401. MsIvoryTower - July 7, 1999 - 6:00 AM PT
And then, of course, when one talks about the insurance industry, even there it requires some differentation between type of insurance coverage (life, health, auto, home, mortgage, travel, etc).
402. MsIvoryTower - July 7, 1999 - 6:24 AM PT
Yet another issue that ought to be fairly clear is unless a huge, and unexpected cost stemming from fraud and corruption is a one time event, at some point the 'unexpected cost' becomes absorbed into the firms planned costs, and will have an effect on prices (or services). Since fraud is a continuing problem in the insurance industry, it does actually raise costs to the consumer because firms include monitoring for fraud and increased security costs at the very least.
403. stostosto - July 7, 1999 - 7:00 AM PT
Re: Passing costs on and all that
I just thought I'd mention one possibility which may arise in a market with perfect competition (i.e. each supplier faces infinitely price elastic demand in the market): The firm hit by unexpected costs goes bust. In this case the cost is passed on to creditors and owners.
Of course, this has been known to happen, even in markets with less than perfect competition (as are most markets).
(Btw, I think Slacks introduced some talk of elasticity of supply as well as demand. When we speak of unexpected cost and the feasibility of passing it on to consumers, I'd say that the relevant issue is demand elasticity. Elasticity of supply would be relevant if the question was how large a change in production and/or price a given shift in consumer's demand would give rise to. This is not what we're discussing, is it? Just to clarify, I don't need an answer if this is a correct perception).
404. stostosto - July 7, 1999 - 7:03 AM PT
.... i have a feeling this discussion is pretty humdrum...
I think we need a boost of pseudoerasmusism. When is he coming back?
As Niner would say: Frankly, I miss him...
405. MsIvoryTower - July 7, 1999 - 7:25 AM PT
Hey!
Believe it or not, we can all handle an economics discussion without the presence of PE.
And yes, it would *appear* to be a fairly uncontroversial topic. When I made my initial comments over in NOTD, I thought them rather standard, but FTC and JJB took great exception to them, as if they were treasonous words. So, while you and I may think this humdrum, others apparently do not.
And I suggested several other topics earlier on, surely we can discuss something interesting without PE having to instigate it. What would YOU like to talk about?
406. stostosto - July 7, 1999 - 7:37 AM PT
MsIvoryTower
You are absolutely right.
My remarks was due to a sudden feeling of, well, boredom over and dismay at my own remarks here as it dawned on me that they were fairly trivial...
And, you know, frankly, I miss PE on a more fundamental level, so to speak, I just took this opportrunity to express it.
What would I like to talk about? Well, the future of capitalism is not such a bad idea. One aspect of which could be the American stock market. Is it a bubble? If so, when will it burst? And how bad will the burst be? Greenspan said something to the effect that a bursting bubble may not be disastrous to the economy - and cited the 1987 experience as an example of this. I'd like to hear some opinions on this.
407. gwindau - July 7, 1999 - 8:32 AM PT
I am reluctant to describe the current level of the US stock market as a 'bubble' but, IMHO, a severe stock market crash over a short period of time would NOT be a 'good thing' for the US economy. If we accept Mr. Greenspan's assertion that the 'wealth effect' of a soaring stock market is strongly contributing to high levels of domestic consumption, then we must conclude that paper losses on the stock market might bring real losses in consumer spending in the real economy. On the other hand, only about 28% of all Americans have invested more than $10,000 in the market, so perhaps a market crash would not affect spending in any real sense. The 1987 crash was precipitated (correct me if I am in error) by a surge in high-volume program trades that perhaps did not reflect the true mood of the human market players. Thus with effective monetary policies, the stock market in 1987 recovered rapidly.
408. Slackjaw - July 7, 1999 - 4:32 PM PT
uzmakk:
thanks for relaying your sister's comments on P&M. I am stunned she actually read it. It may come as some surprise to learn that my interest in game theory lies primarily in applications rather than the advancement of the theory itself. It is not my goal to have a refinement named after me!
However, it is fairly obvious to me, after repeated observation, that one cannot do good applied work without a good grasp of the theory. This is particularly clear if one peruses applications of game theory in political science, where the training in game theory is much weaker than in economics. Lots of wasted ink and breath because simple theoretical results are not understood.
In just the last week I have heard political scientists assert that game theory does not allow for the possibility that players make mistakes (which is false), and that assuming that the utility of some particular outcome is positive is a substantive restriction in and of itself (which is also false).
409. Slackjaw - July 7, 1999 - 4:32 PM PT
Selene:
probably you are gone now, but just in case:
I have actually never been against non-standard techniques like computational modeling. What I am against is the presumption that because somebody used a genetic algorithm instead of Greek letter models, they must have a decent simulation of an evolutionary and/or adaptive process.
As I recall our big discussion was over Axelrod's stuff, and my position still stands--it is provocative but very inconclusive. In particular, if you only allow strategies to condition on the last 2 moves of a repeated game, tit for tat starts to look very good. More general work published in the last two years, allowing for conditioning on a much longer sequence of moves (16) and including just a little bit of noise, finds that successful strategies (in a GA) in the repeated PD are much less forgiving of defection and much less willing to cooperate than tit for tat.
And, what you consider a straightjacket is useful if nothing else as a screening device. I *like* listening to criticisms of game theory, just not wrongheaded ones. At the workshop, a physicist from Los Alamos said, "The problem with game theory is that in order for someone to gain, someone else has to lose by the same amount." Sorry, that is just really wrong. And it's a pretty strong signal about the quality of someone's criticisms.
410. Slackjaw - July 7, 1999 - 4:32 PM PT
stostosto:
no, in general it is not just demand elasticity that is relevant in determining how much of a cost increase gets passed on. Suppose supply elasticity was infinite and the demand curve is almost but not quite horizontal. Who would eat cost increase? If only demand elasticity matters it would be eaten mostly by the supplier. But that clearly doesn't make any sense if supply is infinitely elastic.
The case where the cost increase effects only one firm and each firm faces infinitely elastic demand (ie, the market is competitive) is simply a special case of the usual rule. In that special case, as long as the supply curve is not horizontal or vertical, all the necessary information is contained in the demand elasticity: the firm must eat all of the cost increase. (In the same way, if supply is perfectly elastic but demand is not, then all the necessary information is contined in the supply elasticity.) In a static model of a competitive market the firm will go out of business, but if one considers that the principals have to do something else, it may or may not be more attractive for them to leave the market.
Not sure I grasp the distinction between prospective cost increases and past cost increases. They have to go somewhere either way; why should it matter for the comparative static?
411. thoughtful - July 7, 1999 - 5:56 PM PT
Don't know if any one would find this topic interesting or not, but it was asked of me, and I thought I'd pass it along. Someone asked about the role of rational expectations and whether or not policy is irrelevant.
Any takers?
412. thoughtful - July 7, 1999 - 6:56 PM PT
I could post my response, but I thought maybe I could get some argument first. I think most of you will know what side I come down on.
413. MsIvoryTower - July 7, 1999 - 7:08 PM PT
Thoughtful
Do you mean the implications regarding the role of policy under the rational expections model?
