(FORTUNE Magazine) – IN THIS BATTLE of regulatory titans, Richard Breeden, head of the Securities and Exchange Commission, weighs in at seven pounds, zero ounces. Alan Greenspan, chairman of the Federal Reserve Board and Breeden's opponent in a fierce debate that could lead to sweeping changes in the way companies measure financial performance, tips the scales at a few pounds more -- or less. Ludicrous? Not according to generally accepted accounting principles, or GAAP (pronounced gap). GAAP calls for recording and maintaining most corporate assets and liabilities based on their original cost -- a practice akin to measuring a person's size by his birth weight (as we just have). Should anything be done about this kind of gross distortion of financial reality? Yes, says Breeden. He is campaigning vigorously for a new approach that could ultimately shelve the use of historical costs in favor of current, or market, values. In his camp are Comptroller General Charles Bowsher, who heads the General Accounting Office; Edmund Jenkins, a partner at Arthur Andersen & Co., who chairs a high-level accounting industry group that is studying financial reporting standards; and a number of influential economists. Squared off against them is a powerful crowd that includes Greenspan, Treasury Secretary Nicholas Brady, Federal Deposit Insurance Corp. Chairman William Taylor, and, it seems, virtually every banker and bank regulator in the country. The stakes in this seemingly esoteric argument over the bean counter's art are potentially large and wide-ranging. Greenspan and other defenders of the status quo argue that jettisoning established accounting conventions for new, untested rules depending upon highly subjective estimates of market values would undermine investors' and depositors' confidence in banks. Such a change could also encourage bankers to abandon their traditional role as providers of long-term credit and holders of long-term securities. Those risks must be taken seriously. But the risks of inaction may be even greater. Exhibit A in the case for change: the collapse and bailout of the savings and loan industry, now expected to cost taxpayers at least $150 billion before it's over. Though different accounting rules might not have averted this disaster, greater reliance on market values would have signaled impending doom far earlier and could have forced regulators to shut down thrifts while their losses were still manageable. Says Edward J. Kane, a finance professor at Boston College and an authority on the S&L fiasco: ''Misleading accounting practices allow zombie institutions to conceal the depth of their insolvency.'' Even now, traditional accounting may be allowing banks, which finally seem on the mend, to delude themselves, regulators, and investors about their true health. On average, banks' financial statements show them holding capital equal to about 7% of the historical value of their assets. But Emory University economist George J. Benston, an expert on bank capitalization, maintains that if their balance sheets, which are still burdened with bad real estate loans, were marked to their current market value, many banks and thrifts would appear insolvent or close to it. THE TASK of deciding how best to reform GAAP falls to an unusual organization, the Financial Accounting Standards Board. Its rulings effectively have the force of law. That power is bestowed upon it by the SEC, which has statutory authority over the accounting practices that companies must follow. Yet FASB's seven members are chosen by a private foundation and its bills are paid in part by its main constituents, which are large corporations and accounting firms. Its stated aim is to devise ''neutral'' accounting standards that provide information without regard for any particular policy result. In fact, FASB's dry, academic-minded deliberations, conducted in public in a classroom-like space at its offices in Norwalk, Connecticut, sometimes lead to pronouncements that profoundly affect society (see box). So far FASB appears inclined to side with market value proponents, though just how far it will go remains uncertain. Late last year it finalized a rule, Statement 107, that will require companies with more than $150 million in assets to disclose the ''fair value'' of securities, loans, and a wide range of other financial instruments that they hold, beginning with reports issued for fiscal years that close after December 15, 1992. Smaller companies will have three extra years to comply. Fair value as defined by FASB is the amount at which an asset or a liability can be exchanged in a current transaction between a willing buyer and a seller who is not under duress. Because the forthcoming disclosures are to appear as footnotes to financial statements, they won't alter earnings or the values reported on balance sheets, which will continue to be produced on the basis of GAAP. Even so, the new disclosure rule was strenuously opposed by banks and many nonfinancial companies. They fear it is merely the first step in a radical overhaul of the GAAP system, which already contains some projections about the future alongside its historical-cost foundation. At Breeden's urging, FASB is hashing through a proposal that would go beyond Statement 107 and require companies to recognize -- either on their income statements or on their balance sheets -- fluctuations in the market values of certain securities and related liabilities. Also on its agenda: possible changes in when and how loans are written down once they are deemed impaired and new rules for the accounting treatment of plants, equipment, and real property whose value seems to have permanently fallen. GAAP now gives companies considerable leeway in each of these situations -- leeway they often use to manage the earnings they report. Says Donald Nicolaisen, Price Waterhouse's national director of accounting: ''These proposals are all related. They reflect