The full text on this page is automatically extracted from the file linked above and may contain errors and inconsistencies.
Economic
tizi)
Review
; FEDERAL RESERVE BANK OF ATLANTA
MARCH 1982
SUNBELT "Stealing" Industry from North?
CURRENCY How Much Is Hoarded?
TAX CUTS Shifting the Burden
CHECKS Patterns of Change
BANKS Evaluating Capital Requirements
LIVING COSTS Guide to Sources
I jo
f\
Economic
Review
FEDERAL RESERVE BANK OF ATLANTA
President:
William F. Ford
Sr. Vice President
and Director of Research:
Donald L. Koch
Vice President
and Associate Director of Research:
William N. Cox
Financial Structure:
B. Frank King, Research O f f i c e r
David D. W h i t e h e a d
National Economics:
Robert E. Keleher, Research Officer
Stephen O . M o r r e l l
Regional Economics:
Gene D. Sullivan, Research O f f i c e r
Charlie Carter
William J. Kahley
Database Management:
Delores W. Steinhauser
Visiting Scholars:
James R. Barth
George Washington University
George J. Benston
University of Rochester
Robert A. Eisenbeis
University of North Carolina
Arnold A. Heggestad
University of Florida
John Hekman
University of North Carolina
Paul M. Horvitz
University of Houston
Peter Merrill
Peter Merrill Associates
Communications Officer:
Donald E. Bedwell
Public Information Representative:
Duane Kline
Free subscription and additional copies available upon
request t o the Information Center, Federal Reserve Bank
of Atlanta, P.O. Box 1731, Atlanta, Georgia 30301. The
views expressed are not necessarily those of t h e Federal
Reserve Bank of Atlanta or of the Federal Reserve System.
Material herein may be reprinted or abstracted, provided
this Review and the author are credited. Please provide this
Bank's Research Department with a copy of any publication in *
which such material is reprinted.
Editing:
Gary W. Tapp
Graphics:
Susan F. Taylor
Eddie W . Lee, Jr.
The Federal Reserve Bank of Atlanta also publishes Insight,
a newsletter on economic trends in the Southeast. Insight,
published twice a m o n t h and mailed first class, is designed
to give readers fresh and timely data, analyses, and forecasts on
the Southeast's economy. Free subscriptions are available
from the Information Center, Federal Reserve Bank of
Atlanta, P.O. Box 1731, Atlanta, Georgia 30301.
FEBRUARY 1982, E C O N O M I C REVIEW
The purpose of the Economic Review is to inform the public about Federal Reserve policies and the
economic environment and, in particular, to narrow the gap between specialists and concerned laymen.
Behind the Sunbelt's Growth:
Industrial Decentralization
Explaining the
4 Cash Explosion
Is the Sunbelt really "stealing industry away" from
the North? Are entire plants closing in the North
and moving South? A close analysis finds that
popular conceptions about Sunbeltgrowth maybe
unfounded.
Tax Cuts:
Who Shoulders the Burden?
29
Will the check collection system respond to the
current turmoil (Fed pricing, increased costs, etc.)
with improved speed and efficiency? History suggests otherwise.
Cost of Living Data:
35 A Guide to Sources
How much capital is adequate for a commercial
bank? Should regulators set ratios for the entire
industry, for groups of banks, for individual banks,
or at all? An assessment of the current debate.
Statistical Supplement
Currency held by the public jumped from $160 per
person in 1960 to over $600 per person in 1980.
What factors are behind the sharp increase, and
why does it appear that as much as 70 percent of
total currency is hoarded?
Lumbering at Top Speed:
The Check Collection
19 System, 1952-1979
Will the Reagan tax cut shift the tax burden toward
lower income taxpayers? An analysis of the historical
record of proportional tax cuts suggests some
surprising answers.
Regulation of Bank Capital:
An Evaluation
14
44
Sources for cost-of-living comparisons vary widely
in methodology and consistency. What sources
offer reliable and detailed cost-of-living data, and
what major differences exist among these sources?
51
I
v
VOLUME LXVII, N O . 3
Behind the Sunbelt's Growth:
The Sunbelt's growth is not so much a
result of entire firms moving from North
to South as it is a function of
"decentralization." Manufacturing, in
particular, seems to be seeking out not
only southern locations, but also smaller
cities and non-urban areas.
The economy of the Southeast has been growing
more rapidly than the national economy for
several decades. The Southwest and the West
have also grown considerably, while the Northeast
and the Midwest have experienced a marked
slowdown in their growth. Many commentators
have discussed this pattern in the context of a
Sunbelt-Frostbelt dichotomy, but this is not entirely accurate. The slow growth has been mainly
in the populous industrial states of the manufacturing belt from New York westward t o Illinois,
but this slowdown has not affected other areas of
the North such as the Plains states. The recent
experience of New England has been much
better than that of the neighboring M i d d l e Atlantic states.
The most notable characteristic of the differing
growth of regions has been the shift away from
large metropolitan and small cities. Even in the
Sunbelt, growth has been proportionately greater
outside the largest cities than within them.
The sector of the economy which has been
leading this change is manufacturing. While services, finance, real estate, transportation, trade
and construction employment mostly serve their
local markets, manufacturing tends t o be more
oriented t o the national market, and manufacturers have been far more footloose throughout our
history.
