Full text of Economic Commentary (Federal Reserve Bank of Cleveland) : Commercial Banks' Borrowing from the Federal Home Loan Bank, July 2002

View original document

The full text on this page is automatically extracted from the file linked above and may contain errors and inconsistencies.

July 2002

Federal Reserve Bank of Cleveland

Commercial Banks’Borrowing from the
Federal Home Loan Banks
by James B. Thomson

nactment of the Financial Institutions Reform, Recovery, and Enforcement Act of 1989 was a watershed event
in the financial system policy of the federal government. It represented the first
concrete effort to resolve the 1980s thrift
debacle and to make meaningful financial system and regulatory reforms. It
also changed the role of the Federal
Home Loan Banks by opening them up
to commercial banks.
The Federal Home Loan Banks
(FHLBs) function as special lending
facilities for the housing finance industry. These government-sponsored enterprises increase the liquidity of mortgage
markets by making advances (loans)
against member institutions’ mortgage
portfolios. For the better part of six
decades, membership in the FHLBs
was limited to institutions specializing
in housing finance—largely savings
associations. This limitation established
a strong connection between FHLB
lending and housing finance. The wider
range of assets commercial banks can
hold and the fungibility (interchangeability) of advances weakens the connection between housing finance and
FHLB lending to banks. So, while it
reaffirmed the FHLBs’ housing finance
mission, the Financial Institutions
Reform, Recovery, and Enforcement Act
widened the scope of lending that was
supported, albeit indirectly, by FHLB
advances. A decade later, the GrammLeach-Bliley Act of 1999 contained a
provision that formally extended FHLB
advances to support the small business
and small agricultural loan portfolios
of community financial institutions
(then defined as those with assets of
$500 million or less). In this provision
ISSN 0428-1276

of Gramm-Leach-Bliley, Congress
amended the FHLBs’ mission to include
providing broader liquidity support for
community banks and thrifts.
Today, commercial banks are an important constituency of the FHLBs. Starting
from zero at the beginning of 1990, commercial bank membership had grown to
5,786 (73 percent of all FHLB members)
by December 31, 2001. Moreover, at the
end of 2001, three-fourths of all FHLB
advances went to commercial banks,
which collectively borrowed $198 billion
(nearly 42 percent of the total dollar volume of FHLB advances).
Commercial banks’ prominence as
members of the FHLB system and their
increased use of FHLB advances for
funding is not surprising. After all,
throughout the 1990s the attractive
returns offered by the stock market
and competition from other financial
institutions made it difficult for banks to
attract and retain deposits. Moreover, the
stability of FHLB advances as a funding
source and the below-market lending rate
make borrowing from them attractive.
Given the cost and funding advantages
that FHLBs enjoy as governmentsponsored enterprises (see the Congressional Budget Office studies in the
recommended reading list) and the
increasing share of FHLB assets represented by advances to banks, I am
interested in the effects on the financial
system of changes in access to FHLB
funding. Is the borrowing pattern consistent with the FHLBs’ traditional housing
finance mission? With the FHLBs’
providing liquidity support for community banks? With some other objective?
I shed light on these questions by

Since 1990, when commercial banks
were first eligible to join the Federal
Home Loan Bank System, they have
become an important constituency of
the FHLBs. Currently, seven out of 10
banks are members, and nearly half of
all banks have advances outstanding.
Given the wide range of activities that
commercial banks can engage in, this
Commentary asks whether FHLB
lending to them is consistent with
their traditional housing finance
mission, with the Gramm-LeachBliley extension of their mission to
provide liquidity support to community banks, or with both.

examining the characteristics of banks’
borrowing from FHLBs, including
the distribution of borrowing across
different sizes of banks. I find evidence
of a positive relationship between a
bank’s reliance on FHLB advances for
funding and the share of assets invested
in housing-related credits. Although
community banks are less likely than
their larger brethren to have outstanding
FHLB advances, the data suggest that
the FHLBs are an important backup
liquidity source for community banks.

■

Support of Housing Finance

In opening up the FHLBs to commercial
banks, Congress recognized that the
thrift industry’s post-debacle role in
financial markets would be permanently
reduced. Failures during the 1980s and
early 1990s, combined with ongoing
industry consolidation over the last
decade, brought the number of savings
institutions down from a peak of 3,677
in 1986 to 1,533 at the end of 2001.

