By JJ Hornblass and Boaz Salik
Of all the massive corporate failures that precipitated our current financial crisis, AIG’s has stood out because it endangers the overall economy with such a large quantity of distressed assets. Its risks are so acute that the federal government has felt obligated to bail it out not once, twice or three times, but four. The $173 billion already pumped into AIG will likely not be the last of the government’s rescue funds.
The gravity of the AIG failure shows how institutions that are “too big to fail” are bad for our economy. This week, Ben Bernanke, the chairman of the Federal Reserve Board, Rep. Barney Frank (D-Mass.), chairman of the House Financial Services Committee, and others have proposed that a regulatory body be created to monitor – and presumably regulate – financial institutions that pose a “systemic risk,” with AIG cited as a prime example of such an institution. This proposal is laden with three distinct shortcomings: a) it lacks a definition for “systemic risk”; b) it offers no solution for minimizing such risk; and c) it adds yet another government regulator to the cadre of less-then-successful bureaucrats charged with maintaining order in the nation’s financial services industry.
We have a solution. After long research, we offer a regulatory framework that defines systemic risk and offers a simple methodology for monitoring it without the need for yet another government regulatory body.
But first, some background. The risks still present in our economy are akin to the risks of the trusts of J.P. Morgan and others at the turn of the 20th Century. At that time, President Theodore Roosevelt determined that too much power over the economy was controlled by too few. Back then the “power” came in the form of monopolies, so Roosevelt created the Sherman Anti-Trust Act, which dissolved many monopolies and limited the influence of others.
Today, an overabundance of “power” over the direction of our economy comes from disproportionate concentrations of risk — and a prime example is AIG. AIG’s $377 billion of credit-default swaps (CDS) at the time of the government’s intervention created grave economic stress in the U.S. and globally. Had AIG gone bankrupt, those swaps — which are like insurance policies against the failure of other assets — would have gone with it, and the potential spillover effect could have crippled the economy. Or at least so the government seemed to believe; it intervened before that could happen.
But intervention is, or at least should be, the action of last resort. Similar to the Sherman Anti-Trust Act, there needs to be a sort of “Sherman Anti-Systemic Risk Act” that creates a regulatory framework for preventing such extreme concentrations of risk, and avoids weekend financial fire drills by Treasury Department and Federal Reserve officials. Like the Anti-Trust Act, an “Anti-Systemic Risk Act” would dismantle, or at least limit, risks when they become too concentrated.
Creating such an “Anti-Systemic Risk Act” is a tricky thing, however. How do you decide which firm has too much risk? The Sherman Anti-Trust Act works because tests exist to determine whether an enterprise is “bad” for the economy. Such a test led to the 1984 breakup of AT&T, for example. Any potential Anti-Systemic Risk Act needs a test, too.
The Secret Sauce
The secret rests in the concept of “disorderly liquidation.” There’s a certain point at which the sheer volume of assets being sold into a market precludes liquidation in a normative manner. The Long-Term Capital Management situation in the late 1990s was a prime example of this. LTCM held a massive volume of assets that had greatly depreciated in value after the Russian ruble devalued unexpectedly. Normally, LTCM’s financial loss would be the problem of LTCM and its investors. The only snag was that the federal government and Wall Street faced a quandary, too. If all those devalued assets were dumped onto the market, other assets would devalue greatly as well – and that would have precipitated the collapse of several major financial institutions. The good would get dragged down with the bad. (Sound familiar?)
Research we have conducted shows that an entity that holds more than five times the amount of shares or assets traded daily may lead to a disorderly liquidation should it need to offload its holdings quickly. We inferred this by looking at groups of publicly traded equities, which showed price volatility often increased substantially when the ratio of total shares outstanding to average daily trading volume grew too large. One likely reason is that the liquidations of those large positions often ended up collapsing the price of the public shares by creating a supply-demand imbalance.
This phenomenon occurred during the dot-com bust just a few years ago. Many executives and venture-capital investors held vast quantities of dot-com shares, to the point where their holdings dwarfed the volume of publicly traded shares. These dot-com companies had lock-up periods during which the holders of privately held shares could not sell their stakes. When those periods ended – and they always ended at some point – these private shares flooded the market. This “disorderly liquidation” caused many dot-com share prices to eventually plummet – and then a sock-puppet became a legend.
