On the banking collapse

I’ve been trying for a while now to find someone saying something sensible about the current financial shenanigans. It’s quite difficult. First you’ve got to avoid getting distracted by the various folks who’re going to claim it’s a crisis even if it isn’t — newspapers looking for some bad news to run on page one, politicians figuring they’ve finally found the problem to which they’re the solution or just looking for something to blame on the other guy; people who see an opportunity to reap huge profits by scaring everyone else; and people who’ve got no idea what they’re talking about, but feel obliged to say something anyway.

Speaking of which…

Drilling down to what actually happened helps a little, as far as I can tell.

Fannie and Freddie

This seems to be the most political aspect of the whole mess. Fannie was setup in 1938 by Roosevelt as part of the “New Deal” to help make mortgages work better by providing a secondary market (you take out a mortgage with your bank at 8%, and your bank sells 7.25% of that to someone else so they get (less) money now, but don’t have to worry (as much) if you default). In 1968, the government decided they didn’t want Fannie on their books anymore, so fudged it by “privatising” it as a “Government Sponsored Enterprise”, and in 1970 created Freddie Mac in a similar role so they could say there was competition.

Both those companies had the benefit of (probably) being backed by the US government and being “too big” to fail — so, whereas you might worry that some bank is going to default on its loan if you lend it money, you can be sure that Fannie and Freddie won’t — especially if you’re collecting in US dollars. That means Fannie and Freddie get money more easily than their competition in the US market, and that they thus end up (well, continue) dominating the mortgage market. Which in turn means anyone else who wants to be involved has to play around the margins, by taking on more risk with sub-prime loans, eg. Apparently they even have a term — “NINJA” — for loans made to people with “No Income, No Job or Assets”.

You can then add to that some fairly seedy political practices, including campaign contributions (including illegal ones) from Fannie and Freddy, a Fannie exec sleeping with a member of the House Banking Committee who was encouraging Fannie’s deregulation, and a general push from the political side of the fence to helping poor folks buy homes.

And what do you get? A large risk, that’s been left off the taxpayers’ books for 40 years, that’s been pushed too far and come painfully due.

Mark to market and short selling

Valuing assets is often confusing — if you paid $500 for a bike two years ago, what’s it currently worth? $500 because it’s “as good as new”? $300 because that’s what you’d pay for a replacement bike? $150 because that was the high bid on ebay? $50 because that’s what the guy who stole it got when he sold it from his trailer? $600 because that’s what you had it insured for?

Ultimately you can use any of those prices, or anything in between that you can find some convincing excuse for; and cooking the books like that is a fine art that’s been perfected over the years to avoid taxes, trick co-owners, and befuddle investors. The simplest way to avoid it is to say “it’s what someone else will pay for it” — so $150 in this case, and that works pretty well on fungible assets: commodities, stocks, pineapples, etc.

The downside is what happens when not many people are trading — if you have a million shares outstanding in a company, but only one person is interested in trading at the moment, they can sell at $1, and imply your company is worth a million dollars, or they can sell at $1.50 and make $500,000 appear from nowhere, or $0.50 and make half a million disappear. And the only difference it makes to them is 50c either way.

Short-sellers take advantage of that and the fear that comes with thousands or millions of dollars appearing to disappear, by driving down the prices of shares making it look like the asset is worthless, so others start selling too. By borrowing a share, selling it below market rates, and then buying it back at even lower rates as others start piling on before they have to give the share back, they win big for a tiny initial investment.

It’s all pretty scary and annoying when it happens, but as far as I can tell, it’s not particularly unusual, or poorly understood, or worthy of being called a global crisis just because it happens to houses, banks or loans.

Credit default swaps and complex interdependencies

It gets substantially worse for the banking system for two reasons. One is a lot of their assets are in the form of credit default swaps (reportedly worth more than the entire annual GDP of the US), which tend to not have standardised terms aren’t actually traded on a market. Which ends up meaning that there isn’t a “market price”, and if you’re trying to make up a market price, you’re even more likely to hit the “barely and traders” scenario, with a 50c change in trade price costing you millions.

Worse, the nature of “credit default swaps” is that it ties banks together — if someone defaults on their loan, it doesn’t just impact one bank, it impacts the bank they’ve done the CDS with too. To be fair, that’s the point: rather than a big hit to one bank, it’s a small hit to two banks. But add enough of them at once (because people suddenly can’t refinance their loans, eg) that one bank is hit enough to go under, and they not only can’t sell off their CDS assets (because there’s no market), but the CDS that they’re counterparty too possibly become worthless, increasing the risk for other banks and creating more uncertainty and risk, and an industry wide crisis.

In short, risky, complicated, highly interdependent, non-redundant systems are a bad idea. (What, did we learn nothing from Jurassic Park?)

Fractional reserve banking

As far as I can tell, the way in which banks have a disproportionate impact on the economy compared to any other business is by holding a fractional-reserve when offering loans, and thus increasing the commercial paper money supply. All of which is to say that when you deposit $500 in a US bank, they go ahead and offer a loan of $450 to someone else, while still telling you that you can withdraw your $500 at any time. And when someone accepts that loan offer, and spends the $450, it ends up in someone else’s account at some bank, and $405 is offered in another loan, until your $500 magically becomes $5000, spread across dozens of accounts. Which is fine until you have a bank run, and everyone tries to withdraw all their money, when there’s only $500 actually there, or when people start defaulting on their loans more than was planned for, and your bank doesn’t recover enough money to pay back the $450 of yours it loaned someone. (The $500/$450/$5000 numbers are based on the 10% fractional-reserve that the US requires — other countries have different requirements, or (in the case of Australia and the UK, apparently) none at all)

This isn’t anything special, but it does create inflation, of a form. Rather than thinking there’s $500 in the economy (or $500 billion), people think, based simply on looking at their bank balances, that there’s ten times that amount. And if you’ve got $5000 (or $5 trillion), you’re going to be willing to spend a little more than if you’ve only got a tenth that. But if you suddenly discover you don’t have that much money — because your bank’s failed, because you can’t get a loan you were relying on, or because your currency has deflated; you need to tighten your belt and spend less. And when you spend less, other people sell less, get paid less, and spend less themselves, which gives you deflation — and in so far as people are more annoyed at having their salary cut than finding things have gotten a bit more expensive, deflation is worse than inflation.