If so, I'd say that if all information is processed instantaneously, then the model implies there is no role for policy, since all markets adjust to equilibrium faster than required for policy to matter.
I recall a vertical supply curve under RE assumptions, suggesting that all fiscal stimulus does is translate into higher prices.
414. MsIvoryTower - July 7, 1999 - 7:11 PM PT
I mean, of course, a vertical supply curve under short run conditions.
415. thoughtful - July 7, 1999 - 7:19 PM PT
Well, the question was asked of me just about as I stated it, so it's open to interpretation as to what part of RE the questionner was referring. The Lucas model with price expectations says only unanticipated chanes in the money supply have an effect on output -- which, if you buy it, really puts Greenspan's most recent openness about monetary policy in question, no?
416. MsIvoryTower - July 7, 1999 - 7:26 PM PT
Thoughtful,
Does Luca's model even allow for that? I'm fuzzy on who said what, given that I studied all this more than a decade ago (a lot more, come to think of it), but I remember the RE model suggesting that NOTHING affects output but investment (by the private sector).
In any case, the answer to your last question would be "yes, it certainly calls into question Greenspans openness". However, I never believed in the RE model precisely because of some of its extreme implications regarding the role of monetary and fiscal policies in the economy.
I've always been taken by the notion of lags, it coincides with what I see in the real economy so well.
417. thoughtful - July 8, 1999 - 6:34 AM PT
Mit, y'know, it's been years since I've looked at this stuff too. I was at a Lucas lecture many years ago when he had a chart of US real GDP growth over this century. He carefully selected his data points and showed that US GDP growth averaged 3% over the period -- with or without income taxes, the gold standard, etc. His conclusion being that fiscal and monetary policies were ineffective in altering the long run growth in output.
I, of course, think this is balderdash. I came across an interesting statistic: Since 11/82, the US has spent only 5% of the time in recession vs. from 1853 to 1953, it was in recession 40% of the time. Clearly we've learned something about business cycles and economic policy. It isn't all just luck.
418. DaveCook - July 8, 1999 - 9:25 AM PT
Lucas would say that if we could trade the average real growth between 1853-1953 for the growth between 1982-1999, we would be happy to take 40% recession.
419. MsIvoryTower - July 8, 1999 - 9:28 AM PT
Who is "we" kimosabe?
Lucas might be happy to trade that, but I doubt the majority of Americans would be (including most American businesses).
420. Raskolnikov - July 8, 1999 - 9:34 AM PT
My suspicion is that proving that the two are as interchangable as that would be a difficult task.
I did my undergrad at the University of Minnesota, and my first economics classes were from RE instructors. It left a very bad taste in my mouth for economics that took years to recover from. I couldn't get beyond the problem that they were taking information and rationality assumptions too far.
421. DaveCook - July 8, 1999 - 9:55 AM PT
In any case, thoughtful's statistic's, in fact, support Lucas' theories not disprove them. Lucas believes that long term growth is a much more important concern than business cycle fluctuations. The basis of this is that changes in consumption of the goods that give people welfare are strongly associated with long term growth and move very little over the business cycle. So, then, over a 20 year period we should be willing to give up quite a bit of stability to get a little bit of long term growth. Of course, this is ultimately a prefererence ordering, so maybe some really, really, really, risk averse people would want to emphasize stability.
But anyway, Lucas (and Friedman's and the whole monetarist school's) point about monetary policy was that using monetary policy to target growth or unemployment would be destabilizing unless the monetary authority had much better information than private agents when they set their nominal variables. Now, to some degree, the monetary authority does have some better aggregate information in the short run. But only in the quite short run.
422. DaveCook - July 8, 1999 - 10:06 AM PT
And, so, thoughtful is certainly right we have learned something about monetary policy and what we have learned we have learned from Friedman and Lucas and the Chicago school. The lesson is that a monetary policy which focuses less on the stability of unemployment or growth (and more on the stability of inflation or money growth) is more likely to lead to stable unemployment. So, thoughtful points out the low volatility of output during 1982-1999, then its obvious to point out that this is the period in the post-war when the Fed has minimized the stability of output as a goal. This would come as little surprise to Lucas or other monetarists and a big surprise to Keynesians (not that they exist anymore).
423. thoughtful - July 8, 1999 - 12:18 PM PT
DC, let me point -->
--> First, I can assure you that those of us with strong Keynesian bents still exist.
--> Second, I didn't say Lucas proved his point at all -- and certainly not to me. Rather his conveniently overlooking minor economic disruptions like the great depression and two world wars made his chart look silly.
--> Third, volatility is expensive -- it distorts investment spending as risk premia drive up the cost of capital. It is no accident that strong US growth and rising productivity growth occurs in a period of relative stability in inflation and interest rates. The same was true for Japan. Even if growth was stronger in the 1850s-1950s, it says nothing about the costs of volatility, as we don't know what growth would have been in that earlier period *without* panics and depressions.
424. MsIvoryTower - July 8, 1999 - 2:44 PM PT
DaveCook
On Thoughtful's point #1: Ditto, although I prefer to call myself a neo-keynesian these days.
On #3: This seems to me to be a major flaw in your argument above, AND the fact that the government's fiscal policies during the 80's and 90's provided much of the stability needed for the Feds to focus on inflation control. Thank goodness Greenspan is not a zero-inflation advocate (what was suggested by early RE models), otherwise we might not have seen the sustained growth we have in this expansion.
425. thoughtful - July 8, 1999 - 5:39 PM PT
Neo or not. I don't consider myself a Keynesian in that it implies other economic theories are wrong. I don't believe that. I believe rather that each has had a contribution to economic knowledge. (I confess in the case of the supply siders the contribution was confirming what we already knew wouldn't work.) Monetarists have contributed a lot to understanding inflation and monetary policy; REs have contributed a lot to understanding expectations and imperfect information.
426. MsIvoryTower - July 8, 1999 - 7:10 PM PT
Thoughtful
Well, when it comes down to the theoretical wire, I'm more keynesian than not. Although I agree with you that we've come a long way toward integrating what works well from all theoretical perspectives, in the end, policies that flow from RE models, neo-classical or neo-keynesian constructs can result in very real differences in economic activity and economic opportunities for different groups within the economy.
So, when it comes to policy, I'm more neo-keynesian than neo-classical, and certainly moreso than an RE adherent.
427. MsIvoryTower - July 8, 1999 - 7:15 PM PT
And as I gain a deeper understanding of the real world economy I grow more attracted to problems that analyze the theoretical implications of imperfect information.
However, I've never really associated the growing body of work in this area with the RE school. I'd think it more of a reaction to RE models rather than flowing from them.
428. Amaxen - July 8, 1999 - 7:34 PM PT
Ivory,
A few things reading over the last fifty or so posts that I didn't quite catch:
When you talk about fraud and insurance companies, are you talking about costs of fraud by the companies or by the policyholders?
About about being a neo-k;
Are you saying, then, that you think fiscal policy is a practical tool for controling the economy? Back when I was going to economist school, they told us that fiscal policy is not really a practical tool because its too sticky, and monetary policy is to be used for inflation -related goals only.
I've always associated Keynesianism with the belief that these two tools can be used for more direct control over the economy(i.e. Agg Output, employment level, etc), but then that was about 8 years ago, and I haven't reading theory too closely since.
