The shit is hitting the fan: all those sub-prime mortgages given out so recklessly over the past two years are getting their interest rates re-adjusted. And that, of course, is when the foreclosures begin.
By most measures, sub-prime loans are a bad idea. Look, for example, at the popular 2/28 loan, which consists of a low, fixed-interest rate for the first two years and a much higher, adjustable rate for the next twenty-eight. Most people taking out a 2/28 loan can't afford the higher interest rates that will hit later on. It's not unusual for interest payments on a 2/28 loan to double within four years. (That's why you're seeing such high foreclosure rates in the sub-prime market.)
So why do people take out sub-prime loans? Don't they realize that they won't be able to afford the ensuing 28 years of mortgage payments? I think a big part of the reason sub-prime loans remain so seductive, even when the financial terms are so atrocious, is that they take advantage of a dangerous flaw built into our brain. This flaw is rooted in our emotional brain, which tends to overvalue immediate gains (like a new house) at the expense of future costs (high interest rates). Our feelings are thrilled by the prospect of a new home, but can't really grapple with the long-term fiscal consequences of the decision. Our impulsivity encounters little resistance, and so we sign on the bottom line. We want the house. We'll figure out how to pay for it later.
The best evidence for this idea comes from the lab of Jonathan Cohen. Cohen's clever experiment went like this: he stuck people in an fMRI machine and made them decide between a small Amazon gift certificate that they could have right away, or a larger gift certificate that they'd receive in 2 to 4 weeks. Contrary to rational models of decision-making, the two options activated very different neural systems. When subjects contemplated gift certificates in the distant future, brain areas associated with rational planning (the Promethean circuits of the prefrontal cortex) were more active. These cortical regions urge us to be patient, to wait a few extra weeks for the bigger gift certificate.
On the other hand, when subjects started thinking about getting a gift certificate right away, brain areas associated with emotion - like the midbrain dopamine system and NAcc - were turned on. These are the cells that tell us to take out a mortgage we can't afford, or run up credit card debt when we should be saving for retirement. They are our impulsive pleasure seekers, the hedonists inside our head.
By manipulating the amount of money on offer in each situation, Cohen and his collaborators could watch this neural tug of war unfold. They saw the fierce argument between reason and feeling, as our mind was pulled in contradictory directions. Our ultimate decision--to save for the future or to indulge in the present--was determined by whichever region showed greater activation. More emotions meant more impulsivity.
This discovery has important implications. (A more recent paper by the Cohen lab extends the theory.) For starters, it locates the neural source for many of our financial errors. When we opt for a 2/28 mortgage, we are acting like experimental subjects choosing the wrong gift certificate. Because the emotional parts of our brain reliably undervalue the future - life is short and they want pleasure now - we end up delaying saving until tomorrow (and tomorrow and tomorrow.) George Loewenstein, a neuroeconomist at Carnegie Mellon University and a collaborator on the Cohen paper, thinks that understanding how we make decisions will help economists develop better public policies: "Our emotions are like programs that evolved to solve important and recurring problems in our distant past," he says. "They are not always well suited to the decisions we make in modern life. It's important to know how our emotions lead us astray so that we can design incentives and programs to help compensate for our irrational biases."
Comments
Well, hopefully the financial industry won't get wind of this, because imagine what would happen in they used actual research results to target rubes.
So why do people offer them?
Posted by: Ted | August 9, 2007 10:48 AM
A bird in hand, as they say....
Posted by: writerdd | August 9, 2007 10:55 AM
Ted -
The reason they are offered is because it was (until a few months ago) possible to sell these loans on to the wider market, so it's quite hard to tell exactly who 'owns' the debt. The people who sell the mortgages and the organisations they work for get all their profits 'up-front', and so have no interest in anything but signing more people up.
So if we are to have fully deregulated consumer credit, then we must also have a target inflation rate of over 20%..
Oh, and I recall some research showing that the actual rate at which this whole reward now vs. later thing works corresponds to an average economic discount rate of around 30%.. Which also seems to be the upper limit for what people will pay on credit cards.
Posted by: Andrew Dodds | August 9, 2007 11:03 AM
Good one, Jonah!
