Printer Friendly

Banking conditions and the credit crunch: implications for monetary growth and the economy.

FOR THE FIRST time since 1987, M2 growth did not reach even the lower bound of its target range. What is even more unusual about the weak M2 growth is that, for more than two years, M2 has been below the level that would have been forecasted to prevail, given the levels of gross domestic product (GDP) and interest rates. If changes in the level of income and interest rates -- the dominant explanatory factors in any money demand model -- cannot explain or account for the shortfall or "missing M2," other forces must be at work that are changing the traditional money demand relationships. Simultaneously, the money supply process has been disrupted by regional credit crunches. In this article, we examine the financial health of the banking system -- an important supply side factor -- and a change in bank and thrift closure policy under the Financial Institutions Reform, Recovery, and Enforcement Act (FIRREA) -- an important demand side component -- and find that these are two important factors that help to explain the so-called "missing money."


Duca(1) showed that the effect of the passage of FIRREA and the Resolution Trust Corporation's (RTC's) methodology for closing insolvent institutions reduced the demand for M2. Under FIRREA, when a bank or thrift is closed, individuals holding deposits that paid high interest rates over long maturities are informed that in two weeks their interest rates will be reduced to the prevailing market rate. This policy increased the reinvestment risk of M2 deposits, i.e., the risk that a bank's long-term commitment would be abrogated. This forced the investor to reinvest the funds at current market rates, which, in practice, turned out to be lower than the original interest rate.

The public's demand for M2 balances has been reduced not only by weak income growth and the fall in interest rates during the past several years but by the increased reinvestment risk stemming from the changed bank and thrift closure policy under FIRREA. Reinvestment risk is greater at weaker institutions because the probability of failure is higher. The standard forecast equation for M2, however, assumes that the reinvestment risk is constant and/or negligible. Consequently, for the past two years actual M2 has been below forecasted M2. Duca showed that, once the reinvestment risk was introduced into the forecasting equation by including a measure of RTC activity, two-thirds of the overestimate in forecasted M2 was eliminated.

Klemme and Robinson(2) showed that a shortage of capital at banks and thrifts has been not only an important constraint on lending activity but has also affected the other side of the balance sheet by reducing banks' need for deposits. Banks and thrifts that are capital constrained, i.e., have capital below the regulatory minimum, have sharply curtailed their lending. For example, banks that were not capital constrained in 1991 expanded their loan volume by about 2 percent while capital constrained banks reduced their loan volume by about 4 percent. Similarly, thrifts that were not capital constrained in 1991 reduced their loan volume by a little less than 1 percent while capital constrained thrifts reduced theirs by about 12 percent.

On the deposit side of the balance sheet, capital constrained institutions showed much weaker deposit growth than those that met their regulatory minimums. The most dramatic difference was in the small time deposit category in which capital constrained institutions reduced their balances by nearly 10 percent in contrast with nonconstrained institutions that increased their small time deposit liabilities by about 2.5 percent in 1991. Clearly, the ability of the Federal Reserve to expand M2 is limited to some degree by the existence of capital constrained depository institutions whose ability to book deposits and expand credit is impaired by their weakened financial condition. (For evidence that banks impose their own capital constraint on lending activity that is even tougher than regulators' capital constraints, see Hancock and Wilcox.(3))

Small time deposits are the component of M2 where the largest adjustments have taken place. For the three-year period 1988-90, the small time deposits component of M2 grew at an annual rate of nearly 9 percent. For the seven-quarter period, 1991-1992 Q3, a large portion of the "missing M2" period, small time deposits fell at a 24 percent annual rate. Consistent with the Duca explanation, small time deposits would be the component of M2 most affected by changes in reinvestment risk, because these deposits have a stated maturity and are therefore affected by bank closure policy. The other components of M2 are essentially withdrawable on demand and can be repriced at any time. Comprising about one-third of M2, small time deposits are the largest component of M2.


Because economic conditions are not uniform across the country, regional analysis provides us with an insight into the factors depressing M2 growth and causing M2 forecasts to err on the high side, A cross-section analysis of small time deposits should be unaffected by interest rates set in a national market, but these deposits should be affected by state levels of income and employment. We have also found that two additional variables -- the number of bank and thrift failures and the percentage of healthy banks in each state -- are particularly useful in explaining differences in deposit and loan growth across states. (See Rosenblum,(4,5,6)) The health of depository institutions may play a crucial role in understanding recent M2 behavior.

In our analysis, a bank is considered healthy if it meets all three of the following conditions: (1) it is profitable; (2) it has a troubled asset ratio below 3 percent (approximately the national average); and (3) it has a primary capital ratio greater than or equal to 6 percent (i.e., slightly above the minimum regulatory standards prevailing at the time the studies were conducted a few years ago).

