Yellen for Fed Chief
Details here. And holy cow did you folks know this?!
[Yellen] is married to, and has co-authored a number of papers with, Nobel Prize-winning economist George Akerlof, whom she met in the fall of 1977 when they were both economists at the Fed board.
Must…resist…lemons…joke…
Never understood the market for lemons nonsense. As Wikipedia describes it (which is what I got from the paper):
“So the buyer’s best guess for a given car is that the car is of average quality; accordingly, he/she will be willing to pay for it only the price of a car of known average quality. This means that the owner of a carefully maintained, never-abused, good used car will be unable to get a high enough price to make selling that car worthwhile.”
Does not follow. Subjective preferences and all that. Blows the whole thing out of the water. Whoops. (On a side note, I marvel at how often basic economic concepts like subjective preference, marginal utility, supply and demand, etc. utterly demolish the “spophisticated” reasoning of big name economists.)
Could you explain what you mean? Preferences are subjective in the model as far as I know.
What is “average quality”? Quality is subjective. A car of “average” quality to me is not the same as it is to you.
Also, the statement that an owner of a high-quality car being “unable to get a high enough price to make selling that car worthwhile” implies a third party can evaluate the owner’s subjective preferences regarding the high-quality car vs. cash. It is completely plausible that the owner of a high-quality car may be willing to accept a “low” sale price, as he prefers what he can do with the money more.
Yeah, you can’t take the average of a strictly ordinal metric, like finding the average name in the phonebook.
You can average the prices people pay, but what do you then compare that number against?
You can consider the going rate to have an expert mechanic go through and check everything, they check which percentage of buyers can be bothered doing that ( not many in my experince, suggesting that the going rate for information is lower than the cost of obtaining it).
And don’t forget that preferences change over time and that goods have time value as well.
The biggest gaping hole that I see in the theory is that people are not gods and can not know the correct price of something much less the relative prices based upon quality. In the stupid “Lemons” theory I see the largest hole which is that you can not determine both the supply and demand curves as they are both based upon subjective preferences. Therefore no one knows for sure what the average quality price, or good quality price or crappy quality price is.
Yikes. The lemons hate on here is painful to read.
Look guys. Forget the “people are not gods” thing. Nobody is claiming they are and if you get a whiff of that it’s for modeling purposes and doesn’t really impact the key result.
If there’s asymmetric information and some broad consumer agreement on what constitutes quality then high quality goods are going to be driven out of a market where they can’t distinguish themselves. Of course everyone is on the same page with subjective preferences, so if we were to add the case of consumers that disagree with the majority on what constitutes a high-quality used car (Tanous’s point I think), that consumer would be happy as a big in mud. Fine.
But who cares?
The interesting point here is that while preferences are subjective they seem to be highly correlated across people in a lot of ways. As long as those are the facts on the ground this is an interesting model. And if you add consumers with different preferences, guess what – the model still works out up to a reasonable tipping point where “high quality” becomes “low quality” and then you’ve just the original model only inverted. That’s dumb and uninteresting.
I have to say that I agree with pretty much everything you said here DK.
Perhaps I can shed some light on the “lemon hate”, though I can really only speak for myself. As an anarcho-capitalist, this result is not appealing to me at all on an emotional level. The first time I encountered it, I felt it was attack on market principles and perhaps even my value system. It is of course no such attack, but simply a particular case where the potential gains from innovation (anything that changes “the facts on the ground”, such as reducing information asymmetry through a consumer reports organization or creating a way for firms to distinguish their product etc.) are more obviously identified.
I also think the problem is presentation of the argument. If someone simply stated “If there’s asymmetric information and some broad consumer agreement on what constitutes quality then high quality goods are going to be driven out of a market where they can’t distinguish themselves.” (nicely put by the way) and made sure to precisely define their terms like “asymmetric information”, and then gave an example of market that might be susceptible to this problem (a better example than the used car one), you would not have seen many complaints in the comment section.
The point is that information is a good, and it is perfectly reasonable for there to be a market value of this information.
If you don’t believe that information is a good, then what is the entire IT industry doing? Or the advertising industry? Or a huge range of consulting industries?
If you do believe that information is a good, then the cries of “market failure” over a crappy second hand car must prove that there was no possible way for the buyer to also purchase some sort of inspection that would compensate for the information assymetry ( or for that matter the seller could offer some warranty or other certification ).
I for one find the old formula pretty annoying:
* in theory there might be a market failure if you take the right assumptions
* in practice some people are grumpy for various reasons and feel they could have done better if only something something but never my fault
* government regulation will fix it!
* the regulation is tried and doesn’t fix it
* capitalism is to blame ( once again )
> It is of course no such attack, but simply a particular case where the potential gains from innovation (anything that changes “the facts on the ground”, such as reducing information asymmetry through a consumer reports organization or creating a way for firms to distinguish their product etc.) are more obviously identified.