So does this mean that Max Weber's claim that religious differences between people predisposed to modern capitalism (predisposed, first of all, to delayed gratification) are, in fact, neurological differences? (Or might religion account for neurological differences like the ones described here?) I'm curious to know what you think.
Hope to see you in the HC some time.
Josh
Posted by: Joshua Glenn | August 9, 2007 11:12 AM
But there perhaps is a more rational explanation for at least part of the population taking these loans. If home prices are appreciating at double digit rates and there is an expecation of a better credit situation a couple of years out, then refinancing the note should be easy. The higher interest paid would be offset by the opportunity cost of the price increase in housing that would occur if a purchase was delayed.
Posted by: bwv | August 9, 2007 11:52 AM
I work for CurrentForeclosures.com, a foreclosures site and have seen a huge increase in the number of foreclosures in the past 7 months. I believe it is a combination of not only sub-prime and ARM mortgages, but also the high number of people who have gotten loans with interest rates at an all time low... in addition to the rapid depreciation in some areas and the difficulty some are experiencing in selling their homes.
Posted by: TSmith | August 9, 2007 12:56 PM
All I can say is, buggerit, why can't *I* find neural experiments which offer Amazon gift certificates for my participation?
Posted by: Luna_the_cat | August 9, 2007 1:26 PM
My goodness, but that sounds like a cynical view. Almost makes me think that this bubble was carried out at the expense of rubes on one end, and the treasury on the other end. And that a lot of front line mortgage people willfully sought out these chumps while filling their coffers and knowing they're passing bad paper on.
Of course, I'm all for capitalism and free trade, but I'm starting to think that mortgage brokers have manipulated the value of my house as well, by way of consequence.
You know, there's intelligence, and with it the concept of counter-intelligence.
I wonder if corporations look through research results to determine if it has counter-research market value. If it was my job to find rubes, I'd work extra hard to find and hone the rube-seeking divining rod.
Posted by: Ted | August 9, 2007 1:45 PM
Indeed -- the regulation of the financial markets is (among other goals) supposed to be preventing this sort of "predatory lending" (the legal term for making a loan where the lender could reasonably predict inability to repay). Of course, under the ShrubCo regime, there's not much backing for enforcement of those protections....
Posted by: David Harmon | August 9, 2007 2:19 PM
Temporal discounting is a well-studied phenomenon wherein people in experiments would much rather receive $10 right now rather than a larger amount at a later time. There is a tipping point, of course, if the later-amount is large enough, or if the later date is close enough, that varies from person-to-person, though I'm not sure what the time/money function looks like.
Okay, so here's the deal: Jonathan Cohen's lab does some pretty awesome work, but there's some controversy in the literature right now from the computational folks about what these reward systems are actually encoding.
Given the prevailing model of reward, this synopsis is correct, though you make the assumption that people in an experiment pondering whether they want $10 now or $30 in two months is comparable to buying a house on credit.
The more likely explanation is that people are planning on selling their new houses after just 2-3 years for a hefty profit after refinancing. This is a carefully-planned risk: the housing market is booming, so if I buy a house for $300k, pay only interest, refinance at $400k in 2 years, I'll have made a ton of money without having to really put any money down.
It's not temporal discounting, it's a fallacious assumption that a bullish market will continue to offer similar returns in the future, which is a bit more nuanced.
Interesting little neuroscience reward factoids:
* the neurons that most strongly correlate with reward value in monkeys are not in the brain, they're the neck muscles as the monkey tenses before receiving a reward; but no one is going to argue that the neck muscles are "encoding" reward; (this isn't from any published studies, just personal communication with monkey physiology folks who do reward studies.)
* the midbrain dopamine neurons that encode reward value seem to be somewhat of a computational paradox, because the inputs that they receive from the visual system come from a brain area that only discriminates rough shapes of objects, so they can't possibly determine, from a computational standpoint, which object in the visual field is most rewarding.
All of this midbrain dopamine reward research stems from the original (and many, many, many subsequent) findings by Wolfram Schultz that led to the formation of the reward-prediction error hypothesis, that is, neurons in the substantia nigra (SN, in the midbrain) signal unpredicted rewarding events.
