Full text of Review (Federal Reserve Bank of Dallas) : May 1977

View original document

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

FEDERAL RESERVE BANK OF DALLAS

Offshore Oil RigsOversupply Despite World Energy Shortage

Bank Structure and PerformanceSurvey of Empirical Findings
On the Cost of Checking Accounts

This publication was digitized and made available by the Federal Reserve Bank of Dallas' Historical Library (FedHistory@dal.frb.org)

Offshore Oil Rigs-

Oversupply Despite World Energy Shortage
By Stephen L. Gardner

Offshore drilling rigs are in surplus. Because of the
oversupply, day rates for milling paid some rig owners fell as much as 50 percent in 1975 and, at the
same time, the large backlog of rigs on ~rder at shipYards began to run off. By the end of this year, back
orders are expected to have been largely eliminated.
Paradoxically, the need to dlill seems great.
Energy prices are at all-time highs, and many parts
of the country experienced severe energy shortages
last winter. Nor is there a lack of offshore oil to be
found. Known offshore fields contain about a fifth
of the world's discovered reserves of some 700 billion barrels of oil. And offshore exploration has not
gone on for as long or been cal'ried out in most
areas as intensively as onshore exploration.
The current glut in offshore oil rigs is similar in
SOIne respects to earlier cyclical downturns. But new
factors have made the current surplus of ligs much

Even though demand slackened in 1975,
World supply of offshore rigs kept growing
because of building lags and order backlogs
330 ____________________________________

worse than ever before. Because of the great size of
the current oversupply, it will probably take longer
than in past cycles for price and supply conditions
to return to normal.
Nature of the industry
Offshore drilling originated along the Gulf Coast of
the United States in the late 1940's and has continued to be primarily an American business. Most
of the world's fleet of offshore rigs was produced in
U.S. yards near the Gulf Coast, and it is mainly
owned by U.S. contractors. But foreign builders have
become increasingly active, resulting in a decline in
the share of U.s. production from about half of the
total five years ago to about a third today.
The nature of the offshore rig industry makes it
subject to large cyclical fluctuations in production.
Because of the difficulty of anticipating future needs,
current profitability and the immediate outlook for
prices and drilling strongly influence plans to order
new rigs. But while drilling contractors are unable
to anticipate firmly industry needs beyond a year or
two the custom designing and building of an offsho;e rig may take several years. In addition, when
demand is strong, construction back orders may not
be filled for several years. Once built, however, a rig
may be in service for over 15 years.

300 _

Because increasing the supply of rigs
takes time, those ordering new rigs usually do not adequately foresee the effect
on total supply of the rigs then being built.
The result is often a period of oversupply
like the current one, accompanied by
extremely low day rates for drilling.

270 _

240 _

210 _

180

~Oie:

--,r------r----r-----,---ir
1974

1975

1976

1977

Excludes rigs owned by government-operated firms.
OURCE: Offshore Rig Data Services.

..........
lte .

VIew I May 1977

A short supply of rigs-or of a particular type of
rig-relative to industry exploration nee~s brings
contractors a high rate of return. These high rates
tempt drilling contractors to order more rigs. But
because increasing the supply of rigs takes time,
those ordering new rigs usually do not adequately
foresee the effect on total supply of the rigs then
being built. The result is often a period of oversupply
1

Types of offshore rigs
The first mobile U.S. offshore drilling rig
was built in 1949, as the industry sought to
:find and tap new reserves of hydrocarbons
off the Gulf Coast. Since then, a progression
of innovations has enabled oil companies to
extend their search to geologic structures at
greater and greater ocean depths.
The :first true offshore rig, capable of
working relatively shallow water in the mild
U.S. coastal environment, was a submersible type. This type can be floated to location and then ballasted in place to rest on
the ocean bottom for support while its deck
rises above the water. But submersible rigs
are limited to operating in ocean depths of
less than about 100 feet, and none have
been built since the early 1960's. Though
many of these rigs are still in use, their
numbers have been dwindling as units are
retired or lost through accident.
To extend the search for oil to greater
water depths, the industry had to develop
other types of offshore rigs. One was the
jackup rig, which :first came into use in
1953. Decks on these rigs are supported
above the ocean by long, spindly legs. When
the long legs are jacked up and the deck
is used as a hull, a rig can be towed to a
new site. But they are more awkward and
expensive to move long distances than are
other types. And although recent varieties
stand as tall as 35 stories, some areas of
exploration are in depths beyond their capability. Fully equipped, these are the least
expensive of the modern rigs to build-costing $22 million to $25 million.

like the current one, accompanied by extremely low
day rates for drilling. In the current cycle, rates for
a particular type of semisubmersible offshore rig in
the North Sea fell from $40,000 a day at the beginning of 1975 to little more than half that a year later.
Owners do not :find it profitable to take rigs out
of service, even with low day rates, because they are
given little relief from operating costs. When not
used, rigs must still have constant maintenance
attention to prevent deterioration, so savings on crew
costs are not great. And most other costs have
already been sunk in the heavy capital expenditures
required to build and equip a rig. Therefore, any
2

The drill ship also was first launched in
1953. It resembles a conventional ship but
has a drilling rig positioned in its center.
This type of offshore rig has the greatest
mobility. Most have a complicated anchoring system to ensure stable drilling. The
early 1970's added dynamic positioning to
the ship's design, through which computerdirected propellers placed along the bottom
of the vessel hold the rig on site. Mobility
and stability come at a cost, however. Ordinary drill ships cost $35 million to $45 million, and dynamic positioning doubles the
expense.
In 1963, a decade after the :first jackups
and drill ships appeared, the :first true semisubmersible offshore rig was launched. The
deck of this type of rig is held above pontoons by open support work. Modern varieties are quite large-some with decks as
large as football fields-and often have displacement equal to a large freighter. Most
are towed to site, but some models now
being built are self-propelled. When they
are ready to drill, they are partly submerged so that waves fall through the open
support work instead of crashing against
their pontoons. These rigs are also held in
place with complicated anchoring systems.
Semisubmersible rigs are particularly suitable for exploration in deep water and under
difficult climatic conditions. They are about
as expensive to build as drill ships-costing $35 million to $50 million.

return that helps to meet even a portion of the debt
service justifies continued operation of the rig.
Because of their specialized character, a period of
oversupply in offshore rigs would ordinarily tend to
last longer than for onshore rigs. Offshore rigs are
used almost exclusively for drilling exploration wells.
Once an offshore field has been found and enough
wells drilled and tested to satisfy oil companies and
their sources of financing that the :find is commercial, the mobile offshore rigs are moved away to neW
exploration sites. Production wells are then drilled
and serviced from permanent platforms. Thus, an
excess supply of offshore rigs cannot be absorbed in

Production of offshore rigs characterized by large cyclical fluctuations
RIGS
FROM YARDS
70 - -DELIVERED
- - - - - -________________________________________________________________
__

