rinue | Romie Has an Economics Degree

A few people have asked me what exactly happened to the stock market, and whether anybody predicted it. So here's a quick primer.

You probably already know that this has to do with the subprime mortgage crisis, and maybe something involving a housing bubble. What I'm hearing you don't know is how the crisis happened [Banks loaned money to people who couldn't pay it back, but why? And why all of a sudden?] or how it crashed into so many other areas [Okay, a bank is failing, but my retirement account is not at a bank, so what is going on?]. There's also a sense that it must be someone's fault, but whose?

Basically, it boils down to bad models that mis-estimated risks, and then mechanisms for selling off those risks to people who couldn't see the models. Marketplace and This American Life put together a great explanation here, and it's worth an hour of your time to listen to it.

The problem is housing bonds, which are the things that Fannie Mae and Freddie Mac make. Housing bonds are aggregations of lots of people's mortgages, and buying them has for a long time been thought of as a fairly safe way to store your money and make a little return on it - kind of in the same category as government bonds and certificates of deposit. When you buy a housing bond, what you are effectively doing is giving the bank extra money it can loan to people, and when the bank gets interest on the loan, so do you. It's low risk because the bank is also invested in the loan and you can trust that they've done all the background checks on the people they're loaning to. Aside from that, you, like the bank, are buffered by the fact that this is an aggregate - maybe some people get into trouble and default on their loans, but the majority don't, and so you can count on money coming in.

There is a whole industry based on estimating how many people in a given bunch are going to default. It's like another level of oversight that looks at loans as groups. So you have somebody at the bank whose job is to look at an individual white 30-year-old married couple and decide whether they're a good risk, and then somebody at the ratings agency looks at all white 30-year-old married couples who have ever gotten loans and sees what percentage made their payments, pretty much the same way insurance companies use actuarial tables. The ratings agency then stamps the bond (the aggregation of loans) with a rating that can be anywhere from AAA (virtually no risk of default) to D (almost certain to not pay).

Most people who buy these bonds - including mutual funds - make their decisions based on these ratings. They look at the return on the bond (which is like the average interest rate of the loans) and then they look at the rating to see whether it's worth the risk. This has been going on for ages. It's a great system. It gives banks more money to loan to people (which is good for homebuyers and small businesses), and it provides a lot of information to investors, who can buy something extremely secure without having to do all the research themselves. It's particularly great for, say, retirement accounts - you can buy a triple-A bond and be sure that money will still be there in 20 years, and you will have gotten an interest rate that kept up with inflation. Good stuff.

So what happened? What happened was a software model that caused people to categorize a lot of D-rating loans as A-rating loans. Some analysts looked at an aggregate of high risk loans and realized that they didn't default as often as you might think. A people person would take this to mean that banks' loan officers know what they're doing - they're smart enough to see that some people who look bad on paper are still worth loaning to. Maybe the successful loan applicants had bad credit scores for one reason or another, but something about them showed they were responsible.

However, the analysts were not people people. They were number crunchers. Instead of interpreting the data as an indication that banks were good at finding outliers who beat the odds on the actuarial tables, the analysts figured the tables were wrong. Loans to traditionally high-risk groups (poor, unreliable income, bad credit, no collateral) were suddenly not categorized as high risk. They were still maybe a B-, but that's a heck of a lot better than D. And if they're B-, and you throw a couple of them in with a thousand AAA loans, then the overall bond is still going to be a AA, because we already knew some people were going to default, right?

You can see, I hope, that this makes every single estimate of risk mean nothing. It's like secretly throwing some pebbles in with the M&Ms before you hand them to your cookie-baking friend. On the other hand, if you're throwing a handful of pebbles into a football-field-sized vat of M&M's, it's probably not going to be terrible. It still sucks, but it's not a catastrophe. Something else had to happen.

What happened is that banks listened to the ratings agencies. After all, these guys were speaking with an incredibly high degree of accuracy, and we've always been able to trust them before. They've got to know what they're talking about with all these numbers, or they wouldn't be so specific. It must be true that this category of people is not the risk we thought they were. We should relax and loan them more money, which in turn makes us more money when they pay us back.

So at this point, we've moved on from pebbles to pieces of shit: loans to people who are not outliers, people who were reflected in the original actuarial tables. And even the original actuarial tables were wrong, because they were built out of insufficient data - this was a group of people who had never in history been able to get loans, and so we didn't really know how they were going to behave. End result: at the top of the chain - the investors buying the bonds - people are making confident decisions based on carefully collected financial numbers that are a terrible reflection of reality.

Surely, you think, surely the people at the bottom knew what was going on and should have moved to stop it. And you're right: there were all kinds of loan officers at the bottom who said "wait a second." There were also groups of auditors in internal affairs departments screaming their heads off. Unfortunately, these people at the bottom were no match for a group of evildoers known as the people in the middle. The people in the middle were making a lot of money taking their cut, and wanted to believe in the tables. They wanted to believe with a fervor that was almost religious. They did not want the people at the bottom to spook the investors at the top. They hired new loan officers with less training and more greed, and they buried the internal affairs guys with paperwork.

I'm being unkind to the guys in the middle. I think they deserve it. However, they weren't all horrible - some of them probably did believe in the tables, and some of them probably did think the people objecting were standing in the way of innovation - were the kind of people who refused to use cell phones. More importantly, they saw themselves as normal - they were surrounded by a bunch of people who made the same decision they were making. And that many people can't be wrong, right?

Things got especially scary and stupid when these middle guys got into escalation wars to see who could loan to the most traditionally risky person. They didn't think of it that way - they thought of it like "if my bank doesn't loan to this person, then somebody else will just come along and do it, so it might as well be me who makes the money." On top of that a bunch of sharks moved in to get a piece of the gold-rush action. Easy money.

