Although in the last few weeks there have been some indications that the Japanese economy is bottoming out, this is unlikely to mean that it will soon begin a recovery. There have been many false bottoms to the 10 year Japanese recession, and I believe this is just one more. The reason I have no cause for optimism is that the same problem that has caused the recession remains unsolved--the nonperforming loans on the books of Japanese banks.

The significance of Japan's non-performing loan problem can easily be underestimated. It is, after all, only one element of a syndrome of ills that seem to be driving the Japanese economy ever downward. However, the non-performing loans (NPLs) currently held by Japanese banks--and in effect by the Ministry of Finance as the lender to over 80 government-controlled corporations in the Fiscal Investment Loan Program (FILP)--are in my view the primary cause of the ten year stagnation of the Japanese economy. Accordingly, addressing the NPLs--although it will require drastic steps by the Koizumi government--will eventually bring the Japanese economy back to life.

For almost a year, the Koizumi government has talked about reform, but has not yet formulated--or at least published--a comprehensive program. The Prime Minister himself talks frequently about privatizing the Postal Savings system, the issue that brought him to public attention and labeled him a maverick by Japanese standards, but as difficult as that would be it will not contribute in any significant way to an economic recovery in Japan. Another reform is the limitations place on the government's coverage of bank deposits, and a third is placing a limit on spending. As good as these ideas may be, it is also hard to see how either of them will stimulate economic growth. It is still difficult to discern a government plan in Japan that has any chance of dealing with the most pressing problem the country faces.

In an article in the November-December issue of The International Economy, Harald Malmgren argues that the Japanese economy will not recover until equity capital is once again made available to Japanese firms. He notes that the health of the US economy, especially since tax and financial reforms in the 1980s, rests upon a vast new industry of non-bank financial intermediaries--venture capital funds, hedge funds, pension funds and others--that are willing to make equity or near-equity investments in US companies. If similar vehicles could be mobilized in Japan, the Japanese economy would also thrive.

This is true as far as it goes, but the real reason is much more profound. Certainly there are cultural factors that in part explain the weakness of equity investment in Japan--the traditional Japanese reliance on large banks for financing commercial and industrial enterprises, and the general dearth of entrepreneurial, risk-taking enterprises in Japan--but a more significant reason for the lack of capital and especially equity investment in Japan is the NPL problem. Even if cultural facts did not impose their own limitations, equity investment in Japan would be stunted by the existence of an unknown and unknowable number of NPLs on the balance sheets of the major Japanese banks.

It is important to note in this connection that the real estate economy in the United States did not fully recover until the overhang of S&L and bank defaulted real estate loans and properties was finally sold off by regulators, and Japan's economy will not recover until the NPLs held by banks and government agencies are also liquidated.

The reason is the same in both cases: investors will not come forward as long as their investment in new assets could be immediately devalued by old assets that will eventually be sold at distress prices, or if the new assets they create will have to compete with existing assets that are being subsidized by banks or government agencies that refuse to recognize their true and much diminished value. This subsidy is very real. The companies that owe the NPLs are still in business but are losing money. In a properly functioning economy, they would have been closed down long ago, but because of a government policy that keeps the banks afloat, the debts of these companies are never called and they go on competing with--and weakening--healthy companies that would ordinarily survive them.

In reality, of course, the situation in Japan is considerably worse than the real estate crisis that prevailed in the United States during the late 1980s and early 1990s; the Japanese NPLs extend across the entire economy and not just one sector, and they constitute a far larger percentage of the economy than the bad assets of the S&Ls and banks in the United States. However, I believe the principle is the same. As long as there is a substantial amount of unsold and largely distressed assets overhanging the market, there will be no significant new investment.

Accordingly, while it seems just one of a number of contributing elements of the troubled Japanese economy, the NPL problem lies at the root of Japan's economic stagnation, and must be addressed before any significant capital investment will materialize.

There are four steps that are essential to deal with Japan's NPL problem, and none of them is easy either economically, culturally or politically. Yet it seems unlikely that the NPLs can be fully eliminated--or that the Japanese economy can fully recover--without implementing all of them.

Adopt an Effective Accounting System

Japan simply does not have an accounting system that permits the identification of NPLs. In the US, an NPL is not just a loan that is no longer paying interest or principal. US bank regulators have long been aware that a loan should be classified as nonperforming well before the borrower actually defaults. A borrower may be able to meet its loan obligations by selling off productive assets, reducing product quality, or by failing to invest in new plant and equipment. Eventually, of course, the borrower will actually default, and at that point the lender may have few assets it can rely on for repayment. In Japan, it is even more difficult to determine the financial health of borrowers by considering only their ability to meet their loan obligations, since many Japanese bank loans are "evergreen," and are serviced only by the regular payment of interest.