The current trend is for a decentralizing and
spreading-out of industry. The more rapid growth of
the Southeast is being influenced by the new
location decisions of manufacturers. From 1960
to 1980, total employment in the Southeast grew
56 percent more than that for the U. S. as a
whole; service employment grew only 33 percent
faster in the Southeast, while manufacturing
employment grew 208 percent faster. It is important t o note that these are growth rates relative to
the U. S. average. Manufacturing employment as
a percentage of the labor force has declined in
the Southeast and in the nation. But whereas
other sectors such as construction, trade and
services grow in parallel fashion in every region,
manufacturing growth rates are widely divergent,
w i t h a pronounced shift out of the traditional
manufacturing belt and toward the South and
West.
This development raises t w o questions. One is
whether the Southeast's gain has been at the
expense of the North, in the sense that some
plants have been relocated from the North to the
Southeast or that new plants have been built in
the Southeast which in the past would have been
built in the North. The second question is concerned with the benefits of industry location in
the Southeast. What factors have made this
region so attractive? This article will review some
of the ideas and evidence surrounding these
questions.
From 1970 to 1981, total employment grew in
every state. Yet manufacturing employment declined in some areas, primarily in the manufacturing belt from N e w York t o Illinois, which together
lost over 1.1 million jobs. The Southeast gained
significantly compared t o the total gain of about
5 percent. (Chart 1).
In order to say whether this marked shift in
manufacturing represents a movement from one
region t o another or a process of growth or
shrinkage which is independent within regions,
4 MARCH 1982, E C O N O M I C REVIEW
Industrial Decentralization
w e need information on the experience of individual firms. The only comprehensive source of
this information is David Birch, The Job Generation
Process. Birch reports on the expansion, contraction, birth, death and migration patterns of individual establishments from Dun and Bradstreet's
files on nearly all business establishments. Establishments are classified as independent (single
establishment), headquarters or branch, and
parent or subsidiary. The employment experience
of individual facilities was charted from 1969 t o
1976.
It is clear from the Birch study that entire firms
are not moving from the North t o the South. Even
the states w i t h declining industrial bases lost
only about one-tenth of one percent of their
employment per year from out-migration of
firms. The other popular conception of job movement is that firms are closing plants in the North
in order to open new ones in the South. This is
also contradicted by Birch's findings. The declining states had either the same or a lower rate of
closures than the fast-growing states. It is the
excess of the rate of births of new firms that
determines the difference between the fast and
slow growth areas.
These results have been interpreted by some
t o mean that states should look to the process of
new firm formation in order t o generate faster
growth. However, the case of manufacturing
growth in the Southeast is somewhat atypical. A
large proportion of Southeast manufacturing employment is in branch plants operated by firms
headquartered outside the region. Birch reports
the distribution of net new employment for the
entire South from 1969 to 1976 ownership
status as follows:
Net Employment
Growth
Independent
Headquarters/Branch
Parent/Subsidiary
in
Manufacturing
14.0 percent
66.1 percent
19.9 percent
About 86 percent of new manufacturing jobs
in the South from 1969 to 1976 were generated
in multi-establishment firms. The question that
Chart 1. Manufacturing Employment
(Percent Increase, 1970-1981)
46.6
ili
21.1
18.7
16.5
Source: Bureau of Census, Employment and Earnings, 1970 and 1971.
FEDERAL RESERVE BANK OF ATLANTA 5
14.3
"Most of the manufacturing employment gains in the Sout
Table 1
Percent Distribution of Manufacturing Net Employment
Change in the South Due to Differential Treatment of
Branches and Subsidiaries by Headquarters and Parents.
Location of
Headquarters
Percent of Branch
Employment Change
Location of
Parent
Percent of Subsidiary
Employment Change
Northeast
31%
Northeast
46
North Central
41
North Central
54
South
16
South
West
13
West
100
-18
18
100
Source: David Birch, The Job Generation Process, MIT Program on
Neighborhood and Regional Change, Cambridge, Mass., 1979.
remains, then, is how many of these jobs were
the result of location and expansion decisions
made outside the region? The closest we can
come to an answer is the distribution of branch
and subsidiary employment gains by location of
headquarters and parent, respectively (Table 1).
Seventy-two percent of branch employment
growth from 1969 to 1976 was in plants owned
by firms in the Northeast and North Central, and
86 percent was in plants owned outside the
South. Subsidiaries o w n e d by parents in the
South actually decreased employment over the
period, so that all of the employment growth in
subsidiaries came in firms owned outside the
South.
Most of the manufacturing employment gains
in the Southeast have occurred in firms headquartered outside the region. While single-establishment firms might be considered to be generated from within a region, firms which operate
facilities in several regions are more likely to have
chosen a particular state or region in which t o
locate branches. In this sense the growth of the
Southeast has been at the expense of the North.
W h y has the Southeast proven so attractive to
manufacturing firms?
The Dispersion of
Manufacturing Activity
The rapid growth of manufacturing outside the
old manufacturing belt represents the reversal of
6
a long trend toward the concentration of industry
which dates from the beginning of the Industrial
Revolution. M u c h of the industry grew up in
large urban concentrations whose growth paralleled that of the economy. The major reason for
these urban-industrial concentrations is that related industries such as fabricated metal, industrial
machinery and transportation equipment enjoyed
greater efficiency through the sharing of similar
labor force skills and other resources and through
savings in transportation and communication
costs. Wages, rents, taxes and many other costs
were higher in these industrial centers, but these
were offset by the higher productivity resulting
from the inter-industry linkages.
However, the evidence is growing that a turning
point has been passed. The longtrend of population and employment concentration has been
reversed both in the United States and in many
other developed countries. Many nations which
experienced net immigration t o their major industrial complexes since the beginning of economic development have recently seen a switch
t o net out-migration in favor of less densely
populated regions.
