Over the course of the current U.S. economic expansion, commercial banks have prospered by significantly increasing loans to businesses and consumers. Indeed, since mid-1994, loans outstanding at U.S. commercial banks have risen almost 29 percent. Eighth District loan growth has been even stronger over the period, rising almost 35 percent.1 And the growth shows little sign of tailing off: Anecdotal reports from bankers around the District indicate that loan demand remains fairly strong.

That said, bankers—especially community bankers—also report that they are having increasing difficulty obtaining deposits, traditionally the major source of loan funding. Indeed, over the last three years, deposits have risen about 19 percent—roughly half the growth of loans—at both U.S. and District banks. Moreover, most of the growth occurred early in the period; deposit growth has been anemic and even negative for many banks in recent quarters. To meet loan demand in the absence of strong deposit growth, District banks have had to turn to both traditional and newer funding sources, namely investment securities, fed funds and borrowings from Federal Home Loan Banks.

All Loaned Up

A strong economy and healthy bank balance sheets can be credited with much of the increase in loans during the past three years. Loans have performed so well lately that banks have been able to dramatically reduce the amount of reserves they set aside to cover possible defaults. And record earnings have led to large increases in bank capitalization, further increasing banks' ability to make loans.

As a result of these favorable developments, the loan to deposit ratio at District banks rose from 73 percent in mid-1994 to 82 percent in mid-1997. The growth in loans has been broadly based, with all major loan categories showing substantial gains over the past three years. In the District, for example, commercial and industrial loans have risen 30.8 percent, real estate loans have jumped 42.2 percent, and consumer loans have increased 18.2 percent.

Deposit growth has failed to keep up with the torrid pace of loan growth. Several explanations have been offered to account for this disparity, particularly the notion that retail depositors are leaving the banking system in search of higher yields—a phenomenon termed disintermediation. Some analysts believe this is just a cyclical trend, and that deposits will come back when retail deposit interest rates start rising. Others, however, believe the slowdown is part of a long-term trend, as small investors become increasingly comfortable with putting their money into the stock market and other assets. On the supply side, a number of banks have grown less aggressive in trying to attract deposits, refusing to compete in CD rate wars with other financial institutions, for example. Some analysts have interpreted this behavior by banks as price discrimination in their funding activities, saying that banks have a strategy of paying low rates on retail deposits, exploiting the small depositors who put a high value on federal deposit insurance, and attracting the rest of their funds in other markets.2

Shedding Securities

Banks do, however, have a number of options besides deposits to fund loan demand. One ready source of liquidity is a bank's investment portfolio. Investment securities and loans are substitutes among bank assets; banks can easily sell or decide not to renew most of the securities they tend to purchase to fund loan demand. Since mid-1994, the securities portfolio at District banks that have average assets of less than $1 billion has shrunk 10 percent.

In addition to shrinking the size of the overall portfolio, these banks have also altered its composition, reducing substantially their holdings of both Treasury securities (–38.4 percent) and mortgage-backed securities (–16.9 percent), and replacing them with other types of government and government-backed securities (+28.5 percent), such as debt securities issued by the Small Business Administration and Fannie Mae and Freddie Mac. These composition changes can be explained by a number of factors, including: a need to fund loan demand with liquid securities (those from the Treasury tend to be the most liquid); a desire to increase portfolio returns (Treasury securities tend to yield less than other types of securities); strength in the real estate sector, which heavily influences the return on many government-backed securities; and a low interest rate environment, which made mortgage-backed securities less attractive since low rates tend to increase prepayment of mortgages, thereby lowering returns.

Borrowing from Peter to Lend to Paul

Banks can also create nondeposit liabilities to fund loans and other assets. One of the more popular sources of liquid funds is the fed funds market, in which banks with excess funds on hand lend them to other banks. The funds are generally sought and received on the same day and must be repaid the next at a rate determined daily in the money market. The bulk of loans made in the fed funds market are designed to meet banks' reserve requirements, though many banks also borrow longer-term funds from this market (called term fed funds) to meet loan demand.