60 -

50 40 30 -

20 10 -

--~

____

'49

~~~

'52

__L_
'55

'58

'61

'64

'67

'7 0

'73

'76

'78

NOTE: 1977 and 1978 based on current backlog of orders.
SOURCE: Offshore Rig Data Services.

the drilling of production wells as new discoveries
are made, as would be the case with an oversupply
of onshore rigs.
Sources of current oversupply
he 1973-75 boom in offshore rigs was larger and
sted longer than the two earlier booms in the indusr~'s history. The order bacldog peaked in 1975,
WIth 157 rigs valued at some $4 billion on order.
~hat represented enough offshore rigs to increase
he world's fleet by more than 50 percent and foreS adowed the oversupply soon to follow.
th A.. number of unusual factors were responsible for
e Intensity of the boom-and-bust in orders and
bO~st~uction. First, and perhaps most important, the
oU~di~g period was preceded and sustained by great
ftImIsm about the outlook for exploration drilling.
ino a larg.e extent, this optimism was based on the
ase In the world price of oil exacted by the
go C cartel in 1973. Oil company executives and
o Vernment officials reasoned higher prices would
~en ~lling opportunities by malcing previously
111akinr~ma! areas attractive for exploration, as well as
toa g It economically possible to extend exploration
new areas.

OPE

tr A. second unusual factor was that many conactors ordered offshore rigs on pure speculation.
lle .
"lew I May 1977

Usually, drillers have a contract for a rig's use in
hand when they order it, assuring at least its initial
employment. If such precautions are widely taken,
much overbuilding can be prevented. But with great
optimism prevailing, many contractors and their
backers were willing to order rigs without contract.
Governments encouraged them by guaranteeing
financing and offering generous tax incentives for
rig construction. In the United States, for example,
a change in the Merchant Marine Act in the early
1970's opened the way for Government-guaranteed
financing, resulting in a significant easing in the
terms on bank loans for construction. Of the rigs
on order at the beginning of 1976, over 60 percent
were without contract. Many of these may not find
profitable employment anytime soon.
Foreign owners, particularly Norwegians, took the
lead in placing orders. Even with ownership of less
than 5 percent of the world fleet in 1975, Norwegians
accounted for nearly a third of the new rig orders.
Norwegian tanker owners had made large profits in
the early 1970's that they reinvested to avoid high
domestic taxes. Because tankers were coming into
surplus, oil rigs appeared to be a better investment.
And because semisubmersible offshore rigs were
operating in the North Sea, Norwegians had a definite preference for them.
3

Even on the basis of the soberest of calculations,
however, the demand for offshore rigs has not materialized as expected. High world oil prices, which
seemed so certain a stimulus to exploration, also
helped to plunge economies around the world into
recession. Governments were concerned that policies
encouraging greater energy self-sufficiency might
slow economic recovery. In the United States, for
example, gas decontrol measures were frustrated.
And though controlled prices for new crude oil were
allowed to rise, relatively low prices on old oil held
back profits.
Exploration, with its huge risks, is financed by the
oil companies out of profits. Only after a field has
been found and estimates of oil in place are made
is outside financing from banks and other lenders
possible. More than half of the exploration budgets
of oil companies is spent offshore. From 1970 to
1974, partly because of the increase in profits arising
from higher world oil prices, offshore budgets of the
companies about doubled to around $10 billion. But
higher taxes imposed by oil-producing countries, as
well as relatively low U.S. prices, helped to cause
profits to drop 20 to 25 percent in 1975. Exploration
budgets increased only 8 percent that year, diminishing hopes for a sustained exploration boom.
Finally, Federal leasing of offshore oil acreage
has not been as extensive as generally anticipated,
largely because of the impact of environmental concerns. After the Santa Barbara oil spill in 1969, the
Federal Government drastically cut back its leasing.

This resulted in a sharp drop in oil wells drilled in
U.S. waters. Offshore drilling picked up somewhat
following a resumption of leasing in 1972. But altogether, only 3 percent of the U.S. offshore acreage
has so far been leased. Although the Nixon administration promised to boost offshore lease sales to
10 million acres a year-about as much as the cumulative total to date-leasing at this pace was not
practical. Also, Federal courts have recently raised
doubts, on environmental grounds, about the legality
of over $1 billion of lease sales off the Atlantic Coast.

The companies most affected by the scarcity of contracts and decline in contract
terms are the new arrivals in the industry.
Companies established since 1972-particularly by Norwegians-hold more than
70 percent of the rigs without contracts
for use.

Effect on the industry
The companies most affected by the scarcity of contracts and decline in contract terms are the new
arrivals in the industry. Companies established
since 1972-particularly by Norwegians-hold more
than 70 percent of the rigs without contracts for use.
Established drilling contractors have advantages

Over newer rivals. Experience means a lot in the complicated and dangerous business of offshore drilling,
and oil companies are willing to pay a premium for
it. This explains why the larger American contractors
have few offshore rigs-mostly drill ships-without
contracts. Then too, these contractors operate relatiVely more of the older rigs whose capital costs
have already been paid and, consequently, are in a
better position to maintain profitable operations
even at reduced day rates. But a number of smaller
U.S. companies are less fortunate, and some are
encountering difficulty.
The demand for so-called jackup rigs is fairly
active, and the backlog of these rigs should be
absorbed within a year or two. But because overbuilding has been particularly concentrated in semisubmersible offshore rigs and drill ships, problems
for these deep-drilling rigs might extend into the
next decade. Much will depend on growth of exploration and on the size and location of discoveries.
Another discovery in a difficult-to-drill area on the
scale of the North Sea could quickly absorb many
of the idle semisubmersible rigs.
Despite the current surplus, however, the longterm outlook for the use of offshore rigs is relatively
bright. The offshore potential is huge, as continental
~argins cover some 30 million square miles-exceed~ng the total of all known sedimentary basins on
and. Furthermore, in contrast to land basins, little
?f this vast offshore area can be dismissed as a highly
llnprobable source of oil or gas.

neView
' I May 1977

New member bank

University National Bank, San Antonio, Texas, a newly organized institution
located in the territory served by the San Antonio Branch of the Federal Reserve
Bank of Dallas, opened for business April 28, 1977, as a member of the Federal
Reserve System. The new member bank opened with capital of $600,000, surplus
of $600,000, and undivided profits of $300,000. The officers are: Porter Loring, Jr.,
Chairman of the Board; Ralph Langley, Vice Chairman of the Board; Harold R.
Beckwith, President; J. G. Henderson, Executive Vice President; and Larry G.
Meek, Vice President and Cashier.

Bank Structure and Performance-

Survey of Empirical Findings
On the Cost of Checking Accounts
By D. K. Osborne*

Public policy t oward competition in banking has
assumed a new dimension since 1963, when the
Supreme Court ruled that a merger between the
Philadelphia National Bank and the Girard Trust
Corn Exchange Bank violated Section 7 of the
Clayton Antitrust Act. This section prohibits all
mergers whose effect "may be substantially to lessen