In general, the countries which industrialized
first have decentralized first The group which is
experiencing out-migration includes—besides the
United States—France, West Germany, Belgium,
Denmark and Holland. Countries where immigration t o urban centers has ceased are Japan, the
United Kingdom, Italy, Sweden, Spain and Canada
Countries still in the process of industrializing,
such as South Korea and Taiwan, continue to
experience immigration to their urban complexes.
The forces behind this dispersion of manufacturing come under the general heading of industrial maturity. Many industries pass through a life
cycle in which the product no longer experiences
frequent design changes and can be turned out in
similar form year after year. The production process
for the product is also refined and standardized,
principally by substituting machinery and lowerskilled production workers for hand operations
and skilled workers and designers.
These developments make firms less dependent
on the specialized labor force and other firms in
the large urban areas. They can reduce their cost
of production by locating in smaller cities or rural
M A R C H 1982, E C O N O M I C REVIEW
|
1
have occurred in firms headquartered outside the region.".
areas where construction costs, wages, and taxes
are lower. In addition, most industries today have
a wider range of products, individual products are
more complex, and more stages of processing are
required. This encourages large firms to create
individual plants specializing in particular products,
processes or components. These plants are sited
in cities or regions suited t o their particular
resource needs.
The Southeast has benefited from the maturing of
products and production processes since the end
of the nineteenth century. The textile industry
saw much of its most labor-intensive production
move from New England to the South Atlantic
between 1880 and 1920 after the development
of automatic machinery in spinning and weaving
replaced many skilled workers. Similar factors
facilitated the southward movement of chemicals,
paper, rubber and furniture. Today many multiplant firms in a wide variety of industries are
finding the Southeast a favorable location for
producing some of their products or components
of products.
In a study of new plants built by Fortune 500
firms during the 1970s, Roger Schmenner identified
the characteristics of Sunbelt plants which distinguished them from Frostbelt plants. In declining
order of significance, Sunbelt plants were found
to be:
»
• non-union
• not independent (that is, Sunbelt plants are
more often single-function facilities rather
than producing the complete product)
• purchasing more inputs from other company
plants
• not engaged in product innovation
• using more raw materials (that is, they are at
the lower end of the stages of processing)
• more capital-intensive (that is, more automated)
This and other studies suggest that it is the
relatively standardized production processes which
are moving out of the older industrial areas, while
research and design, engineering and management
tend to remain behind. This division of production is
most noticeable in the broad field of electronics
and related equipment New product development
and highly-skilled production such as medical
electronics and scientific equipment is located in
FEDERAL RESERVE BANK OF ATLANTA
Chart 2. Manufacturing Employment
Change 1959-1969
75 —
SMSA
Counties
0-50 Miles
from SMSA
Over 50 Miles
from SMSA
Source: Table 2.
the Northeast and California. Plants outside the
manufacturing belt produce high volume components and sub-assemblies, while low-skill laborintensive parts such as circuit boards are made in
low-wage, underdeveloped countries.
Foreign competition must be mentioned as a
contributing factor t o the accelerated movement
of industry in the United States. The share of
foreign trade in the economy has doubled in the
last 10 years, and many more industries now find
themselves competing in world market In autos,
rubber, textiles, shoes and consumer electronics,
lower foreign wage rates have sent domestic
producers searching for lower-cost locations for
mass-produced goods. The result has been a
more rapid growth of industry in the Southeast,
Southwest, Plains states and in general away from
large, high-cost metropolitan areas.
This pattern can be seen in Table 2, which
presents the growth of all employment and
manufacturing employment in the South by size
of urban area and distance from the nearest
Standard Metropolitan Statistical Area (SMSA).
The manufacturing growth rate was highest for
counties at the greatest distance from SMSAs
(see Chart 2) and it was higher for counties up t o
50 miles from SMSAs than for the metropolitan
areas. In general, the rule seems to be, the smaller
the city size, the higher the growth rate of
manufacturing. While both total employment
and manufacturing employment grew by 49
7
Table 2
The Extent of Rural Industrialization in Thirteen Southern States:
Total Nonfarm Employment and Manufacturing Employment Changes of
Southern Counties, 1959-1969, by Distance from the Nearest
SMSA and by Size of Largest City
Number
of
Counties
Total
Nonfarm
Employment
Total Nonfarm
Employment
Change
1959-1969
Number
Per
Cent
Manufacturing
Employment
Manufacturing
Employment
Change
1959-1969
Number
Per
cent
SMSA counties: total
Population of SMSA more
than 1,000,000
Population of SMSA
250,000-999,999
Population of SMSA less
than 250,000
153
23
5,660,076
1,384,134
2,811,677
965,325
49.7
69.7
1,604,903
340,123
701,916
221,338
43.7
65.1
63
2,996,093
1,270,317
42.4
897,999
335,210
37.3
67
1,279,849
576,035
45.0
366,781
145,368
39.6
Counties 0-50 miles from
SMSA: total
Main city population more
than 10,000
Main city population
2,500-9,999
Main city population less
than 2,500
595
2,050,630
989,771
48.3
963,604
505,508
52.5
127
1,190,025
579,051
48.7
558,786
286,617
51.3
287
717,530
344,467
48.0
344,060
182,501
53.0
181
143,075
66,253
46.3
60,758
36,390
59.9
Counties Over 50 Miles From
SMSA: Total
Main city population more
than 10,000
Main city population
2,500-9,999
Main city population less
than 2,500
553
1,379,489
674,345
48.9
505,585
308,972
61.1
85
630,248
296,225
47.0
230,469
113,034
49.1
244
582,561
291,506
50.0
214,662
148,738
69.3
224
166,680
86,614
52.0
60,454
47,200
78.1
"Texas Oklahoma, Arkansas, Louisiana, Mississippi, Tennessee, Alabama, Georgia, Florida, South Carolina, North Carolina, Virginia, Kentucky.