That second point comes from some awesome work out of Peter Redgrave's lab (summed up in a 2006 review in Nature Reviews Neuroscience). Redgrave argues that the timing and nature of the afferent (incoming) signals to the SN don't contain enough information for the SN to be able to discriminate rewarding from non-rewarding stimuli. Rather, they are most likely responding to novelty or saliency.
Posted by: bv | August 9, 2007 3:01 PM
My concern -- I hope someone's looking at the patterns.
My guess -- these were sold disproportionately to poor and non-white families. Of those sold, more poor and non-white families will default. Thereafter, the lenders' "credit scoring" systems will be able to show statistical evidence that poor and non-white applicants have a higher default rate, and so justify .... selling them higher-priced, higher-fee, mortgages in the future.
Anyone got data?
Posted by: Hank Roberts | August 9, 2007 3:47 PM
Why is it everyone is trying to find someone to blame for this "subprime" issue. Everyone is to blame, but no one held a gun to the borrower's head to sign the mortgage. Okay let's blame someone:
1. Congress for the tax laws which make mortgage interest tax decuctable, but not rent;
2. The President for not enforcing fair lending laws, which make it a crime to lend money on a mortgage if the borrower does not have the ability to repay;
3. Real Estate Broker - Selling too much house to the borrower who cannot afford it;
4. The mortgage broker for selling the borrower the mortgage product which the borrower cannot afford;
5. The aggregator (Countrywide, Washington Mutual, Option one, and many many more who created a conduit to funnel the mortgages to Wall Street Dealers;
6. Wall Street Dealers who have no idea what a mortgage is other than a commodity to be securitized;
7. Leverage Firms who purchased the securities with minimal cash, leveraged their balance sheets to squeeze out the last bit of yield, and, note last or least, the ones with the highest degree of blame is;
The Rating Agencies, i.e. Standard and Poor's and Moody's. These agencies are the one's who are suppose to be the gate keepers. To add sanity to the market, but,,,,,, they collected their rating fees and can sit and say they did their job, but market conditions changed and they cannot be held accountable.
Prudent lenders, Banks, Savings Banks and Credit unions did not put these mortgages on their books. Do you know why, because their regulators only allow minimal speculative lending.
All the blame above is useless as they all have pocketed their money. The rating agencies will not rate a mortgage issue today, regardless of the quality of the loans. Now, because of their situation, they are hurting good quality borrowers who have managed their credit historically.
Again, most homeowners, allow a broker to make the decision on the homeowners largest investment. Shame on the homeowner for not reading up on what is available to them.
Does anyone care?
Posted by: TAK | August 9, 2007 6:07 PM
very interesting article.
Posted by: kim | August 9, 2007 7:26 PM
"The more likely explanation is that people are planning on selling their new houses after just 2-3 years for a hefty profit after refinancing. This is a carefully-planned risk: the housing market is booming, so if I buy a house for $300k, pay only interest, refinance at $400k in 2 years, I'll have made a ton of money without having to really put any money down."
Which brings us to another one to blame: congress for making gains on home sales tax-free if you hold the house for two years. That is the sole reason for the 2/28 loan. If the law said you have to hold the property one year, the loan "product" would feature a one year teaser rate and 29 year libor-plus-7 percent or whatever ridiculous interest rate they're using now.
The entire bubble was predicated on tax-free capital gains, currently available only to hedge fund owners and doggy-brained house flippers.
Posted by: stuntmidget | August 9, 2007 9:44 PM
"no one held a gun to the borrower's head to sign the mortgage"
AKA the drug dealers' argument.
Posted by: Chris O'Neill | August 9, 2007 9:54 PM
bv -
Yes, the promise typically was that house price increases would go on forever, so the rubes would make a whole lot of cash selling even if they couldn't refinance.. which just meant that even riskier loans would be accepted.
Remember that subprime loans are *by definition* made to people with poor money management/math skills, and I'd guess these would be the people in the population most likely to favour immediate reward (A new house!) over delayed gratification.
This hits everyone else, because house prices are based on the relatively small number of houses sold vs. the total number of houses. So a relatively small number of people taking on unrealistic mortgages can upset the whole market.