States with unhealthy banking industries have been concentrated in the Northeast and Southwest. It is in these states that we would expect to see some of the greatest weakness in deposit and loan growth. This is indeed what occurs. At least on a superficial level, credit crunches, financially troubled banks, and small time deposit shrinkage all seem to be interrelated phenomena. The remainder of this article explores these relationships in greater detail.


It is worth exploring on a somewhat intuitive level the reasons why and the ways in which banking behavior has changed in recent years. It is not at all unusual for credit growth to slow to a near crawl during a recession and during the first quarters of a recovery. During this phase of the business cycle, the weakness in sales is accompanied by depletion of inventories and by postponement of capital investments and durable goods purchases, all of which reduce business and household demand for credit.

Even during recessions, however, not all firms are contracting their activity; a few are experiencing growing demand for their products and, as a consequence, have an increasing demand for bank credit to finance their growing operations. These businesses generally find little difficulty finding banks willing to lend to them. Indeed, during a period of macroeconomic weakness and weak loan demand, creditworthy businesses are generally solicited by a large number of banks wanting their loan business. Over the past few years, however, businesses in several regions of the United States perceive they have experienced the opposite scenario, i.e., a strong reluctance to lend at every bank they turn to. To the extent that these perceptions accord with reality and to the extent that these unsatisfied borrowers are creditworthy, this phenomenon is suggestive of a very different set of supply-side forces that are driving the lending behavior of the banking system.

Those who deny the working of these supply side constraints on credit behavior argue that, in a competitive marketplace, some firms are always looking to gain market share at the expense of their competitors. If some banks have gotten themselves into regulatory hot water and are constrained from continuing their aggressive asset expansion, then other, presumably healthier, banks should fill the credit vacuum created by the financially weak banks.

But what happens if a very large percentage of banks in a region have significant financial problems of their own, i.e., when few banks are healthy enough to be in a position to expand their lending base and most banks are unhealthy enough to feel compelled -- on their own or due to regulatory pressure -- to contract their loan volume? In these circumstances, a regional credit shortage becomes possible, at least for those businesses whose loan demand must be satisfied by local banks. Those businesses that have access to the national money and capital markets would scarcely notice any difference in their access to credit. But those smaller businesses that are more geographically constrained in their credit relationships (and often locally limited in their banking arrangements) would feel the pinch of reduced credit availability.

Texas provides a perfect example of a breakdown in the provision of credit. In the three-year period, 1986-88, severe loan losses caused outstanding loans at Texas banks to fall by more than 50 percent in real terms. Unhealthy banks, either voluntarily or at the insistence of bank regulators, contracted their lending sharply. Healthy banks, at the time, controlled only 20 percent of the banking assets in Texas and did not have the capacity to expand lending to offset the severe decline in lending occurring at the unhealthy banks. (For a more complete development of this explanation see Clair and Yeats.(7)).

Even though unhealthy banks were not lending, they made an effort to appear to be willing to lend. In an effort to appear healthy, these banks considered loan proposals but always found a reason not to approve the credit. When a significant percentage of banks are unwilling to extend credit, we have the necessary preconditions for a regional credit crunch. The Texas credit crunch began in 1986 and was still going on in 1992, but to a considerably lesser extent than was the case a few years earlier.

What impact does such a reduction in credit availability have on the economy? In Texas, the sharp reduction in bank lending probably reduced the rate of growth of the economy to a rate that was lower than would have occurred had the Texas banking system been increasing credit, even at a modest pace. As measured by employment, the Texas economy contracted sharply throughout 1986 and the first quarter of 1987 following the precipitous decline in oil prices in January 1986, when the price tumbled from $28 to $12 per barrel and then stayed at the lower end of that range for the rest of the year. Given the magnitude of this shock to one-sixth of Texas' economic base, the decline in Texas' economy that followed was not surprising. The Texas economy would have declined even if bank credit had been expanding. The Texas economy turned up in spring 1987 and continued to grow until early 1991, when it declined modestly for a few months. During that same period, bank credit was shrinking. So the economy can grow even while bank credit contracts, though the growth is probably somewhat slower than otherwise.

While the overall economic situation in Texas was difficult, it was less severe than the downturn that took place in the New England states. The percentage declines in employment in Connecticut, Massachusetts, Rhode Island, and New Hampshire were much greater and occurred over a much longer period of time than what took place in Texas. This is probably due, at least in part, to the fact that New England's banking sector was made more unhealthy by the New England downturn that, in turn, reinforced the negative forces impinging on the New England economy.