Source, Thomas Till, "The Extent of Industrialization in Southern Non-Metro Labor Markets in the 1960s," Journal of Regional Science, Vol. 13, No. 3,1973.
percent over this period for the South as a whole,
manufacturing growth was below total growth in
the SMSA counties and above total growth in the
non-SMSA counties. Manufacturing seems to be
seeking out not only southern locations for its
facilities, but it is growing at a disproportionate
rate in small cities and non-urban areas.
What Factors Have
Made the Southeast
So Attractive?
The primary attraction that a region can offer is
lower production costs. However, a region needs
more than low labor costs to attract industry. Also
important are the availability of efficient labor,
access to national distribution, and establishment
of a favorable business climate. An additional
8
factor that will enhance a region's relative attractiveness is the agglomeration of firms within an
industry, or among related supplying and purchasing firms. This concentration of industry
increases the productivity of each firm and
encourages more firms to follow. So the initial
attractions of labor cost and other factors eventually draw enough industry that firms begin t o
come because of the other firms that are present
as well as for the region's natural advantages.
Cost of Labor
The Southeast has generally been acknowledged
as a low labor cost region. A study by Lynn
Browne demonstrated that this cost differential
applies to all industries in the Southeast and is
not simply due to the predominance of low wage
M A R C H 1982, E C O N O M I C REVIEW
Table 3
Growth Rates in Manufacturing Hourly Earnings*
(annual percentage rates of change)
Northeast
North Central
East
West
Nor.
Nor.
Cent.
Cent.
South
East
So.
Cent.
West
West
So.
Cent.
U.S.
New
Eng.
MidAtl.
1960-70
4.0%
4.3%
4.0%
4.1%
4.2%
1970-75
7.4
6.8
7.6
8.1
7.4
7.4
7.5
7.6
6.9
6.7
1960-75
5.2
5.2
5.2
5.4
5.3
5.4
5.3
5.2
4.5
4.8
So.
Atl.
4.4%
4.2%
Mtn.
Pac.
3.4%
3.9%
4.1%
Source: Lynn Browne, "How Different are Regional Wages?" New England Economic Review, Jan.-Feb., 1978, p. 42.
'Geographical divisions in this article are based on U.S. Census regions.
Table 4
Estimated Rates of Growth in Average Earnings Standardized for Industry Mix
(For each region, hourly earnings by industry are averaged with each industry figure weighted according to that industry's relative
importance nationally.)
NE
MA
ENC
WNC
SA
ESC
WSC
MT
PAC
1960-70
4.3%
3.9%
4.0%
4.2%
4.2%
3.9%
4.1%
3.5%
3.8%
1970-75
6.8
7.4
7.8
7.3
7.3
7.3
7.5
6.6
7.0
1960-75
5.2
5.1
5.3
5.2
5.2
5.0
5.2
4.5
4.9
Source: Lynn Browne, "How Different are Regional Wages?" New England Economic Review, Jan.-Feb., 1978, p. 42.
industries. In the study, Browne identifies the
wage differential among the regions of the country
by standardizing the industrial make up, or mix
within each region's economy. This standardization
removes the regional and industrial biases that
arise from geographic specialization and differences among industrial wage rates. After standardizing for industrial mix, the South Atlantic
and East South Central states still have the lowest
industrial wages in the nation.
Despite the relatively lower rates, wages in the
South are growing faster than the national average
(see Table 3). However, there is little evidence
that the South's relative wage position is changing.
The unadjusted wages have grown more rapidly
than average. Yet when wages are standardized
for industrial mix it becomes apparent that the
increase is primarily due to a shift toward higher
FEDERAL RESERVE BANK OF ATLANTA
wage industries, rather than faster growing wages.
This is evident because the adjusted wages for
the South are equal or below the national average.
Only if the adjusted wages were greater than the
national average could it be concluded that the
South was losing its relative low wage position
(see Table 4).
In addition t o wages, total labor costs are also
lower in the South. Workmen's compensation
and unemployment insurance are lower in the
South than elsewhere; eligibility restrictions and
benefit levels are also less. Combined with lower
levels of insured unemployment over the last 30
years, these contribute to lower employment
taxes in southern states.
I n order t o take full advantage of the relatively
lower labor cost, business must also employ
laborers w h o are equally as productive as the
9
TABLE 5
Comparison of Average Hourly Earnings in Manufacturing Unadjusted and Standardized for Cost of Living Differences —1975
Northeast
U.S.
king's
0
~~NE
MAT
North Central
ENC
WNC~
SAT
South
ESC
WSC
MTN
Wfcirt
PAC
$4.81
$4.42
$4.98
$5.60
$4.92
$3.95
$4.07
$4.45
$4.70
$5.31
4.81
4.03
4.73
5.54
5.07
4.11
4.47
4.94
4.95
5.09
Standardized for
Uv1ngf
Source: Lynn Browne, "How Different are Regional Wages?" New England Economic Review, Jan.-Feb., 1978, p. 43.
national average. Without at least equal productivity, the reduction in output could offset the
wage advantage and potentially make production
more expensive than in other regions. However,
studies indicate that southern workers are equally,
if not more, productive than northern workers.
So lower labor cost does translate into lower
production cost.