Posted by: Andrew Dodds | August 10, 2007 4:48 AM
Subprime loans are not "by definition" made to people with poor money management skills (although that is often the case). Many find themselves in the category due to business failures, job losses or for medical reasons. There is a legitimate function for the industry to exist. However, the lending standards did go to just short of nil (I think a pulse was still required to qualify for a mortgage in 2006, but perhaps that is debatable). Bernake's last congressional testimony (http://www.washingtonpost.com/wp-srv/business/transcripts/bernanke_071807.html) and the interchange with congress is actually a pretty good overview of the situation
Posted by: BWV | August 10, 2007 1:15 PM
I actually have some background in both neuroscience and finance. All of this raises really interesting issues about who and what is "rational".
The mid-brain dopamine system may not be as "irrational" as we think. It is standard in financial theory that immediate cash flows are worth more than future cash flows; the classic way of expressing this is by "discounting" future cash flows.
The subjects in the experiment were only "wrong" if they had reason to believe that the more valuable Amazon gift card had a some threshold probability or greater of being delivered (and the books or whatever subsequently ordered and consumed). To put it another way, they applied a high discount rate to the future card. They probably were wrong, because in modern society, the near term future is far more predictable than under the conditions prevelant during most of the evolution of the human brain. However, the future is never certain, and a preference for something right now rather than later may be seen as "bio-rational". (Note that choosing between a card now and a bigger card later is a very different problem from having no card at all, and needing to plan to get one.)
Despite somewhat biased conjectures about the social class of sub-prime borrowers, it was not just the borrowers who made the mistake in this situation. A mortgage is a collateral-backed loan. The collateral asset is the property being purchased. If the property retains its value or appreciates, there is little risk of foreclosure, because the borrower can sell the property to pay off the mortgage. Furthermore, although foreclosure is usually suboptimal, if the property greatly appreciates and the borrower somehow fails to sell before foreclosure, the lender can conceivably benefit.
Both the "lowly" "rube" subprime borrowers and the "educated", "upper class" lenders actually made the same mistake - they overvalued the collateral asset, the properties being bought.
Unfortunately, the borrower of a mortgage with a balance of more than the market value of the underlying property, is put at a very strong perverse incentive to default. Imagine if someone's making payments on a 300k mortgage but the house has a market value of 200k. There is a perverse incentive to default and leave the lender holding the 100K bag, credit score notwithstanding.
This whole thing is a very interesting example of behavioral finance in action, and it isn't over yet.
Posted by: harold | August 10, 2007 4:27 PM
An interesting comment hoisted at Brad DeLong's site:
Posted by: Ted | August 10, 2007 5:20 PM
I beg to differ with the sentence, "... was determined by whichever region showed greater activation." As I understand this study, this is just as bv pointed out with neck muscles. It is correlation, not cause.
If someone asked me to do a task and watched to see if I did it with my left hand or my right hand, the observer could say that which hand I used was determined by the one I moved first. Yet that would be ignoring whatever decision making went on before that. It may be that by the time one region shows greater activation in this study, the decision has been made long before.
I doubt that there is a war in most of us between reason and emotion. We each have established values involving the two. We have established priorities, ones that differ from person to person based on experience and whatever else. Researchers can probe those by manipulating rewards, but it's not like the final decision of the immediate emotional payoff vs. the rationally mediated deferred gratification tells the whole story of the process. For most of my life my emotional self has been shackled and walks three paces behind my rational self, but in certain situations my rational self and my emotional self agree that it's better for him to have his fun now, not necessarily for the same reason.
I've never been a subject in a study of financial rewards, but I suspect my strategy there would be to go for the largest reward possible, regardless of time. I'm in no rush to get yet another book from amazon. For whatever reasons I see getting the biggest payoff whenever as winning and smart. By the way, that's also emotional, isn't it? It's just different emotions than those paid off by instant gratification. If one can see that in my fMRI, you're seeing something very late in the process, not an instanteous calculation.
On the other hand if the offer was that I could have sex with Pamela Anderson once right now or twice in a month, I'm sure something in my brain would light up like a Christmas tree, and everything rational in me would say that sex twice really isn't that different from sex once. Sex once might actually be better. I know I could argue that for the sake of my emotional side. So there would be unanimity, but I suppose my emotional side would be especially excited. Off with the shackles. I have a lot of stories about why they need to be on my emotional self ordinarily, but we do learn to be flexible about such things.