competition or tend to create a monopoly" in any
"line of commerce" in any area of the country. Its
application to bank mergers, which began with this
rUling, ext ended the scope of public policy beyond
the traditional concerns with solvency and liquidity
of banks and the "convenience and needs of the com~Unity" to competitive aspects. Moreover, later rul~gs of the Supreme Court applied to banking not
Just Section 7 of the Clayton Act but the entire body
of antitrust law.
This change in banking policy soon stimulated
empirical research into banking competition, and the
Past dozen years have yielded a substantial flow of
research findings. The studies mainly follow the
Structure-Conduct-Performance Approach developed by economists working in the field of industrial
~hganization: the structure of a market determines
.e conduct of the participants, which in turn determInes how well the market performs. 1 To use this
approach one must identify a banking market, mea~~'e its structure, and try to find a relation between
th s .measure and some aspect of performance, say
th e Interest rate on real estate loans or the cost to
1 e consumer of checking account services. But martets and market structures are theoretical con~ructs, the observable counterparts of which are not
WayS obvious. This presents the researcher with
a nUmber of difficult problems.
e First, what is the relevant market? Elementary
conomic theory seems to suggest an answer: as the

;---'I'he author is grateful to Adrian W. Throop for pointing
Out a conceptual error in a previous draft.

~ctu~lJy, the "conduct" member of this trio-which

hscrlbes, for example, how firms set their prices and
~knge them- is usually neglected, as most researchers
n e It fo r granted that a competit ive structure is
ecessary and sufficient for satisfactory performance.

ne'View I May 1977

rational behavior of buyers and sellers will tend to
keep the price of a given product everywhere equal
in a given market, it follows that where prices are
different the markets are different. But this proposition ignores the cost of searching for the best offer,
and in banking markets this cost might well exceed
the payoff. Furthermore, it is a proposition about
equilibrium, not actual prices. Hence, no matter how
good the approximation of a theoretical banking
market, prices will probably differ from bank to bank
within it. Consequently, equality of prices cannot be
a criterion of actual markets.

Studies of banking competition mainly
follow the Structure-Conduct-Performance Approach developed by economists
working in the field of industrial organization: the structure of a market determines the conduct of the participants,
which in turn determines how well the
market performs.

In fact there is no simple economically defensible
criterion for delineating banking markets, and in its
absence the practice has developed of separating
them by political boundaries. Most investigators
assume that towns, cities, counties, and standard
metropolitan statistical areas (SMSA's) constitute
banking markets. Such an assumption is understandable, for it sidesteps a vexing conceptual problem,
but obviously it begs what is usually the main question in public policy proceedings. In the Philadelphia
National Bank case, for example, the question was
whether the relevant market included New York
City and much of the northeastern United States, as
argued by the defense and held by the lower court, or
only a four-county area contiguous to Philadelphia,
as argued by the Justice Department and concluded
by the Supreme Court. Clearly, it sheds no light on
such issues to assume that banking markets coincide
with local political units.
7

A second problem is measuring the structure of the
banking market. By market structure we mean those
characteristics that are thought to affect the forms
and strength of competition. One such characteristic
is the number of sellers: generally it is believed that
the more numerous the sellers, the stronger the competition between them. Other obvious structural
characteristics are the number of buyers, the amount
of heterogeneity in the products, the cost and availability of information, and the conditions of entry for
new firms. Clearly, market structures differ in a number of dimensions. But the common practice in banking studies is to represent structure unidimensionally,
usually by the so-called concentration ratio that
measures the percentage of sales supplied by the
largest sellers.2

With respect to markets, structure, and
prices, empirical studies of banking competition are a series of compromises
between theory and tractability, and their
findings constitute persuasive evidence
only so far as the compromises are
acceptable.
The measurement of prices can present yet a
third problem. Not only do prices differ within markets, but often within firms as well according to the
amount purchased. It is remarkable how many familiar goods and services are priced in part according
to the quantity purchased-electricity and other
public utility services, most grocery items with their
"large economy sizes," transportation, and, in particular, checking account services. But the StructureConduct-Performance Approach does not lend itself
well to analyzing arrays of prices, so researchers have
tended to resort to the use of one-dimensional proxies of bank price schedules.
So with respect to markets, structure, and prices,
empirical studies of banking competition are a series
of compromises between theory and tractability, and
their findings constitute persuasive evidence only
so far as the compromises are acceptable. In this
article we survey the published studies of the relation between market structure and the cost to consumers of checking account services. These studies,
with one exception, describe banking markets in
terms of political or statistical units, and all of them
2. This and other commonly used structural measures are
defined in the accompanying appendix.
8

represent market structure one-dimensionally. But
even accepting these compromises, the price proxies
are so demonstrably far from actual prices that the
findings cannot be accepted as good evidence.
Price schedules for checking accounts
The depositor's monthly cost for a checking account,
or in other words its price, consists of an explicit service charge and an implicit opportunity cost. The
explicit charge usually varies with the minimum (or
in some cases the average) monthly balance in the
account and the monthly number of checks written
on it. The implicit cost is equal to the interest that
would have been earned had the minimum balance
been invested in interest-earning assets. For the typical "regular" checking account, the explicit part of
the price is zero if the minimum balance is sufficiently
large but varies with the number of checks (and in
some cases contains a fixed service charge) if the balance is smaller. The required minimum balance and
the charge per check vary greatly from bank to bank.
For example, a bank in the Washington, D.C.,
area-call it Bank A-offers a regular checking
account without explicit service charge to those
whose balances do not fall below $300. The bank
levies a service charge of 15 cents per check on
accounts in which balances are below $300. Those
who maintain a $300 minimum balance do not, of
course, get free checking, for they bear an implicit
monthly cost of $300 times the monthly interest rate
on funds as liquid as demand deposits. If, for ease
of computation, we assume that this interest rate is
a relatively high 0.5 percent per month, then the
implict cost of the $300 minimum balance is $1.50
per month. Those who write fewer than ten checks
per month minimize their costs by keeping no minimum balance and paying the service charge of 15
cents per check. Those who write more than ten
checks per month minimize costs by keeping a $300
minimum balance and "paying" the implicit cost of
$1.50. Hence Bank A's price is 15 cents per check for
ten or fewer checks per month and $1.50 per month
for more than ten checks.