Cost of Living
Over and above the benefits derived from the
region's low wages, southern businesses receive
a surprisingly large real wage cost advantage (see
Table 5). Before adjustments for cost of living,
southern wages are between 7 percent and 18
percent below average. After adjusting for cost of
living, the wages are between 15 percent below
and 3 percent above average. This represents a 3
to 10 percent improvement in purchasing power
which businesses receive as a windfall. In addition,
even though industry pays lower wages, the
workers receive a relative cost of living bonus
that brings them close to the national average.
The lower real cost has t w o advantages for
industry. It not only enables the payment of
lower nominal wages, but it also reduces the
costs of materials purchased from local markets.
Local prices for products and especially services
will be relatively lower due to the local labor cost
component. The production cost saving from
lower locally purchased goods and services has
not been measured, but it is potentially quite
important.
10
Labor Supply
I n addition t o labor cost factors, businesses are
concerned with the availability of labor when
they consider new sites. W i t h an inadequate
supply of labor, business would be unable t o
sustain the advantage of the relatively low wages.
The excess demand would tend to bid the wages
higher, and thus, reduce the relative advantage.
However, population increase in each southern
state exceeds the national average, and as an
aggregate exceeds the national average by 10
percentage points (see Chart 3).
Included in the population increase is the
resurgence of migration from other regions. Lynn
Browne 1 has analyzed the migration trends from
1958 t o 1977 in her 1979 study, "The Shifting
Patterns of Regional Migration."
Since the 1950s, three discernible trends have
affected labor supply in the South. First, from
1957 to 1967 there was an out-migration from
the rural East and West t o South Central. The
exodus was primarily attributable to the poor
economic conditions of the 1950s and early
1960s. The late sixties comprise the second
period, a transition period when the South Atlantic
attracted a large share of the migration from the
Mid-Atlantic and East North Central. The seventies
'Lynn Browne, "The Shifting Pattern of Regional Migration."
New England Economic Review, November-December 1979, pp. 17-32.
MARCH 1982, E C O N O M I C REVIEW
1
Source: Census Bureau.
,
are the third period, during which there has been
a large out-migration from the industrial belt. The
southern regions attracted the largest share and
are expected t o continue attracting a large share
into the 1980s.
With the combination of the natural birth rate
and the net migration trend, the Southeast provides
an expanding labor pool that exceeds the average
national growth rate. Availability of such a labor
pool provides insurance that an ample supply
will be available as business activity expands in
the South.
In the previous section, w e argued that the
Southeast is the lowest production cost region.
Along with the wage benefit, the region also
provides real purchasing power savings and has
an available supply of workers w h o are as productive as northern workers. However, these
factors have been present in the Southeast
during past decades. Businesses could have
taken advantage of these regional benefits before,
but few did. W h y are businesses presently responding to the regional advantages of the Southeast?
FEDERAL RESERVE BANK OF ATLANTA
Transportation Factors
Decreased transportation costs, increased speed,
and increased accessibility have all contributed
to the ability of businesses to locate in the lowest
production cost areas. With these transportation
economies, the relative importance of transportation has been decreasing when compared with
total costs. This has induced firms to become
more sensitive to changes in other production
costs and more flexible in relocating to regions
that provide such cost advantages. Without the
transportation economies, movement or expansion of production facilities would be hampered
by the relative cost of transporting to and from
established markets.
From a historical perspective, the primary
mode of transportation for commerce has changed
from water (inland and oceanic) t o rail, and then
to highways and air. Each step in the transition
has increased the flexibility of transport and has
produced economies that have changed the
structure of transportation costs. The resulting
economies can be divided into the effects on
11
"The result of all this clustering of related firms is that a new
terminal costs and line haul costs. The shift from
rail to highways, while it increased line haul costs,
reduced the terminal cost. Despite rail's relatively
lower line costs, trucking has a relative advantage
with short hauls and partial shipments, which
makes trucking especially effective for shipping
higher value t o weight products. Furthermore,
trucking's added flexibility enables it t o reach
virtually all domestic markets faster than rail.
Rail's relative advantage remains in the shipment
of lower value to weight products such as raw
materials. It is the combination of the structural
changes in the transportation costs and the shift
from predominance in lower value t o higher
value to weight products that have elevated
trucking to the primary mode of transportation.
The development of the interstate highway
system enhanced the Southeast's ability to attract
industry. The highway system has provided access
t o national markets at competitive costs, which
has enabled southeastern businesses to produce
in a region with lower costs while maintaining
distribution to the national markets they serve.
Local highways in the Southeast have also been
improved. This improves accessibility t o regional
labor and product markets. W i t h the improvements in local roads and the increase in personal
automobile ownership, the radius within which
business can attract workers and source supplies
for rural plants has dramatically increased. This
allows business to take advantage of the lower
production cost of a rural location without restricting access to local labor and supplies.
Concentration of Industry
The complex network of trade between businesses results in a magnified growth effect as
industry begins to locate in a region. Textile
producers have been in the Southeast for many
decades, but it was not until they reached a
critical level of concentration that textile machinery makers and chemical firms tied t o the
textile industry found it advantageous t o locate
near their customers. More recently, the furniture
industry has drawn the makers of machinery,
leather and textiles into its orbit. Electronics is
another industry with many small componentmakers which are attracted by each other's
12
Chart 4. Share of U. S. Manufacturing
Employment 1939-1976
30
Percent
Trad. Southern Industries*
OL
1939
_L
1947
1958
1976
•Textiles, tobacco, food, paper, lumber
"Plastics, leather, primary metals, electrical equipment, instruments,
miscellaneous
Note: 1939 and 1947 proportion based on production workers; 1958
and 1976 based on total employment.
Source: Census of Manufacturing, 1939, 1947,1958, and 1976.
presence in clusters such as those in Florida,
Georgia and North Carolina.
