This article is a synopsis of a paper prepared by John H. Boyd, senior research officer at the Federal Reserve Bank of Minneapolis, and Mark Gertler, professor at New York University. The full paper appears in the summer issue of the Quarterly Review, published by the Research Department of the Minneapolis Fed.
It is widely believed that commercial banking is a declining industry. Two factors are often cited to support this contention. First, nonbank credit alternatives have grown rapidly over the last 15 years. Second, in the late 1980s, banks experienced record levels of failures and loan losses, symptoms of an industry in distress.
The view that banks are shrinking in importance is held by banking executives, academics and high officials in many branches of government. For example, William Isaac, former chairman of the Federal Deposit Insurance Corp. and now a prominent banking consultant, said in The Wall Street Journal last year that "... the banking industry is becoming irrelevant economically, and it's almost irrelevant politically." Carter Golembe, the dean of bank consultants, in a 1993 paper similarly noted the "... major problems faced by the bank industry, most notably its eroding competitive position in the financial community and the crushing burden of regulation ..."
The purpose of our study is to check the accuracy of the consensus position that U.S. banks are in decline. We define the U.S. banking industry to include all commercial banks, branches and offices located in the country—whether they are American or foreign-owned. From the perspective of this study, what is most relevant is line of business, not ownership. Our conclusion, based on an analysis of a variety of data, is that there is no evidence of a significant decline in the U.S. banking industry. After correcting for a number of measurement issues, we find that commercial banks' share of total financial intermediation in this country has been roughly stable over the last four to five decades. At most, banks may have suffered a slight loss of market share in the late 1980s and early 1990s. A case can be made, further, that this slight loss in market share was mainly a transitory response to a series of shocks to the banking industry that occurred over this period. On the other side of the coin, commercial banking has actually risen in importance relative to aggregate economic activity, even over the last 15 years. While banks have maintained a relatively constant share of intermediation, financial intermediation has been growing steadily relative to gross domestic product.
Should we care about banking?
Why should anyone care whether banking is in decline? Many industries naturally expand or contract with the passage of time. The demise of the buggy whip and the Hula Hoop industries were hardly cause for great concern. Commercial banks, however, have traditionally played a central role in the financial system. They have been key providers of liquidity. They have also been important conduits for credit flow to households and small- and medium-sized businesses. The regulatory policy that comprises the financial safety net (deposit insurance, discount window lending, capital requirements and so on) stems from the premise that banks are critical to flow of credit, particularly the flow of short-term credit.
If technological change and financial innovation were making banks irrelevant, then, at a minimum, it would be necessary to rethink regulatory policy. Perhaps it would no longer be correct to focus the financial safety net around banks. Perhaps a safety net would no longer be necessary. On the other side of the coin, if "excessive regulation" was producing a decline, then it may be desirable to relax some of these restrictions. Under either scenario, knowing the facts—is banking declining or not?—is essential. This is what this paper is about.
Why conventional wisdom is unwise
Why do our results run counter to conventional wisdom? Formal evidence for the traditional view comes from analyzing the ratio of bank assets to other forms of credit. There are, however, two major problems with this metric. First, traditional measures of bank assets fail to account for banks' off- balance sheet activities. Over the last 15 years, banks have increased the extent to which they do business off the balance sheet. The combination of deregulation and financial innovation has permitted banks to increasingly decouple the various functions involved in intermediating lending. For example, banks now sell some of the loans they originate to other financial institutions. They have also increased the extent to which they indirectly support lending by providing backup lines of credit and guarantees. They now facilitate risk-sharing through the provision of derivative instruments. The moral is that industry share measures based on on-balance sheet assets understate commercial banks' contribution. We show that a good fraction of what appears to have been a decline in commercial banks' share of intermediation by traditional measures instead reflects a relative movement of bank activities from on to off the balance sheet.
The second measurement consideration involves the expansion of lending by foreign commercial banks to U.S. firms that occurred over the 1980s. In addition to providing increased competition for domestic banks, the increased foreign involvement has also contributed to mismeasurement of commercial banks' share of domestic credit flows. The official measures have significantly understated the rise in loans supplied by foreign commercial banks. After correcting for both the mismeasurement of foreign bank loans and the exclusion of off-balance sheet assets, any evidence of a substantial decline in commercial banks' share of intermediated assets vanishes.
