Network News

X My Profile
View More Activity

Explaining financial regulation: Leverage and capital requirements


If you watched the Sunday shows this weekend, they were thick with politicians and administration officials arguing over financial regulation. But the gladiators were getting ahead of the audience, and they knew it. On Meet the Press, Treasury Secretary Timothy Geithner began explaining the need to "move derivatives out of the dark" only to be cut off by host David Gregory. "We have to stop and tell people what a derivative is," Gregory said.

Indeed we do. This week, I'll be trying to explain the basic elements of financial reform. And we're going to start with the most important bit: Leverage and capital requirements.

What you need to know, first, is that the words "leverage" and "capital requirements" are different ways of talking about the same thing. Leverage refers to how much money you've borrowed compared with the capital you have on hand. Let's say you're buying a house, and you put 20 percent down. Your leverage is 5:1. For every dollar you had in capital, you've borrowed four more to complete the sale.

If you're a bank, the upside of leverage is that it gives you a lot of money that you can use, well, to make more money. It's the difference between investing $1 in the stock market and $40.

The downside is that it makes your firm fragile. If your leverage is at 2:1 -- that is to say, you've borrowed one dollar to add to the dollar you already had -- you could lose a full dollar and still be able to pay your creditor back. If you're at 10:1, anything beyond a 10 percent decline in your assets means that if your creditors want repayment, you can't pay them back (as you've lost more than your original dollar). At 20:1, a 5 percent decline will put you underwater. At 40:1, a mere 2.5 percent decline can finish you off. You can see the ratios in the graph atop this post. The more leverage you have, the less bad luck you can survive.

Leverage also helps firms become too big to fail. A firm with a billion in capital can have $40 billion in liabilities, which means that if it goes down, there are other banks and lenders who need $40 billion in repayments if their balance sheets are going to add up. The banks that survived the crisis best, like J.P. Morgan, had the lowest levels of leverage.

That's led many observers to focus on capital requirements -- regulations limiting the amount of leverage firms can take on -- as the most important regulation going forward. "The reason I raise the capital issue so often," Alan Greenspan said in testimony to the Financial Crisis Inquiry Commission, "is that, in a sense, it solves every problem." Timothy Geithner was no less complimentary. “The top three things to get done are capital, capital and capital,” he told David Leonhardt.

The question, of course, is how you get "capital, capital, and capital" done. The most basic approach is to look at what's called "gross leverage ratios." That's what we've been talking about here: The amount you've borrowed for every dollar you've got in capital. But the worry is that companies can skirt those rules by purchasing riskier assets with the leverage they do have. That is to say, they'll respond to low capital requirements by buying riskier stuff in order to keep their profits high. It would be a little like someone responding to gastric bypass surgery by only eating butter and chocolate truffles.

The way to handle this, in theory, is "risk-adjusted capital limits." In this telling, regulators go in, look at the stuff you're buying, and then set capital requirements that are responsive to the realities of your balance sheet. In other words, if you've decided to load up on truffles and butter, you won't get to have as much of them.

These ideas are not either/or. You could have gross capital requirements plus the option for regulators to impose harsher capital requirements in response to firm behavior. "There is no perfect solution," says David Moss, the John G. McLean Professor at Harvard Business School. “But lower leverage should definitely help. It’s like washing your hands. It may not totally prevent a cold. But it apparently reduces the odds quite a bit."

The financial regulation bill that passed the House of Representatives includes an amendment that Moss helped craft and that Rep. Jackie Speier sponsored that forces the financial sector to do a lot of hand-washing: It sets a 15:1 gross leverage limit and then gives regulators the ability to clamp down further if they so chose.

That's not the administration's preferred route. They'd like to see the capital requirements left entirely up to regulators. There's also an international conference of regulators called Basel III that's expected to set some new capital requirement rules.

That said, when something is left to the regulators to decide, it is also there for the regulators to un-decide. So many believe capital requirements should be written into law at a moment when the dangers of leverage are fresh in our minds rather than left to regulators who might loosen them when memories of the meltdown fade.

By Ezra Klein  |  April 19, 2010; 11:33 AM ET
Categories:  Explaining financial regulation , Financial Regulation  
Save & Share:  Send E-mail   Facebook   Twitter   Digg   Yahoo Buzz   StumbleUpon   Technorati   Google Buzz   Previous: Names you should know: Neil Wolin, Michael Barr, Diana Farrell
Next: Lunch Break


You missed the obvious comparison here- to the housing market and foreclosures! A 20% down mortgage is the same as 5 to 1 leverage, thus somebody who bought a house in the middle of 07 and saw the value decline by 20% had their entire investment wiped out. Of course many people put 5% down, which is 20 to 1 leverage or leveraged their home purchases even further.