Thus, it is impossible to determine whether a loan is in fact nonperforming without an effective accounting system--beginning with a set of agreed accounting principles and backed up by an auditing function in which auditors bear direct financial responsibility for failure to discover accounting deficiencies. But the accounting system utilized in Japan does not meet these standards. Although improvements have been made in recent years, some of the Japanese accounting rules are not determined by the accounting profession but by the Diet through the commercial law. Needless to say, the legislature cannot determine by law what is a true or accurate presentation of a company's financial condition. This may be adequate, or even necessary, for tax collection or for regulatory purposes of some kind, but in order to gain and hold the confidence of financial investors--or to create a true picture of financial reality--accounting principles and a resulting accounting system must be privately developed and administered.

Because of the weakness of Japan's accounting system, it is probably accurate to say that no observers in Japan--not the government, not independent analysts, not even the banks themselves--have a clear idea of the size of the NPL problem. That is why the numbers keep changing from month to month and year to year, and why--when bankruptcies actually occur--the losses are always so much greater than would have been predicted from the defunct borrower's last financial statements.

Accordingly, the first step in addressing the NPL problem must be the development of an accounting system that will be able to differentiate between companies that are operating profitably and those that are not. Until such a system is in place, the proper disposition of NPLs will always be something of a guessing game.

Obviously, an accounting system cannot be created overnight, although importing the US GAAP system or the International Accounting Standards used in Europe would enable Japan to leapfrog years of arduous intellectual work in other developed countries. But the absence of an effective accounting system complicates the task of resolving NPLs because it makes the problem of valuation much more difficult. Indeed, the first objection faced by those who recommend a rapid sell-off of NPLs is that their value cannot be established. This brings us to the second step for resolving Japanese NPLs.

Sell NPLs Promptly, at Whatever Price They Will Bring

In the absence of an accounting system that will assist buyers and sellers to establish values, NPLs must be sold for whatever they will bring. This can perhaps be done most effectively through auctions, with the widest possible world-wide crowd of bidders. The Internet may provide the most efficient way to accomplish this, permitting the selling banks to provide all the necessary bid-package documents to buyers electronically--establishing in effect an electronic document room on a server in Japan or elsewhere. If bidders are given sufficient time to consider the values involved--including the value of intangibles such as patents, trademarks or trade names--and if the NPLs are properly packaged in sets that offer diversification or other advantages to bulk buyers, the selling banks should be able to realize the maximum value from the sales.

This is not to say that, at the outset, the sale prices will approach the true value of the assets. The likelihood is that they will not. For one thing, without an effective way to estimate values, buyers will demand high risk premiums. For another, many potential bidders will stay out until they see how well the process functions. The first bidders, accordingly, as occurred when US regulators began selling off defaulted real estate loans, will get enormous bargains. However, these windfall profits will attract more buyers to the next auctions, and prices will eventually rise to levels that approach the real value of the assets. In order to create an orderly process, a government agency could sell the NPLs for the account of the banks, but it is essential that the resulting losses be the responsibility of the banks and their stockholders, and not the government. A bailout of the banks' managements and stockholders through the government assuming their losses will only guarantee similar NPL problems in the future.

Obviously, these sales will create political problems for the government. For one thing, the buyers of these loans will be seeking to profit by either closing down the companies that are or should be in default and selling off their assets or taking control of and recapitalizing them, reducing their costs, and seeking to return them to profitability. In either case, this will cause rapidly rising unemployment in Japan.

I have had government representatives in Japan tell me this simply cannot be done within the Japanese culture--that unlike the United States workers in Japan are not accustomed to changing jobs or upgrading their skills. The laid off workers will become a major social as well as a political problem. Although this sounds suspiciously like an excuse for inaction, I am obviously not in a position to contradict it. But if it is true, it is still not a reason to temporize. First, the slow deterioration of the Japanese economy that will result will hurt everyone, not just the laid off workers, and second, the laid off workers will eventually lose their jobs anyway. The policies the government has been following allows the banks gradually to close off support for failing borrowers--it's just a matter of time. If laid off workers are going to be a social and political problem in Japan, they will become one anyway if the government continues its current policies on NPLs--it will just happen more slowly and with greater damage to everyone else. This will also mean that any recovery--if it takes hold--will also be slower, because more of the economy will have been weakened. It would thus be much better for the government to get it over with quickly, get the economy moving again, and hope that a growing economy will quickly re-employ those who have been laid off.