Along the lines of what Tel said, there are already entrepreneurial entities dedicated to reducing the information asymmetry, for a fee. The entire consulting industry comes to mind immediately, as well as Consumer Reports.
So if this paper is simply identifying another way in which the market “cares” for its participants, then fantastic! It sure seems like everyone wants to talk about market failure.
“if we were to add the case of consumers that disagree with the majority on what constitutes a high-quality used car (Tanous’s point I think)”
My point was actually to add the case of sellers that either disagree, or don’t actually care about taking a “lower” price than otherwise.
And all this, of course, exempts that part of the used car market wherein people that know each other well (say friends or relatives) are selecting buyers. My grandmother, for instance, selling her used car to my cousin – it’s a high quality car, but she was willing to take a lower price to (a) help my cousin out as she went to college while (b) obtaining some amount of money to use towards a newer car.
Anything that wasn’t written by an Austrian and seems (because anybody who has read the paper knows the following isn’t necessarily true) to undermine some unsophisticated free market ideology obviously doesn’t assume something as basic as subjective preferences.
Guys, you can cut down the snark some, I think. Clearly interventionists have used Akerlof’s approach to justify more government intervention in markets. So people aren’t inventing phantom menaces to the free market.
The comments, which led to snarky reactions, were not merely objections to the misuse of the result in order to justify government intervention. They were instead claims that the result was obviously and stupidly wrong because it relied on the denial of fundamental economic principles.
^
I subjectively prefer the “average quality” price more than my “high quality” car. The paper presents that as not only against the “general rule of thumb”, but as downright impossible. That’s a failure to properly account for subjective preferences.
It, in short, presumes that “average quality” maps to the market price for every single individual – that every consumer will guess that the “average quality” car is worth $X (and not be willing to pay more) and that sellers aren’t willing to take what is deemed the price for an “average quality” car. The assumptions involved don’t hold – and can’t if we account for subjective preferences and inverse valuations.
The model presented is that Consumer A assumes that the car is of “average quality”, which is associated with Price X. Seller B knows it is of “high quality”, associated with Price X+Y. Thus, we are told, “high quality” sellers like B won’t sell their cars – only “average” or worse cars will be sold in the market. Problem is, there is no single function between price and “quality” (itself subjective)! Each person has a different evaluation! Consumer A may believe the “average quality” car is worth $5K, while Seller B thinks his car – despite its “high quality” – is only worth $4.5K. In such a case, the whole model disintegrates.
My objection to it has nothing to do with it not being Austrian, but by its ignorance of basic economic truths. It’s the same problem with Keynes’ nonsense in the General Theory – it utterly ignores supply and demand and other basic considerations in describing individual behavior.
Inasmuch as there is a tendency for people to value their high quality car more than price of an average quality car low quality cars will tend to price out high quality cars.
I think you’re getting hung up on the example a little too much (which is a poor example by the way).
However, consider a market where firms produce statistical models in order to help retail chains choose better locations for their new stores:
Let’s say a high quality model takes 10 days to produce and a low quality model takes 9 days to produce. Now consider the fact that most people know very little about statistical modeling and furthermore a company producing a high quality model may be using their own innovative technique that they don’t want their competitors to know about. In such a scenario, it would be difficult for consumers (the retail chains) to distinguish between high and low quality models because little information about the products would be available to consumers and what little information consumers do have they may not even fully understand. The result is that the higher quality models which take longer to produce are priced out.
“Inasmuch as there is a tendency for people to value their high quality car more than price of an average quality car low quality cars will tend to price out high quality cars. ”
There’s about 4 different subjective evaluations involved here. Not only must we assume that the consumer is picking the price of the “average quality” car, due to a lack of information, but we are also making drastic assumptions regarding the subjective valuations of both consumer and seller.that don’t generally hold.
“In such a scenario, it would be difficult for consumers (the retail chains) to distinguish between high and low quality models ”
That’s decidedly wrong, as the industry in which I work (hard drive production) clearly demonstrates. It is quite easy to distinguish between high and low quality models, despite the lack of understanding of statistical modeling and feature set function by consumers and competitors.
“That’s decidedly wrong, as the industry in which I work (hard drive production) clearly demonstrates.”
There are easily measurable features of hard drives. There are not easily measurable features for statistical models. My girlfriend works in this industry and there are significant pressures for her to produce lower quality products for just the reasons I stated.
Again, you’re simply getting hung up on the particular example and how it might not neatly fit the characteristics of the market described in the thought exercise. Such a market doesn’t even need to exist in order for the result to be true.
“There are easily measurable features of hard drives. There are not easily measurable features for statistical models.”