Posted by: tmorgan2 | April 19, 2010 12:44 PM | Report abuse

You mention Basel III as though Basel I & II hadn't been around for years. Aren't there capital requirements now? based both on basel pillars and the fed's tier 1 capital reqs?

Moreover, there is still the question of what assets are subject to capital requirements. When a bank funds commercial paper vehicle (under ninety days, ultra low riks) is this going to be subject to the same capital requirement as an RMBS scratched together from re-financed ARMs in Fresno? If so, why? Part of leverage requirements also entails being able to keep everything on the balance sheet, which would require further reforms to the GAAP (unless you conceive of some other way of enforcing/conceptualizing leverage requirements). I imagine if you subject to everything to this 15:1 requirement, you'll just see more things moving off into the ether.

Posted by: dgs290 | April 19, 2010 12:56 PM | Report abuse

What about the fact that bank capital can't be reliably measured, particularly as bank balance sheets grow more complex?

Capital requirements are fine, but you need a second line of defense. Require contingent debt which converts to equity on top of those capital requirements, and you'll have a private bailout / market discipline built in as well.

Hattip to tmorgan2 - have maximum leverage thresholds on both homeowners and banks. If mortgages were 20% down, there would have been no housing bubble. It's hard for a house to go from $300,000 to $600,000 over a few years if the downpayment soars from $60,000 to $120,000. If the downpayment goes from $10,000 to $20,000, that's not as big of a barrier.

Posted by: justin84 | April 19, 2010 1:26 PM | Report abuse

"The way to handle this, in theory, is "risk-adjusted capital limits." In this telling, regulators go in, look at the stuff you're buying, and then set capital requirements that are responsive to the realities of your balance sheet. In other words, if you've decided to load up on truffles and butter, you won't get to have as much of them.

These ideas are not either/or. You could have gross capital requirements plus the option for regulators to impose harsher capital requirements in response to firm behavior."

Is the risk-adjusted ratio a "harsher" requirement? I've always thought of it as a softer requirement. Say a bank must a 5% tier 1 leverage ratio and a 10% total risk-based ratio, and say the bank has $5 in capital against $100 in assets (5%). The risk-based ratio basically says we're going to count a lot of these assets as zero, and count only 20% of these other assets, and so on, so now it looks like you got $5 in capital against $50 in risk-adjusted assets (10%). In effect, risk-adjusted ratios take your high leverage and, with the same capital, makes it seem like you're not really that leveraged. That works great, of course, until all that AAA-rated stuff that got a 20% risk-adjustment ends up being worthless.

So I've never thought of risk-adjusted ratios as requiring higher leverage because you have riskier assets. If you could explain your thinking and how I'm wrong, I'd greatly appreciate it.

Posted by: JamesCody | April 19, 2010 1:48 PM | Report abuse

When attention turns from regulation of banks to regulation in general, the numbers are equally important. A typically mortgaged home purchaser, with 20% down (5:1), is driven 25% of the way to insolvency by a 5% tax increase. I rarely hear the numbers discussed when it comes to taxes, though: for example, the 2% health penalty tax eats 12% of a mortgagee's liquidity. Or do taxes "not count"?

Posted by: rmgregory | April 19, 2010 2:12 PM | Report abuse

A small quibble with the description of leverage - everyone I knew in my working years used the debt-equity ratio, not assets-equity. So, in the homeowner example, leverage would be 4:1, not 5:1, and 5% equity would be leverage of 19:1. Like I say, a minor thing.

More substantively, risk weighting is absolutely vital and should be non-negotiable. I worked for a major auto finance company, and our internal risk assessment recognized that not only different classes of asset (e.g., retail financing versus leases versus dealer floorplan), but different levels of risk within classes require different amounts of equity. Levels of risk are defined not just by customer characteristics (e.g., FICO score), but also by the characteristics of the loan (e.g., down payment, payment-income ratio, fixed versus floating rate).

If you get the equity requirements right, it eliminates (or at least vastly reduces) the problem that occurs when riskier loans look more profitable than less risky ones. Give those zero-down, floating-rate, zero or negative amortization loans leverage requirements of no more than 2 or 3:1 and they'll disappear overnight.

Now, there remains the huge issue of how to define equity (I'd exclude any intangible assets related to securitizations, for one thing), but that's a topic for another post.

Posted by: donkensler | April 19, 2010 3:08 PM | Report abuse

Ezra ...

You're a pontificating little fool . Aside from that , I ocassionally find you're columns wildly funny in a dumb sort of way . However , you're parent's money would have been better spent had they sent you to a Trade School .

Posted by: lagnafrah | April 19, 2010 3:38 PM | Report abuse


The risk adjustment is necessary because if all assets are treated as equally risky, an obvious strategy to drive earnings is to load up on high yielding risky assets.

