Words of wisdom, from Kevin Drum

I agree with this:

It's not practical to micromanage risk-taking in the financial sector, nor is it feasible to eliminate bubbles and bank crises entirely. But I really do believe that we could very substantially reduce the risk of bank crises without affecting the efficiency of legitimate banking operations. The way to do it is with very simple, very blunt leverage restrictions that apply to all financial actors over a certain size: banks, insurance companies, hedge funds, private equity, you name it. If you have assets over, say, $10 billion, then the rules kick in. Strict leverage limits (say, 10:1 or maybe 15:1) based on conservative notions of both assets and capital would be a pretty effective bulwark against excessive risk taking but wouldn't seriously interfere with the basic asset allocation function of the financial industry.

It wouldn't be perfect. Nothing is perfect. But if we got obsessed with leverage the same way that, say, the Fed is obsessed with inflation, we could all sleep a lot easier at night.

I would add, however, that obsessing over leverage is likely to prove an electoral disaster.  It means limiting credit growth, money supply growth, raising the cost of consumer borrowing, and putting the housing market at a further disadvantage.  It also means staring down financial interest groups.

Insert obligatory St. Augustine quotation here.


Maybe we need MORE frequent crises.

Wouldn't it be better to simply remove the incentives that exist to bias the markets towards excessive borrowing and excessive risk?

Rather than regulating leverage, how about phasing out the home mortgage interest tax deduction, making bankruptcy more severe, pledging not to bail any company out in the future, getting Fannie and Freddie out of the mortgage securitization business, re-privatizing student loans and getting rid of the government subsidies of them, reducing FDIC limits and publicizing the hell out of that to make sure people understand that insolvent banks are a bad place to put their money?

Unwind the credit card legislation that has resulted in lower interest rates but higher annual fees - rewarding borrowers and punishing people who pay off their balances. Let's let credit card companies charge however much interest they think the market will bear.

Along the way, we can increase the incentive to save by cutting back on Social Security and Medicare funding, and using the money saved to allow more tax-free savings instruments.

It would be nice if we could also stop the fed from holding interests rates artificially low for an extended period of time, but that's a tough problem to solve, since you often don't know that the interest rate was too low until the bad effects start showing up.

If we need some additional legislation, how about transparency laws that simply require banks and other lenders to post their current leverage status very visibly in branches, on ATM machines, or in print and TV ads?

If we can get the public to understand that certain investments carry risk, then we should be able to set up an environment where banks compete with each other to show stability.

That actually used to be the case a long time ago, which is why bank branches used to look like imposing fortresses rather than DMV offices. They were projecting an air of stability and safety, because the public valued that aspect of their banks. But when the government 'protects' you from possible harm when banks fail, or protects banks from failing, that incentive vanishes.

It also means people will realize their income growth is fairly anemic.

Completely agree with him. Both in substance and in form. Legislatures should focus on passing simple clear bills.

The other thing would be to pay every bonus in kind. You loan money, your commission is a slice of the payments on it. You do advisory work for cash, you get paid in cash. You manage assets, you get paid with a stake in the fund.

If there is anything the last 40 years has taught us, it's that non-dynamic, arbitrary rules on bank leverage work like a charm.

From what I understand before the FDIC banks kept 30% capital, why not make that the rule now.

Though sensible, and you could say "we will only bail out the first 10x" or so, it's not really the one firm 30 or 40 times leveraged that was the problem, it was the millions of individuals and the firms financing short-term cash flow, and all the carry trades.


What mattered was the risks inherant in the assets and liabilities.

But doesn't this risk determine what the sensible limit on on leverage is? You can go broke overleveraging on any asset.

Gee, I'm sure no one will *ever* think of splitting their holdings across multiple "banks"...


Can't wait for some hedge fund to crater and threaten a bank.

On the same subject of regulation, however, we sometimes forget to talk about the "dog that didn't bark".

More specifically, given the severity of the financial crises, it seems to me there were fewer corporations that went into bankruptcy with allegations of accounting chicanery than we had in 1999-2001 ala Enron and MCI.

After Sarbanes Oxlecy I have watched CFOs sweat over the financial disclosures and also watched them get more backbone in responding to the CEO because both of their tails would be in the guillotine. Sarbanes-Oxley might be the correction of the last cycle that reduced the amplitude and number of accounting failures this cycle.

Maybe Sarb-Ox muzzled some potential worse crises that never barked.

@ Dan Helphrey:

Isn't that in broad strokes the idea behind Glass-Steagall (R.I.P.)?

The risky entities and the taxpayer-guaranteed entities used to have to be separate. After 1999, no longer so.

It's probably impossible to get there, but taking your idea and expanding it, making pure lending a highly regulated, low risk, low ROI utility and putting everything else in unregulated, 'take all the risk you want but no bailouts for you' system is appealing.

The problem there is of course financial innovations...borrowers and lenders will find ways to transact plenty in the risky sandbox, and nothing has really changed...

Not sure on why the large-cap requirement before regulation kicks in: If incentives matter, then systemic risk through several medium-sized firms making similar, correlated high-risk choices because they face similar incenitves is as much a problem as systemic risk through a single large firm being too large to fail.

