Risk-based capital issues are emerging from the shadowy world of asset- liability technocrats and entering the forefront of financial industry policymaking-or at least the frontal lobes of policymakers.

Capital is, after all, the cornerstone of every financial institution. How much is held against which products is the beginning of the complex calculation of which line of business is more profitable and whose company will post the highest returns.

It is not an accident that, after implementation of the Basel Accord in 1988, Japanese banks began their long slide from the top. U.S. banks, under heavy pressure during the late 1980s and early 1990s, quickly discovered that the Basel risk weightings dictated bottom-line decisions.

Much of the wrangling in the financial modernization legislation is over capital. Whether an activity is assigned to an operating subsidiary or a holding company affiliate is an important issue principally because of the different capital treatment these options may allow. In short, capital has mattered a lot, and it is likely to matter still more as pending initiatives advance.

There are two new approaches to risk-based capital in the works. One is being prepared at the Committee on Bank Supervision in Basel, the same body that introduced the first version of what are universally dubbed the Basel standards. The second initiative is strictly a U.S. affair, focused on the two huge government-sponsored enterprises in the U.S. mortgage market.

Both approaches point not only to the benefits of a coherent and disciplined approach to risk-based capital but also to the many pitfalls of designing a system that can make sense for different companies in different countries engaged in different lines of business.

The Basel committee is working on what some at the Federal Reserve call Basel 1.2. Using this software-industry nomenclature makes it clear that the revisions will constitute not a major rewrite but an improvement that addresses flaws in Basel 1.0. (Basel 2.0 is years away. If software evolved at the pace of bank capital rules, the world would still be on DOS.)

Basel 1.2 addresses one of the most widely recognized problems with Basel 1.0: The 100% risk basket includes assets with wildly different characteristics. These discrepancies were noticed a decade ago when the rules were first issued.

At the time, however, policymakers were principally concerned with the prospect that assigning risk baskets would result in credit allocation. The more banks are favored by regulators with lower risk weightings, the more likely they are to put certain assets on their books. After more than 10 years of experience with the rules, however, regulators are less concerned that capital requirements will allocate credit than that credit will be misallocated.

As a result, the pending proposal goes beyond some of the tinkering with risk weightings of the past few years. It suggests a wholesale restructuring of the 100% basket. Because the current rules treat high- and low-risk commercial loans the same, a capital incentive for taking on higher-risk assets has been created.

The regulators are looking for a way to rely more on internal risk management models or ratings agencies to refine this basket and minimize the capital arbitrage that has rapidly become an established art at larger, internationally active banks.

The second capital initiative also tries a refined approach to credit and interest rate risk. The Office of Federal Housing Enterprise Oversight has issued a long-awaited risk-based capital rule for Fannie Mae and Freddie Mac.

The law mandating the regulation demands a stress test, not a specific statutory minimum as in the system applicable to U.S. banks. As a result, OFHEO has built a complicated set of models designed to ensure that the mortgage companies have enough capital to withstand wide interest rate shocks and market downturns over a 10-year period.

The OFHEO proposal weighs in at about 700 pages, making it obvious that this approach to risk-based capital is even more complicated than the Basel approach. OFHEO does, however, draw on work at the bank regulators by relying to some degree on ratings agency determinations to assign "haircuts" for its capital charges. Though the OFHEO approach is based on what is probably Freddie Mac's internal model, the agency decided to develop its own approach rather than rely on its ability to supervise models used by the mortgage companies.

Getting risk-based rules right is wretchedly difficult. The initial Basel scheme tried to compensate for all the noncredit risks it left out of the weightings, first, by shoving assets into certain categories without real regard to risk and then by adding risk criteria (but not placing assets into the right categories, as had originally been anticipated).

OFHEO's job would be, one might think, far easier. It has only to come up with a risk-based rule for two companies-not thousands- and for two companies restricted (more or less) to one line of business. Nevertheless, its capital rule is more than five years late and far from brief.

The capital issue becomes even more complex when one considers that capital counts not only within industries but also among them. Any capital rule for banks that is substantially different from comparable standards for comparable risks held by securities firms has immediate competitive implications.

At present, domestic and international regulators are dedicated to the proposition that, with enough time and money, they will come up with a capital system that accurately captures risk, prevents perverse incentives, avoids regulatory arbitrage, and is elegant enough for examiners to enforce.

Maybe so, but the decades-long efforts in Basel and the more recent struggles at OFHEO should give one pause. It looks very much as if the more one tries to get capital rules right, the worse they get.

One wonders if a better approach to risk-based capital might be to focus less on the capital and more on risk.

In this case, the risks would not be those taken on by a bank or any other financial services firm but the risks that firm poses to the public. One would calculate not how much capital a bank needs to absorb its risk but rather how much capital it needs to compensate the Federal Deposit Insurance Corp. for deposit insurance and the Fed for discount-window access.

It may well be that returning to the bad old days of simple leverage rules is the best way through the capital conundrum. Any leverage standard would, of course, now have to apply both to on- and off-balance-sheet risk if it is to have any meaning. It would have to apply internationally within each industry, but it need not be an effort at cross-industry capital parity.

If banks engaging in a certain line of business found that the securities rules for a comparable activity were more advantageous, they could switch charters. Capital rules should be posited on the risk an institution poses by virtue of its charter, not on the risks it may take on in the abstract.

One recognizes that all of these alternatives to the current approach are probably pebbles in the big pond of risk-based capital revisions.

However, the longer it takes and the harder it is to fix the current system, the easier it may be to turn to an alternative that has-at the least-the advantage of creating new problems instead of compounding the old ones.

Subscribe Now

Access to authoritative analysis and perspective and our data-driven report series.

14-Day Free Trial

No credit card required. Complete access to articles, breaking news and industry data.