While Bank A has a single price for a given number of checks, it does not have a single price for a
checking account. The bank is quite typical in this
respect but not necessarily in respect to the precise
manner in which its price varies with the number of
checks. Another Washington bank-Bank B-provides six "free" checks for accounts with a minimUJll
balance of $200 and three additional "free" checks
for each additional $100 minimum balance; it
charges 5 cents for each check beyond those numberS
and $1.50 per month if the balance falls below $200.

Price Schedules for Checks at Two Banks
PRICE PER MONTH (DOLLARS)
3.00 - - - - - - - - - - - - - - - - - -

2.50 BANK B

2.00-

#
###

##
##
##

••••••••••••••••••••••••••••••••••~....
..""."'" :

1.70

1.50 -

BANK A

.50-

12
16
20
24
CHECKS PER MONTH

Still assuming that the opportunity cost is 0.5
percent, the implicit cost on a $200 minimum balance is $1.00 per month, which is less than the $1.50
lUonthly charge for a smaller balance. So all customers
of Bank B should keep the $200 minimum balance.
:ach additional $100 in the balance implicitly costs
ocents per month while saving only the 15 cents
corresponding to the three additional "free" checks
Which would have cost 5 cents each; therefore, no '
~hstomer should keep a minimum balance larger
an $200. Hence Bank B's price is $1 per month
plus 5 cents for every check in excess of six per
lUonth. The accompanying diagram gives the price
~chedule of Bank B and of Bank A as well, illustratIng the difference in the way the price varies with the
nUlUber of checks at these two banks.

l' Now suppose that every bank in a certain town,
thown 1, has the same price schedule as Bank A and

13 at every bank in Town 2 has the same schedule as

thank :S.B In which town is the price higher? Clearly,
tn ere IS no single answer. For 7 to 16 checks per
beon~h,. the price is higher in Town 1; for other numl' rs It IS higher in Town 2. Suppose further that
wn 1 has fewer banks or a greater concentration
h accounts in its banks. Do our findings confinn the
Ypothesis that fewer banks or more concentrated

3
. 1ton fact, of course, price schedules vary greatly within
Wns.

lleView
. I

May 1977

deposits mean a higher price, or do they contradict
it? The answer is by no means clear. So the price
schedules present a problem, and the value of
research findings on the relationship between market structure and performance depends, among other
things, on how this problem is handled. A number
of approaches are possible.
The best approach would be to take several different numbers of checks-say 10, 20, 40, and 80 per
month-and use each of the corresponding prices in
a separate analysis. Thus, we would try to uncover
the relation between market structure and the price
of 10 checks per month, the price of 20 checks per
month, and so on. The relation might vary according
to the number of checks; indeed it might be strong
for some numbers and absent for others .
The next best approach would be to select the
average number of checks written by depositors and
look for a relation between the corresponding price
(paid by the average depositor) and market structure. Thus, if the average depositor writes 30 checks
per month, we would look for the relation between
market structure and the price of 30 checks. This,
like any approach that substitutes a single price for
the entire price schedule, will waste infonnation, so
it is inferior to the first approach.
The preceding approaches will be explored in a
future article examining the observed relationship
between market structure and actual price schedules
for checking accounts. Either one would appear to
be greatly superior to the three approaches actually
taken in past studies. These completely neglect the
implicit part of the price, and proxy the price schedule by a number based on explicit service charges
only. They differ merely in the proxies chosen.
Summary evaluation of earlier findings
One approach taken in past studies is to proxy price
by the explicit monthly service charge for a hypothetical account on which 20 checks are written and
which contains a $100 minimum balance and $200
average balance. The two studies following this
approach uncovered no statistically significant relation between the market structure and the chosen
price proxy.
The problem with this approach is not only that
it is arbitrary-why 20 checks rather than 10 or
40?-but also that it specifies a minimum balance
($100) that may not minimize the cost of writing 20
checks, with the result that the chosen price proxy
is off the actual price schedule. In Banks A and B, for
example, depositors writing 20 checks on accounts
with a $100 minimum balance would pay not the
9

required amounts of $1.50 and $1.70, respectively,
as shown in the diagram, but $3.00 and $2.50. This
proxy actually reverses the true ranlting of these
particular banks with respect to their prices. It may
not perform quite so badly in every case, but its
large error loads the dice against the discovery of
any relation between price and market structure.
The second approach taken in these studies is to
proxy price by the ratio of the bank's total service
charge revenue to its total or average deposits. This
ratio is a poor proxy, even for the explicit service
charge. Not only does it vary with the number of
accounts and their activity, as well as with the service charge per account, but it also varies with the
mix of personal and business accounts. As is well
known, businesses keep substantially larger minimum balances than individuals do, partly to compensate their banks for various services performed
for no explicit charge. 4 Two banks with the same
price schedule but different proportions of business
accounts will have different ratios of service charge
revenue to deposits, thus appearing to have different
prices. Clearly, this approach ends up off the true
price schedule and, like the preceding one, has a poor
chance of finding whatever relation there may be
between price and market structure. It therefore
means little that most of the studies talcing this
approach found no such relation or that the relation
found by one study, while statistically significant,
was practically negligible.

From the studies here reviewed we have
findings but no good evidence. The relation between market structure and the
price of checking accounts remains to be
discovered.
The third and final approach is to estimate for
various banking markets average explicit service
charges per account, adjusted for activity and average balance. This estimate assumes that the explicit
charge varies with account characteristics in the
same way at all banks in the sample. As Banks A
and B illustrate, this is not necessarily so. But even
if these estimates of average explicit charges are
roughly correct, they bear no ascertainable relation
4. The relation between the bank and its business customers
is the foundation of Hodgman's (1963) theory of bank
behavior. This theory. especially as extended by Wood
(1975). is by now well established.

to the average of the actual price schedules in any
market. Though the one study talong this approach
found that market structure greatly affects the price
of checIcing services, it inspires no confidence because
of the poor price proxy and several other problems
with its methodology as well.
From the studies here reviewed we have findings
but no good evidence. The relation between market
structure and the price of checIong accounts remains
to be discovered.
Details of the studies
All the studies here reviewed used cross-section multiple regression equations, with a proxy for the price
of checIcing accounts as the dependent variable. In
every study, the proxy is related to a bank's explicit
service charge but not to the implicit part of its
price. The studies fall into three groups according to
the proxy.
Group 1. This group consists of two studies
(Heggestad and Mingo 1976; Stolz 1976) that
proxied price by the explicit service charge for a