429. MsIvoryTower - July 8, 1999 - 7:39 PM PT
Amaxen
Well, how far did you go in your studies?
430. Amaxen - July 9, 1999 - 11:39 AM PT
Well, I got my M.A from the George Washington U, and then went on and took a job for the Commerce dept as a technology economist. My training focused on trade, and while I was there I did some work on post-communist economies.
431. MsIvoryTower - July 9, 1999 - 11:56 AM PT
Amaxen
Are you saying that you don't think fiscal policy acts as a stablizing force in the economy? If so, I suggest you look at Defense spending in the 80's and infrastructure spending in the 90's.
Re: neo-keynesians,
I'd say they back away from validating all forms of government spending, they recognize that it *can* and does (to a certain extent) crowd out private investment if it's not carefully structured, and tend to support government spending on infrastructure and other sorts of "investment" related spending (education, for one) rather than just spending per se. And I think neo-keynesians have pretty much backed away from using tax cuts/increases as a mechanism to control economic stability (ie trying to stimulate consumer spending).
They also recognize the need for monetary policies to focus on low inflation (not zero-inflation), however, Greenspan has been quite willing to use the discount rate as a mechanism to further support the current expansion versus a total focus on JUST inflation control.
So, neo-keynesianism, from what I understand, is now more tempered by the theoretical investigations of both monetarists and RE folks, but still sees a role for active involvement of government in the economy.
432. MsIvoryTower - July 9, 1999 - 12:22 PM PT
Btw, there's a two volume series of papers edited by Robert Hall (an ex-professor of mine whom I couldn't stand, but don't let that stop you) called _The New Keynesianism_, published in '91 or '92, that maps out some 'neo-keynesian' positions fairly well.
433. bubbaette - July 9, 1999 - 12:25 PM PT
Aw right folks. Please take the time to vote in today's Fray Presidential Elections. It's your Fray and your future, so vote early and vote often. Remember, democracy doesn't work unless *you* participate.
Vote vK and MsIT!
434. FreeToChoose - July 10, 1999 - 8:46 AM PT
In Message #428 Amaxen says:
"When you talk about fraud and insurance companies, are you talking about costs of fraud by the companies or by the policyholders?"
Actually, neither.
The incident that spawned the discussion was a corporation headed by Martin Frankel, that is alleged to have defrauded some insurance companies. If the allegations of fraud turn out to be correct, there will fraud against the insurance companies, but not by policyholders.
435. FreeToChoose - July 10, 1999 - 9:04 AM PT
Slackjaw
I had hoped you would find time to respond to Message #389 and Message #390. Is it possible that you missed them?
436. FreeToChoose - July 10, 1999 - 9:23 AM PT
In Message #391 MsIvoryTower says:
"Only in the short run, in the longer run, the company will shift the costs onto consumers, either by raising prices, or reducing standard coverage, or possibly both."
I suspect, as if often the case, that the disagreement can be resolved by careful definition of terms. I accept that, in the long run, virtually all costs must be passed onto the consumer or the company fails. This is pretty basic economics, and I have said nothing to disagree with this point.
If a significant proportion of insurance companies were the victim of some type of fraud year after year, this amount would have to be passed onto consumers (though not necessarily in strict proportion to the fraud incurred.) I haven't seen anyone in this thread, or elsewhere, for that matter, disagree with this notion.
However, the issue in question is a specific fraud against a small subset of companies. The question is to what extent this cost will be passed onto consumers.
When you say that costs might be borne by the company in the short run, but not in the long run, did you mean that this *specific* cost might be borne by the companies over the short run but would be recouped over the long run? My guess is that you are not saying this, but I would like to confirm.
437. FreeToChoose - July 10, 1999 - 9:23 AM PT
Continued
"Even in the short run, it's a big assumption to assume that demand is elastic with insurance, it depends on the type of insurance. If it's auto insurance, then the demand is likely highly inelastic because of state laws that often force people to have insurance or be banned from driving (no license). In those states, demand becomes inelastic, since the consumer MUST purchase insurance to begin with, they don't have the option of leaving the market."
You are confusing the elasticity of the broad insurance market with the elasticity of the specific policies issued by the companies subject to fraud. Mankiw (Principles of economics p.90) notes that "[n]arrowly defined markets tend to have more elastic demand than broadly defined markets." For the reasons you cite, the overall market for auto insurance probably has low elasticity, but the market for a specific auto policy for a specific driver is much more elastic. Given that the affected insurance companies comprised a tiny percentage of all companies, and none had any sort of monopoly power, the elasticity of demand is likely to be high. Consequently, economics suggests that the costs are borne by the company, not the policyholders.
438. FreeToChoose - July 10, 1999 - 9:35 AM PT
In Message #393 MsIvoryTower says:
"This is somehow remarkable in your view?
Nothing different about this from many other situations producers face."
If you notice, I was asking the question whether economics generally cares about the distinction. In other words, my question to you was whether economics finds this remarkable. You did point out that manufacturers operating under contacts to deliver goods face some of the same issues. I suspect that there will be an increase in such arrangements (as witnessed by GM locking up long-term contracts with suppliers) so it will become more important, but I still contend that, for most manufactured goods, changes in costs precede the decision as to the price. If you can show me that this changes nothing, I will drop the argument that it is a distinction, but it is generally an article of faith that the economics of insurance differ from the economics of manufactured goods due to this difference.
439. FreeToChoose - July 10, 1999 - 9:40 AM PT
In Message #395 MsIvoryTower says:
"B tw, supply is always elastic in the long run, in competitive equilibrium (and I'd argue even with imperfect markets). The firm always has the ability to rearrange factors and to shift costs to consumers."
Of course (with the obvious exception that some firms are unable to shift the costs and go out of business). However, this does not answer the question whether the specific one-time cost of fraud incurred by a few insurance companies can be passed onto consumers. I'd say that the general discussion has supported my view that this cost will not be passed onto consumers. I'm guessing that you haven't yet been persuaded.
440. FreeToChoose - July 10, 1999 - 9:50 AM PT
In Message #396 thoughtful says:
"Admittedly knowing very little about it, I would suggest a big factor allowing insurance companies to pass along costs is very imperfect information for consumers."
I agree. This is an area where newspapers, general interest magazines, financial magazines, and consumer oriented magazines, such as Consumer Reports, could do a better job. I do see quite a few articles along this line, but there is lot's of room for improvement.
"I would imagine the possibility of dealing with strong-arm insurance salespeople also discourages consumers from switching."
I haven't had the experience of strong-arm salepeople. Is this comment based upon personal experience?
"Perhaps the internet will change that."
It will take time, but I think you are right.
here is one example of an insurance source on the web. Many companies are actively looking into it. In some cases it is existing companies wanting to have a presence on the web; in other cases it is companies not in the insurance business who think that web based insurance will revolutionize the industry. Brian Arthur believes that the latter may prevail. It will be interesting to watch.
441. FreeToChoose - July 10, 1999 - 9:56 AM PT
In Message #397 MsIvoryTower says:
"I was thinking of the role of imperfect information in this process as well. It works mainly to the advantage of the supplier, IMO."
