Citi risk measurement scheme warrants closer look

November 8, 2011

By Jeffrey Goldfarb
The author is a Reuters Breakingviews columnist. The opinions expressed are his own.

Walter Wriston, the vaunted former chairman of Citibank, famously said risk is a four-letter word. Vikram Pandit, chief executive of the larger successor organization, Citigroup, reckons it has nine letters: benchmark.

Pandit is championing an idea to make it easier to compare the way banks assess risk. To accomplish this, he wants to start with a standard, hypothetical portfolio of assets agreed by regulators the world over. Each financial institution would run this benchmark collection through its risk models and spit out four numbers: loan loss reserve requirements, value-at-risk, stress test results and the tally of risk-weighted assets. The findings would be made public.

Then, Pandit wants the same financial firms to run the same measures against their own balance sheets – and to publish those results, too. That way, not only can investors and regulators compare similar risk outputs across institutions for their actual portfolios, but the numbers for the benchmark portfolio would allow them to see how aggressively different firms test for risks.

Such a scheme would be a giant leap from current practice. Broad value-at-risk figures have been largely discredited. And while in theory there is some standardization dictated by Basel capital accords, such as the zero weighting ascribed to sovereign debt, in practice banks run their own numbers using different assumptions.

There are also gaps between European and American banks – for instance in the calculation of risk-weighted assets – that make comparison unreliable. Few consider Bank of America’s Tier 1 ratio of capital to risk-weighted assets an apples-to-apples comparison with, say, Santander’s.

Regulators have tried to square the circle with stress tests. The U.S. effort in 2009 helped draw a line under the worst of the financial panic, though details of the results were largely kept under wraps and therefore provided little extra help for investors. Later European exams proved insufficiently stressful.

An experiment conducted in Britain two years ago lends some credence to the value of Pandit’s plan. The Financial Services Authority asked banks to estimate the likelihood of default on a hypothetical portfolio of corporate, bank and sovereign exposures. Though the median expectation was in line with the long-run Standard & Poor’s default rate, there were big variances in the amount of capital banks said they would need. The Bank of England’s head of financial stability, Andrew Haldane, said the inconsistencies were enough to be the difference between life and death for a bank.
Similar insight would come from Pandit’s expanded benchmark portfolio. If a bank’s analysis showed losses in a stress scenario that were less than those of all its peers, for instance, it would indicate a tendency to underestimate risks on its balance sheet. Regulators could decide to scrutinize the bank’s risk systems while investors might discount the firm’s stock until it became more conservative about risk.

The idea isn’t, of course, without complexities. Choosing the collection of assets in the benchmark portfolio would be controversial, even if they numbered as many as 1,000. And a big range of assets and careful weighting would be necessary so that the benchmark would be as meaningful for, say, BNP Paribas as it was for Mitsubishi UFJ. Most bank bosses will also resist revealing as much as Pandit wants. And getting a raft of regulators to agree on all this extra transparency and what would amount to a big shift of their oversight role to the market wouldn’t be an easy sell, either.

Even so, the plan warrants attention. The blowup at MF Global helps show why. For one thing, Pandit’s plan would extend to the so-called shadow banking system, where regulation tends to be less hands-on. The now-bankrupt brokerage’s relatively low level of concern about European sovereign debt might have shown up in the benchmarking exercise and raised red flags. Moreover, Pandit envisions quarterly disclosures of the modeling, which potentially would have hoisted those flags early enough to ward off investors or allow MF Global to save itself.

Pandit must think Citi has a distinct edge where risk is concerned to be advancing the idea. Yet achieving worldwide consensus looks a distant dream. Even getting U.S. financial institutions to agree would be tough – and might not be sufficient to get the plan off the ground. At least Pandit is trying to convince the financial industry that risk really isn’t a four-letter word.


This is never going to happen, but Pandit could at least publish Citi’s investments which might go a ways towards informing potential investors of the wisdom of investing in Citi. Frankly I would never invest in a bank.

Posted by MaryWilbur | Report as abusive

Post Your Comment

We welcome comments that advance the story through relevant opinion, anecdotes, links and data. If you see a comment that you believe is irrelevant or inappropriate, you can flag it to our editors by using the report abuse links. Views expressed in the comments do not represent those of Reuters. For more information on our comment policy, see