personal checlcing account with an average balance
of $200, a minimum balance of $100, and monthly
activity of 20 checks and two deposits. Even if this
proxy were on the price schedule for 20 checks per
month, its use would uncover only a part of the
relation between price and market structure. But in
fact it is off the price schedule of every bank at
which $100 is not the optimal minimum balance for
the writing of 20 checks, so it proxies those banks'
prices with an error. If this error is random (and, of
course, it may not be), it does not bias the estimated
coefficients, but it does reduce the calculated t statistics and thus the statistical significance of the
coefficients. With this proxy it is difficult to uncover
any part of the relation between price and market
structure.
Unexplained variation in prices is further increased
by the omission of a measure of cost from the regression equations. To the extent that the cost of providing checlting accounts differs among markets, its
omission reduces the statistical significance of the
estimated coefficients.
Table 1 summarizes the nature and results of
these studies; it shows the low R2'S (less than 20 percent) that we expect from poorly measured dependent variables. Indeed, the only coefficients significantly different from zero at the 5-percent level are
those relating to the banks' total deposits and the
market's income growth (in the Heggestad-Mingo
study) and the presence and entry of a multibank
holding company (in the Stolz study) .

Table 1
STUDIES PROXYING PRICE BY THE EXPLICIT SERVICE CHARGE ON A HYPOTHETICAL ACCOUNT
Market structure
Study
and
sampl e

Heggestad and
Mingo (1976)
236 banks In
69 SMSA's
in states
permitting
little or no
branching ;
some data
from 1970,
others from
1972
Stolz (1976)
333 banking
offices in
75 rural
markets
in Iowa

Minnes~ta

Wisconsin :
1975
'

Market
definition

SMSA

" Area of
convenience"
(See text.)

Measure

Two measures:
Herfindahl
index of
total
deposits
Reciprocal
of index

Three measures:
Herfindahl
index of
total
deposits
Reciprocal
of index
Cubic function
of index

Effect
on
service
charg e'

None
(Reciprocal
significant at
10-percent
levei)

None

Additional expl anatory vari abl es

Total deposits of bank "
Ratio of market personal income
in 1970 to that in 1967"
Per capita personal income
Ratio of demand to total deposits
Ratio of demand deposits in
accounts smaller than $1 ,000
to total deposits
Market share of bank in terms of
total deposits

.11

Dummies for:
Presence of a multibank holding
company in market *
Entry of such during 1970-74*
Affiliation of bank with such
Market share of bank in terms of
total deposits
Total deposits of bank
Population in market and percent
change of such from 1960 to 1970
Weighted per capita income in
major towns in market
Dummies for presence of:
Thrift institution in market or
Production credit association
in market

.17

~ ·Clf

significantly different from zero at 5-perc ent level.
oettlolent slgnlfloantly different from zero at 5-peroent level.

. Stolz's study introduces a new approach to bank-

~~g ~a.rkets. Eschewing the ready-made political or

atlstIcal unit (city, county, SMSA) used by most
~tudents of banking competition, Stolz identifies a
ocal banking market with a geographic "area of
convenience" within which local residents and busi~eshes transact most of their business. He delineates
'puc areas according to shopping and commuting
attems and various other economic and demo~raphic data and finds that they do not correspond
t ery well to the ready-made units. This approach
e~ tnarket determination is not, of course, without
fur~hr, but it appears to deserve both emulation and
r er refinement.

Ii Both studies measure market structure by the

st e~~n?ahl index or a function of it. Neither obtains
stadshcally significant coefficients, though Heggesi; . and Mingo find the reciprocal of the index to be
nlficant at the lO-percent level. But as argued

ne·Vlew I May 1977

above, these results only show that market structure
has no significant effect on a particular price proxy;
they are silent about the effect of market structure
on actual prices.
Group 2. This group consists of studies that proxy
the banks' price schedule by the ratio of total service
charge revenue to total or average demand deposits.
This ratio varies with the number of accounts and
their activity, as well as with the explicit service
charge. It would be a reasonable proxy for the explicit
service charge--though not for the price, which
has an implicit part-if the number and activity of
accounts were held constant or, alternatively, placed
in the regression equation as regressors. Neither is
done in these studies.
Furthermore, variations in the mix of personal
and business accounts must be controlled if this
ratio is to be a good proxy for the explicit service
11

Table 2

STUDIES PROXYING PRICE BY THE RATIO OF TOTAL SERVICE CHARGE REVENUE TO TOTAL OR AVERAGE DEPOSITS
Market structure
Study
and
samp le

Fraser and
Rose (1971)
All banks in 78
Texas cities or
SMSA's, each with
less than 250,000
population; 1966
and 1967

Fraser and
Rose (1972)
All banks in 72
relatively isolated
towns in Eleventh
Federal Reserve
District ; 1965 and
1966
Ware (1972)
All banks in 57
non-SMSA counties
in Ohio; 1969 and
1970

Rose and
Fraser (1976)
704 banks in 90
Texas counties (68
counties belonged
to no SMSA, and
78 had fewer than
100,000 residents);
1970

Market
defin ition

City or SMSA

Town

Measure

Cost measure

1-bank CR
(concentration ratio,
total deposits)

None

Ratio of yearly wages and salarl~S
to average assets during year

Dummy variable:

None

o if town had 1 bank
1 if town had 2 or 3
banks

County

2-bank CR

None

Ratio of total operating expenses
to total assets·

County

Nine measures:
(1) number of banks
(2) 1-bank CR
(3) 2-bank CR
(4) 3-bank CR
(5) entropy
(6) Herfindahl index
(7) Hall-Tideman index
(8) relative entropy
(9) Gini coefficient
(plus three
measures of
market structure
changes)

(2), (5), (6), (7) :
increase of 1,000 basis
points in structure
measure increases
service charge by
3 basis points.

Ratio of wages and salaries
to total assets

1. If significantly different from zero at 5-percent level.
• Coefficient significantly different from zero at 5-percent level.

Effect
on
service
charge'

All others: none

l- - - - - - - - -

f-----i

Add ition al explanatory variables

I)
Bank d b
pelts per capita and yearly
Du ercent change In same
a~dY for presence of S&L (savings
Aver loan association)
, Rau age deposit size of banks In market
C~s to total loans of:
. R rnrnerclal and Industrial loans
Ceal estate loans
Ao~surner loans
gncultural loans

.30 (1966)
.26 (1967)

Durn

Road;r for presence of S&L
POPUI t',stance to nearest Reserve city
t P a Ion in 1960 *
I ercent
' Percent change In population, 1950-60 *
( 19 60 6Change In total deposits,
P
- 5 or 1960-66*

,30 (1965)
,32 (1966)

ercent

1960'7~hange

in county population,
incorn' and per capita personal
Perc
e, 1959-69
,
ent of
manuf covered employment In
Numb acturlng
Ratl er of S&L's
a of
AVerag consumer loans to gross loans
[
e deposit size of banks in market"

r Ratlas t

Not

reported

,35 to ,SO

Real 0 total loans of:
~arm ~state loans
Con oans
BUSlsurner loans
I AVeragness loans
~ Ratio o~ ~sset size of bank
r Urban P emand to total deposits*
, Retail s~rcentage of populatlon *
es per capita *

charge. Banks with the same price schedule but
different proportions of business accounts will have
different ratios of service charge revenue to deposits.
One way of controlling these differing proportions
is to include regressors that are likely to be correlated with them. But the lack of statistical significance of all such regressors that were tried (such as
the ratio of various loan categories to total loans)
indicates that the control is inadequate.

As Table 2 shows, these studies are weak in a
number of other respects. It is not clear, for example,
how the ratio of wages and salaries to assets corresponds to the cost of servicing demand deposits. But
