Credit and Credibility:

Are banks crazy, or a cartel?

previous   next

21 June, 2011

When a government (let us say, in Athens) could possibly renege on promises made to banks who loaned them money or bought their bonds, which that government is unable to fulfill without draconian cuts to public services, all right-thinking people attack the feckless politicians and threaten a collapse of confidence and the world economy. This is a DEFAULT! Other governments and the IMF might jump in to pour money into the state coffers, on the condition that they flow out the other end into the investors' pockets.

But when a government (could be in Athens, or in London, or for that matter Madison, Wisconsin) has made promises of pensions to government employees, but has failed to fund them adequately, it is short-sighted and greedy for these civil servants to insist on these promises being honoured.

Why is a default such a terrible thing? Because, they say, if the country defaults on its debts it will be shut out of the credit markets. Hmmm. Let's suppose it is true. Why? Suppose you own a bank, and your thinking of lending to one of two countries, let's call one of them Piigsia and the other Sameria. Both are heavily indebted. Piigsia introduces crushing austerity measures, while Sameria repudiates its sovereign debt. Which of those countries would you rather loan your bank's money to? The one that's shown a great willingness to pay off its debts but is financially crushed, or the one who may be more likely to try to weasel out of its debts, but is eminently capable of paying. Solvency is not merely (or even primarily) a state of mind. I mean, what good is it to have the current government express a willingness to pay off its debts, knowing that it's likely to be punished by voters for these "good" intentions? Maybe they just don't want to be serial defaulters, so having avoided defaulting this time will encourage them to default on the next batch of loans.

As for Sameria, it sucks for the other banks that have lost their money, but why should I give up a chance to make a good profit for the sake of punishing Sameria for hosing my competitors? In fact, in a competitive market, why shouldn't I be happy that my competitors have made a loss, and just try to get better conditions for my loan?

That's why I wonder if there isn't something of a cartel or class solidarity in the way bankers behave, rather than a free market. Yes, they compete, but fundamentally they see themselves as collaborators in enforcing each others interests. No one is going to jump in to

The alternative is that bankers just aren't very smart. It's not really an alternative... There's no reason why bankers couldn't be dumb and criminal. I used to think that bankers were pretty sharp, even if they were dull... Stolid, to be sure, perhaps grubby or conservative or philistine or narrow-minded, but within the constraints of their worldview and their interests, I figured they knew what they were doing, how the world works (in its most dreary and quotidien aspects), and how to turn that understanding into a profit. Events of recent years have forced me to reconsider. In a country whose economy is comprised of financial institutions to (approximately) 99.7%, and whose most unwavering political conviction is that all institutions of the nation would be improved by making them more like banks, this is not unlike the position of a Vatican native who has begun to doubt the existence of the pope.

Now, it may be that the problem is the business journalists who interpret the bankers than the bankers themselves. At least as presented in public, their whole conception of risk seems bonkers. Consider a typical sort of comment -- hardly the most egregious -- on the current unpleasantness, by BBC business editor Robert Peston:

Here's the thing. If any eurozone member were to leave the euro, if Greece or anyone else were to adopt its own independent currency, the cost of borrowing for pretty much every other eurozone member - with the exception of Germany, Luxembourg and the Netherlands - would rise.

The reason is that euro membership would no longer be forever. So anyone lending to Spain, Italy or even France would have to be compensated for the risk - however remote - that their euro-denominated debt would one day convert into something tied more directly to the health of their respective economies and the strength of their respective public-sector balance sheets.

Now looked at by the usual standards of logic, this doesn't make any sense. Imagine a man coming sitting down with his wife and telling her "I've been having an affair, and my life with you seems unbearably dreary, but on the balance I've decided that staying with you is better than leaving." Would she then breathe a sigh of relief and feel reassured that their marriage is "forever"? Or would she decide that it's only a matter of time until a bigger crisis comes. Or you live on the top floor of a skyscraper. Along comes a gale-force wind that cracks the foundations and twists the girders. But the wind subsides, and the building is still standing. Whew, you say. This is a tough building. Glad I won't have to worry about it falling over. (I'm reminded of the way seventeenth-century French investors are reported to have bought annuities on the lives of greybeards, presumably on the theory that they'd already shown a propensity for long life.)

If Greece leaves the euro, I can see that this might be a template for others; on the other hand, it would depend a lot on how Greece fares on its own. If it does badly, that might frighten the others. In any case, it seems inconceivable that someone could look at this sorry mess and think, well, Greece has stayed with the euro, and crushed its pensioners rather than repudiate even the meanest of its bonds, so I guess I'll never have to worry about Portugal again?

Folk probability theory: The zero-one law for default

There seems to be an alternative form of probability theory in effect, where all events have probabilities that are either 0 or 1, but we need to figure out which it is. It's not as though the reality where Athens descends into near civil war and emerges with an austerity package passed by a couple of votes, and the alternative reality where the package fails by a couple of votes, are vastly different in terms of what they predict about the future willingness of "Greece" to honour its debts. But at least the financial commentariat seems to treat them as though they were. (Obviously, the difference matters more to current holders of Greek debt, but even there, it seems that there's a continuum of meanings and projections that could follow from, say, a single missed coupon payment.)

