« Stock markets, bubbles, and dividend growth | Main | Keynes's General Theory, chapter three »
A Remarkable Question
Our colleague, Richard Wagner, a leading light in the public choice revolution, wrote the following remarkable question for the 2005-2006 graduate political economy preliminary exam at George Mason.
Joseph Schumpeter claimed that capitalism would give way to socialism largely for ideological reasons. This does not seem to have happened, at least directly. But might it be happening indirectly? Consider, for instance, a significant change that has occurred in the economic organization of debtor-creditor contracts. Not too long ago, lenders held their loans in their portfolios. They would lose if the borrower defaulted, which gave the lender a strong incentive to monitor the borrower, particularly for large loans. Now, lenders split their loans into numerous small pieces and disperse them throughout the economy. (For instance, many people who hold mutual funds and retirement accounts will find that they are holding small pieces of large loans made by commercial banks.) The burden of non-performing loans is thus dispersed throughout the economy rather than residing with the original lender. Does this development weaken the incentive of lenders to monitor borrowers and thereby weaken overall economic performance? That is, can market transactions generate institutional arrangements that impair the market economy? However you address this topic, do so clearly and cogently.
I am sorry to say that none of the students got the answer right. Of course, very few of their teachers, here or elsewhere, got the answer right, either. Kudos to Dick for his prescience.
I thank David Levy for the pointer.
Posted by Alex Tabarrok on December 11, 2008 at 07:10 AM in Economics | Permalink
Comments
So essentially concentrated ownership is a public good and there are lots of diversified free riders?
Posted by: nelsonal at Dec 11, 2008 7:28:15 AM
Yes.
Posted by: Andrew at Dec 11, 2008 7:29:32 AM
offenders -> of lenders.
P.S. "However you address this topic, do so clearly and cogently."
Shouldn't it be unnecessary to say that?
Posted by: dearieme at Dec 11, 2008 7:34:49 AM
Hey! I'm not currently taking an econ grad seminar (nor am I likely to). A hint would help, or at least a link.
Posted by: Jeff Burton at Dec 11, 2008 8:09:21 AM
Yes.
Posted by: Brian at Dec 11, 2008 8:13:51 AM
so what is the answer? nelsonal's sounds good to me.
this has always bothered me about the argument for investing in index equity funds. if enough investors are doing valuation analysis to price individual equities efficiently, then index funds are the optimal investment. but if everyone invested only in index funds, there would be no mechanism to keep the price of individual equities efficient. instead prices would be driven by (1) what the index funds pay for IPO's and (2) premiums for stocks whose low floats give them limited supply. you can then get cycles where the index funds keep buying dot-com stocks, thus driving up the price of the limited float, thus driving up the indices, thus attracting more investors into the funds, etc.
Posted by: DK at Dec 11, 2008 8:22:12 AM
Alex, two points are in order here:
Does this development weaken the incentive offenders to monitor borrowers and thereby weaken overall economic performance?
Yes, but why would that not be reflected in the price? If I'm buying part of a loan from you, weakening your incentive to monitor it in the first place, wouldn't I be offering a lower price for it than if your incentives were perfectly aligned? Shouldn't I be able to realise that the asset I'm buying has a higher probability of default and thus change my bid price accordingly (or choose not to bid)? Furthermore, loans these days are not awarded on the basis of personal knowledge (Good ol' Bill's farm is doing OK, let's lend him some money), they are screened and monitored on the basis of credit-worthiness information that will generally be available to all financial companies willing to request it - so there's nothing that the issuer of the loan knows that cannot, in principle, be found out by whoever is purchasing the loan. Yes, they own only a small part of the loan so it may not be worth the hassle of monitoring the loan, but that’s a point that’s well known ex ante and should affect the decision of whether to buy the CDO (or whatever) from the primary issuer in the first place.
Of course the whole point here is that this information regarding weakened incentives was not taken into account by the buyers’ of these securities, hence the current mess. But if it was not taken into account by the buyers, it sure as hell wasn’t (and couldn’t really be expected to be) taken into account by governments and regulators operating with even less information. Perfect information is not some natural state of the world, and complaining about the lack of it in the abstract is akin to complaining about how the existence of gravity greatly reduces our freedom of movement.