There’s an additional wrinkle to this. The mark-to-market accounting required by public companies amplifies the risks associated with liquidation. As market prices for an asset climb, mark-to-market accounting lets an investor in that asset recognize greater paper profits. However, such profits may quickly disappear, or even turn into huge losses, when the owner tries to liquidate the assets. Unfortunately, this change in fortune may not happen for months or even years, depending on when management decides to liquidate (a serious moral hazard in itself). Or, as in AIG’s case, it could happen at the worst possible time, when other investors decide to liquidate their own holdings.
A Sherman ‘Anti-Systemic Risk’ Act
So beyond the right to pursue a Sherman-like breakup, the buildup of such holdings can be discouraged by requiring public companies to discount mark-to-market valuations when their holdings of a given asset class exceed the average daily trading volume by too much. For example, a company might be required to take a 50% discount to market prices for asset levels exceeding five times the trading volume, and a 75% discount for asset levels exceeding 10 times the trading volume. Clearly this would create a massive disincentive for public companies to accrue too much of a certain asset class—but then that is the desired effect.
This would have worked at AIG, for example. AIG’s balance sheet at the end of 2005 showed that it owned $387 billion of CDS risk, known as net notional exposure. In that year, the International Swaps and Derivatives Association estimated that $20 billion to $30 billion of CDS indices were traded daily. Under the solution we propose, AIG would have been prevented from accumulating CDS exposure beyond $150 billion to $200 billion, or recognize a liability on its balance sheet of $150 billon to $200 billion — possibly sidestepping their eventual government bailout completely.
What is best for economies is that they exhibit healthy asset distribution with risk diffused accordingly – which is exactly why we are bewildered by the federal government’s actions since the credit crisis intensified last September. Rather than scramble to pair troubled firms with another company, what the federal government should have done (and the Obama administration could do still) is limit the risks to the economy by diffusing them. Don’t mash Merrill Lynch with Bank of America, for example; break Merrill up so that its composite risk is dispersed across the economy and investors.
To resolve the current predicament, we need a concrete, workable framework for minimizing a buildup in risks that endanger our markets and our economy. By limiting concentrations of asset ownership and preventing “disorderly liquidations,” the economy will be better protected and markets will function with greater fluidity.
Mr. Hornblass is the Executive Editor and Publisher of BankInnovation.net, a blog and social network for the banking industry. Dr. Salik is the Managing Principal of FischerJordan, a financial services consultancy.
I disagree with the logic that you guys are using but at the end of the day your result is valid. AIG CDS exposure decreases from roughly $400 billion to $200 billion. Assuming that AIG capital is unchanged, its leverage decreases by half (exposure to capital), which is really what you want. What I would suggest is an FDIC-like program for insurance companies that sell insurance (credit default swaps) to federally regulated financial institutions. The FDIC requires banks to maintain a capital balance sufficient to cover roughly 99% or more of the possible risk scenarios out there. The FDIC is in second loss position and it covers the remaining 1% of really bad scenarios that blow through the bank’s capital and leave the FDIC at risk. The problem with insurance companies such as AIG is that the rating agencies do this risk analysis. Now we see the problem. If it turns out that capital is insufficient at the rated institution then the rating agency says “My bad!” and walks away. It is not at risk so its rating can be bought for a price. If things go well I collect my fee. If things go bad I walk away. If the FDIC stepped in and replaced the rating agency then we can expect that the now federally insured institution will maintain sufficient capital to cover its risk because the FDIC is at risk and will cover losses that exceed the insured institutions capital balance. The result of this will be less leverage, which is the result that you guys are going for.
Does the FDIC really need to step in? It does because as we have seen, the federal government does in effect insure these institutions (i.e. when the insurance company capital is insufficient to cover losses then the government steps in to dump capital into the company). If the government “substance over form” does insure these companies then the government should regulate them in a manner very similar to the way the government regulates banks.
My two cents.
Actually, I love your entire premise! And wasn’t it cute the way they called the CDS products “swaps” instead of the insurance products they actually were. Insurance products falls under much more stringent regulation! But what’s in a name?
And where is Elliott Spitzer when we need him?
Credit default swaps are pretty slick. In one of my past lives I was a quant at a CDO shop and it got so nuts at the end of the housing boom that you couldn’t buy enough “cash” bonds to fill up a CDO so you had to do it synthetically. You would do this by investing at Libor (risk free) and then selling insurance (short a CDS) such that the Libor “risk-free” rate plus the the premiums you receive for selling insurance = the yield on a risky bond. You in effect create the risky bond (because you can’t buy them with cash) by investing in Libor and selling insurance. Pretty neat stuff but boy did it ever blow up!
While I typed my comment before I read Monica’s comment, my comment still applies.