At year-end 1990, only one-fourth of banks in the Boston Federal Reserve District (all of New England except Fairfield County, Connecticut) were healthy, and these banks controlled only 8 percent of the banking assets in the region. This is clearly a situation where creditworthy borrowers had few alternative sources of bank credit, especially because the majority of banks in the next nearest Federal Reserve District (New York) were unhealthy and were unlikely to be seeking out-of-market loan customers.
Table 1
Distribution of Healthy Banks by Federal Reserve District
 Percentage of
 Percentage of Assets in
 Healthy Banks Healthy Banks
 1992:Q3 1990:Q4 1992:Q3 1990:Q4
Boston 42 26 39 8
New York 59 46 38 20
Philadelphia 78 69 81 62
Cleveland 90 82 92 58
Richmond 81 68 73 51
Atlanta 80 61 85 53
Chicago 92 85 78 77
St. Louis 91 79 94 80
Minneapolis 91 74 89 69
Kansas City 85 64 86 64
Dallas 72 48 77 59
San Francisco 57 65 33 75

The regional aspects of this explanation of a credit crunch are supported by the data in Table 1. During 1990 and 1991, healthy banks tended to be concentrated in the Midwest, which also is the region where the fewest cries of "credit crunch" were heard. This table also reveals the degree of recovery in the Texas banking industry over the past few years. The figures shown for the Dallas Federal Reserve District are dominated by Texas and demonstrate marked improvement over the situation that prevailed in 1989 when less than half the banks in Texas were healthy and these banks controlled only 20 percent of the banking assets in the state. Also worthy of note is the serious deterioration of the health of banks in the San Francisco district, primarily as a result of deterioration of the California banking industry. As of the third quarter 1992, only 10 percent of the banking assets in California were held by healthy banks, which represent 42 percent of all California banks. This could represent a potential credit crunch developing.


Having explained the nature of regional credit shortages, we now return to the implications for monetary growth. Figure 1 shows the annualized rate of change of small time deposits at both banks and thrifts combined over the seven-quarter period from year-end 1990 through the third quarter of 1992, roughly the period of the "missing money" for which data are available. The two regions that experienced the greatest financial turmoil -- the Southwest and New England -- also show the greatest decline in small time deposits. A similar map for loan growth at banks and thrifts over the same period shows that declines in lending are substantially greater in New England and the Southwest than in other regions(**).

The relationship between the relative healthiness of banks in the nation's fifty states and their loan and deposit growth were investigated by regression analysis for the 1986-91 period. This analysis shows a significant positive relationship between average bank healthiness in each state and the growth rate of loans and deposits; on average, states with a healthy banking system tend to expand credit and fund that expansion with time deposits, while states where sick banks predominate tend to experience contracting credit and time deposit run offs.


We have discussed two different explanations for weak M2 growth: (1) the RTC's imposition of a de facto call premium that has raised the reinvestment risk on time deposits at banks and thrifts; and (2) the effect of a bank's financial health -- including its capital and other regulatory constraints -- on depository institutions' demand for interest-sensitive deposits. Both of these explanations of weak M2 growth have the same origin -- the massive loan losses that banks suffered in the late 1980s. Both hypotheses are valid, and it is difficult to separate the decline in small time deposits that might be attributable to each. For example, increased reinvestment risk will depress M2 demand, with the largest decreases in M2 demand taking place at institutions that are weak because of low capital positions. At the same time, weak capital positions make it difficult or impossible for these institutions to increase their lending activity. The credit crunch did not cause the "missing M2." And the "missing M2" has not exacerbated the credit crunch.

The post-1989 behavior of M2 has been different from its behavior in prior years, even though credit crunches have been a factor during both periods. The important difference between the two periods is the reinvestment risk associated with bank closure policy. We can see this by examining total deposits at Eleventh District banks and thrifts during the past nine years. Banks and thrifts in the Eleventh District (Texas, the northern portion of Louisiana and the southern portion of New Mexico) were clearly experiencing financial difficulties by 1986, if not earlier. Small time deposits (shown in the dark portion of the bars in this chart) continued to rise until 1988, and they showed little or no decline in 1989. Consequently, the abundance of capital-constrained banks does not explain the "missing M2." Small time deposits were growing, in part, because many depositors were reducing their large time deposits to levels that were fully insured by the federal deposit insurance funds. At the time, there was no corresponding "missing M3" story because these deposits were shifting from one component of M3 into another. (For additional evidence on the shift out of large time deposits at New England commercial banks and savings banks, see Peek and Rosengren.(8))

It wasn't until the passage of FIRREA in 1989 and the beginning of bank and thrift failures in a manner that introduced reinvestment risk that small time deposits began to fall in the Eleventh District. Although a very large contraction of loans took place at Eleventh District banks and thrifts in 1986-88, there was only a small contraction of deposits in those years and no sign of a "missing money" problem until after FIRREA came along.