When thoughtful mentioned imperfect information, I was thinking information asymmetry. I suppose there might be a technical distinction, but broadly speaking the issues are similar.
Both parties to an insurance transaction have information asymmetry. Both exploit it to their benefit. It might be an interesting exercise to determine whether the supplier gains more than the customer does in general. You might well be right, but I don't think the issue is as clear-cut as you suggest.
442. FreeToChoose - July 10, 1999 - 10:03 AM PT
In Message #399 stostosto says:
"I don't quite follow this insurance cost discussion. Is the question specific to insurance or is it a general question of whether producers can pass unexpected costs on to consumers?"
There really are two (or more) questions. The original question related to a specific one-time cost borne by a few insurance companies. The question is whether this cost would be passed onto policyholders. JJB said no, I largely agreed, the Ms disagreed. I brought the narrow question over here to see if some of the economic regulars would weigh-in and also asked the broader question regarding what circumstances costs can be passed on.
"If consumers can easily choose a different supplier, the producer has to bear the cost."
I agree. This may still leave the question unanswered if there is disagreement about whether a consumer can change insurance companies, but evidence shows that they can (at least in property casualty; the issue is stickier with life).
443. FreeToChoose - July 10, 1999 - 10:11 AM PT
In Message #400 MsIvoryTower says:
"FTC would like this conversation to be exclusively about the insurance industry, but I've been floating between a general discussion and one directed at this industry specifically."
????
As I said, there are two questions, one about the specific fraud and the other about how costs are passed on in general. I've tried to discuss both. I have more knowledge about the insurance industry, so I find it easier to construct examples using insurance, and I can speak with a bit more authority about insurance, but I think I was the one who tried to broaden the discussion to other areas.
"I don't think it matters that much to the points I've been making anyway, although it seems to be a barrier to understanding for FTC."
Are there points you don't think I understand?
"In any case, I'd hardly label the insurance industry as highly competitive, given that the number of insurance firms is not that large in any one state."
There are over 2 thousand insurance companies in the US, albeit many operating in some, rather than all states. Articles I've read that were written by economists about the insurance industry tend to conclude that the industry is highly competitive. There are some obvious exceptions, but many of the exceptions are in the commercial arena, not the personal lines. Can you quote an article that concludes differently?
444. jayackroyd - July 10, 1999 - 10:48 AM PT
Message #420
When was that Rask? I was in the graduate program at U of Minn in the mid-80s. Had both the best graduate assistant I'd ever had there, and the worst as well--all in one micro theory course.
On the value of RE, in general, I found myself persuaded by Sargent, Sims and their fellow stochastic processors that:
1) Simple policy regimes are better than complex ones.
2) Interventions often have unanticipated consequences.
3) The multiequation macro models subject to the Lucas critique were in fact terribly flawed by not incorporating the effect of the models on the actors.
4) Keynesians place an awful lot of weight on money illusion, which seems to imply that people are pretty stupid. People aren't stupid.
On thoughtful's comment that we've clearly figured something out about managing the economy--that the recent extended expansion is not just luck, I have my doubts. It could well be luck. There are some screwy things going on right now. Transactional friction is reducing, really fast, which has to be a deadweight gain to the economy. Inflation, even at current low single digit levels, seems overstated to me. The only thing in my life that's gone up in price is movies. Everything else is even to down. There was, and continues to be, waves of corporate downsizing, after which the companies output doesn't drop, and all the downsized go get jobs elsewhere.
Nobody planned any of this. Nobody predicted any of this. No policy maker created this, because nobody believed it was possible.
Is it technical change? Europe and Japan are technically changing too, and they're in the dumpster. Luck sounds good to me.
445. MsIvoryTower - July 10, 1999 - 7:18 PM PT
"I haven't seen anyone in this thread, or elsewhere, for that matter, disagree with this notion."
Well, then, you should go back to the original thread (NOTD?) and read the exchange again. JJB clearly said that companies would eat their losses. As I clarified moreso the idiocy of this position, you came along and told JJB he was correct.
"However, the issue in question is a specific fraud against a small subset of companies. The question is to what extent this cost will be passed onto consumers."
This may be the issue you want to play with now, but it is not now, or has it ever been the issue I discussed. Just to be clear here, my original point to JJB was that fraud and corruption in the private sector can, and does, end up being as costly to the public (consumers) as does fraud and corruption in the public sector.
That was, and is, the only point I was making in my original example coming from Wexxfords post.
Personally, I doubt you can estimate how much fraud that occurs in a 'specific subset of companies' will be passed on to consumers without the empirical data to analyze it, and even then, one would want to know at what point the full costs are passed on (when the long run arrives). This would change for each incident, and even for the same subset of companies given different market conditions.
446. MsIvoryTower - July 10, 1999 - 7:39 PM PT
"When you say that costs might be borne by the company in the short run, but not in the long run, did you mean that this *specific* cost might be borne by the companies over the short run but would be recouped over the long run?"
Why does this matter to you? At some point, if the fraud is more than a one time event (which it typically is), the company will begin fully recouping these *anticipated* losses, one way or another.
The only scenario I can think of where the company might not ever recoup these costs would be if it was a single event, not repeated for a significant number of years, then the company could choose to take a one time loss and take lower profits rather than raise prices.
"You are confusing the elasticity of the broad insurance market with the elasticity of the specific policies issued by the companies subject to fraud."
I'll assume you're joking here.
The comment was meant to illustrate that the type of insurance bought is subject to different elasticities of demand, regardless whether they are issued by one parent company or not.
Howver, this does raise an interesting issue. If an insurance company issues policies across a broad range of products (auto, life, mortgage, medical, etc) can they more easily shift costs across consumer groups depending on what types of insurance have more inelastic demands than others? I wouldn't be surprised if insurance companies did this to some degree.
447. MsIvoryTower - July 10, 1999 - 7:47 PM PT
"For the reasons you cite, the overall market for auto insurance probably has low elasticity, but the market for a specific auto policy for a specific driver is much more elastic."
This doesn't follow at all. Nor does it flow from what Mankiew suggests.
Actually, the comment doesn't make any sense to me at all. You're defining elasticity of demand at the individual level, which will always be MORE elastic than the aggregated market, because individual tastes and preferences vary quite significantly. If taken to the extreme this would result in an infinitely elastic demand curve, precisely the conditions of a competitive market.
However, even in a competitive market, this only defines the demand curve facing each 'individual' firm, not the industry as a whole. Nor does it suggest that prices are never increased as a result of higher costs. What it suggests is that this new cost will drive competitors out of the market until supply (at the industry level) falls enough for prices to reach a new, higher equilibrium. The new prices then would reflect the higher costs.
"Given that the affected insurance companies comprised a tiny percentage of all companies, and none had any sort of monopoly power, the elasticity of demand is likely to be high. Consequently, economics suggests that the costs are borne by the company, not the policyholders."
No. And for the reasons I stated above. This can only be true if the cost is a one time event, otherwise, if companies cannot pass costs onto consumers, some will begin exiting the market, and overall, supply will contract, reducing the amount of the product available at the old price, and eventually driving price up to a new level (given that demand doesn't change).