The difficulty, as you identified, is figuring out how risky everything is (in fact, if this could be easily done, we probably wouldn't need capital requirements as banks).

I think part of the solution is to be very picky on what sort of risk weights to apply to supposedly Aa or Aaa assets. I think your very low risk weights ought only to apply to large liquid sovereigns - USTs, German bunds, etc.

The other part is that the financial system needs to be robust to failure. We don't want to be sitting around again in 2014 being like "oops, we got the risk weights wrong. didn't have enough capital again." So we should require banks to partially fund with contigent debt, which would convert to equity if regulators determined a recap was necessary. The bondholders here would be very attentive to risk, and would require a high yield if a bank was looking shaky. Of course, market discipline doesn't always work, in which case the contigent debt allows for a privately funded bailout. Futhermore, to remain a going concern, once the debt converted to equity, the bank would have to go out and raise new contingent debt, or else it gets wound down by the regulators.

Posted by: justin84 | April 19, 2010 4:14 PM | Report abuse

Why do we continue to have this argument? What will it take to LEARN the hard lesson of what has failed in the past, why and accept it must be written into law in order to REALLY bring about change. Not just any change, but the change we NEED to never re-visit this mess again.

Posted by: moonglowsun | April 19, 2010 6:51 PM | Report abuse


Thank you for your response.

A question to you or anyone else: what does this bill do in terms of setting up a contingent debt scheme? My memory is that it basically sets up some group to study it, but doesn't actually create any contingent debt mechanism that would go into effect upon the bill passing.

Posted by: JamesCody | April 19, 2010 7:17 PM | Report abuse

..."Excellent article, truthful, "This is what Republicans hate to read "FEDERAL REGULATION OVER THEM STEALING EVERYTHING THEY CAN CARRY!


..."The Republicans can thank their "LUCKY STARS PRESIDENT BARACK DIDN"T PUT THEM IN JAIL!

..."The Record of the "Blew it Bush/Republican/Admin to this registered Voter/Vet USAF, is CRIMINAL/FACT!




Posted by: ztcb41 | April 19, 2010 7:34 PM | Report abuse

Great series!

One aspect that would be worth covering: What impact to SPEs and SPVs have on measuring leverage? How would a new system of leverage requirements account for off-balance sheet transactions?

Posted by: JPRS | April 19, 2010 8:51 PM | Report abuse

Actually there is a big difference between the mortgage version of leverage and the wall street version of leverage. If your house on which you paid 5% down declines in value by 20%, you're 15% underwater, but as long as you keep making the payments nothing much happens. On wall street, if it actually gets out that your 30:1 leveraged portfolio is down 10%, lots of your creditors will call in their loans, in the desperate hope that they will be the first carrion crow to reach the corpse. Hence the suspension of mark-to-market and so forth.

Posted by: paul314 | April 19, 2010 8:55 PM | Report abuse

Captial ratios are written into law. For national banks the statutes are commonly know as "Prompt Corrective Action"

Posted by: hitaxpayer | April 19, 2010 9:21 PM | Report abuse


I have only heard about the idea being floated around, from the likes of Greg Mankiw and Alan Greenspan. I instantly liked the idea as it doesn't depend on the wisdom of regulators (or even for that matter markets) - if all else fails, at least you have a private bail out built into the capital structure. I hope there is more than just a study group being created.

Anyone else know more about this?

Posted by: justin84 | April 19, 2010 10:24 PM | Report abuse

You have a lot to learn about banking, banking practices, risk and financial regulation.

I hope you sit down with a professor of banking or a regulator before you attempt to "educated" us further.

Posted by: onehanded | April 19, 2010 10:41 PM | Report abuse

But why should the Fed Govt determine capital requirements? The regulators determined them the last several years and before. Let the marketplace do it. I have far more confidence in bankers to mind their affairs effectively that for regulators who are subservient to the political authority. This ignores the elephant in the room,i.e., the Fed Reserve which has continuously "inflated" in recent times. This was mainly done through credit creation rather than monetary inflation. Same result, just sneakier.

Posted by: wpcharowhas | April 19, 2010 10:59 PM | Report abuse


Effectively, the private markets did determine cap requirements in 2004. Private investment banks said to Bush's SEC "we can handle much higher debt levels" and the Bush SEC obliged allowing investment banks to go from 15 to 1 ratios up to 30+ to 1 ratios.

The higher debt levels amplified the crisis, because the large banks didn't have sufficient reserves to cover losses when the market turned.

On the other hand, we had a period of relative financial stability over a few decades and fairly robust economic growth when there were tighter cap requirements.

Effectively, saying "the market should determine the correct level" is a little like saying we should let monopolies and oligopolies flourish. The natural condition of markets is towards consolidation. First you gain economies of scale, which are good, but eventually you gain a captive, or near-captive market because the barriers of entry become too high for new entrants (e.g. even though there is undoubtedly a market opportunity right now for an alternative to the big investment banks, no individual or group of individuals has sufficient capital to start a $500 bill. plus enterprise from scratch -- few nations have that capacity).