The result of all this clusteringof related firms is
that a new advantage for the Southeast has been
created. In the past, the low production cost in
the region was offset by its distance from the
major industrial areas as well as consumer markets
of the North. But industrial location surveys
today find, somewhat surprisingly, that the Southeast is beginning to reach the level of industry
concentration necessary to give it all the attractions
of the North, in addition to having lower costs.
Business Climate
An important yet highly elusive factor in industry
location decisions comes under the broad heading of business climate. This term is used to
compare states on such characteristics as labor
legislation (primarily right-to-work laws), business
taxes, political attitude toward industrial development (which may be highly subjective) and
the general regulatory posture of the state. Indexes
of business climate are compiled regularly and in
industrial location surveys this factor ranks in the
MARCH 1982, E C O N O M I C REVIEW
^vantage for the Southeast has been created."
top five for all industry and as the number one
consideration for some industries.
It is difficult to sort out just how important
business climate actually is in attracting industry.
W o u l d a favorable business climate in one state
ever induce a company t o locate a plant there if
production cost and distance to markets were
greater than elsewhere? It is hard to answer this
question because the states with favorable business climates are also those with the other
advantages being sought. But the pattern of
industry location in recent years appears to be
oriented more toward production cost and markets than toward business climate. And northern
states which have attempted to improve their
image with business have not been notably
successful in attracting industry due t o their
higher cost structure.
The Southeast7s advantage is that many attractive
features have come together in recent years. The
lower cost structure has always been present,
but it has been joined by favorable developments
in transportation and communication, by the
decrease in transport cost sensitivity of many
products and by the maturing industrial technology
which can make effective use of the labor and
other resources of the Southeast.
—john Hekman
and Alan Smith
FEDERAL RESERVE BANK OF ATLANTA
13
Inflation alone does not account for the sudden surge in currency. Evidence
suggests that cash being held in hoards increased from 56 percent of total
currency in circulation in 1960 to almost 70 percent in 1980.
Explaining
the
Cash
Explosion
The amount of U. S. currency in circulation
today amounts to $121 billion, a four-fold
increase over the $30 billion circulating in
1960. In 1960 there was sufficient currency
in circulation for each person in this country
t o hold $162; today each of us could hold
more than $600 (Chart 1). 1
Let's examine the composition of this
currency measure, by denomination. The
most striking aspect is the spectacular increase in the value of $ 100 notes outstanding: an increase from $5.9 billion (18 percent
of the value of all currency outstanding) in
1960 to $49 billion (almost 36 percent of
the value of all currency) in 1980 (Chart 2).
In terms of value, the $100 note actually
replaced the $20 note in 1978 as the largest
denomination outstanding. Twenty dollar
notes grew slightly less rapidly than $100
notes.
The value of $50 notes grew from 8.6
percent of the value of all notes and coins
outstanding in 1960 to 10 percent of tne
total in 1980, while the stock of coins grew
in importance from 7.4 percent of currency
outstanding in 1960 to better than 9 percent
in 1980. The relative importance of ten and
one dollar notes declined from 1960 to
1980.
'The difference between currency in circulation and the currency
component of M1 is currently held in the vaults of commercial
banks. This vault currency has increased in direct proportion to
the public's increased demand for currency. The ratio of
currency held by commercial banks to the currency component
of M1 has varied within a four percentage point range from
1965 to 1980. This stability confirms that commercial bank
holdings simply reflect the demand for currency by the population.
14 M A R C H 1982, E C O N O M I C REVIEW
Chart 1. Total Currency in Circulation
Outside Treasury and Federal Reserve Banks
140 r
Chart 2. Currency Outstanding by Denomination
$ Billions
$ Billions
$100
120
50
20
10
5
I--
100
80
2
1
60
Coins
40
20
o hrrriUTÏÎ
30
40
50
60
70
80
The Reasons Behind Currency Growth
knows the proportion which has been destroyed
or lost. Also, very little is known about the
amount of ourcurrency in foreign hands. In many
countries, U. S. bills circulate as a second currency,
and foreigners hold an unknown amount of bills
as a portable hedge against economic or political
turmoil in their own countries.
W e know that business firms hold a relatively
small portion of total currency outstanding. Most
currency is held by individuals, and estimates of
that proportion run as high as 90 percent 3
Currency as a percent of overall economic
activity (measured by nominal GNP) fell from 5.7
percent of GN P in 1960 to 4.4 percent in 1980,
for several reasons. Interest rates are higher, so
holding currency costs more in terms of interest
foregone. Also, a number of financial innovations
have appeared as functional substitutes for cur-
Currency is demanded by the population for
two reasons only, to facilitate cash transactions
and to serve as a store of value. In recent years a
number of studies have attempted to explain
growth in currency demand. 2 The conclusions of
this literature are less than satisfying, primarily
because of the difficulties in obtaining good
information on how people hold and use currency. People w h o use more currency for off-thebooks tax avoidance or in illegal transactions, for
example, are unlikely t o admit to it. Few of us
keep track of cash transactions.
Data on individuals' use of currency is scarce.
Researchers are confined basically t o analyzing
the aggregate data on currency outstanding.
Even this data is an estimate. W e know how
much currency has been issued, but no one
'See for Example: Phillip Cagon, "The Demand For Currency Relative to Total
Money Supply", National Bureau of Economic Research, Occasional Paper
62 (New York: National Bureau of Economic Research, 1958). George G.
