Is it Time for a National Institute of Finance?

Posted by: Theo Francis on August 18

For all that reporters, politicians and financial gurus like to talk about “the market,” no one has a clue it looks like in all its glory — a comprehensive overview of stocks, bonds, credit-default swaps and commodities.

Even if today’s regulators or financial firms were to suddenly get a bird’s eye view, even just a snapshot in time, they’d have no idea what to do with it. The tools to analyze that breadth and depth of data just don’t exist.

And if that’s the case, “does anybody have the data in place to really deal with systemic risk?” asks John Liechty, an associate professor of marketing and statistics at Penn State University.

That’s a problem — in fact, it’s a problem at the heart of plans for putting a systemic risk regulator in charge of ensuring that a cascade of smaller failures don’t bring down the entire financial system.

And so the Committee to Establish the National Institute of Finance was born.

The group -- Liechty is among its founders -- wants to establish the institute to gather, standardize and analyze market data for the government. It would serve as a kind of National Weather Service for everyone from the Federal Reserve and Treasury to the FDIC and the SEC -- and whichever new or existing body gets the role of systemic-risk regulator. Backers include dozens of academics, analysts, former regulators and current and former market participants. Supporters say some of the banking and financial regulators have been receptive; some key lawmakers have been cooler, waiting for the regulators' reaction. A spokesman for the Treasury says he isn't aware of the effort.

The arguments are straightforward: Without a vast array of data on derivative contracts, trades, aggregate positions, leverage and more, regulators have little or no hope of understanding where systemic risks lie.

Policymakers expect to endow the new systemic-risk regulator with the ability to gather that data, but without standardized reporting from industry and a dedicated group of researchers to crunch the data, it will be difficult, if not impossible, to develop the tools necessary to make sense of the flood of information. With such an institute in place, regulators last fall could have modeled the repercussions of Lehman's collapse, using actual portfolio data from financial institutions.

"There's nobody who holds the bag on research in finance," Liechty says. Existing regulators have too narrow a view, academics have trouble getting funding and comprehensive data, and industry is too competitive -- and too centered on the bonus cycle to allow the kind of years-long effort required to get such an institution running full-throttle. "They can't put a Bell Labs together," Liechty says. "There are problems that take longer than eight months to figure out."

The group is firm on a few points, less doctrinaire on others: Any institute must be funded by fees on the financial industry, to avoid dependence on the appropriations process; it should be a government agency, not an industry organization; it should be outside the usual government pay scales, to attract top talents. It would work best as an independent entity, a service bureau for regulators. But, Liechty adds, "to some degree, I don't really care where it fits, as long as we actually get it somewhere."

Will financial institutions be willing to give up the mystery and opacity that surrounds many corners of the financial markets? Yes, says one executive at a large investment bank. For one thing, only the government would see all the data. But or another, it stands to save firms money in very concrete ways by standardizing a host of vital data.

Currently, different firms track transaction terms, deal descriptions, even corporate and subsidiary names differently, causing considerable confusion, the executive says. Systematic reporting to an Institute of Finance could standardize the financial industry in much the same way Medicare's reporting requirements in the 1990s standardized medical billing.

That could help solve a problem that costs the industry hundreds of millions of dollars a year: derivative and swap transactions that are DK'd -- "don't know'd," or unwound -- because discrepancies in how they were described or recorded leave it unclear just what was agreed to in the first place; one side thinks a transaction centered on a subsidiary, for example, while the other thinks it involved the parent corporation. Big institutions "have an entire staff that does nothing but fix these things," the banker says.

"Once you get the standardization of data structures, you'll significantly reduce operational risk -- the risk that the transaction breaks," the banker says. "These things cost a lot of money."