the same gnawing concern that historical cost accounting doesn't work anymore.'' Bankers, the group most directly affected by these new notions, concede that a few current practices need reform. The most notable of these is known as ''gains trading,'' the practice of selling your winners (bonds on which you have paper gains) while holding your losers (bonds with unrealized losses) in order to boost reported earnings. Such cherry picking is made possible by rules that allow bonds to be accounted for at cost rather than market value as long as the company claims it intended to hold them until they matured. BUT THE MONEYMEN argue vehemently that trying to mark assets and liabilities to market is a cure far worse than the original disease. Their first line of defense: It is virtually impossible to come up with verifiable estimates of what many financial instruments are worth. The assumptions and guesstimates required, bankers contend, make the end result meaningless. Take commercial loans, which account for about a third of the assets on many banks' balance sheets. ''Our portfolio has tens of thousands of highly customized loans, mostly to small and midsize companies,'' says Susan Bies, chief financial officer for First Tennessee National, a Memphis bank holding company with $7.9 billion in assets. If such loans traded on a well-developed secondary market, determining their value would be a snap. But most trade infrequently or not at all. So calculating each loan's present value depends on how you answer a battery of tough questions. Will a borrower pay off the loan early if interest rates go up by two (or three or four) percentage points? Under what conditions might he convert from an adjustable-rate loan to a fixed-rate one? How vulnerable is his collateral to an economic downturn? Says Bies: ''In good conscience, different banks can come up with different forecasts and arrive at vastly different estimates of value for similar loans.'' Valuing other crucial items on a bank's balance sheet is even slipperier. Banks profit by lending money held in low-interest checking and savings accounts to other customers at far higher rates of interest. The stability and long-term nature of these so-called core deposits endow them with considerable worth beyond their face value. But they don't trade; nor is there any good way to forecast future cash flows, because they can always be withdrawn at any time. Economist Benston, who is a market value enthusiast, admits no one has come up with a satisfactory way to value core deposits. Others raise an even more basic objection to market value accounting. They argue that showing the price a 20-year loan could fetch were it sold today is simply not an appropriate way to measure the performance of a bank that fully expects to be around 20 years hence when that money is due to be repaid. Weighing in last year against what became FASB's Statement 107, Lester Stephens Jr., Chase Manhattan's corporate controller, wrote: ''These values represent a 'fire sale' valuation that will gyrate up and down as interest rates and credit factors change. This doomsday portrayal can be very damaging to the uneducated user.'' Or perhaps even to educated users. For Stephens and other bankers, the doomsday scenario is built upon a belief that Wall Street -- and maybe regulators as well -- will severely penalize them if the more volatile earnings that market value accounting would bring show up on their income statements. ''A 1% change in interest rates would cause a 3% flip-flop in the value of our bond portfolio. That would have a significant impact on our earnings,'' says P. Michael Brumm, chief financial officer for Cincinnati's Fifth Third Bancorp. In that case he predicts banks will try to dodge such swings by stuffing their investment portfolios with far more short-term bonds, whose market values are less susceptible than long-term securities to big up and down movements. Banks might also be far less apt to hold bonds issued by small municipalities, speculates Gary Anderson, CFO at Zion Bancorp. in Salt Lake City. This debt is often relatively risky, not actively traded, and thus more prone to wide swings in value. Since banks are major buyers of municipal bonds, such a shift would make it harder and costlier for cities to raise capital for urgent needs such as rebuilding bridges and upgrading schools. Treasury Secretary Nick Brady contends that market value accounting ''could even result in more intense and frequent credit crunches.'' In an insistent letter to the chairman of FASB, Brady maintained that temporary declines in the market value of assets, if recognized on the balance sheet, ''would result in immediate reductions in bank capital and an inevitable retrenchment in bank lending capacity.'' NOW HERE'S THE CASE for making a change, despite these worries. It is undeniable that interest rate shifts send the value of a bank's loans, say, spiraling up and down, though current accounting practice allows such changes to remain undisclosed to the outside world. While no one can be certain how investors and creditors might respond to a clearer picture, it seems doubtful that the reality hidden by GAAP would come as a complete surprise to Wall Street analysts, who supposedly earn their big salaries by parsing balance sheets in search of the truth -- and who presumably find, if not the whole truth, at least a good deal of it. True, some banks with unexpectedly spikier earnings probably would get punished. But a former Fed economist, who asks not to be identified because he remains part of the banking industry, maintains that on the whole, Wall Street will accept with equanimity financial statements that look more volatile. Another closet market value enthusiast, a respected economist at a major bank, agrees and adds, ''If you're not sufficiently capitalized to withstand the true volatility inherent in your balance sheet, taxpayers are better off knowing that