Kaufman, "The Demand For Currency", Board of Governors of the Federal
Reserve System, Staff Economic Studies, (Washington: Board of Governors
of the Federal Reserve System, 1966). J. Carl Poindexter, Jr., "The Currency Holding Behavior of the Public and the Strength of Monetary Controls",
Graduate School of Business Adminstration, Institute of Finance, New York
University, The Bulletin, No. 67 (New York University, November, 1970).
Robert D. Laurent, "The Growing Appetite For Cash", Federal Reserve Bank
of Chicago, Business Conditions, (Chicago, Federal Reserve Bank of
Chicago, April, 1971). Donald L. Kohn, "Currency Movements in the United
States." Federal Reserve Bank of Kansas City, Monthly Review (Kansas
City, Federal Reserve Bank of Kansas City, April, 1976). Paul S. Anderson
"Currency in Use and in Hands", Federal Reserve Bank of Boston, New
England Economic Review (Boston, Federal Reserve Bank of Boston,
FEDERAL RESERVE BANK OF ATLANTA
March/April, 1977). Robert D. Laurent, "Currency and the Subterranean
Economy," Federal Reserve Bank of Chicago, Economic Perspectives
(Chicago, Federal Reserve Bank of Chicago, March/April, 1979). Norman N.
Bowsher, "The Demand for Currency: Is the Underground Economy Undermining Monetary Policy?" Federal Reserve Bank of St. Louis, Review,
Volume 62, No. 1 (St. Louis, Federal Reserve Bank of St. Louis, January,
1980). Peter Gutmann, "The Subterranean Economy," Financial Analyst
Journal, November/December 1977. Edgar Feige, "The Irregular Economy:
Its Size and Macroeconomic Implications," Social Systems Research Institute,
Working Paper 7916. And Charles J. Haulk, "Thoughts on the Underground
Economy,"Federal Reserve bank of Atlanta, Economic Review. (Atlanta,
Federal Reserve Bank of Atlanta, March/April, 1980).
3
See Paul S. Anderson "Currency in Use and in Hoards", New England
Economic Review, Federal Reserve Bank of Boston (Boston, Federal
Reserve Bank, March/April 1977) p. 23.
15
rency: cash management services, negotiable
certificates of deposit, authorization to offer
savings accounts to state and local governments
and businesses, telephone transfers from savings
accounts, repurchase agreements, preauthorized
third party transfers, N O W accounts and share
drafts, and money market funds. 4 The surprising
thing, in view of the past 20 years of innovation
and interest rates, is that currency did not fall (as
a percentage of CNP) more than it did.
Measuring the "Hoards"
One obvious explanation is that people are
hoarding increasing amounts of currency. To
approach this question, w e need to separate the
amount of currency held for transactions purposes
from the amount of currency "hoarded" as a
store of value. One ingenious method for doing
this was developed by Paul S. Anderson. 5 He
reasoned that every piece of currency in circulation
could pass from hand t o hand only so many
times before it was worn out and had to be
pulled from circulation and replaced by the
Federal Reserve.
He further reasoned that one dollar notes
were the least likely t o be hoarded and the most
likely to experience constant use in transactions.
(To the extent $1 bills are hoarded, the hoarded
components below are underestimated.) He
therefore calculated the average life of a one
dollar note (1.81 years) by dividing the total
number of one dollar notes redeemed in 1980
($1.77 billion) i n t o t h e t o t a l n u m b e r o f one dollar
notes outstanding at that time ($3.26 billion). So
the average one dollar note was replaced every
1.8 years.
The higher denomination bills use the same
paper, and the durability of that paper has not
changed significantly. This implies that every
currency unit—$5 bills, $20 bills, or whatever —
would also wear out in 1.8 years if it were used
entirely in transactions. If higher denomination
"See Marvin Goodfriend, James Parthenos, and Bruce Summers "Recent
Financial lnnovations:Causes, Consequences for the Payment System and
Duplications for Monetary Control." Federal Reserve Bank of Richmond,
Economic Review, March/April 1980.
5
See Paul S. Anderson, "Currency in Use and in Hoards", Federal Reserve
Bank of Boston, New England Economic Review, March/April 1977, pp.
25-28.
16
bills last longerthan 1.8 years, then the longer life
must mean they were at rest, or hoarded, for the
time in excess of 1.8 years.
It then is an easy step t o estimate the portion
of currency in use for transactions and held in
hoards, by denominations. "Currency in active
use" is equal to the quantity of notes of any given
denomination redeemed every year times the
average life of a currency unit in constant use for
transactions (1.8). The difference between currency in active use and the amount of currency
outstanding, in a given denomination, must be
the quantity " n o t in active use"—resting in
hoards. For example, $1.4 billion in $50 notes
were redeemed in 1980. If w e multiply that
figure times 1.8 (the average life of a currency
unit in 1980), w e obtain $2.5 billion, which is our
estimate of the amount of $50 notes required for
transactions, not hoarding. The amount of $50
notes actually in circulation was $13.1 billion in
1980, which means that $10.6 billion worth of
$50 notes ($13.1 - $2.5) were at rest, or hoarded.
Similar calculations for each denomination yield
the total amount of currency in active use and
the total hoarded.
Table 1 shows the estimates for currency in use
and currency in hoards calculated at five-year
intervals for the period 1960 through 1980.
Table 1
Estimated Currency in Use and in Hoards
.(in $ Billions)
Currency Currency Currency
Outstanding
in U s e in H o a r d s
1960
1965
1970
1975
1980
$29.1
35.5
47.7
72.3
117.4
$12.8
13.4
20.0
25.4
36.2
$16.3
22.1
27.7
46.4
81.2
Cur. in H o a r d s
as p e r c e n t of
Cur. O u t s t a n d i n g
55.9%
62.3
58.0
65.8
69.1
During the period 1960 through 1980, the average
length of life for a Federal Reserve Note was 1.7
years, so w e used that figure for average transactions life rather than 1.8 (the 1980 figure).