In addition, to add to the political problems, the bargains realized by initial buyers, including foreign buyers, will set off complaints from taxpayers, who know that eventually they will have to bear the cost of these losses, and from those who object to foreign ownership of Japanese assets. There is no good solution here; if, as is likely, many banks are insolvent after the sale of their NPLs, their losses will have to be covered somehow. The question is whether that cost is borne by the taxpayers or the private sector. When US regulators faced this problem in the late 1980s and early 1990s, their frequent choice was to sell the banks themselves, often together with their nonperforming loans, and to provide assistance to the buyers that would compensate them in part for the losses already imbedded in the balance sheets of the institution they were acquiring. In this case, the taxpayers at least realized some of the going concern and other values imbedded in the failed institutions.

However, since the likely acquirers of the Japanese banks will be solvent foreign banks or other foreign institutions, this is probably not a policy the Japanese government will want to pursue. Neither Japan nor any other country wants to have its banks controlled by foreign interests, so the best policy for Japan would be the sale of the NPLs and their associated assets, with a temporary government takeover of the banks in the event that they become insolvent. This will reduce the ultimate cost to taxpayers, and--as described below--save the banks as Japanese-controlled entities.

Recapitalize the Banks

By selling off their NPLS at distressed prices, many banks will become insolvent. This is probably the principal underlying reason that the Japanese banks and the government have made no aggressive moves over the years to follow this course, with a resulting depressing effect on the Japanese economy as a whole. But, if handled properly, bank insolvency is not an insurmountable problem.

Faced with a substantial number of potentially insolvent banks, the government should make clear to depositors that it will recapitalize the banks and honor their obligations. Although the stockholders of insolvent banks will lose their investment, depositors should stay in place because of the government's commitment. Again, the US experience provides a useful model. In the United States in the late 1980s and early 1990s, despite widespread knowledge that many banks and thrift institutions were insolvent, there were few if any depositor runs as the government gradually closed failed of failing banks. US depositors understood that if the bank or thrift in which they held their funds was declared insolvent the government would honor the obligations represented by their deposits. With a sufficiently clear government pledge, the same calm should prevail in Japan.

To be sure, this course contains some significant risks. Assuring depositors and other bank creditors that the bank's obligations will be honored by the government will create moral hazard if it is understood by depositors and creditors to be a permanent government policy. It will be important for the government to make clear that this step is a temporary expedient--necessary to address the NPL problem--and to back this declaration with a statutory limit on the size of the deposits that financial or regulatory authorities are authorized to reimburse in the future. One of the reasons for the development of the NPL problem in the first place was the absence in Japan of the market discipline that comes in part from a concern by bank depositors and other creditors that their investments will be lost if a bank becomes insolvent. Up to now, the Japanese government's policies have encouraged depositors to believe that their deposits will be protected by the government, and this has eliminated whatever market discipline depositors and creditors might have exerted on Japanese banks.

In recapitalizing the insolvent banks, the Japanese government will become the controlling stockholder of the insolvent banks. Having taken control, the government should install new management, new accounting systems, and new operating standards. When the banks are operating profitably again, all the government's shares should be sold off to Japanese investors, a process through which the government will be able to recoup some of the money it invested in the banks' recapitalization. The healthy banks that have then been reintroduced into the private capital markets should be able to operate profitably over the long term.

Sell the FILP Assets

The private sector Japanese banks are not the only repositories of NPLs in the Japanese economy. The Fiscal Investment Loan Program (FILP) operated by the Ministry of Finance has invested more than Y414 trillion ($3.35 trillion at Y120=$1) in government owned or controlled companies, most of it over the last 10 years. Needless to say, this enormous investment--made principally with the savings of Japanese citizens in the Postal Savings System (58%) and the Public Pension System (33%)--has not revived the Japanese economy. Indeed, it can be argued that the squandering of these funds on uneconomical infrastructure projects--bridges, roads, housing and airports among many others--has actually held back economic growth.

What seems clear is that a large proportion of the loans made to the FILP companies are nonperforming. Recently, a group of experts reported to MOF that they thought the NPLs in the FILP companies totaled at least $91 billion. Because of the deficiencies of the Japanese accounting system, it is likely that the total is much larger. These NPLs, too, must be sold, and the FILP companies themselves--except to the extent that they are performing actual government functions--should be privatized. Again, private investment will not materialize as long as subsidized assets overhang the market, or private capital must compete with government-subsidized capital.

The most important result of all these moves, difficult as they are, will be the elimination of the NPLs that are today impeding private investment in the Japanese economy. Once the assets that now overhang the market have been assigned a realistic value and placed in private hands through sale, new capital investors will materialize and the Japanese economy will begin to grow again.

Peter J. Wallison is a resident scholar at AEI.