to criticize the other weaknesses would be analogous
to charging a convicted murderer with disturbing the
peace. The gross imprecision of the price proxyeven as a measure of only the explicit part of the
price schedule-is enough to destroy our confidence
in the results, whether they indicate a O.003-percent
increase in the price for a 1-percent increase in some
measures of the market structure, as in the Rose and
Fraser study of 1976, or no relationship, as measured
in other studies.
Group 3. This consists of a single study, which
developed a price proxy for the explicit service
charge through regression analysis. Bell and Murphy
(1969) began with the regression equation:
log R, = ao + alA, + a2 S, + asN, + a4 Wi,
where for bank i (i ranging over the entire sample):
R, = service charge revenue per account
A, = log of check-writing activity per account
Si log of average balance per account
Ni = log of number of accounts
Wi log of a wage index. 5
They estimated the a'S and then, holding activity per
account and balance per account constant at their
mean values A and S, respectively, they replaced
bank i's observed R i by the calculated value R /,
where:
(1)
log R{ = ao + alA + a2 S + asN" + a4W"
and the a's are the ordinary least squares estimates
of the a'S.

=
=

R/ is what bank i's service charge revenue per
account would have been had it varied with the
regressors in the same way the typical bank's did and
had its accounts shown the same average balance
and activity as the typical sample bank's. It is a
price proxy from which all market structure effects
have been removed. One bank's R' can differ from
another's if (and only if) it has a different number
5, The equation also contained five branching dummies,
which we neglect for simplicity,

· I May 1977
aeView

of accounts or faces a different wage index. These are
cost factors, not market structure factors. Conceptually, R' is more akin to a bank's cost than to its price.
The authors then regress, not each bank's R' on
measured market structure and variables specific to
the bank, but the average R' in each market on the
measured structure and the average of the market
banks' estimated marginal costs.6 With Pj being the
average value of R' in market j, Cj being the average
value of marginal costs there, and r j being the threebank concentration ratio there, the regression equation was
(2)
log Pi = f31log C j f32log r j . 7

6. There were 14 markets, all in New England, 13 of them
SMSA's and the 14th Suffolk County. The sample was
taken in 1966.
7. The authors actually estimated eight sets of f3's, one for
each of eight different three-bank concentration ratios
defined with respect to various classes of checking
accounts. The results differed hardly at all.

This method eliminates all of the within-market
differences in price-cost ratios, helping to raise the
statistical significance of any estimated effect of
market structure on price without, however, biasing
the estimate. But since all market structure effects
were removed from the measure of explicit price
in the first place, the calculated market averages
bear no ascertainable relation to average explicit
service charges, and, of course, even less to average
price schedules, in the various markets. Though the
estimated value of f32in equation (2) is .21, indicating that a I-percent increase in the concentration
ratio would raise the price of a checking account by
0.21 percent-by far the largest such effect ever
found-the method that produced it inspires so little
confidence that we cannot regard it as evidence.

REFERENCES
Bell, Frederick W., and Murphy, Neil B. "Impact of Market Structure on the Price of a Commercial Banking Service." Review of Economics and Statistics 51 (1969) : 210-13.
Fraser, Donald R., and Rose, Peter S. "More on Banking Structure and Performance: The Evidence from Texas. "
Journal of Financial and Quantitative Analysis 6 (1971) : 601-11.
.
- - -. "Banking Structure and Performance in Isolated Markets: The Implications for Public Policy." AnUtrust Bulletin 17 (1972) : 927-47.
.
Hall, Marshall, and Tideman, Nicolaus. "Measures of Concentration." Journal of the American Statistical Assoctation 62 (1967): 162-68.
Heggestad, Arnold A., and Mingo, John J . "Prices, Nonprices, and Concentration in Commercial Banking." Journal of Money, Credit, and Banking 8 (1976) : 107-17.
Hodgman, Donald R. Commercial Bank Loan and Investment Policy. Champaign, Ill. : University of Illinois, Bureau of Economic and Business Research, 1963.
Rose, Peter S., and Fraser, Donald R. "The Relationships Between Stability and Change in Market Structure: AIl
Analysis of Bank Prices." Journal of Industrial Economics 24, no. 4 (June 1976) : 251-66.
Stigler, George J. The Organization of Industry . Homewood, Ill.: Richard D. Irwin, 1968.
Stolz, Richard W. "Local Banking Markets, Structure, and Conduct in Rural Areas." In Proceedings of a Conference on Bank Structure and Competition, pp. 134-48. Chicago: Federal Reserve Bank of Chicago, 1976.
Ware, Robert F. "Banking Structure and Performance: Some Evidence from Ohio." Economic Review, Federal
Reserve Bank of Cleveland, March 1972, pp. 3-14.
Wood, J . H. Commercial Bank Loan and Investment B ehaviour. Wiley Monographs in Applied EconometricS.
edited by A. A. Walters. London and New York: John Wiley & Sons, 1975.
14

,
J

Appendix
Some measures of market structure

Let there be n sellers in a market. Let Si be
the market share of seller i. Number the
sellers from 1 to n according to their market
shares, with Sl :::::.. S2:::::" ... :::::.. s" . Evidently,
Sl
s" 1. Most of the commonly
employed market structure measures can
be expressed in terms of these s's.

+ ...+ =

1. The k-bank concentration ratio, R k :
R k = Sl
S k (k L n).
The minimum value of R is ki n, attained
when all market shares are" equal; the maximum value is 1. This measure is usually assumed to vary inversely with the strength
of competition.

+ ... +

2. The Herfindahl index, H:

+ ... + S,,2.

The minimum value of H is l i n, attained
when all market shares are equal, and the
maximum value is 1, attained when there is
only one seller. This measure, like the concentration ratio, is assumed to vary inversely with the strength of competition.

3. Hall-Tideman index, T:
1
T =-n
2 1 is l -1
i= 1

The minimum value of T is lin, attained
when all market shares are equal, and the
maximum value is 1.l T is assumed to vary
inversely with the strength of competition.

(

4. Entropy, E:
E = - [sllog2(Sl )
S,JOg2(S,, )].
The minimum value of E is 0, attained when
the market has only one seller; the maximum value is log2(n), attained when all
sellers have the same market share. This
measure, unlike the preceding three, is assumed to vary directly with the strength of
competition.

+ .. .+

5. Relative entropy, Er :

E
Er = '10-g-2(' n' )

This measure, which can range from to 1,
is also thought to vary directly with the
strength of competition.
6. The Gini coefficient, G:
Put ql = S"
q2 = S,,-l

q,,= Sl ;
thenG = 2(ql+ 2q2+ ... + nq,, ) -n-1
n
This measure ranges from
to (n - 1) In,
inversely, it is assumed, to the strength of
competition.
A useful discussion of the strengths and
weaknesses of these things, as measures of
market structure, may be found in Stigler
(1968).

1. The formula for T is incorrectly printed in
H all and Tideman (1967) and incorrectly
printed (and incorrectly calculated?) where
used in the studies reviewed here.

ne'View I May 1977

The REVIEW is sent to the mailing list each month without charge, and
additional copies of most issues are available from the Research Department, Federal Reserve Bank of Dallas, Station K, Dallas, Texas 75222.
Articles may be reprinted on the condition that the source is credited. It
will be appreciated if copies of the reprinted material are sent to the authors.

Federal Reserve Bank of Dallas
May 1977

Eleventh District Business Highlights

DRILLING RECOVERY PICKS UP
~rilling activity has been increas-

Ing, and active rigs in the Eleventh