Moreover, the thrift industry’s share of
total credit intermediated by depository
institutions fell from 34 percent in 1986
to less than 17 percent at the end of
2001. On the flip side of the coin, the
commercial banking industry emerged
as an important housing finance lender.
At the end of 1990, banks held nearly
$399 billion in home mortgages, a
number that rose to more than $966 billion by the end of 2001. In contrast,
savings associations’ mortgage portfolios showed anemic growth, rising
just 10 percent (from $543 billion to
$598 billion) over the same period.
At the margin, the ability to pledge
mortgages as collateral for FHLB
advances should increase the attractiveness of investing in these assets. After
all, this collateralization option
increases the liquidity of mortgage
portfolios and reduces the cost of funding additional mortgages, especially
when advances are made at belowmarket rates of interest. Giving banks
the option of using housing-related
assets as collateral against advances is
consistent—in principle—with the
FHLBs’ housing finance mission. But
is lending to banks also consistent with
this mission in practice?
To answer this question, I compare the
housing-related assets held by FHLB
members against those held by nonmember commercial banks as of
June 30, 2001. I use data from the
June 2001 Call Report exclusively
because banks did not report FHLB
advances as a separate item before
2001, and some data of interest are
reported once a year, in June. I also look
for differences in portfolio shares of
these assets, comparing FHLB members
that have advances outstanding with
members that choose not to borrow. If
access to advances promotes housing
finance lending by banks, one would
expect to find higher concentrations of
housing finance loans for FHLB member banks than for nonmembers, and
for FHLB borrowers than for nonborrowers. For the most part, I find exactly
that. The average FHLB member’s
housing-related assets make up
19 percent of total assets, while the
average nonmember bank holds roughly
14 percent of its assets in housingrelated loans. Moreover, banks with
FHLB advances outstanding invest
40 percent more in housing-related
assets than do nonborrowing banks.

Another interesting question: Is greater
reliance on FHLB advances associated
with a higher portfolio share for housingrelated assets? As noted earlier, for
commercial banks the fungibility of
funds, coupled with asset powers that are
wider than thrifts’, weakens the link
between FHLB advances and housing
finance. If access to FHLB advances
increases banks’ holdings of housingrelated assets, we should observe a
positive relationship between a bank’s
reliance on FHLB advances and the
share of its assets in housing finance
loans. Regression analysis shows a
significant relationship between FHLB
advances and housing-related loans for
banks: For a 1 percent increase in the
ratio of advances to assets, the share
of assets invested in housing loans
increases 0.74 percent. Overall, the
evidence suggests that admitting commercial banks to the FHLB system is
consistent in practice with the system’s
traditional housing finance mission.

■

Support of Community
Banks

For commercial banks, much of the
value associated with FHLB membership comes from access to FHLB
advances—a low-cost (subsidized),
stable source of funding. Advances give
banks an additional tool for managing
their liabilities so as to minimize interest
costs and increase liquidity. Advances
potentially reduce a bank’s interest costs
by increasing its flexibility in funding its
assets. Banks are likely to borrow from
the FHLB whenever the marginal cost
of raising additional deposits or the
interest costs of other nondeposit funding sources exceed the cost of advances.
In addition, access to the FHLB credit
facility may allow a bank to supply
credit to customers that might otherwise
go unfunded; credit access may also
help a bank cover an unexpected shortterm deposit outflow (or other temporary losses of funds), thereby avoiding
potentially costly short-term borrowing
or asset liquidation.
By expanding the collateral that community financial institutions can pledge
against advances, this provision of the
Gramm-Leach-Bliley Act represents a
potentially important expansion of the
FHLBs’ mission by providing additional
support for community financial institutions. By making a larger part of their
balance sheets eligible as collateral (in
comparison with larger-sized institutions),
the Act substantially increases these

institutions’ ability to use FHLB
advances for funding assets and asset
growth. In June 2001, community banks
(those with total assets not exceeding
$500 million) held nearly 19 percent of
their assets in housing-related loans and
investments that could be used as collateral for FHLB borrowings. Under the
terms of Gramm-Leach-Bliley, the
additional 19 percent of community bank
assets held in the form of small business
and small agricultural credits also qualifies as collateral.
A recent study by Craig and Thomson
(2001) questions the need to expand the
pool of collateral that community banks
can pledge against advances. The
authors find no evidence that these institutions need greater access to FHLB
advances to meet their customers’ credit
needs. Valid concerns have been raised
in rural areas about the possibility that
shrinking deposit bases could cause
community banks to ration credit to
small businesses and farms. But Craig
and Thomson find no evidence that
community banks’ access to funding has
damped credit access in rural areas. On
the contrary, they find that the deposit
base of the average rural community
bank is more than sufficient to fund its
customers’ credit needs.
Although this study looks at whether
community banks need expanded access
to FHLB advances to fund assets, it does
not consider the role FHLB advances
might play as a source of liquidity.
With outstanding advances to commercial banks approaching $198 billion,
it is natural to ask whether, in addition
to funding housing finance assets,
borrowing from the FHLBs is an
important source of liquidity for community banks. One way to get at this
question is to look at how borrowing is
distributed across different size classes
of banks. For that purpose, I break the
industry down into three size classifications: large banks (total assets greater
than $10 billion), medium-sized banks
(total assets greater than $500 million
and less than or equal to $10 billion),
and community banks (total assets no
greater than $500 million).
Table 1 shows that large banks account
for about $104 billion (49 percent) of
total dollar advances outstanding to
commercial banks. Medium-sized banks
account for more than $74 billion,
representing 35 percent of the total
outstanding to banks. Community banks,
which borrowed less than $34 billion