unsettling fact sooner rather than later.'' CONSIDER AGAIN the lessons of the S&L mess. In 1981 the S&L industry's collective balance sheet showed $28 billion in capital, according to retrospective figures gathered by Richard Pratt, former chairman of the Federal Home Loan Bank Board, the main overseer of thrifts in the 1980s. Yet, on a market value basis, the industry's liabilities exceeded assets in 1981 by a staggering $178 billion. If financial statements had given off even a whiff of such gross insolvency, regulators -- and legislators too -- would have had little choice but to force sick institutions to find additional capital or, failing that, to close them before their difficulties multiplied. Three years ago the Office of Thrift Supervision, the successor to the Home Loan Bank Board, began requiring the thrifts it supervises to submit detailed cash flow and interest rate information about their loan portfolios and other balance sheet items. OTS feeds the data through an elaborate model that estimates how sudden changes in interest rates might affect the market values of assets and liabilities. The model was designed primarily to help thrifts better understand and manage their interest rate exposure. As a byproduct it also produces approximations of the market value of an institution's capital, the crucial margin of safety for uninsured creditors and the federal deposit insurance fund. Though OTS officials remain harshly critical of shifting to market value accounting, some grudgingly admit this new information could prove useful. Says one: ''It's potentially an important red flag.'' While FASB ponders how far to push market value accounting, a few companies are moving ahead on their own. Two years ago Roosevelt Financial Group, a St. Louis-based thrift holding company with $2 billion in assets, began including in its financial statements what it calls net market value, in essence the difference between the market value of its assets and the market value of its liabilities. Interestingly, Roosevelt's most recent net market value is 9% lower than the company's book value according to GAAP. Despite such negative tidings, CEO Stanley Bradshaw praises market value accounting with the fervor of a preacher delivering the gospel. He gives the back of his hand to the argument that such calculations are impractical, acknowledging that might have been true when interest rate swaps, mortgage- backed securities, financial futures, and other innovations didn't exist or were not widely traded. But the proliferation of such products over the past decade, he says, along with the development of complex mathematical pricing models, now makes it feasible to estimate market values for most financial instruments. Though this depends on a lot of assumptions, Bradshaw argues that the judgment required is no more extensive or imprecise than what GAAP currently expects of bankers in estimating, say, loan loss reserves. Some investors also see big benefits from publishing market value estimates. Says Thomas Jones, chief financial officer at Teachers Insurance and Annuity Association, which provides retirement and insurance plans for colleges and universities: ''Policy holders and regulators will have more insight into whether liquidity and safety concerns are warranted, and investors will get a clearer fix on possible hidden values.'' One such financial Spindletop burbles on the books of SunTrust Banks, a $35 billion bank holding company headquartered in Atlanta. SunTrust has long disclosed market value information for its investment securities, though not its loans. The annual eye popper: the revelation that its vaults contain shares of Coca-Cola Co. that a subsidiary received as an underwriting fee when Coke went public in 1919. That stock, which remains on SunTrust's balance sheet at its original value of $110,000, is now worth almost $1 billion. Among companies committed to market value accounting, none is more sold on the idea than the Federal Home Loan Mortgage Corp., a New York Stock Exchange- ! listed company created by Congress to buy home mortgage loans from banks and other lenders. Freddie Mac first began experimenting with market value management concepts in the mid-1980s and since 1989 has issued a full-fledged market value balance sheet as a supplement to its regular quarterly financial statements. It is probably the only company whose present disclosures satisfy the requirements of FASB's Statement 107, which all big companies will have to start complying with eight months from now. Freddie Mac Chairman Leland Brendsel believes the real appeal of market value accounting lies in the managerial insight it provides. He insists, for example, his company's computerized market-value model gives managers more realistic guidance about how to limit exposure to interest rate changes than could ever be gleaned from historical numbers. How is FASB likely to proceed? Slowly, if the past is a guide, and that's probably not a bad speed. While the value of the insight offered by market value accounting is compelling, the nitty-gritty of adopting such a new approach on a comprehensive scale will require a lot more hard work and thought. John Spiegel, chief financial officer at Atlanta's SunTrust, strikes what seems the right balance on this contentious issue. He recognizes the power of analyzing balance sheets through a market value lens, an exercise that he and his colleagues perform regularly for the better ''feel'' it gives them in hedging interest rate risks. And unlike most bankers, he doesn't object to FASB's Statement 107, which limits market value information to footnotes. But before going further, Spiegel argues, it's important to find out whether such information will really help users of financial statements make better decisions. In other words, let the disclosures FASB has already decreed serve as a laboratory to test the feasibility, and ultimate value, of market values.