448. MsIvoryTower - July 10, 1999 - 7:56 PM PT
"If you can show me that this changes nothing, I will drop the argument that it is a distinction, but it is generally an article of faith that the economics of insurance differ from the economics of manufactured goods due to this difference."
I don't see how it changes anything. And I don't see why I should show you otherwise, given your article of faith. Seems like you're the one who should indicate how this is such a big deal to the insurance industry wuch that it operates differently.
"The original question related to a specific one-time cost borne by a few insurance companies."
No, it didn't. My comment, which JJB took such exception to was not restricted in such a way. JJB's comment suggesting that companies don't pass such costs onto consumers was not restricted as you suggest.
How you interpreted the conversation which you leaped into is your problem. However, you're engaging in revisionism at this point.
"Are there points you don't think I understand?"
Yes, there are. I've identified them repeatedly.
449. MsIvoryTower - July 10, 1999 - 8:01 PM PT
Jay Message #444
Well, I don't think those were revelations attributable only to RE theorists, particularly #2. But I grant you we operate in a different theoretical world now than we did in the era prior to the RE explosion. I don't know many mainstream economists who are not more flexible in what they use theoretically than those from the pre-80's era.
Thoughtful is correct in that no one really adheres to one rigid theoretical school anymore, however, I think economists still have theoretical 'bents' which are more neo-classical (including RE) or more neo-keynesian.
450. FreetoChoose - July 11, 1999 - 3:53 PM PT
FTC "However, the issue in question is a specific fraud against a small subset of companies. The question is to what extent this cost will be passed onto consumers."
MsIT “This may be the issue you want to play with now, but it is not now, or has it ever been the issue I discussed. Just to be clear here, my original point to JJB was that fraud and corruption in the private sector can, and does, end up being as costly to the public (consumers) as does fraud and corruption in the public sector.”
Now I am confused. I thought it was clear that we were talking (inter alia) about a specific incident. Here is why I thought so:
You said, “You pay for it even though you don't have to confront it. Who do you THINK pays for that 100Million bilk?”
This sounds like a discussion of the particular fraud in question.
You also said, “Lets just take the 100 Million dollar internet insurance scam discussed in the article I linked above.”
This also sounds like a discussion of the particular fraud in question.
And I said, “This particular example is one that is more likely to be passed onto parties other than consumers.”
I tried to make it clear that I was talking about the particular fraud in question. Did you think we were talking generically about fraud?
I tried hard to make it clear that I was interested in both questions: the question of whether a particular fraud would be passed onto consumers, and the extent to which costs in general are passed along. My apologies if I wasn't clear enough.
Can you reiterate your position with respect to the specific fraud in question? I content it will be largely borne by the companies, not by the consumer. Do you have a position? (I thought you had clearly stated it, but now I see that you claim to have been talking about something else.)
451. MsIvoryTower - July 11, 1999 - 8:59 PM PT
FTC
Is this a one time event in the insurance industry? Other than this one event, has the insurance industry never had other losses due to fraud and corruption?
If so, then the answer is that in this *specific* case, the costs will most likely be eaten by the companies involved. However, the reality is that fraud and corruption are a large part of the insurance industry. There have been numerous reports regarding the annual losses to the industry *as a whole* from fraud. These have already raised costs to consumers significantly, in the form of higher security costs, higher claim verification costs, more paperwork for documentation of claims, even more labor intensive demands on the system of verification and prevention of fraud itself.
There is no reason to believe that an event like this won't produce higher costs to firms throughout the industry, as they move to shore up fraud coming from this new source (over the internet, now). For this reason, I would argue that this loss will only be absorbed by the firms involved in the short run, over the longer run, all firms in the industry will raise prices as a result of these continued losses due to fraud.
452. MsIvoryTower - July 11, 1999 - 9:00 PM PT
As for my position outlined in the other thread, I started with a comment pulled from Wexxford's post, stating that fraud and corruption result in higher costs to consumers not only from the public sector (wasted taxes) but also the private sector, through higher prices (or reduced services).
JJB then came back and said that this wasn't so, that private industry doesn't pass costs onto all consumers, and as a consumer he could avoid individual firms that conducted bad business. That's when I brought in the Wexxford insurance scam example, stating that even if he thought he could avoid the costs of such fraud in the private sector, that we all bear those costs as a result of higher prices for those goods (services).
As far as I'm concerned, the insurance example was always just an illustration of the larger point I was making.
453. cdm1110 - July 12, 1999 - 2:05 AM PT
Dropping in briefly just to endorse the thoughtful/MsIT view that the schools of thought in macro are much less sharply defined than they used to be. I think almost all macroeconomists now basically endorse RE, at least as the best starting point for modeling. But equally, I don't think many macroeconomists still subscribe to the strong policy ineffectiveness results of the early RE revolution. Things seem to be rather more eclectic now, with many macroeconomists willing to think both in terms of the dynamic equilibrium models associated with RE and (particularly) real business cycles and also in terms of the distortions and price stickiness emphasized by Keynesians (new, neo, whatever).
BTW, MsIT mentioned a two volume collection of papers on new Keynesianism, saying it was edited by Robert Hall. I think that the editors were actually Mankiw and Romer -- at least, they did have a publication that matched the description, while I don't know of anything similar edited by Hall.
454. jayackroyd - July 12, 1999 - 5:46 AM PT 456. FreetoChoose - July 12, 1999 - 6:08 AM PT 457. MsIvoryTower - July 12, 1999 - 6:31 AM PT 458. MsIvoryTower - July 12, 1999 - 6:38 AM PT 459. MsIvoryTower - July 12, 1999 - 6:50 AM PT 460. FreetoChoose - July 12, 1999 - 6:58 AM PT 461. MsIvoryTower - July 12, 1999 - 7:47 AM PT 462. FreetoChoose - July 12, 1999 - 8:10 AM PT 463. MsIvoryTower - July 12, 1999 - 8:22 AM PT 464. FreetoChoose - July 12, 1999 - 8:34 AM PT 465. Raskolnikov - July 12, 1999 - 10:29 AM PT 466. jayackroyd - July 12, 1999 - 10:54 AM PT 467. jayackroyd - July 12, 1999 - 11:01 AM PT 468. thoughtful - July 12, 1999 - 1:48 PM PT 469. thoughtful - July 12, 1999 - 1:56 PM PT 470. thoughtful - July 12, 1999 - 2:00 PM PT 471. thoughtful - July 12, 1999 - 2:02 PM PT 472. FreetoChoose - July 12, 1999 - 2:12 PM PT 473. thoughtful - July 12, 1999 - 2:28 PM PT 474. FreetoChoose - July 12, 1999 - 2:34 PM PT 475. thoughtful - July 12, 1999 - 2:44 PM PT 476. MsIvoryTower - July 12, 1999 - 2:45 PM PT 477. thoughtful - July 12, 1999 - 2:50 PM PT 478. thoughtful - July 12, 1999 - 2:52 PM PT 479. MsIvoryTower - July 12, 1999 - 2:53 PM PT 480. FreetoChoose - July 12, 1999 - 2:57 PM PT 481. FreetoChoose - July 12, 1999 - 3:01 PM PT 482. MsIvoryTower - July 12, 1999 - 3:05 PM PT 483. FreetoChoose - July 12, 1999 - 3:07 PM PT 484. MsIvoryTower - July 12, 1999 - 3:07 PM PT 485. Slackjaw - July 12, 1999 - 3:10 PM PT 486. Slackjaw - July 12, 1999 - 3:15 PM PT 487. FreetoChoose - July 12, 1999 - 3:33 PM PT 488. FreetoChoose - July 12, 1999 - 3:58 PM PT 489. FreetoChoose - July 12, 1999 - 4:00 PM PT 490. Slackjaw - July 12, 1999 - 4:06 PM PT 491. MsIvoryTower - July 12, 1999 - 4:13 PM PT 492. Slackjaw - July 12, 1999 - 4:15 PM PT 493. MsIvoryTower - July 12, 1999 - 4:16 PM PT 494. MsIvoryTower - July 12, 1999 - 4:20 PM PT 495. Slackjaw - July 12, 1999 - 4:24 PM PT 496. MsIvoryTower - July 12, 1999 - 4:38 PM PT 497. AdamSelene - July 12, 1999 - 4:41 PM PT 498. FreetoChoose - July 12, 1999 - 4:52 PM PT 499. Slackjaw - July 12, 1999 - 5:21 PM PT 500. MsIvoryTower - July 12, 1999 - 6:23 PM PT
cdm and MsIT--
Maybe the differences before and after RE are more striking to me because I went from a mainstream Keynsian undergrad program to the REers in Minnesota. And I happened to hit Minnesota just about at the apex of the academic conflict.