■

TABLE 1 FHLB BORROWINGS BY BANK SIZE
Small banks
(total assets less
than $500 million)

Number of borrowers
Borrowers as s percent of total banks
Total advances (millions of dollars)
Percent of advances to banks
Average advance (millions of dollars)
Advances as a percent of assets

3,549
45.68
33,520
15.83
9
6.41

Medium-sized banks Large banks
(total assets $500–
$10 billion)

(total assets more
than $10 billion)

528
73.54
74,389
35.13
141
8.42

50
65.79
103,866
49.05
2,077
6.51

SOURCE: Federal Financial Institutions Examination Council, Reports of Condition and Income, June 2001.

from the FHLBs, account for less than
16 percent of total bank advances. A
similar picture emerges if we look at the
average size of advances outstanding—
and the percentage of banks—in each
category. Less than half of small banks
had advances from the FHLBs; for those
that did borrow, the mean size of the
advance was $9 million. Nearly threequarters of medium-sized banks and
almost two-thirds of large banks borrowed from the FHLBs. The mean
advance for borrower banks was $141
million for medium-sized institutions
and $2,077 million for large ones.
Clearly, medium-sized and large banks
have been the most active users of
advances.
The distribution of advances by bank
size does not necessarily mean that
FHLBs are not an important source of
liquidity for community banks. After all,
large banks hold 66 percent of total
commercial bank assets, while mediumsized banks hold 20 percent and small
ones hold 14 percent. Table 1 shows
that, controlling for asset size, the use of
advances looks very similar for large
banks and for small ones. For banks
with borrowings from the FHLBs, the
ratio of advances to assets is 6.41 percent for small banks and 6.51 percent
for large banks—and the difference is
not statistically significant. However, for
medium-sized banks with outstanding
borrowings, the ratio of advances to
assets is 8.42 percent, which is significantly higher than for either of the other
two sizes. Overall, banks that borrow
from the FHLBs do not show wide,
size-related disparities in terms of
reliance on them for funding.
Finally, this analysis does not consider
the value for banks of the option to borrow from the FHLBs, that is, the
FHLBs’ role as a backup source of liquidity. For small banks without access to

capital markets, this option to borrow
may be particularly valuable, whether or
not it is ever exercised. Remarkably,
while about one-third of FHLB member
banks had no outstanding advances,
nearly 96 percent of these nonborrowers
are small banks. Because FHLB membership is voluntary, it must be the case
that for these nonborrowing banks the
value of the option to borrow exceeds
the costs of membership. A reasonable
inference is that banks, especially small
ones, value access to FHLB advances as
a backup source of liquidity.

■

Conclusions

FHLBs have become an important
source of funding and liquidity for
commercial banks. Seven out of 10
banks are members, and nearly half of
all banks have advances outstanding.
In fact, more commercial banks are
members of the FHLB system than of
the Federal Reserve System. FHLB
advances currently fund more than
3 percent of the banking system’s assets.
Moreover, as noted earlier, banks constitute the largest single membership group
in the FHLBs and are the second-largest
group of borrowers.
I find little evidence that commercial
banks’ admission to the FHLB system
and their increasing prominence as
members and borrowers is inconsistent
with the FHLBs’ traditional housingrelated mission. The FHLB borrowing
provision of the Gramm-Leach-Bliley
Act increases the FHLBs’ capacity for
providing liquidity to community banks.
Today, membership and borrowing
patterns across banks are largely consistent with small banks’ use of FHLBs as
a backup source of liquidity. It remains to
be seen what effect, if any, the expanded
collateral provision of the act will have
on community banking activities in
years to come.

Full text of Economic Commentary (Federal Reserve Bank of Cleveland) : Commercial Banks' Borrowing from the Federal Home Loan Bank, July 2002

FRASER