Given that statistical models are all hypothetical information, anyway, that’s to be expected. There isn’t even a way to measure “quality” there.
That sounds like a contractual failure, if the modelling company is confident that its result improves sales then it should be sharing in the risk somehow like getting paid a percentage of sales above some point ( their quality models will tell them how to put together a very attractive offer and still make a good buck ).
Once the model becomes directly connected to the bottom line of the same company that is producing it, suddenly the boss starts wanting good models.
It doesn’t. Read the paper.
I read the paper before, and that’s the picture I got. Can you explain what it is that I am missing here? Because it is a perfectly plausible scenario that the vast majority of holders of “high quality” cars value the cash they get from selling more than the car, even at the “average quality” price point. Either this scenario is discounted explicitly or implicitly, but it certainly must be in order for the conclusions of the article to hold.
Just to be clearer, in the paper, Akerlof assumes heterogeneity of preferences.
But the higher quality goods can always distinguish themselves, by brand name, service logs, contractual warranties, third party expert inspection, and dozens of other ways. Point is, most buyers show their preferences by not wanting all that stuff ( or not being willing to pay for it ).
>>>high quality goods are going to be driven out of a market where they can’t distinguish themselves.
Possibly one of the most charmingly naive statements on this thread. Congratulations!
(BY DEFINITION, a “high quality” good is one that has ALREADY distinguished itself in the mind of a consumer, or the minds of many consumers, as being “high quality.” There is no objective criterion of “high quality” any more than there is an objective criterion of “low quality”. It’s always “quality as evalulated by someone in the market.” It’s not high or low quality objectively apart from someone’s evaulation. That you simply don’t get this, Kuehn, shows that you simply don’t grasp the concept of subjective preference.)
Wonder if they have kids. Would love to drive by a neighborhood where there is a sign with “Akerlof’s lemonade stand” and ol’ George sitting there with his grandkids.
I’ve never read the original “lemon” paper. Alex Tabarrok argues that we don’t seem to see that effect in the used truck market. Is there empirical support in Akerlof’s paper, or is it just theoretical?
In his original paper, Akerlof has a short discussion of various institutions that help remedy information asymmetries.
Voluntary institutions or statutory imposition?
For example, the “lemons” theory has been used to justify all sorts of expensive compulsory building codes on the basis that the consumers must be protected from their own stupidity.
Voluntary.
A little bit of self-promotion. Because my above comment was (I admit) snarky, I try to explain what’s useful in Akerlof’s paper: “How to Read ‘Market for Lemons’.”
“At this price, p, sellers have an incentive of getting rid of their worst quality goods first.”
Do people generally have multiple 2002 Civics sitting in their driveways that they want to get rid of? Certainly, if I had two of them, I’d want to get rid of the one in worse shape (subjectively) were I to sell only one… but why would I even have two to start with?
“Furthermore, sellers of good quality goods, even if they don’t have bad quality goods, have an incentive to not put their goods for sale on the market.”
This is my main dispute here. They have such an incentive if they value keeping the car more than $p. But why would we assume that? Most sellers of used cars, from my experience, are not really looking to keep the car – they want the highest price they can get, but like houses, the decision to sell is not simply, or even mainly, a result of price, but more of a need or desire for money instead.
I disagree that you need Akerlof’s paper to understand the creation of institutions that reduce or eliminate information asymmetry. It simply requires that the seller speculate that an increased profit potential exists – even if they are willing to take $p, they could see that if they spend $x, they could sell it for $(p+x+y), which would be worth the extra investment. Or perhaps the consumer wants to spend $p, but is willing to spend $x more on ensuring the product he gets is of good quality. Or some combination thereof. It’s not necessary to postulate or demonstrate that higher quality products can’t or won’t be sold.
I’d note also that my disagreement with Akerlof’s conclusions and method are not, as supposed in your article, due to any stance that they could be used for intervention – I acknowledge that “Group 3” exists, and if I am wrong about his conclusions, I’d be happy to exist in a “free market will fix it” stance. I just happen to think his simplified model, like many “macro” models including Keynes’ (as Hazlitt showed in great detail), unintentionally (or intentionally) neglects or does not give proper due to basic economic concepts, leading to error.
Think of a used car dealership. More importantly, it’s a hypothetical, to explain one side of why the average quality of the car will fall.
It’s an externality argument. Some of those sellers will sell their cars. But, we’re talking about margins, i.e. there will be a number of sellers who don’t enter the market.
Trust me, there is no neglect of basic economic concepts.
“But, we’re talking about margins, i.e. there will be a number of sellers who don’t enter the market.”
Sure, but then that’s the same the other way around – there will be sellers of “low quality” vehicles who hold that their vehicle is still worth more than p, and thus refrain from selling.