It is also important to emphasize that proponents of the consensus view have tended to incorrectly use market share of intermediation numbers to draw inferences about banks' importance to the economy. As we implied earlier, market share numbers fail to account for the relative growth of financial intermediation. Indeed, we find that even the unadjusted balance sheet measures indicate no decline in bank assets relative to gross domestic product. And our adjusted measures indicate a clear increase.
As noted, we define the U.S. banking industry to include all commercial banks, branches and offices located in the country—whether they are American or foreign-owned. Our analysis does indicate that foreign banking firms have gained substantial market share in this country, particularly in commercial lending. U.S. bankers might argue, therefore, that our analysis misses the point. They're still losing market share; if not to other types of intermediation, then to foreign banks' commercial lending departments.
That has been true over the last decade or so. However, this is a rather different story than the consensus view that banks are losing market share in the United States. (And, potentially, with very different policy implications.) In any case, the point of this study is to try to get the facts straight. As we discuss in the full paper, it seems likely that foreign lending inroads in the United States were largely due to differential regulatory treatment favoring foreign banks. However, significant inequities were eliminated in the early 1990s. Consistent with this change, recent data suggest that the foreign banks' loan market share has levelled off and (very recently) begun to decline.
Evidence of robustness
In our full paper, we construct measures of bank assets that are designed to properly account for off-balance sheet activities and for total lending by foreign banks. We then analyze the behavior of this newly constructed aggregate relative to other forms of credit and to GDP. For robustness, we construct credit-equivalents of off-balance sheet assets using two quite different methods. The two approaches yield very similar results.
To obtain further evidence on the robustness of results, we also present a completely different approach to measuring banks' importance, using data from the National Income Accounts (in place of balance sheet data). Here the idea is to use value-added numbers to measure banks' contribution to economic activity. Computations based on this approach give an impression that is very similar to that provided by the augmented balance sheet data: There is no evidence of a secular decline. Because of possible measurement problems with the value-added data, we also do computations based on input usage. Again, there is no evidence of secular decline.
It certainly is true that a number of disturbances jolted the banking industry in recent years. These disturbances included increased competition, loan losses and the phasing in of new regulatory requirements. In our study, we assess the impact of these shocks on the condition of commercial banking. We conclude that these factors may have accounted for the slight loss in banks' share of intermediation over this period. But there is no evidence to suggest that these disturbances have pushed the industry into permanent decline. Indeed, in the last two years the fortunes of banks have steadily improved, along with the overall economy, and there are other reasons to be optimistic about the future of banking.
Public policy should be reevaluated
We do not dispute the notion that the banking industry experienced severe difficulties in the late 1980s. What we are calling into question is whether the poor performance over this period signals the beginning of a permanent decline. Both the balance sheet data (adjusted for a variety of measurement issues) and value-added and input data from the National Income Accounts fail to reveal any striking decline in the role of commercial banks.
Clearly, banks have faced increased competition from nonbank alternatives. They have responded, however, by changing the way they provide traditional services and by developing new products. The rising importance of off-balance sheet activities, ranging from credit-lines to derivative products, are symptomatic of these developments.
If we are right that banking is not a declining industry, then more than academic interest is at stake. Important public policy decisions have been and continue to be based on the consensus view. One such policy is in the area of bank mergers. Consolidation in banking (largely via mergers) has been encouraged, partly on the grounds that it is a way to mobilize resources out of a declining industry. If the industry is not declining, only changing, this argument loses force. A second such policy is the expansion of bank powers. One common argument is that banks are declining because with current powers limitations they cannot compete. This argument also loses force (although there may be other, perfectly valid, reasons why bank powers should be expanded). Along the same lines it is often argued that banks can't compete because of excessive regulatory burden; or, that interest should be paid on required reserves to help out this troubled industry. Related policy proposals abound, all based on a premise that is questionable. If public policy is based on bad assumptions, it is unlikely to be good policy—except by accident.
