What's the Best Way to Measure Systemic Risk?

Now that the Obama Administration has released its proposed Financial Regulatory Reform: A New Foundation for updating the regulatory structure of the financial system (pdf file of reform, WSJ article), the public debate will begin in earnest regarding the details and proposals in the document. Ironically, while the public seems open to new financial regulation, the release of the document is coming at a time when some citizens are starting to worry about growing deficits and government intervention (see WSJ/NBC poll article), with almost seven in 10 people surveyed saying that they had concerns about federal interventions into the economy. Nonetheless, finding ways to limit risk and prevent another credit crisis through added regulation of banks and hedge funds still rings a populist tone, and is likely to continue to receive support.

Like many others, I feel that there are aspects of the new regulations that seem appropriate and make sense, while there are others that seem counterproductive. As for those that concern me, I agree with some who point out that it seems odd that the Fed is going to be given greater power (and responsibility) to fix some of the very problems it may have caused or contributed to (see Larry Kudlow article). I also worry that the new proposed Consumer Financial Protection Agency could end up just adding new and potentially unnecessary regulations and red-tape paperwork, along with producing counterproductive limits on rates and fees that will do less to protect individual consumers and more to reduce the availability of needed products that are offered to such individuals. Such restrictions, if not properly crafted, are likely to reduce the earnings of financial services companies, and even worse, limit their ability to effectively manage risk (through fee and rate changes).

The idea of requiring companies to retain 5% of all structured product offerings also has me concerned, even though this specific proposal seems to be generating some of the most support, at least initially. Recently I wrote that I have worries about forcing companies to retain a stake in each securitized product they develop since I believe it could actually make companies more risky (since they cannot off-load and manage all their risk, see previous post). Furthermore, while I agree that forcing companies to have some "skin-in-the-game" would make it less likely that they would offer risky products, it also makes it more likely, in my opinion, that they would offer less structured products. While this may be a desired outcome for some, the impact of this would be less liquidity and available credit at a time when the country can least afford it.

Another area that is generating support involves the idea of controlling systemic risk. I too believe that this is a good idea in theory, but am unclear exactly how this would be measured and acted upon. As recently reported in the WSJ (see article), one potential area to start with is leverage. As the chart below illustrates, financial sector borrowing increased steadily between 2004-2007, before dropping in 2008 as companies began unwinding leveraged positions. It is now well known that banks, hedge funds, and average citizens were carrying too much debt, thereby helping to increase systemic risk, and trigger the credit crunch.

Source: WSJ, Fed data

Yale economist John Geanakoplos, who has studied leverage in the economy, believes that regulators need to gather daily data from all market participants and then publish aggregate data. The feeling is that if market participants knew that leverage values were getting to extreme levels, they would be more likely to begin backing-off their own debt and leverage levels in anticipation of eventual market corrections, or restrictions imposed by regulators. Focusing at least in part on debt and leverage seems to make sense.

Unfortunately, measuring system-wide daily leverage changes at banks and hedge funds may be difficult at best, and suspect at worst. Yet, maybe the focus does not need to be on the three person hedge funds that are investing $10-50 million in capital. Getting comprehensive data which includes smaller players may not be necessary. Risk manager Richard Bookstaber - whose book "A Demon of Our Own Design" is recommended reading - believes that focusing on just the largest financial firms and hedge funds would cover roughly 80% of the risk, allowing you to see systemic trends.

At this point it is unclear if such trending information on leverage levels would be enough. Other related areas that could also cause potential cascading effects, such as specific derivative use (i.e., CDS) and counter-party risk, should also be examine - although regulating that which has not yet been developed is obviously difficult. Nonetheless, looking for new ways to measure leverage might be a good place to start for spotting worrisome trends. Such a signal could allow investors, funds, and the Fed (or other regulator) to take action. Of course, what action they take is another area of debate, and potential new area of concern.

Comments

[max12345:]

Your headline question (What's the best way to measure systemic risk?) is quite a good question, I think. At this stage the Obama administration has made some "proposals". These are not final and will only be "finalized" (and they won't be "final" even then) after going through both the House and the Senate.

Both of which will be sure to ask your own question as well as several other much more specific ones. Although "systemic risk" has not been managed properly recently (or maybe ever) "The System" does in fact work!

That is, within about a year or two of a major set of problems arising, our Democratic System has already come up with some proposals to address the problem(s).

And these proposals will now be debated and fine tuned by gathering up all kinds of opinions for and against them -and their specifics-, in order to improve them.

After that they will become laws that will provide for new (or modified) authorities, institutions, institutional configurations, institutional mechanisms, and policies. (and maybe also improved "institutional cultures" and "accountabilities")

The impact of all these things will then be assessed and probably (or at least possibly) some further fine tuning will take place. We will then "finally" have in place a "reasonably effective" new regulatory, legal and institutional framework designed to also deal with "systemic risk"

But at this stage we need to be patient and let The System work.

And of course it's not impossible that in another 20 years such a revamped framework of institutions, policies and laws also may have become obsolete or not up to some of its newly emerging tasks. (just like the last one that we had, had done)

Even though it's been a bumpy road, we avoided a Great Depression and we seem to be haltingly coming out of recession.

Overall this is a more than acceptable performance for "The System", even if fraught with "Systemic Risk"

That is, the Overall System has proven itself to be broader (since it is not only economic and financial but political and institutional as well) and definitely more capacious than "mere" "Systemic Risk"

I remain more than optimistic on "The U.S. System"

Jun 19 11:33 AM

[Spartacuss:]

And just who in the hell is supposed to pay for these new regulatory agencies? We see an enormous growth in the size of government, again as usual. Hell, why don't we all just quit our jobs and go and get a government job?

Don't you guys ever get it? We are headed down the yellow brick road to fascism: total government control over everything. The new future for employmnet means funded by government! All the large corporations will become GSE's now to be controlled and operated by the FED? New Czars appointed to lead this fiasco down the yellow brick road and we know where that leads to: The great Wizard who knows all and controls all, and I have a suspicion I know who that is, his name starts with an O.

America is being sold down the river and it seems to me that the press is complicit as there are a number of taboo subjects such as the US is bankrupt and will never be able to pay it's bills. How about the off balance sheet rot in the big banks? How about the baby boomer problem? How about the resets of Alt-A and Option ARM's and the Commercial paper. How about government taking over industry and the 'too big to fail' claptrap.

No, of course you guys mustn't report on these subjects, why they are just too negative and pessimistic and might precipitate a run on the markets and even a run on the dollar. Can't have that and besides we cannot tarnish the brass halo around Obama's head.

Jun 19 04:01 PM

[nym:]

We know now that we should have been watching the ratio (home sale price)/(annual household income), which FHA set at 3/1 for their mortgages, but rose to 5/1 in many markets since 1998. The rise from nowhere of home equity borrowing should have rung bells. Also, chronic trade deficits were there for all to see. This is not rocket science.

Jun 20 12:44 AM

[Gregman2:]

The best way? Quality of Earnings, which is a function of income relative to debt. When earnings start to slip, chance of default go up.
Then, examine why earnings are slipping: cancelled orders? higher supply chain costs? etc. For the financial (lending) sector, its customers are businesses that borrow for inventories, new projects, etc.

Jun 20 07:08 PM