I’m presuming, though so far I haven’t seen any actual evidence, that the problem with all these sub-prime mortgages and such is impacting the reserve banks are holding, which is in turn reducing the loans they can make, and causing a cascading contraction of the money supply. It might be enough that they’re simply not lending to other banks due to trust issues (or fears their current outstanding loans won’t be repaid on time) — effectively increasing their reserve from 10% without being officially required to, and that causes deflation too.

Naked shorting

You can also get inflation (and hence deflation) with shares too: all you need is the ability to make shares appear from nowhere. Which is what happens in naked short selling: rather than borrowing a share, selling it, then buying another back to return in its place, you just sell a share you don’t have, and buy one later, and hope either that the computer system doesn’t complain, or that you don’t have to deliver the share until you’ve bought it.

But for the time between the sale and the purchase, you’ve tricked everybody else into thinking there are more shares around than there actually are — if there were a million shares outstanding, and you’ve sold ten thousand shares you don’t have, then suddenly there’s a million ten thousand shares outstanding, and since the company’s still worth just as much as it used to be, each share’s suddenly only worth 99% of what it used to be. And if someone offers to sell for that, you get to cancel your short, and claim 1% of 1% of the company’s entire net worth. That only works if people don’t factor in the fact that you own a negative number of shares, but not only is short selling all about profiting from other people’s confusion anyway, but it’s hard for people to tell how many shares are outstanding in total, but easy to tell how many are currently up for sale, and that’s exactly what you’re influencing.

I haven’t seen any evidence that’s had anything to do with anything — at least other than the recent knee-jerk banning of short selling, both naked and otherwise, but it seems interesting and somewhat worrisome in and of itself.

The bailout-come-rescue

Well, it’s not like there’s just one — there’s the Fannie/Freddie buyout, the AIG buyout, the TARP programme, and various “liquidity injections” around the rest of the world too. It’s not at all clear to me what any of that will actually achieve; it could go anywhere from making a profit for taxpayers by buying currently underpriced assets that will recover in the near future, to being used to inflate prices of worthless assets in a grand maneoveur to transfer billions of dollars directly to bankers and financial portfolio managers and the preferred projects of the current political elite for nothing more than an assertion of “increased confidence” from those same folks.

My guess is it probably really is necessary — without it, there likely would be a real contraction in the money supply, and the changes in prices necessary to accommodate that (ie, making cars cheaper because people can’t get car loans; lowering incomes because people aren’t paying as much) would cause a lot of difficulty. Worse, deflation exaggerates bank balances; so if you’re already rich with a big bank account, you’ll be able to buy more with that; if you’ve already got a big credit card debt, that’ll be harder to pay off. While that might be all nice and anti-consumer, it’s also a pretty regressive transfer of wealth from the poor to the rich.


The US (and apparently most other countries except Australia) offers deposit insurance. That is, if you put money in a bank backed by the FDIC, and it goes bust, the government will make sure you get your money back (up to $100,000 anyway). There’s been talk of increasing that to $250,000 as part of the bailout, but I haven’t paid enough attention to see if (or when) that’s actually happening.

That sounds all nice and rosy, but like everything it has a downside. Here’s one way to exploit it.

First get a few patsies to run the bank for you — it’s designed to fail (otherwise the FDIC isn’t going to matter!), so you need someone (else!) for people to blame, and who can do any jail time or have their assets confiscated, etc.

Second, setup a high-risk, high-return venture, and use that to offer ridiculously high interest rates on deposits, say 15%.

Third, make sure that it’s easy to explain why what you’re doing is good, say “we’re helping the poor and underprivileged start businesses”, but hard to explain why what you’re doing is high risk, say by hiring a lot of clever econometricians to create complicated derivative products. That way you can keep politicians on your side, and avoid the public getting wise.

Fourth, get lots of folks to invest up to the FDIC limit, and make high-risk, high-return loans, right up until you get too many defaults and your patsies have to declare bankruptcy.

Finally, pay off all your creditors, issue a press release about how some people just can’t be helped no matter how generous you try to be, and collect your original investment plus 15% direct from the government.


So, as near as I can tell, the lessons are probably these:

  • don’t quasi-privatise things — you just get the worst of both worlds
  • don’t leave things off the books — they’ll only get worse for the lack of attention, and they will reappear eventually
  • systemic failures happen, internal redundancy is important
  • banks are important to the economy and need special care
  • the bailout is probably unjust, but doing nothing is probably even worse
  • just because things sound like a good idea doesn’t mean they’re not hurting more than they help

Of course, don’t expect anyone to learn any of these lessons: the Federal buyout of 80% of AIG is both a quasi-nationalisation and specifically scaled to stay off the government’s books; redundancy is the same as inefficiency and no one likes that; everyone knows banks are evil; the bailout’s just in time to be fodder for the most important election ever; and that certainly means no one’s got time to work out what is good rather than what sounds good.

Leave a Reply