Another source of ready liabilities that is available to an increasing number of banks is the Federal Home Loan Bank (FHLB) System. Banks first became eligible to join this previous thrift-only, government-sponsored enterprise in 1989. In the past several years, commercial bank membership has soared, with commercial banks currently making up about two-thirds of the system's 6,000 plus members. Through the 12 FHLBs, member institutions can obtain either overnight, short-term (one week to six months) or longer-term loans—all of which are called advances—to meet a variety of funding needs. Long-term advances, available in maturities of up to 20 years, are used by many banks to fund residential mortgages. Advances are available at both fixed and adjustable rates and are almost always cheaper than other wholesale funding sources.3

District banks have clearly become big fans of the FHLB System: District-wide, almost 50 percent of them are members (see table).4 Within the District's borders, FHLB membership ranges from 37.9 percent in Illinois to 71.9 percent in Indiana. Overall, almost two-thirds of District member banks had outstanding FHLB advances as of June 30, 1997, and in Kentucky and Tennessee, all District member banks were borrowing at mid-year 1997. In several District states, especially Mississippi (34.7 percent) and Indiana (25.7 percent), FHLB advances make up a significant portion of banks' total nondeposit liabilities, indicating how much banks—especially community banks that have experienced actual declines in deposits—have come to rely on other funding sources.

The bulk of FHLB loans made to District banks are longer term in nature, with almost three-quarters of total District borrowings carrying maturities of a year or more. There is quite a bit of variance across states, however. In Missouri and Indiana, for example, more than 50 percent of borrowings are for terms of less than one year, compared with the District-wide average of 27.3 percent.

Liquidity Trap?

To date, District bankers have been able to satisfy a surprisingly sustained level of loan demand, despite a significant amount of disintermediation. Although many banks continue to maintain flexibility by holding large amounts of liquid securities that can be sold to meet loan demand, an increasing number are also finding that they can gain "instant liquidity" through wholesale funding sources, like the Federal Home Loan Banks.

Although this strategy has certainly helped banks post record profit levels in recent years, it's not without risks. So far, loans have performed extremely well, and loan losses have been minimal. However, if credit quality deteriorates unexpectedly, and loan portfolio returns start to fall, the burden of interest payments from FHLB advances and other borrowings could pinch earnings significantly. In a worst-case scenario, a significant downgrading of a bank by supervisors could force it to immediately pay back all FHLB advances and halt all further borrowing, potentially causing a liquidity crisis. Although the industry is exceedingly well-capitalized at this time, and liquidity is readily available, there may be more than a few District banks that will wish they hadn't made that last loan when the next economic downturn arrives.

Thomas A. Pollmann provided research assistance.

Endnotes

1. District loan and deposit figures for 1994 through 1997 exclude the former Boatmen's banks, which were merged into NationsBank of Charlotte, N.C., in mid-June 1997. [back to text]
2. See Gilbert (1997). [back to text]
3. See Newkirk (1996) and Faulstitch (1997) for further detail on the operation of FHLBs, membership requirements and program offerings. [back to text]
4. Banks in Missouri are eligible to join the FHLB of Des Moines; banks in Kentucky and Tennessee belong to the FHLB of Cincinnati; banks in Arkansas and Mississippi join the FHLB of Dallas; banks in Illinois belong to the FHLB of Chicago; and banks in Indiana join the FHLB of Indianapolis. [back to text]

References

Faulstitch, James R. "Banks Get More Adept at Using Home Loan Bank Funding," ABA Banking Journal (June 1997), pp. 79-80.

Gilbert, R. Alton. "Banks Profit From Low Rates on Time and Savings Deposits," Monetary Trends, Federal Reserve Bank of St. Louis (June 1997).

Newkirk, Kristine M. "Discovering a New Funding Partner," Independent Banker (June 1996), pp. 17-22.

Rose, Peter S. Commercial Bank Management, 3rd edition (Chicago: Richard D. Irwin, 1996), chapters 8, 9 and 12.