In terms of answering Richard’s question, yes capitalism can produce scary outcomes but this is just a manifestation of the high risk/ high average return paradigm. If society lacks the stomach for such variations it can well move in a more socialist, low risk/ low average return direction, perhaps even all the way to communism (no risk at all, nice and steady decline)
And by the by, CDOs and CDSs would not even be the example that first came to my mind if you want to ask the question Richard is asking. How about employing someone when you can’t monitor their effort and even their output perfectly? Corporations are creations of the free market, and they ‘weaken incentives and overall economic performance’ considerably. Compared to the labour market, the whole securitisation business and its effect on incentives is truly minimal.
Posted by: datacharmer at Dec 11, 2008 8:35:42 AM
Ways not to answer a grad student exam:
"Only the greedy people try to get paid the same for new products of lower quality or that push the cost onto others. And even then, they only do it on days that end in 'y'."
"First, we take the standard assumption of economists and believe all consumers act rationally, but bureaucrats don't."
"No problem, we have Moody's!"
"I have a lot less riding on this than the geniuses who have been pondering the same question every day for the last two years and still haven't a clue, so, pass."
Posted by: Andrew at Dec 11, 2008 8:58:58 AM
So does this mean that the government should begin subsidizing Gordon Gecko types? Greed for lack of a better word, is good, indeed.
Posted by: nelsonal at Dec 11, 2008 9:48:12 AM
The answer is yes, but I can understand how so many people might have gotten it wrong. We are so used to thinking about central planning and socialization of private losses as an act of PUBLIC agents that some people still can't wrap their heads around how the worst offenders this time around were PRIVATE agents. I called it "Private Socialism" in a post I did a while back.
http://brokensymmetry.typepad.com/broken_symmetry/2008/09/private-socialism.html
Posted by: Michael F. Martin at Dec 11, 2008 9:52:31 AM
How can you grade a question like this right or wrong? There is only the well thought out answer or one that isn't.
Posted by: asiequqnq at Dec 11, 2008 10:24:43 AM
So institutions become willing to originate loans with negative expected rates of return if there is a market of investors who can't be bothered to monitor the mosaic of loans they are purchasing.
How do the ratings agencies make their money again?
Posted by: Drucker at Dec 11, 2008 10:36:11 AM
If the entity buying the risk knows that they will not be responsible for the risk (either through selling it again or through bailout) then yes, they will not monitor the borrowers, as they have no exposure. It falls to the final holder of the security to properly monitor the borrowers and to bear the risk of the securities that they have purchased. And 3rd party intervention (either through bailout or by making money available cheaper than the market rate) disrupts this market.
Posted by: Chris at Dec 11, 2008 11:36:42 AM
"can market transactions generate institutional arrangements that impair the market economy?"
Yes.
"However you address this topic, do so clearly and cogently."
See the video behind the ad What Now? at the top of the page.
Posted by: Zorro at Dec 11, 2008 12:00:17 PM
Yes, but why would that not be reflected in the price? If I'm buying part of a loan from you, weakening your incentive to monitor it in the first place, wouldn't I be offering a lower price for it than if your incentives were perfectly aligned?
Yes, but that destroys value. It meas that some sound loans will not be made, precisely because they cannot be sold at their actual value. Lemons.
Two other issues arise.
One is the built-in conflict among holders of pieces of the loan if renegotiation becomes necessary. Loans are forced into default/foreclosure that might be renegotiated to the aggregate benefit of the lenders, but because the benefits are unevenly distributed no renogiation can take place.
The other is that when buyers rely on the coarse categories used by ratings agencies to judge creditworthiness, which runs on a continuum, sellers will inevitably game the system.
Not that I would have thought of all this in 2006.
Posted by: Bernard Yomtov at Dec 11, 2008 12:14:56 PM
Does this development weaken the incentive offenders to monitor borrowers and thereby weaken overall economic performance?
That is, can market transactions generate institutional arrangements that impair the market economy?
For the first question, datacharmer has the right response. A loan fragment sold by a "hot-potato-passing" lender is on average a less vouched-for loan and thus less valuable (obviously this was not sufficiently appreciated before the crash). And if loan-eligibility-determination is so heavily data-driven, the personal touch may not matter as much (I expect it still matters some, but I don't have relevant expertise).
For the second question, according to clean models this case should be self-correcting, but in reality, yes, of course this can happen. Innovation and complexity create novel market decisions for participants. If too many participants, faced with a novel situation, don't think it through, there may be systematic error in the pricing decisions, leading to inflated or deflated asset values. Once these have been established, even "rational and thorough" market players may have no choice but to follow suit.
