Wednesday, October 28, 2009

Thursday, October 8, 2009

Total Bollocks from Alex Pollock

The scholars up the street at American Enterprise Institute are at it again. Now they're blaming the crisis on deposit insurance, specifically the $384 billion in fully-insured brokered deposits in the banking system. In total, just over 1% of the nation's total mortgage lending at the height of the bubble may have been funded by fully-insured brokered deposits. While this may account for some growth on the margin, it hardly seems to make deposit insurance the most likely cause of the crisis.

Mr. Pollock conveniently forgets that according to the Federal Reserve, commercial banks and savings institutions held only 30% of non-revolving consumer credit outstanding at the close end of 2006, on the eve of the crisis. Let's assume that these institutions held the same percentage of outstanding mortgage debt. Banks with $10-100B in assets were the most egregious users of brokered deposits. At these regionals roughly 15% of their total deposits were brokereds, however only 10% of their domestic deposits were fully-insured brokered deposits. We can estimate that these institutions held 7% of domestic real estate loans outstanding at the end of 2006. As for their funding mix, 24% of their funding came from non-depository sources such as the capital markets which means brokered deposits made up 11% of their total funding mix, and just over 7% from full-insured brokered deposits.

I don't disagree that some banks which have subsequently failed relied too heavily on brokered deposits to grow but it seems Mr. Pollock has spotted an idiosyncratic cause of bank failures and presented it as systemic. Much more sophisticated investors, who conducted the sort of due diligence on the bank that the average consumer is relieved of by deposit insurance, provided more than twice the funding of brokered deposits. As for AEI's previous scapegoat du jour, the GSEs held only 6% of outstanding non-revolving consumer credit at the end of 2006, compared to the combined 46% held by unregulated finance companies and in securitized pools. I wouldn't be surprised if their share of outstanding mortgage credit was significantly higher. Given the perverse incentives present in the originate-to-distribute business model, which is the most likely source of funding to fuel the real estate bubble?

Tuesday, March 3, 2009

"It works in practice, but does it work in theory?"

Willem Buiter provides an explanation of the joke about economists:
What this shows, not for the first time, is that models of the economy that incorporate the [Efficient Markets Hypothesis] - and this includes the complete markets core of the New Classical and New Keynesian macroeconomics - are not models of decentralised market economies, but models of a centrally planned economy.

Friday, February 13, 2009

Flawed Assumption Underlying Fair Value

The Insitutional Risk Analyst identifies the flawed assumption underlying the notion of fair value.
First, FVA relies on efficient market theory, namely that short term price = value and that consequently income, assets and liabilities should be adjusted in real time to reflect same.... We think that the collapse of all of the other market efficiency based constructs, from structured assets to hedge funds, ends the discussion of derivative notions such as FVA.

Second, FVA fails to recognize the historical role of depositories, pensions and insurance companies as repositories for long-term value and consequently as havens from swings in short-term market pricing and economic trends.
While Geithner's announcement didn't include any explicit plans for a further relaxation of mark-to-market accounting, a read between the lines suggest that's the direction he's heading. In my mind, this paragraph from the fact sheet is the tell:
The Treasury Department will work with bank supervisors and the Securities and Exchange Commission and accounting standard setters in their efforts to improve public disclosure by banks. This effort will include measures to improve the disclosure of the exposures on bank balance sheets. In conducting these exercises, supervisors recognize the need not to adopt an overly conservative posture or take steps that could inappropriately constrain lending.
Call it "Increased Transparency and Disclosure" while signalling a willingness to give banks some breathing space. Orwellian, isn't it?

On Untread Ground

I think this sums it up quite nicely. More cartoons, courtesy of USA Today, here.

Thursday, February 12, 2009

Is Geithner's Plan Really a Disaster?

A lot of black pixels have been splashed across our computer screens bemoaning the inadequacy of Geithner's bank bailout plan as he presented two days ago. I think the problem is mostly one of public relations, not substance.

First, what Wall Street was actually looking for was a plan that allowed major players to unburden themselves of all their toxic exposures at taxpayer expense. Luckily, they didn't get this from the Obama administration. Geithner's plan is a subtle evolution of TARP rather than a radical revision. I suspect the market would've reacted badly no matter what, and as the White House Press Secretary Robert Gibbs explained, "yesterday's speech and framework were not designed for a one-day market reaction." Can you imagine the markets reaction had Geithner invoked the N-word? In my mind, a good plan should really disappoint almost everyone: "Mission Accomplished".