From a probabilist's point of view, the reports of credit agencies also seem batty. I mean the ones of the form: "Consolidated Counterfactuals bonds are rated BBB-, but we have a "watch" on it, and we are likely to downgrade it soon". I take a credit rating to be a way of describing a range of probabilities of the company or country defaulting on its loans. What can it mean to say, "The probability of a default is currently 0.1, but there is a significant chance of it increasing to 0.2 in the next month. This could only make sense if there is some event X that we are waiting for, and if X happens the probability of default goes up, and if it doesn't the probability will go down, and it all averages out to 0.1. But then a "watch" would be a purely neutral statement, something like, we expect the default probability to be volatile over the next month, equally likely to go up or down. That's certainly not how it gets reported in the financial press. (I'm reminded of the Robert Heinlein novel, The Moon is a Harsh Mistress, which I read as a teenager. Lunar colony secessionists go to war against Earth, guided by a sentient computer that compulsively computes odds of their ultimate success based on current status. At one point there is a discussion that even then troubled me: The computer has  just lowered the estimate of the probability of success to one in nine. "Getting worse?" the protagonist asks. The computer replies: "They’ll get worse for months. We haven’t reached the crisis." Later the success probability drops to one in thirteen after the rebels initiate a propaganda effort directed at the Earth public. The protagonist protests that the computer had agreed that this was a necessary step toward victory. The computer "explained patiently, “it increases risk. That it is necessary risk does not change the fact that risk is increased.”" If a step is necessary to success, it can't be that carrying out that step lowers the probability of success.)

An extreme example is this recent report: "Standard & Poor’s would cut the U.S. credit rating to its lowest level and Moody’s Investors Service said it will probably reduce its ranking if the government fails to increase the debt limit, leading to a default. S&P would lower its sovereign top-level AAA ranking to D, the last rung on its scale if the U.S. can’t pay its debt, John Chambers, chairman of the company’s sovereign rating committee, said today." So, how can it be that the debt is today AAA -- which is generally understood as a declaration of close to zero risk of default, at least as low as possible -- but that they also can declare that they believe that the US might plausibly default on its debts next month. And if you think what they're really saying is that the "default" isn't a real default, but only political theatre, then what's the point to lowering the rating to D after the fact?

There is one reasonable interpretation that makes sense of all this: When the credit ratings agencies rate sovereign debt -- at least, for wealthy countries -- they are doing something similar to what Gallup is doing when it predicts the outcome of the presidential election, namely, doing publicity work for their real money-making business. (For Gallup, that's market research; for S&P it's rating private debt and equities.) No one really looks to S&P to tell them whether US treasury bonds are a safe investment -- or German or British or even Greek sovereign debt, for that matter. The goal of these announcements is simply to provoke headlines, and create the impression that this ratings agency is a big player. It's a bit like the rooster who threatens that the day won't start if he withholds his crowing.

Retail banks and the zero-one law

I've long found the practices of US lenders peculiar, particularly the apparent belief that credit should only be extended to those with a credit history. One imagines an 18th century banker being leery of lending to to someone for the first time, someone who has had no practice of borrowing and repaying. Will they understand what it means to pay interest? Can they count? But today, dealing with money and basic banking functions is not a specialised skill. It seems strange that someone who has declared bankruptcy will be able to obtain some sorts of credit, whereas someone who has not taken out a loan before often will not be considered for a credit card or loan under any terms. (Obviously, there must be some route into the system. For myself and my friends it was special offers for credit cards to students. Student loans count as well, I guess. I'm not sure how the non-students manage it. Perhaps with a co-signed loan, or a deposit-secured credit card.

In the UK, I found that there was basically only one lender (Natwest) who would even consider us for a mortgage, because I applied with less than 2 years remaining on my work permit. It seems odd that they would have loaned me money when I had first arrived in the country (I have a 5-year residency permit), but not in the two year interval between the end of year 3 and the end of year 5. (Actually, well into year 7, since it takes well over a year to get a work permit renewed.) It's zero-one: All other banks, apparently, will not even consider an application on any terms from a foreigner who has been here too long, until he's gotten permanent residency. It's hard to imagine what the risk is that they're trying to exclude. It's not as though a sneaky foreigner can take out a mortgage and then abscond with the house. In our case it was particularly risible, since my co-applicant is German, so doesn't need a visa, but that changed nothing in the status of our joint application. Again, one would think that in a free market there would be more competition for the custom of respectable foreigners. (It would probably not be reasonable to link this to Britain's never-flagging xenophobia. In Canada, the most foreigner friendly country I've had dealings with, we couldn't even get any sort of credit card without permanent residency.)

Cover page