I don't think fMRI tells the whole story.
Posted by: DavidD | August 10, 2007 9:24 PM
Interesting post -
Home ownership has long been a performance statistic for politicos.
Although this situation isn't idea, people who couldn't afford to buy homes and did can not fault the "system" for preventing them. Many of these folks put very little accumulated savings into the houses. Their credit may be ruined but they haven't actually lost that much although their expenses are high.
Posted by: AnnR | August 11, 2007 10:03 AM
I think there's an uniquely American problem featured into this as well - the tendency of the American lower classes to consistently overestimate their future prospects.
There are multiple manifestations of this issue - the opposition to estate taxes and taxes on higher income brackets, for example. The toleration of relatively poor social mobility statistics. And also direct survey data showing that Americans overestimate their chances of getting into the top income groups.
I don't think that people fail to understand long term finance, or that depreciation of value over time is a flaw. In something as majorly impactful on the future as a subprime loan, the primary driver, it would seem to me, is hope. If you are poor, the only thing that makes your situation bearable is this hope, and loans on your future take advantage of this protective self-image.
Posted by: FhnuZoag | August 11, 2007 12:05 PM
Very interesting, who benefits from these foreclosures?
Posted by: Loc | August 11, 2007 2:46 PM
I'd say that Realtors and mortgage loan officers benefit most from foreclosures -- as they get to sell the place again and collect their fees.
Posted by: AnnR | August 12, 2007 8:52 AM
The larger question is how does all this play out?
Experts in finance are saying that 'this is much worse than we expected' so, as far as I see it, any prediction from them that 'everything will be all right' is just as clueless and misguided.
It will be interesting to see what happens when the 'rush' begins...
Glad I don't live in the city!
Posted by: wifiwaves | August 15, 2007 2:19 AM
Everyone is blaming those poor suckers who signed these mortgages but that's not fair. There's been a massive brainwashing. I've said for the past couple of years that I would't buy a house for the ridiculous prices offered and I was laughed at because a house is a valuable asset, and renting a waste of money, yada yada and interest rates were going to stay the same, the market could at the most stabilize, but fall, no. C'mon.
Now, nobody admits to saying anything like this. (As they look away and whistle sheepishly and run for the hills)
If analysts said this for years, what's a poor person to do?
I feel sorry for all these people who've been lied to.
Posted by: Anni | August 28, 2007 4:55 AM
Pity they didn't get to run these experiments before this 'subprime-related problem'. Also, how hard-wired is this effect and are there ways to reconfigure your brain?
Posted by: Garry Williams | August 29, 2007 4:29 PM
Today's news
Wall Street Journal
Subprime Debacle Traps Even Very Credit-Worthy
As Housing Boomed, Industry Pushed Loans To a Broader Market
By RICK BROOKS and RUTH SIMON
December 3, 2007; Page A1
----excerpt----
Many borrowers whose credit scores might have qualified them for more conventional loans say they were pushed into risky subprime loans. They say lenders or brokers aggressively marketed the loans, offering easier and faster approvals -- and playing down or hiding the onerous price paid over the long haul in higher interest rates or stricter repayment terms.
Sales Pitch
The subprime sales pitch sometimes was fueled with faxes and emails from lenders to brokers touting easier qualification for borrowers and attractive payouts for mortgage brokers who brought in business. One of the biggest weapons: a compensation structure that rewarded brokers for persuading borrowers to take a loan with an interest rate higher than the borrower might have qualified for.
---- end excerpt ----
Put this kind of pressure on old people and what do you get?
$$Profit$$$$
Posted by: Hank Roberts | December 3, 2007 1:27 PM
Most people taking out a 2/28 loan can't afford the higher interest rates that will hit later on. It's not unusual for interest payments on a 2/28 loan to double within four years.
Posted by: Sue | January 2, 2008 9:41 AM
they were pushed into risky loans, that is for sure.
Posted by: hgh | March 19, 2008 4:14 PM
Nice information
Posted by: Patchouli | March 27, 2008 1:18 AM