Hm. Sounds suspiciously like Austrianism with government controls.

Wow, this is the dumbest idea I have heard all year. If there was a way to encourage regulatory arbitrage this is it.

If you have an FDIC and you have any sort of capital restrictions or requirements you end up causing massive regulatory arbitrage; this is even stronger if you have a bailout authority for creditors. Well, if there is inefficiency and regulatory arbitrage, more room for me to collect rents.

10:1 for a pharmaceutical start-up? That's insanity. 10:1 for Euro/USD trades? You'll kill the forex. Unfortunately, you can't have a leverage cap without micromanaging.

leverage does NOT equal risk. A book that is long 10 year notes, for example, can reduce its risk by selling the 5 year, this increases its leverage. In more complex cases there will be no simple rule mapping leverage to risk so no formula created by regulator will do a good job of assessing risk by looking at leverage. There's no simple solution, the best approach is to not have government backstops thereby increasing the incentives to the institution and its investors of creating prudent safeguards against too much risk taking.

bobg, leverage doesn't necessarily create risk, but it magnifies the gains (and losses). So it puts more capital at risk.

And if EVERYONE is making levered deals on the same assets, that's a pretty good contributing factor to a price bubble which creates systematic risk. I agree though that there is no simple formula and a bad formula or a one-size-fits-all formula can be disastrous.

The problem with a broad brush approach to regulation is that it coordinates behavior. If the regulator makes a mistake with a particular requirement, it either mandates all the regulated firms to make the same mistake or all to seek a common loophole. E.g. Choosing the riskier asset with a higher return when the two assets are treated equally by the regulator.

Cyrus has it right. Too big to fail isn't the problem. Too systemically important to fail is the problem.

@Dan: we already have your two options. You can get low returns on fully-insured deposits by spreading your funds out to several banks. You can get higher returns by investing in money market accounts, mutual funds, or junk bonds. We can have an unregulated market for uninsured investments, but then we'd have lawsuits over fraudulent financial reporting every time an investment vehicle went belly up. Do we want lawyers and judges to play ex-post regulators? The principle-agent problem is just too huge for a completely unregulated system, as much as I would like to believe otherwise.

We talk a lot about "bailouts", but we often only bailed out the team franchise, not the players. The bondholders and stockholders have been essentially wiped out. That's not to say taxpayer money hasn't been siphoned off for low interest loans and the criminals allowed to walk free.

@Dan: The commercial banks that got into trouble were primarily NOT linked to any investment banks. Investment banks which got into trouble were NOT generally affiliated with commecial banking. Those banks where IB and CB were linked generally did much better to weather the crisis. Glass-Steagall is a red herring designed to blame "deregulation" for the crisis.

The failures of EXISTING regulation served as an enabler or facilitator of the crisis. Few people in Congress or regulatory agencies were clamoring to tip the grog over the rails, until it was too late. Indeed, many of them were breaking out fresh casks.

Yo ho ho!

Kill the interest deduction for corporations (and lower the corporate income tax rate accordingly) ? One incentive less for excess leverage.

Also, why the hell do we need any financial innovation in the first place ?

"Cyrus has it right. Too big to fail isn't the problem. Too systemically important to fail is the problem."

Actually, that's not the problem. The problem is bankruptcy is a government program and they have not provided the correct process.

One problem is a failure of gov't to provide clear, simple, and fast bankruptcy process for Too Big To Fail "systemically important" financial firms. This was a forgivable failure with Long Term Credit Management, but not after. See http://www.erisk.com/learning/CaseStudies/Long-Te...

To avoid the threat of a systemic crisis in the world financial system, the Federal Reserve orchestrated a $3.5 billion rescue package from leading U.S. investment and commercial banks. In exchange the participants received 90% of LTCM's equity

A rescue package instead of a bankruptcy.
A "market solution" might be 100% equity loss, and fast bond/ debt to equity conversion, with top management fired and new debt=equity owners choosing new managers, or loaning their own personnel as managers.

Second problem, too much AAA rated financial instruments. Monetary expansion of AAA rated junk was fed into more junk loans on house prices, continuing to jack up the bubble. $10 trillion? $20 trillion? $60 trillion? How much AAA rated junk was there? (I keep looking, still don't know.)

There should be a maximum total limit of last year's GDP for AAA rated debt, with new AAA debt forcing downgrades of old AAA debt -- or not getting the SEC/ bank regulator required AAA rating needed to be included as Tier 1 capital.

It wasn't the leverage which was the problem, it was the wrong AAA rated financial tranches which were really junk. Transparency and restrictions on ratings of total amounts of debt is more needed than leverage control.

The failure of the rating agencies is the biggest almost undiscussed issue. When was the last time you saw a graph of total AAA rated issues? (Where? I've never seen one.) But this is the M1 base money which is used for 10:1 or 65:1 leverage.

And, yes, obviously, if there is more leverage on misrated "AAA but really junk", there is a bigger problem.

sixounces - in the example I gave you create more leverage and put less capital at risk, that's of course what is important. Increased leverage can actually reduce risk in some cases so it is not a useful metric.

Comments for this post are closed