But I think you're understating how much things have changed. I once heard one of Kennedy's CEA guys say "Back when Jim Dusenberry and I were running the economy..." Nobody today would say that. This was, of course, in the context of evidence that command economies could engender rapid growth. The failure of the Nixon ("We're all Keynsians now.") administration to deal with the oil embargo cast a lot of doubt on the effectiveness of managing the economy.
The RE folks capitalized on that doubt, with persuasive arguments about how agents know about the models the Keynesians were using, but that knowledge wasn't reflected in the models. That _was_ a new idea, and it pointed out to a fundamental problem with macro theory. In Sargent's words: "Where are the people in this model?"
MsIT, you say that the Keynesians also knew that interventions had unintended consequences. But that's not how I was taught by the Keynesians. ISLM. Phillips curve. When you do this, the economy does that.
Yes, I agree that there are fewer clear divisions across policy economics lines. The RE folks had good criticisms, but their bets on multiplayer dynamic games turned out to be bad bets, for the same reasons that there are no people in the Keynsian models.
Final thought in this particular ramble is that the predominant view in most policy areas by most economists is that intervention is a bad idea. It's pretty much always been true that economists have been less interventionist than the politicians they advise. But we're at a level of non-intervention today that I've never see before. I really think you have to credit (or blame) the RE folks for that.
Message #357
This is a nice segue to computational economics and agent based models. I don't know much about that stuff, although I just got introduced to bootstrapping samples. What's happening these days, Slack?
jayackroyd
Interesting timing. Less than half an hour ago, a colleague dropped in to tell me about progress he is making on a bootstrapping model he is developing. I don't have many details at the moment, but I am supposed to review it later this week.
Jay
"It's pretty much always been true that economists have been less interventionist than the politicians they advise. But we're at a level of non-intervention today that I've never see before. I really think you have to credit (or blame) the RE folks for that."
Well, its true that we are now in a period of strong consensus for non-intervention unless shown likely to generate minimal side effects (those nasty unintended consequences), and certainly the most non-interventionist in the post-WWII period.
However, I'm hesitant to issue full credit (or blame) to the RE folks. I prefer to think of it as a resurgence of the neo-classical perspective. And, lets face it, we've been collecting much better data on the economy since the 1950's, with better and more precise modelling which has shown just how costly those unintended consequences can be.
And I'm not a fan of total non-intervention because this too can have as many negative unintended consequences as too much intervention.
CDM,
Most start from RE modeling? Are you assuming RE models are fundamentally different than the older neo-classical equilbriums? I've always just thought of them as improvements, or tweaks to the classical models.
"MsIT, you say that the Keynesians also knew that interventions had unintended consequences. But that's not how I was taught by the Keynesians. ISLM. Phillips curve. When you do this, the economy does that."
Well, I concede a certain degree of rigidity in the teaching of economics both from the keynesian and neo-classical models prior to the RE work. I don't think working economists ever thought of the models that rigidly, however, at least I don't remember them being as confident as one would suggest from the formal models themselves.
By the way, the neo-classical economists were just as rigid in their formal models as the keynesians, so I think this was more a function of how economics was taught prior to the 80's rather than their lack of understanding of unintended consequences. I think one must credit the new generation of economists with much greater willingness to *admit* to the uncertainty surrounding their models than prior generations ever were, regardless of their theoretical preferences.
CDM
"BTW, MsIT mentioned a two volume collection of papers on new Keynesianism, saying it was edited by Robert Hall. I think that the editors were actually Mankiw and Romer -- at least, they did have a publication that matched the description, while I don't know of anything similar edited by Hall."
Well, this is entirely possible, Hall may have only been one of the contributors. I know Mankiew is one of the contributors. I have the set in my office, I'll check when I get there (I was almost positive Hall was ONE of the editors).
It's Mankiw, not Mankiew.
(my apologies in advance for the nit-picking, but this is the second time, so I'm guessing it wasn't a typo, but a misrecollection.)
Yeah, but I like Mankiew better, so that's how I remember it.
We are talking about Gregory M, correct?
This guy
Incidentally, you are in good company. Someone at Duke prefers a different spelling:here
cdm
You are entirely correct regarding the editors of The New Keynesianism, they are Mankiw and Romer. I bought it because Hall was a contibutor.
No excuses, not even old age.
Anyway, I recommend the volumes, both contain interesting papers.
BTW, there was a discussion on what the Japanese should do about there economic woes. It appears that Mankiw subscribes to the same prescription as Krugman here although I don't know whether they arrived at their conclusions independently.
Jay:"When was that Rask? I was in the graduate program at U of Minn in the mid-80s. Had both the best graduate assistant I'd ever had there, and the worst as well--all in one micro theory course."
I took Macro and Micro in 87, and the 5000 level International class in 89. The International and Micro classes were rather straightforward, but the Grad student teaching the Macro class was very dogmatic. I never had a professor in any of my econ classes at the U. I have slowly become convinced that there aren't any - only names on doors and paychecks to Grad Students. I don't remember the Instructor's name (dark hair, and skinny is all I remember). The TA was a big guy named Dan Grout or something (I only remember his name because he had a kid in the daycare that my girlfriend worked for, and I ran into him a few times there). The TA tended to amend a lot of the instructors comments, pointing out how he was using a straw man version of Keynesianism to make his point.
My more mature understanding of economics since then has led to believe that RE did add a lot of value to the understanding of economics, but that it did so by staking a very extreme position to contrast itself with Keynesianism. The Keynesians may have excessively assumed that the public was capable of being perpetually fooled, but the REs excessively assumed that the public was informed.