That's one argument for establishing a fixed rule at a more manageable level. Unlike individuals, big firms can grow so big, and so interconnected that their failure poses systemic risk to the entire financial system. That's one of the lessons of the very recent past. Letting the leverage ratios float to a higher level again is basically ignoring the lessons of the recent past.

Posted by: JPRS | April 20, 2010 12:15 AM | Report abuse

Isn't this post a little patronizing? It is possible that your readers didn't know what a derivative was in 2008 or even 2009. But let's get on with it.

Posted by: chase-truth | April 20, 2010 8:13 AM | Report abuse

Klein, you can start by explaining why this legislation DOES NOT COVER Fannie Mae and Freddie, AIG or the five largest banks. Then, you can try to showcase your bit of intelligence with numbers, as if anyone cares.

Posted by: RonKH | April 21, 2010 2:05 AM | Report abuse

Klein, you can start by explaining why this legislation DOES NOT COVER Fannie Mae and Freddie, AIG or the five largest banks. Then, you can try to showcase your bit of intelligence with numbers, as if anyone cares.

Posted by: RonKH | April 21, 2010 2:05 AM | Report abuse

Paul314 hits the nail on the head. The real culprit here is the insane demand for liquidity. Liquidity turned out to be a myth (it ain't liquid unless it can be sold at full value at ANY time regardless of circumstances). Bankers haven't demanded that homeowners "balance" their loans because the loans are primarily to individuals who use their homes as backup collateral. (At least that was the way it used to be!) If the Wall Street boys hadn't been concerned about converting their investments/loans/paper into cash immediately, Lehman and Bear Stearns would still be around. And if the Fed and the FASB wouldn't require companies to mark to market items companies deem to be investments rather than items of inventory, there wouldn't be nearly the problem either.

At the end of the day, no matter what you legislate or regulate, you can never fully protect against a run on the bank, and a run on any sizable bank or financial institution will result on a run on the system. The Fed and the Treasury EXIST for this reason if no other. They will ALWAYS be the lender of last resort. Who was it that saved George Bailey? All his friends and neighbors, the public. Also known, in some circles, as the Fed!

I really don't understand the problem. The Fed and the government did their jobs, the system was "righted", the public made billions on their bailout funds and everything is getting back to normal. What's there to correct? Everything worked!

Posted by: patPatterson1 | April 21, 2010 2:16 AM | Report abuse


Interesting. I remember someone mentioning Keynesian economics in a post, and at that time it was not something I understood in the least way.

I took the time to educate myself a little, and it seems that you may be correct.

Given my limited understanding, it would seem that under the current prevailing Keynesian economic theory... The government, its treasury, and its bank, are there to bail out the economy during downturns/recessions in order to keep the market stable and unemployment at a somewhat "natural" level while not deflating the economy.

The main problem, as I'm seeing with these other posts, is the fact that the market was allowed to leverage itself into an untenable position.

Logically, I see the merit in JRPS's argument. Where smaller organs of the market could be allowed to fail, these very large corporations pose a threat to the market in such a fashion that if they are allowed to fail, they systemically affect the market and thus must be kept solvent in order to prevent a much larger crash.

Thus it could be argued that caps for leverage must be put into law and that the regulators continue to enforce said cap as well as other measures to prevent too-big-to-fail operations from existing.

Our U.S. government has enough debt without having to bail out large companies from bad decisions. Our social programs soak up much of our revenue, and they also need to be reformed lest we suffer a bankruptcy in the public sector, our Government.

Posted by: OfConservativeMind | April 21, 2010 11:56 AM | Report abuse

OfConservativeMind -

I think the suggestions of what I consider appropriate government intervention in the free markets, namely providing monetary incentives and penalties through tax policy, should be instituted to try to "persuade" banks/financial institutions to reduce the risk levels of their portfolios, taxing them at higher or lower rates depending upon the risk measurement of their portfolios. The devil is in the details of how you do that, but who better than the Fed, supposedly apolitical and independent? But at the end of the day, there is no way we will ever be able to prevent a run on a bank, and there is no way that institutions can be small enough so that what happens to them won't impact almost all of their competitors. I think we just need to suck it up, acknowledge that the Fed and the Treasury are there to be lenders of last resort (thus "earning" the tax dollars they collect by creating a stable environment in which capitalists can operate). Then we can make a case for the legitimacy of regulation of such things as risk taking.

Posted by: patPatterson1 | April 21, 2010 12:29 PM | Report abuse

The comments to this entry are closed.

RSS Feed
Subscribe to The Post

© 2010 The Washington Post Company