Just as the credit crunch did not cause the "missing M2," the credit crunch is not being exacerbated by the "missing M2." While it is likely that increased reinvestment risk depresses small time deposit growth, bank balance sheets show that liquidity is not a constraint on bank lending. Banks in the nation as a whole and banks in regions that have experienced credit crunches have a very large percentage of their assets in cash and government securities, at least by the standards of recent decades. At the end of first quarter 1992, commercial banks had just short of $1 trillion of investments in cash and securities, a substantial portion of which could be converted to loans when their financial condition improves, when loan demand increases, and/or when the perceived regulatory environment returns to "normal."


The growth rate of M2 is not only low but it has been below its forecasted value from 1990 through 1992. Both supply and demand factors are depressing M2 growth. The new procedures established in FIRREA for resolving failed banks and thrifts have increased the reinvestment risk associated with small time deposits, a major component of M2. As a result, demand for M2 is lower; furthermore, traditional forecasting equations do not take reinvestment risk into account and are therefore overpredicting M2. Simultaneously, the money supply process is restricted by the numerous banks that are constrained in their lending activity by their low levels of capital and other manifestations of financial weakness. Traditional money multiplier models assume that bank capital levels will not constrain the deposit creation process.

Corresponding to these effects on bank liabilities, bank assets, and in particular bank loans, have been depressed. The deposit creation process requires bank lending. In response to their low capital levels, many bankers have curtailed lending either voluntarily or because of regulator insistence. Both of these phenomenon are closely related to regional bank conditions because they originated from the large loan losses experienced by banks in economically depressed regions of the United States. Reinvestment risk is greatest at banks that are likely to fail, and the money supply process is most disrupted at banks whose capital levels have been severely reduced by loan losses. A state-by-state analysis of contraction in small time deposits supports the above hypothesis that M2 has been most depressed in the regions afflicted by credit crunches, i.e., the Southwest and the Northeast.

Though related by common origin, certain aspects of these issues are unrelated. The credit crunch is not being exacerbated by weak M2 growth, because no amount of liquidity will solve a shortage of capital. The "missing M2" problem is not a function of the credit crunch as demonstrated by the Texas experience. The Texas credit crunch, which began in 1986, predated the first signs of "missing M2" by nearly five years.

Harvey Rosenblum is Senior Vice President and Director of Research and Robert Clair is Senior Economist and Policy Advisor, Federal Reserve Bank of Dallas. The authors would like to thank Kelly Whealan and Kevin Yeats for excellent research assistance and helpful comments. The views expressed are those of the authors and not necessarily those of the Federal Reserve Bank of Dallas or the Federal Reserve System.

** In the past two quarters, banks in Texas, Oregon and Arizona have reported increased loans outstanding, but further inspection indicates that the growth has been concentrated in a few institutions and is likely the result of asset transfers across state lines by interstate bank holding companies.


1 Duca, John V. (1992), "The Case of the Missing M2," Economic Review, Federal Reserve Bank of Dallas, Second Quarter.

2 Klemme, Kelly and Kenneth J. Robinson (1992), "The Transition to Healthier Banks and Thrifts: Has Money Growth Been Affected?" Financial Industry Issues, Federal Reserve Bank of Dallas, Second Quarter.

3 Hancock, Diana and James A. Wilcox (1992), "The Effects on Bank Assets of Business Conditions and Capital Shortfalls," in Proceedings of the 28th Annual Conference on Bank Structure and Competition, Federal Reserve Bank of Chicago, pp. 502-520.

4 Rosenblum, Harvey (1990), "The Texas Economy: Conditions and Prospects for Recovery," Testimony before the Committee on Banking, Finance and Urban Affairs, U.S. House of Representatives, Field Hearings, Houston, Texas, June 22, 1990.

5 Rosenblum, Harvey (1991), "The Pathology of a Credit Crunch," Southwest Economy, Federal Reserve Bank of Dallas, July/August.

6 Rosenblum, Harvey (1992), "The Macroeconomic Impacts of Bank Regulatory Policies," in Proceedings of the 28th Annual Conference on Bank Structure and Competition, Federal Reserve Bank of Chicago, pp. 434-444.

7 Clair, Robert T. and Kevin J. Yeats (1991), "Bank Capital and its Relationship to the Credit Shortage in Texas," presented to Federal Reserve System Committee on Financial Structure and Regulation, San Antonio, Texas, January 1991.

8 Peek, Joe and Eric S. Rosengren (1992), "The Capital Crunch in New England," New England Economic Review, Federal Reserve Bank of Boston, May/June.
COPYRIGHT 1993 The National Association for Business Economists
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1993 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Rosenblum, Harvey; Clair, Robert T.
Publication:Business Economics
Date:Apr 1, 1993
Previous Article:Health insurance derivatives: the newest application of modern financial risk management.
Next Article:The growth of small firm jobs by state, 1984-88.

Terms of use | Copyright © 2016 Farlex, Inc. | Feedback | For webmasters