This behavior was incredibly predictable. It reminds me of a classic Economics exercize, one which is more commonly used to illustrate the tragedy of the commons:

Take any group of people. Take a group of kindergardeners. Have them stand around a chalk circle and drop a few hundred pennies into it. Then set a timer. Tell them that any penny they pick up in the first minute is just a penny, but you'll give them a nickle for every penny they pick up after that minute is up.

That group of kids will stand there for a while, waiting for the minute to pass, treating it like a ready-set-go. Then one of them will break - he'll figure out that he can start picking up pennies while everyone else is waiting. He'll drop to the ground and start scooping.

Instantly, all the other kids will do it too. It's like a wave breaking. Nobody will make it to one minute. Nobody will get a single nickle. They can't wait for a penny to mature into something worth five times as much, because if they wait, somebody else will take the penny and they'll get nothing at all.

If you do the same setup, but divide the chalk circle into pieces and give each kid a piece from which they and only they can take pennies, every single one of them will wait out the minute.

The banks were like kids around the first circle. They were competing like a market, but they didn't understand the risks they were taking and were worried they'd lose out on all the money if they played it safe. The result was that everyone wound up with something, but something that was practically worthless. Then they put the pennies in a sack and handed it to investors who knew how many coins the sack held, but not what kind of coins they were. Risk didn't stay with the people who made the bets.

I hope this explanation makes it clear that a lot of people messed up. A lot of people. There was mass hysteria. The good news, if you want to think of it that way, is that almost all of these people are now out of work, because the banks all collapsed. The bad news is that the people at the bottom (people who want loans, people who want to be trusted to know which loans are good, companies that need credit for their day-to-day operations) and the people at the top (investors - both individuals and large businesses) are still screwed. It's not just people losing their houses because they took out loans they couldn't possibly pay. It's pretty much everybody. That's why Congress has had to step in. (Incidentally, I trust Bernanke, the Fed chairman. I'm not so sure about Paulson, the treasury secretary. And Cox, the SEC chairman, is some kind of ghost.)

The word "regulation" is getting thrown around a lot, because this was clearly a regulation failure - internal oversight broke down at the banks, ratings agencies, and mutual funds. Then nobody at the federal level stepped in until just now, even as it became clear things were going crazy. The problem is, well, how are you going to do that kind of oversight from outside without freezing everything up? The smartest people are already working in the industry, and the reason the government hasn't bothered the industry guys is that the industry guys have done their job well for a long time. Moody's, Standard & Poor, and until just now Fannie Mae - these are names we actually trust more than we trust the government. We trust them enough to give them monopolies, and in this case a monopoly is the most efficient way to do things. This stuff is complicated. It takes a lot of people. You could have the government check every loan and every bond, but they'd just be doubling the work of the guys who are already doing it.

You may have noticed something sneaky in the above paragraph, which is that I've said monopolies are good. (It's really oligopolies, but the point remains.) But that's crazy, because when you're dealing with a monopoly, you have no recourse when something goes wrong. You can't go to somebody else. And you can't let the company go under. Which leaves you in a conundrum if it's an industry that works better as a monopoly - as an authoritative single source.

We can't let these companies fail because they have taken on jobs that they do better than the government does (since we refuse to fund it), and we can't afford to leave those jobs undone. Effectively, unofficially, they are a branch of the government, kind of like air traffic control. Other countries deal with this by making it official - by keeping these few big everyone-needs-them only-one-entity-can-do-them tasks in the hands of an elected, accountable body - or at least people appointed by one.

But America is uncomfortable with the word socialism. Instead we have to tiptoe around it and do things that are less efficient than simply taking over. We need a bailout. The basic idea is that the government will take on the risks of the riskiest bonds and spread that risk out across more people and a longer timeline - kind of a reflection of what mortgage bonds are meant to do. We don't hand Wall Street seven hundred billion dollars so much as move some negative numbers from their balance sheets to ours. Of course, since the ratings agencies never rated the bonds accurately, we don't know how much risk we're taking on, and how many of those negatives are really negatives. But at least we know we don't know, which is more than individual investors knew when they did the exact same thing a few years ago.

The Republicans are voting against the bailout as socialist and the Democrats are against it as insufficiently punishing and people on both sides are pissed that it's inefficient and lacks oversight, even though it's probably the best compromise we can manage in this political climate, and it's something that certainly needs to happen. That said, the most "no" votes came from Representatives in contested, uncertain seats. Senators - who have 6-year instead of 2-year terms - and Reps in secure seats were more likely to vote yes. Which tells us that above all, the bailout is politically unpopular - unpopular with the American people.

Basically, we have put ourselves in a ridiculous position where the only possibility is anarchy, which we also don't want. We don't want government to mess with stuff, because that's socialist and inefficient. (Hint: it is not actually inefficient.) We think we want private industry to do everything, and to take its licks when it fails, because that keeps our taxes low. But when you leave regulation to the same industries that profit from being unregulated, this kind of crisis is inevitable, and we can't absorb it. Wall Street is Main Street. When we punish the banks, we are punishing ourselves - not just the people who took out mortgages, but the good credit risks who can't get loans, the retirees who put all their money in guaranteed safe investments, and the family members who suddenly have to care for them.

Or, in Jimmy Stewart speak, mixed with a little Wizard of Oz: That money's not in the bank. It's in your house, and your house, and yours.

Government at its most basic has two jobs: to provide physical security and to provide economic security. Without the ability to do those two things, it is not a state. It is not functional. And in a democracy, the government is us. We fucked up. We handed the bar keys to a bunch of alcoholics on the grounds that they know more about liquor than we do. We've got to have the balls to take them back, because we're the ones losing money regardless.