I remember one RE argument about deficit spending. The guy argued that government deficit spending would make no difference on the savings rate, since rational people would know that tax raises to make up for it would have to be made in the future, and would therefore squirrel away enough money in savings to pay for the future tax raises, thus negating the impact of federal borrowing on total savings. The idea, of course, is hooey.
That almost overlaps with my time there. There really were professors, some of them quite good, as both researchers and teachers. Sargent was demanding and evangelical, but also amusing and clear. Marcel Richter was the best econ teacher I've ever had. Sims was reputed to be a mess.
The grad assistant who I couldn't stand was assisting James Jordan in the second trimester of Micro theory. Thought I'd forgotten his name, but association suddenly reclaimed--Mr. Lahiri was his name. The first trimester was taught by Richter, and his ta was great--Mr. Kim. (They were never referred to by their given names for some reason.)
Yes, when the RE folks tried to make positive contributions, like reviving the real bills doctrine, or claiming that there was no difference between deficit spending and taxation, they approached self parody. However, it seems to me that economic models without decision makers or other agents are themselves self-parodies. Or stories, and nothing more.
BTW, what is also interesting is that the monetarists are simply gone, not even worth discussing.
MsIT:
"I think one must credit the new generation of economists with much greater willingness to *admit* to the uncertainty surrounding their models than prior generations ever were, regardless of their theoretical preferences."
This is true. The discussions have lost much of their religious tenor as well.
"And I'm not a fan of total non-intervention because this too can have as many negative unintended consequences as too much intervention."
Here's where the New Classical guys make their point. Stable policy regimes are in many ways better than any particular set of changing regimes.
FTC
"When thoughtful mentioned imperfect information, I was thinking information asymmetry. I suppose there might be a technical distinction, but broadly speaking the issues are similar."
Not information asymmetry which is present in most transactions, but imperfect information --y'know the tweaks they do to make sure policies are not strictly comparable so price comparison becomes impossible. Y'know the "value-added" changes they make -- it adds value to the firm. }:-)
jaya, re luck, of course some of what's happening in today's economy is a result of luck -- but it's an unquantifiable amount. I also suspect that a lot of what's happening (you may call it luck, or maybe just a confluence of events) is a result of decades long learning about economic systems including the failure of the centrally planned economy, the benefits of free trade, globalization of capital flows, and microeconomic learning by OPEC so they stopped trebling and quadrupling prices.
I also think policy makers have learned that sustained fiscal deficits in a growing economy are harmful, volatility in interest rates is harmful, and over regulation is harmful (to wit the taxless Internet environment -- at least so far). I also think that, perhaps unwittingly, the growth in federal spending, especially social security has added an element of stability to the economy.
So, in my view, there's a lot more going on than just luck.
"The only thing in my life that's gone up in price is movies. Everything else is even to down. "
If that's the case, you must be very, very young!
Re "religious tenor" of economics, I wonder if some of that has to do with how it is taught vs. how it is "practiced". I think economic thought is difficult to learn and requires a good deal of discipline, which is hard to achieve if at every turn you have to repeat assumptions or state exceptions or counteract arguments.
Of course, once you get the basics under your belt, then argue away. (I think the willingness to argue is a prereq for studying economics!)
In defense of monetarists (as I said, I'm eclectic) I think we have learned a lot from them about inflation and the money supply -- more than we knew with Keynes alone. I think the error was pushing the basic tenet further than it was intended. Friedman fundamentally said there is a long run relationship between money supply growth and inflation, and that the lags are long and variable. To go from that to trying to build a short run economic forecasting model....
YIKES!
thoughtful
I understand the difference. Do you understand why I said they are broadly similar?
If you issue a policy with a set of features, and I issue a policy with a different set of features, I am likely to know exactly how much those features are worth, but you are not. We can call it imperfect information, in the sense that you have to make a guess as to the relative value of the feature sets, or we can call it information asymetry, because I have information about the relevant costs of the various features that you do not possess. In either case, the differences in the policies creates an advantage for me.
Yes, I understand -- but they are not the same thing. Asymmetry can exist just because you don't know how much money I have or how much I'm willing to spend, even if we are talking about buying a perfectly competitive commodity. Imperfect information to me implies that whatever information is available is insufficient to make the "right" decision.
Thoughtful
(I think) information asymmetry is irrelevant if we are talking about buying a perfectly competitive commodity (but I will be happy to be corrected if I'm wrong).
As a producer, I have no control over the price of a perfectly competitive commodity, so why would it matter that I do not know how much money you have or are willing to spend?
I didn't say it was relevant -- only that it existed.
(Don't mind me -- I've been up since 2:30 a.m. and can't see straight let alone think straight!)
Rask
re: teaching assistants and profs.
I went to a large undergrad institution (UCLA), majored in Economics, and never once had a class where the professor didn't teach it. TA's were simply that, teaching assistants, they led seminars attached to the class, assisted with questions and out of class interactions, and helped with the grading.
I went to a small, elite private school for my graduate work, in Economics, and never once had a class that wasn't taught by the Professor, even with graduate assistants. Something wrong with a program where undergrads only have contact with TA's and not the big name profs.
Thoughtful,
In defense of Monetarists we wouldn't be focused on low inflation rather than economic stimulus as the primary purpose of the Feds without them. Well before the RE folks, monetarists were talking about zero-inflation as the most important means of stimulating investment, and hence economic growth, which is essentially the RE message (see why I think they're just repackaged neo-classicals?).
As for where the monetarists are now, I'm thinking they've reinvented themselves as RE folks. That is, those two perspectives seem so compatable as to have absorbed one another.
RE: imperfect information - I agree with Thoughtful on this, it's not the same thing as asymmetric information.
Or moneterrorists as I used to call them! Yes, I think they are largely responsible for the focus on low inflation. Fortunately, I think Greenspan is smart enough to recognize and accept the dual function of the Fed of maximizing employment *and* minimizing inflation. If not, the Fed would surely not have moved to a neutral bias at the last FOMC meeting.
FTC, further on that, it won't matter at the macro level, but at the micro level it could matter as the producer of a good in a perfectly competive market cannot set the price of the good, but will still face variation in costs re inventories, spoilage, pilferage, credit risk, etc.
Yeah, but you've got to admit, he's first and foremost concerned with low inflation, TH, and although he's been more relaxed in the last few years, he was a bit edgy on the trigger (discount rate) for the first four years of the recovery.
MsIvoryTower
“RE: imperfect information - I agree with Thoughtful on this, it's not the same thing as asymmetric information.”
Good. We all agree they are different.
But with respect to the issue that started the discussion (of imperfect information) I don't believe the distinction is important.
MsIvoryTower
BTW, I may have to take back my general agreement with the notion that costs can be passed onto consumers in the long run. It may well just be that there is a valid economic principle regarding the incidence of costs, and people who claim that all costs are passed on are guilty of sloppy phrasing, but it is pretty clear that it is false.
I need to do a bit of homework to see if I can translate it into economic language.
Well, it is wrt my comments.
Asymmetry in information simply means that there's uncertainty wrt the transaction, will it take place? Does the buyer know something the seller doesn't, or visa versa? But enough information exists to make risks on both sides.