Let's look at where Geithner is going. At its simplest, a bank basically has three legs: capital, liquidity and assets. The Treasury Capital Purchase Program is strengthening the capital base of individual institutions when neccessary and liquidity is ensured by the expansion of deposit insurance and the alphabet soup of debt guarantees and Fed lending programs. The problem that Paulson's original Tarp proposal foundered on how to deal with toxic assets and this remains the most difficult to address. Geithner's approach is somewhat more nuanced and while only subtly different is much more promising.

There's a chicken-or-egg problem here which prevents the start of any resolution of bad assets. Banks are holding hard-to-value assets at prices above what they would fetch on the market. The lack of investors is something of a myth - there are vulchers willing to buy these securities at the right price and these securities do trade. The reality is that there are no investors willing to pay the prices banks must sell them for to stay in business. Accounting rules mean that selling any portion of these assets mean similar assets should be carried at the sale price. With any sale, banks would have to declare all their losses at once, shred their capital base and put themselves out of business, so they are unable to sell. Their only option is to unload assets at a carrying price only loosely associated with reality. With no private investors is willing to buy at the carrying price the banks can only hope the government will take them off their hands. By temporarily softening mark-to-market and allowing banks to pay for losses out of earnings over time, Geithner's plan could potentially start getting some of the bad assets out of banks without destroying their entire capital cushion with mark-to-market losses.

Ideally the purchase of toxic assets would best be left to private participants willing to take on a high level of risk. However the scale and urgency of the problem necessitates government intervention. For such a plan to be fair to the taxpayer and politically viable, government must not overpay for such assets and thereby preserve the potential for upside. Creating a hybrid public-private vehicle to make these purchases mean the government will not pay more for toxic assets than its private partners are willing. A clear line in the sand may jump-start asset sales when banks realize they are waiting in vain for the government to pay inflated prices. Pushing a ward of the state like Citi to sell to the vehicle while simultaneously easing mark-to-market accounting rules will encourage price discovery without making it poisonous.

The primary problem with suspending mark-to-market is that you have to trust banks to eventually value all such assets correctly, rather than paying out bonuses and dividends as if the problem didn't exist. This is clearly the reasoning behind restrictions on executive pay and dividends for institutions receiving extraordinary assistance. Such restrictions ensure that it is primarily shareholders and management who pay for the losses over time. This is also the reason behind the "stress test". The "stress test" has been universally panned as just another bank exam but the true intention is to create an inventory of bad assets, assess the true scale of potential losses and maintain a "real" balance sheet in parallel to the public balance sheet.

This plan is really one where we see if the banks can bail themselves out given a little time and minimal support. The fed ensures access to low cost liquidity through ZIRP and allows banks to earn larger spreads. Those $100B institutions which require 2 years or more of earnings to cover potential losses will probably be nationalized. Obama is right that magnitude is a problem. It's important not to forget we're talking about $6 trillion assets in the 4 largest BHCs now. Unless you'd like the financial system to collapse and start from scratch, nationalization means taking both these assets and $5.7 worth of liabilities. This is what the FDIC does when resolving a smaller bank failure. That's a lot of risk on the nation's balance sheet, and if you think the system is insolvent - ie. liabilities > assets - nationalization charges losses in excess of capital right to the taxpayer.

Anticipating an outcry about addressing the foreclosure crisis and stabilizing house prices, I will point to a legislative solution moving through Congress: Mortgage Cram-downs. The issue has been thoroughly addressed over at and the solution has a lot going for it including it imposes a fairly severe cost on consumers, reducing moral hazard. The Conyers bill is imperfect but it affords homeowners with severely underwater mortgages a chance to force lenders to share in the loss of collateral value and remain in their homes. The bill also preserves some upside for lenders should property values revert to their recent aberration from historical trends.

Wednesday, February 11, 2009

Too Busy to Blog

Sorry about the lack of updates.

So much going on that's worthy of comment, but it means I'm too busy to find a spare moment for blogging.