District in late April numbered 37
percent more than a year earlier.
The number of rigs in Texas and
Oklahoma were at 17-year highs.
Louisiana had a 12-year high, and
New Mexico a one-year high.
There is usually a seasonal
decline in drilling activity in the
first quarter of a year. However,
such a decline did not occur in the
Eleventh District this year.
A slump in drilling in early
1976-largely due to bad weather
and price and tax uncertainties surrounding the oil and gas industryled to a modest buildup of idle rigs
and a slowing in new orders for oil
~eld equipment. However, drilling
hgan to pick up last July, when
t e Federal Power Commission
announced that the price of newly

discovered interstate natural gas
would be boosted to $1.43 from
$0.52 per 1,000 cubic feet.
This action, along with greater
crude oil price incentives, accounts
for most of the rebound in drilling
activity. During the last half of
1976, the number of active rigs in
Texas, for example-the top-ranking state in U.S. drilling-rose 22
percent above the first half. Growing demand for drilling rigs reduced
the number of idle rigs to low levels.
This led drilling contractors to
begin ordering new oil field equipment by year-end.
With an increase in new bookings, production of oil field equipment is strengthening as drilling
backlogs continue to build up. But
equipment shortages are not
expected to hamper drilling operations. Moreover, oil field equipment
producers have ample capacity to

RISE IN RIG ACTIVITY IN ELEVENTH DISTRICT STATES
HAS NOT HALTED DECLINE IN CRUDE OIL PRODUCTION
~
PRODUCTION
NUMBER
BILLION
BARRELS
1,200 _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
_ 2.6
1,100 _

2.5

1,000 _

2.4

900 _

2.3

2.2

~-~-~ -

2.1

800 _

"" "\.
"

600 -,.---.,.---r--.........--.--.....:;.---.---...---r--T'""'--r- 2.0
1976
1974
1966
1968
1970
1972
SOURCES: Hughes Tool Company.
U.S. Bureau of Mines.

meet any expected surge in demand
for rigs. The continued recovery in
onshore drilling should lead to
steadily improving sales for rig
manufacturers this year.
Along with the increase in the
number of active rigs, the number
of wells and footage drilled last
year in the Eleventh District rose
4.8 percent and 4.4 percent, respectively, above 1975. Oklahoma experienced the largest percentage gains
in both, while New Mexico showed
the only declines.
Despite the pickup in drilling
operations, production of crude oil
and natural gas has continued to
decline since 1972 and 1973, respectively. Oil production in the District in 1976 fell 3.3 percent from
the previous year, and gas production fell 1 percent. All four states
posted setbacks in crude oil output,
but Texas was the only District
state to register a drop in natural
gas output.
Proved reserves of crude oil and
natural gas in the District slumped
8.4 percent and 6.8 percent, respectively, from 1975 to 1976. However,
New Mexico had a slim gain in
natural gas reserves. The drop in
District reserves was much greater
than the decline in output.
Production of oil and gas may
decline further unless additional
reserves are found, particularly if
efforts are made to conserve the
prospective supply. There were no
large new discoveries of crude oil
and natural gas last year. Exploration was limited by uncertain policies, especially those that affect
prices for energy.
The drilling outlook in the District for the remainder of this year
is for a high level of activity. It will
be affected, of course, by the direc(Continued on back page)

INDUSTRIAL PRODUCTION
(SEASONALLY ADJUSTED)

TOTAL PRODUCTION

MANUFACTURING

MINING

1967=100
150 - - - - - - - - - - - -

1967=100
130 - - - - - - - - - - - -

140 -

140 _

TEXAS

,.. ....

(

130 -

/ .... -1

/?"./\./"'.

130 -

120 -

135.0
110 -

- :'
/.:

\.
J .... :
:::'~./ ",'

120 -

110 -

,.. 140.9

, ,136.5
J I 135.1

1",1

1'/1,

.........../'\:'\...

100 -

.....

...... !"\.
'0, :

".:

'........:

......:..f··.•.••.!::.

110 -

90 -

100 """'T---,-----,,---....,1977
1975
1976

80 '"""'T---...---......,~---r"'
1975
1976
1977

......

LOUISIANA
100 -'---~~---r---T1975 I 1976
1977

·Estimatlon of Louisiana industrial production index discontinued pending receipt of revised employment data .
SOURCES: Board of Governors, Federal Reserve System.
Federal Reserve Bank of Dallas.

....

EMPLOYMENT AND UNEMPLOYMENT

PRICES RECEIVED BY TEXAS FARMERS

FOUR SOUTHWESTERN STATES 1
(SEASONALLY ADJUSTED, BY FRB)

1967=100
260 - - - - - - -_ _ _ _ _ _ _ _ _ _ ___

EMPLOYMENT
UNEMPLOYMENT
MILLION PERSONS
PERCENT
8 .6 - - - - - - - - - - - - - - - - - - - - 8

8 .4 -

8 .2 -

UNEMPLOYMENT RATE

............ ....
......
........•.•..:.......................
••
'
........

8.34

-7

CROPS ../\.-, ".,.,_, 225

210 -

,,,,. -"/ '_-_-1
.... -..

...

175
160 -

LIVESTOCK AND LIVESTOCK PRODUCTS

-6

110~--------r--------~-----~

8.0 -

-5

7.8 -

-4

1975
SOURCE:

TOTAL EMPLOYMENT

7 .6 --y-------,-----~r-------..- 3
1975
1976
1977
1. Louisiana, New Mexico, Oklahoma, and Texas.
SOURCE: State employment agencies.

1976

1977

U.S. Department of Agriculture.

SAVINGS AND LOAN ASSOCIATION ACTIVITY
AND HOME BUILDING IN TEXAS
(SEASON ALL Y ADJUSTED, BY FRB)

SAVINGS

RATIO

MILLION
DOLLARS
1,000
__
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _--100

900-

CONSUMER PRICES

700-\
600-

190 -

180 -

170 -

-- HOUSTON
U.S.
..... DALLAS

800-

1967=100
200 - - - - - - - - -_ _ _ _ _ _ _ _ __

,._.183.6

//~.~.~.~ ..•.......••.••..•..••..

160 - _ -

150 I--=~-..,..------r------T"""
1975
1976
1977
SOURCE: U.S. Bureau of Labor Statistics.

~~
400 ---r-----~----___,r_----rl 40
1975

178.2

1976

1977

THOUSAND
12-----------------------

:=" ~;;;iiiI~Iiiii11111111111
I

1975

1976

'0 3

.
1977

SOURCES: Bureau of Business Research . University of Texas.
Federal Home Loan Bank of LIttle Rock .

CONDITION STATISTICS OF ALL MEMBER BANKS

RESERVE POSITION OF MEMBER BANKS

ELEVENTH FEDERAL RESERVE DISTRICT
(CUMULATIVE CHANGES)

ELEVENTH FEDERAL RESERVE DISTRICT
(MONTHLY AVERAGES OF WEEKLY DATA)

______________________________________
_
B5ILLION-DOLLAR
CHANGE

MILLION DOLLARS
200---------------------------------------

LOANS

150 -

""
"
1976 ,,/"

...,

11977 _

,..'"

..: ... .,,_--'"
•••••
.••-:-...."
1975
•• '
•••• ~'i: •.•.••.•.. ••••••• ••..•••••••....•.
Ii'
I
i
I
I
i
BILLION-DOLLAR CHANGE

100 50 -

........

.:
i \ r. .:

~:::EXCESS

~
\
:

\.\! _V

RESERVES
BORROWINGS
FROM FRB

\l\,j ~

! \ ~ A f:
.

~
.,..

O_~

::··.f

\./.

\ . .=

-50-

- 100~----------~----------~(~--------~-

1975

1976

1977

INVESTMENTS

2 _

1'977

_--:7.:7;~::::.":'>,.......:.:..:.:.::..:::::~::

_ .....::::::·········,975

O-~

1976

iii

LOANS AT WEEKLY REPORTING BANKS

BILLION-DOLLAR CHANGE

ELEVENTH FEDERAL RESERVE DISTRICT
(CUMULATIVE CHANGES)

TIME DEPOSITS

,,-

'" "

_..................

1976 ,,"

Z,977 ----------::. . . . . . . . .
---*'

.,;;_.c.-=:............................

BUSINESS LOANS
400 --

--1977

1975

_",--

------' ......

..",...".."

1976

........

MILLION-DOLLAR CHANGE
800

iii
i
i
3-_____________________________________
__
BILLION-DOLLAR
CHANGE

o ~~-~-_-~-~-~-~'~-~-~.v.~."'~.~~~~~~~~~~,t~
...................... '
.........................
1975
-400----~--~~--~-r.--~~--~~--~~--~

DEMAND DEPOSITS

--1977

........~""'/

•

' . . --~-----«=.. ...... -----_
...
1976

~~I;DING

SON

.,.,..-- ........--_-------

1976

o --....~-:;:;;::;:;;;;:-;;.:--:?;..'..:':'"'
.."F:'ii.~;;;:;;;-5....~"'.~:::...-...-..-..-..-..-...-..-..-..-...-..-..-..-...-..-..-.. -...-:....

-l-----ir~--~~-,--T__r~~,__T--r_._
I
Iii
I
i
I
Iii
I
i
J

CONSUMER LOANS
100 -

1975 .••••••..:::.--

0---~"~··~·~··,~~~~··~
.. ·~~~··.:.:···~..-··-.. -.. ·-..7••~~~-_-_________

....

MILLION-DOLLAR CHANGE
200

-100----~I--TI--TI--TI--TI--TI--TI--TI--TI--TI--TI--TI-

SON

FOREIGN TRADE

CONTRACTS

HOUSTON CUSTOMS REGION
(SEASONALL Y ADJUSTED, BY FRB)

(SEAS SOUTHWESTERN STATES 1
....... ONALL Y ADJUSTED, BY FRB)
Billi

1.2~OLLARS

BILLION DOLLARS
1.7
1.5 -

.9_
.89

1.3 -

EXPORTS

1.18
.58

.93

.5 -..........

1976

1977