Imperfect information is precisely what TH suggests, that not enough information is available, or known to make the right decision.
thoughtful
I agree that at the micro level a supplier will have a variation in costs. (And the supplier must bear the costs, to the extent that they exceed the “cost” inherent in the market defined price. I suspect I am not using accepted economic language, hence my comment to the Ms. Incidentally, this is why the supplier bears 100% of unexpected costs.)
But why does it matter to the supplier whether it knows the willingness of the consumer to spend? The amount of profit will be a function of the price and the supplier's costs, but not in any way a function of the consumers cash position. Am I missing something?
"It may well just be that there is a valid economic principle regarding the incidence of costs, and people who claim that all costs are passed on are guilty of sloppy phrasing, but it is pretty clear that it is false."
False? By what reckoning? In what situation? By what evidence?
FTC:
no, I didn't miss those posts, but I haven't had the time/inclination to post lately. Was celebrating a birthday last week and I was out of town since Thursday.
One thing that occurred to me re. insurance is the definition of the commodity. What is a unit of insurance? There seems to be an implicit definition running through some of the posts on this subject that the commodity is a whole insurance contract, but I think that definition may cause trouble for the application of ready-made models because of nonconvexity of the commodity space.
In regard to your post above, what is a "perfectly competitive commodity"?
Glad to see everyone agrees that imperfect info is not the same as asymmetric info.
"Imperfect information is precisely what TH suggests, that not enough information is available, or known to make the right decision."
this sounds like pre-expected utility talk to me. For any amount of information and suitably defined preferences over final outcomes and suitably defined (i.e., compact) outcome space, there will *always* be a utility maximizing decision, i.e., a "right" decision.
Slackjaw
Belated Happy Birthday. Sorry I didn't acknowledge it at the time.
Wrt imperfect information and asymmetric information, I would define the terms very differently that the Ms.
The net is failing me today. I did a search on information asymmetry, and got plenty of hits, but no definitions. I recognize you as an expert; perhaps you can weigh in.
Crudely speaking, I think of information asymmetry as when one (or the other) party knows something that the other doesn't know. I think of imperfect information as when one party (or the other) *thinks* they know something, but they are wrong.
How would you define the terms?
MsIvoryTower
“False? By what reckoning? In what situation? By what evidence?”
All in good time.
Slackjaw
“One thing that occurred to me re. insurance is the definition of the commodity. What is a unit of insurance? There seems to be an implicit definition running through some of the posts on this subject that the commodity is a whole insurance contract, but I think that definition may cause trouble for the application of ready-made models because of nonconvexity of the commodity space.”
Yes, I agree that the implicit unit is a whole contract.
“In regard to your post above, what is a "perfectly competitive commodity"?”
It was thoughtful's term.
FTC:
Your notion of asymmetric information is basically correct.
Information is imperfect when a decision maker cannot distinguish between two different states of the world. Formally, some information set is not a singleton. It is essentially the same as uncertainty.
Incorrect knowledge, your notion of imperfect information, is specifically disallowed by the modal logic underlying Bayesian decision theory. (There is an explicit Truth of Knowledge axiom.)
Also...
"It may well just be that there is a valid economic principle regarding the incidence of costs"
There is certainly such a principle. The claim about cost incidence in the long run then simply amounts to a statement of belief that long run supply is perfectly elastic and long run demand is not.
Slack
"this sounds like pre-expected utility talk to me."
Well, it wasn't what I was thinking about, even if it sounds that way. And the term right was not precise, I should have said maximizing, but then I was in a hurry and didn't stop to change the word.
"For any amount of information and suitably defined preferences over final outcomes and suitably defined (i.e., compact) outcome space, there will *always* be a utility maximizing decision, i.e., a "right" decision."
This is tautological.
I am used to thinking of a unit of insurance as a dollar payoff in some state of the world, and some (less than a dollar) cost incurred in every state of the world.
If a unit of insurance is the contract, what does a firm's supply schedule tell you? Number of insurance contracts offered at a given price per contract? What exactly then goes into a "contract"?
alrighty then...
Thoughtful: what is a perfectly competitive commodity?
"The claim about cost incidence in the long run then simply amounts to a statement of belief that long run supply is perfectly elastic and long run demand is not."
Don't know of any product where this is the case. It's easier to believe that long run supply is not perfectly elastic rather than that long run demand is not. In the long run, consumers can always find substitutes.
Oh shit,
That was a mistake.
Forget that last post.
it can be a tautology if you abuse "suitably" or take that to mean "whatever makes the conclusion true," which I do not. I mean preferences should be a continuous weak order and the prize space compact--entirely standard condtions which I presumed obvious from the context.
Then it is no more tautological than any mathematical proposition. It is a simple consequence of the Expected Utility theorem and the Weierstrass theorem.
In short, there is no level of information, which is to say, there are no beliefs about the relevant state space, that prevent maximization expected utility, or determination of the "best" choice.
I do not see how the word "maximizing" would have helped.
Slack
I've always thought of imperfect information as resulting in imperfect markets. Imperfect markets are, by definition, inefficient.
Hi all.
I have a question. Given the current debate around HMO legislation, what is the economic perspective? My wife (an ex-hospital CFO) thinks that the republican plan of regulated-but-HMO-chosen standard-bearers is the way to go because individual doctors will always err on the side of expensive tests. I (a libertarian, if there is any doubt) think that the liberal plan whereby the ultimate responsibility lies with the patient's choice of doctor/2nd opinion is the way to go.
The crux of this is: Can HMO's effectively regulate doctors at contract renewal time, or do they have to do it on a case-by-case basis?
Have at it, me hearties.
AdamSelene
Howdy stranger.
Unfortunately, HMO's bore me to tears. So I will watch, rather than contribute.
msit:
Imperfect information doesn't appear to me to have any necessary relationship to imperfect markets (by which I presume you mean markets that are imperfectly competitive).
I do not see why imperfect markets would cause uncertainty. Consider Cournot oligopoly, for example (or Stackelberg, or whatever you like). There is no uncertainty anywhere in that model, either as a primitive or a result, yet markets are certainly imperfect.
Conversely, Arrow-Debreu results carry over to environments with uncertainty (except possibly for uncertainty over preferences), so it need not cause market imperfections. I mean, firms can be price takers and still have their output determined in part by the realization of a random variable.
Unforeseen contingencies or unforeseen elements of the state space would doubtless cause problems, and probably cause markets to be incomplete in the Arrow-Debreu sense; perhaps this is what you are thinking of. But, there is really no good way to deal formally with unforeseen contingencies and hence no systematic treatment of their impact.
Slack
Okay, you've raised the situation where imperfect information doesn't necessarily result in imperfect competition. However, imperfect information can (and usually does) cause imperfectly competitive markets. The latter are not always the result of the former, I agree, and didn't suggest that in my comment. I don't know of very many examples where there IS imperfect information that doesn't result in imperfect competition, though.
"Unforeseen contingencies or unforeseen elements of the state space would doubtless cause problems, and probably cause markets to be incomplete in the Arrow-Debreu sense; perhaps this is what you are thinking of."
No, I'm thinking of imperfect information creating barriers to maximization, or conversely cost minimization. Not just unforeseen contingincies or elements, but uncertainty wrt things like actual productivity (in labor), or lack of information (by consumers) about competitive prices (or wages), because of high costs associated with collecting that information.