In other words, when novelty and complexity are injected into a market, you may want the "well-trained elites" to be the very first market participants. Of course, elites can make systematic mistakes too...
Posted by: mk at Dec 11, 2008 12:30:49 PM
As an addendum: related issues with instituting novel markets may pose a challenge to imposition of a CO2 cap-and-trade system. It might be related to what happened in Europe, where the initial carbon permits were trading at absurdly low prices.
Or the problem in Europe could have just been too much supply.
Posted by: mk at Dec 11, 2008 12:38:29 PM
"Joseph Schumpeter claimed that capitalism would give way to socialism largely for ideological reasons."
I think this is incorrect -- most of Schumpeter's argument in Capitalism, Socialism, and Democracy gives functional, structural and psychological reasons.
Further on in the question, a "development [that] weaken[s] the incentive offenders to monitor borrowers and thereby weaken[s] overall economic performance" is not, in itself, an "ideological" reason.
Posted by: Lee A. Arnold at Dec 11, 2008 12:43:14 PM
The answer: no.
Clear and cogent reason: If lenders split securities up and they make loans into that system that are unwise, then a profit opportunity will emerge for other more careful lenders who could deliver better returns to investors via lower default rates. Dividing the mortgages up merely helps manage risk in the long run.
Posted by: liberalarts at Dec 11, 2008 1:49:39 PM
If this were a once off transaction, then yes, having effectively sold off the risk for the loan, the originating lender would have no incentive to monitor the borrower. The people buying the loan, only bearing a small fraction of the default risk, would have little incentive to monitor the borrower (if they even knew who he was).
However, in a dynamic world where there is both competition from other lenders and reputational effects over time, an originating lender who has lax standards will have trouble selling off a loan. Because he will pay a market penalty for his laxity, he will have an incentive to carry out an efficient level of monitoring of the borrower.
At least that's my best stab at the answer.
Posted by: Bill H at Dec 11, 2008 3:12:45 PM
"Furthermore, loans these days are not awarded on the basis of personal knowledge (Good ol' Bill's farm is doing OK, let's lend him some money), they are screened and monitored on the basis of credit-worthiness information that will generally be available to all financial companies willing to request it - so there's nothing that the issuer of the loan knows that cannot, in principle, be found out by whoever is purchasing the loan. "
The credit-worthiness information you speak of: turns out that "information" was all garbage.
Posted by: meter at Dec 11, 2008 3:31:52 PM
Chris sums it up simply:
If the entity buying the risk knows that they will not be responsible for the risk (either through selling it again or through bailout) then yes, they will not monitor the borrowers, as they have no exposure. It falls to the final holder of the security to properly monitor the borrowers and to bear the risk of the securities that they have purchased. And 3rd party intervention (either through bailout or by making money available cheaper than the market rate) disrupts this market.
And according to that Michael Lewis article in Portfolio (http://www.portfolio.com/news-markets/national-news/portfolio/2008/11/11/The-End-of-Wall-Streets-Boom), the ratings agencies were obscuring (lying about?) the quality of the mortgages behind the securities, so the buyers couldn't know the risk they were assuming. The market will always have bad results if this sort of thing is allowed. While we'll never be able to get rid of unscrupulous behavior, hopefully in the future we'll be on better guard against behavior that will cause a crisis this big. But can we be, given the (usually positive) fact of innovation? I take the Chuchillian line: The market is the worst way of organizing an economy, except for all the others that have been tried.
Posted by: Bill C at Dec 11, 2008 5:44:06 PM
MBSs are priced in the market. If that price does not reflect the cost of increased default resulting from the reduced incentives on the part of the originator to maintain a certain standard of underwriting then it is an arbitrage opportunity for the bank. This was certainly the case in the past, partly facilitated by the rating agencies. But otherwise if the market price of the MBS is set so that it reflects a higher default risk then by definition there is no impairment.
Posted by: surge at Dec 11, 2008 5:57:50 PM
The market is the worst way of organizing an economy, except for all the others that have been tried.
"If it isn't worth doing at all, then it isn't worth doing well."
If the market is the worst way to do it except for all the others, maybe we should not do it. Retrench to a simpler economy that doesn't need so much organising.
Posted by: J Thomas at Dec 11, 2008 5:57:57 PM
Adam Smith asked this same question (and unlike Richard's students and most economists today) got the answer right several centuries ago. Check out the last 10 pages of Chapter II Book II of The Wealth of Nations.
Posted by: Will at Dec 11, 2008 6:04:04 PM