But, the number of sellers of poor quality cars will still be greater than what it should be (compared, that is, to a perfectly competitive market at price p).
>>>But, the number of sellers of poor quality cars . . .
Sorry, I don’t understand that term. Please provide an objective definition of “poor quality cars.”
Does it mean,
1) Cars that the seller doesn’t like;
2) Cars that the buyer doesn’t like;
3) Cars that you don’t like.
For the first question, actually, think of a private seller with multiple used cars. Assume there are no brands, such that the only way to distinguish between the cars is to rely on what the seller tells you. This isn’t a realistic assumption given what we know about brand names, but that’s actually Akerlof’s (explicit) point: the reason brand names exist is to allow consumers to better judge the quality of the car, so that they can discriminate between different types of cars.
> the reason brand names exist is to allow consumers to better judge the quality of the car
That’s a rather backward way of putting it. That’s not the reason they exist — brands are carried by vendors for the vendor’s benefit rather than emptor. In fact, many brands are sold from higher-quality producers to lower-quality producers, in order to fool consumers into paying the higher-quality price.
I would say that what you are describing is a valuable role that brands play, analogous to, say, the role of speculators in commodity markets (the reason they are there is not to provide a counterparty to farmers).
Nice writeup. It sure seems to me like this only applies to goods for which quality is impossible to distinguish. Using the 2002 Civic example, there are tons of indicators of quality available to the emptor — mileage, bodywork & paint, tire wear, engine noise, rust, leaks, etc. Only someone completely unconcerned with quality will refrain from ascertaining quality.
So sure, asymmetric info. But everything you and Akerlof seem to describe are situations where the emptor literally cannot evaluate what he aims to purchase. Rather than 2002 Civics, I think you need to talk about lottery tickets.
I have no doubt there is a “market” for lottery ticket scams. Extending the Lemon conclusions to markets beyond literally indistinguishable (yet variable) goods is where I get uneasy.
FYI, here’s where you lost me, Jon
> Because I can’t tell the difference between the various 2002 Honda Civics on the market — as far as I know, they are all of similar quality —, I have to accept a more-or-less uniform price for this good
Also, what about Group 5 — those who, based on the conclusions’ champions or not, examine the premises, find them wholly unsatisfying, and reject the conclusions? How safe is it to say that Group 5 doesn’t exist?
http://en.wikipedia.org/wiki/The_Market_for_Lemons
George Akerlof 1970 paper
http://en.wikipedia.org/wiki/File:Kovacs_special_1968.JPG
Ernie Kovacs as a used car salesman
This is a riot. The Wikipedia article on Akerlof’s paper posts a photograph that it captions, “A used car dealer with a low-priced used car.” But the photo is actually of television comedy genius Ernia Kovacs, probably a still taken from some skit he performed on his show.
Talk about information asymetry! Try learning a bit about American popular culture and television history before pretending to be profound and relevant.
Kovacs’s most famous skit, of course, was his “Nairobi Trio”, linked here:
http://www.youtube.com/watch?v=DoLTFQsFswM
Like the mysterium of Keynesian economics, this performance of the Nairobi Trio requires that we put ourselves into a Zen state of “Prana”, or perhaps “Beginner’s Mind”, to grasp its meaning fully.
Note the philosphically profound lyrics:
Mi, Sol, La
Re, Fa, Re, Sol,
Do, Mi, Do, Fa, Re
Sol, Sol!
Mi, Sol, La
Re, Fa, Re, Sol,
Do, Mi, Do, Fa, Re
Sol, Sol, Do!
Do, Mi, Fa
Si, Re, Si, Mi
La, Do, La, Fa, Re
Sol, Sol!
Do, Mi, Fa
Si, Re, Si, Mi
La, Do, La, Fa, Re
Sol, Sol, Do!
(Repeat as many times as necessary until you see the light)
And like the mysterium of Keynesian economics, to those who “get” the profound meaning of all this, no explanation is necessary; to those who do not “get” it, no explanation is possible.
Hint for the uninitiated: Think of the trio members as Krugman, Krueger, and Kuehn, and all will be clear.
http://blogs.telegraph.co.uk/finance/ambroseevans-pritchard/100025713/rejoice-the-yellen-fed-will-print-money-forever-to-create-jobs/
On the Lemons topic:
Note the oblique reference to market failure, from a guy who is intelligent, articulate and nominally conservative. Bob’s point: “Clearly interventionists have used Akerlof’s approach to justify more government intervention in markets,” is supported right here.
Anyhow, on the QE and money printing topic, AEP is solidly Keynesian, thinks Yellen will do a wonderful job of money printing and thinks the job situation is just about to turn around. Almost every comment on his blog is expecting the opposite.