~~~~---------------------------------
. . W. Dodge, McGraw-HIli, Inc.

.4-T.---------,r---------~---------r-

1975

OUlsla

1.OURCE'
l
~a , New Mexico, Oklahoma, and Texas.

SOURCE:

1976

U.S. Department of Commerce.

1977

tion of impending energy legislation, although the major impact of
this will likely come in 1978 and
later years.
BANK LOANS TO DEVELOPING
COUNTRIES WELL DISPERSED

Commercial banks in the Eleventh
District had about $2.5 billion in
loans and other claims to foreigners
in developing countries at the end
of 1976. Over 90 percent of the total
was accounted for by the five largest District banks, but a few small
banks also held claims on foreigners
in less developed countries (LDC's).
District banks had an additional
$2.2 billion in loans and other
claims to foreigners in developed
countries. The total claims on
foreigners were about 6.4 percent of
total assets-up from 5.6 percent a
year earlier but still a fairly small
proportion. Insofar as the rise in
foreign loans of U.S. banks has
become a matter of widespread
interest, it has focused largely on
the growth of loans to the developing countries.
Direct claims by District banks
on foreigners in developing countries were widely distributed,
involving a total of 56 countries.
The amounts ranged from a low of
about $1 million in each of five
countries to a high of around $400
million in Mexico. Approximately
$560 million of total claims represented debt from the Bahamas.
But a major portion of the claims
on the Bahamas was investment
funds shifted from District banks
to their branches in the Caribbean-funds that probably were
subsequently loaned to foreigners
in various other countries.
The five largest banks in the District hold three-fourths of the total
claims on foreigners in Mexico, with
the remainder held mainly by
several smaller banks located near
the Texas-Mexico border. The Mexican debt held by these border
banks-approximately $100 million-represents well over half of all
claims on foreigners in LDC's that
are held by smaller District banks.
As in the United States as a
whole, District bank claims on
foreigners in LDC's are fairly well
dispersed. In addition, in countries
where District banks have the big-

gest direct commitments-Mexico,
Brazil, the Philippines, and Spainover a fourth of the debt is irrevocably guaranteed by one or more
parties outside the LDC in which
the borrower is located.
Banks have a good record in
terms of loan losses in LDC's. So
far, only a few borrowers have
found it necessary to reschedule
loans because of changing economic
conditions in their countries. Moreover, it would appear unlikely that
many countries would allow widespread default on foreign borrowings, since such default probably
would severely dampen a country's
credit rating for some time. Even so,
Congressmen, banking authorities,
and others have expressed concern
that adverse economic conditions in
some LDC's might call into question the repayment of a portion of
foreign loans.
District banks undoubtedly will
continue to make loans to foreigners in LDC's, but it appears that
growth in the loans may be slower
in the near term. For one thing, the
ratio of external debt to exports has
risen sharply in many oil-importing
LDC's. As a result, additional credit
requests from those countries may
appear less creditworthy.
For another thing, part of the
growth in foreign lending by District banks probably has been attributable to relatively weak loan
demand from domestic businesses.
A significant pickup in domestic
demand for bank funds could moderate the growth in foreign lending
by District banks.
Finally, the increasing public
attention being directed toward
the growth in foreign lending may
cause some banks to go slow in
extending additional credit in
LDC's.
OTHER HIGHLIGHTS:

• Preliminary figures show the
Texas industrial production index
rose in March at a seasonally

adjusted annual rate of 4.7 percent.
The growth centered in the mining
sector, as manufacturing output
declined slightly.
• Total employment in the four
southwestern states increased in
March for the sixth consecutive
month. The unemployment rate
was 5.6 percent of the total labor
force-an improvement from 5.8
percent a month earlier.
• Total credit at member banks in
the Eleventh District increased
moderately in March. Most of the
increase reflected further acquisitions of U.S. Government securities.
Total loans rose slightly. But
borrowing by businesses declined
further, indicating no strong need
for credit to finance additional
inventories or additions to plants
and equipment. Reflecting the
strength in District construction
activity, however, real estate loans
again rose sharply. And consumer
loans to finance automobile and
credit-card purchases continued to
increase, but at a slower pace.
Net acquisitions of U.S. Government securities continued to
account for most of the growth in
investment portfolios. However,
District banks also increased their
holdings of municipal securities
more than in any other month
since June 1976.
• The value oftotal building contracts in the four southwestern
states rose 11 percent in March, as
a sharp increase in residential contracts more than offset a small
decline in nonresidential contractsd
Housing starts in Texas increase
to 10,288 units, seasonally adjusted,
in March. That is the highest level
since last September.
• According to the mid-decade
population estimate by the U.S ..
Bureau of the Census, San AntoniO,
Texas, is now the tenth largest city
in the country. Ranked 14th in the
1970 census, San Antonio joins
Houston, ranked 6th, and Dallas,
ranked 8th, in the top ten cities.

This issue of the Highlights, published monthly by the Research
Department under the general supervision of Edward L.
McClelland, Senior Economist, was prepared by Jonathan
Euseary, Mary G. Grandstaff, and Jean T. Adeler.

Full text of Review (Federal Reserve Bank of Dallas) : May 1977

FRASER