Adding these calculations across denominations
indicates that the percent oftotal currency at rest
in the economy, or in hoards, increased to almost
MARCH 1982, E C O N O M I C REVIEW
70 percent of total currency in circulation in
1980, up from 56 percent in 1960. This suggests
that the largest part of recent growth of currency
is not a result of increased demand by the public
for transactions, but instead increased demand
for hoarding purposes.
This inference is further supported by similar
calculations in Table 2, which suggest, sensibly,
Table 2
Estimated Currency in Use and in Hoards
by Denomination in 1980
Currency in
Active Use
$1
$2
$5
$10
$20
$50
$100
outstanding are hoarded seems plausible. So
Tables 1 and 2 suggestthat currency hoarding is a
major part of the answer as t o w h y currency has
grown rapidly during the 1960-80 period.
Our estimate of the component of currency in
active use for transactions, relative to nominal
GN P, actually declined from 2.5 percent in 1960
t o 1.3 percent in 1980 (Table 3). The residual
amount of hoarded currency, however, appears
to have remained a relatively constant percentage
of GNP, ending the period just one tenth of a
percent lower than it started. It appears from our
estimates, therefore, that the demand for currency
Currency in % of Denom.
Hoarded
Hoards
$3,015,601.1 5;
248,883.9
72,743.0
596,792.0
3,785,512.4
548,310.6
7,290,349.7
3,823,569.3
15,241,055.9
22,933,160.1
2,448,805.8
10,622,279.2
4,379,183.0 42,374,940.0
Table 3
Currency in Circulation and
Estimated Currency in Use and in Hoards
as a Percent of GNP
Total Currency Currency in Currency
in Circulation
Active Use in Hoards
7.6%
89.1
12.7
34.4
60.1
81.3
90.6
1960
1965
1970
1975
1980
that in 1980 the large denomination notes were
the ones w i t h the largest hoarding component.
The higher denomination notes are easier t o
store (and possibly to hide) than lower d e n o m i n ation notes. Therefore, the demand for currency
for hoarding purposes would more likely involve
the use of high denomination notes. Except for
the t w o dollar denomination, the percentage of a
given denomination of currency hoarded, according to our calculations, increases as the size of the
denomination increases, rising from only 7.6
percent of the one dollar bills t o just over 90
percent of the $100 bills. (The hoarding component of the $1 bill in 1980 reflects our use of the
1.7 year average life over the 1960-80 span
which is lower than the 1980 average life of 1.8
years.)
The only inconsistency is in the t w o dollar note
denomination, 89 percent of which was hoarded.
However, the public's limited acceptance of the
t w o dollar note has limited the number of t w o
dollar notes circulating for transactions. They are
more attractive to collectors. Therefore, the estimate that 89 percent of the t w o dollar notes
FEDERAL RESERVE BANK OF ATLANTA
5.7%
5.1
4.8
4.6
4.4
2.5%
1.9
2.0
1.6
1.3
3.2%
3.2
2.8
3.0
3.1
for transaction purposes has not kept pace w i t h
growth in GNP, while the demand for currency for
hoarding purposes has grown at about the same
pace as GNP.
Implications for Measuring the Money Stock
Currency outside banks is included in the M l
definition of the money stock. But the work in
this article suggests that currency includes both
an active and a hoarded component. The active
component logically belongs in both M l and
M2, but the hoarded component should logically
be excluded from M1, because M 1 is a measure
of transactions balances. The motives behind
hoarded currency are probably more closely
related to those of time and savings accounts
than t o currency in use for transactions. If this
reasoning is correct, removing the hoarded com17
Chart 3. M1 and M1 Adjusted
Table 4. M1 and M1 Adjusted
(less currency in hoards)
(less currency in hoards*)
1960-1980
(in Billions)
Year
p o n e n t o f currency from M l should increase the
quality of the M 1 series.
Chart 3 shows M l and M l adjusted (less
currency in hoards) for the period 1959 through
1980. M1 adjusted tracks M l very closely up to
1972, and from 1972 forward the gap between
the t w o widens steadily. Since 1972 the directional change has been almost totally in a down-
1960
1961
1962
1963
1964
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
M1
M1 Adjusted
141,358
144.208
147.842
152.317
158.267
165.025
172.558
179.392
191.967
203.358
211.042
225.333
241.467
259.000
271.850
284.200
299.258
320.692
345.342
363.342
379.758
122.943
127.399
131.080
134.290
139.936
146.873
161.322
154.733
168.357
178.997
186.134
195.422
212.429
224.586
231.658
240.297
251.271
265.317
287.290
298.172
307.443
*ln making these calculations, we reduced our estimated of hoarded
currency in circulation by the proportion of currency held outside banks,
recognizing both the stability of the ratios of vault cash to currency and
the fact that M1 excludes vault cash.
ward direction. This suggests that hoarded currency has become a larger and larger percentage
of M l .
—David D. Whitehead
18
MARCH 1982, E C O N O M I C REVIEW
Tax Cuts:
Who Shoulders
the Burden?
The Reagan tax cuts have been criticized as a "welfare for the rich" approach.
Yet analysis of the tax cuts of the 1920s, the 1964 tax cut, and the 1932 tax
increase suggests that the Reagan program will shift the tax burden toward
the rich.
Economists have long recognized that taxes on
economic activity distort prices and reduce
the size of the gains derived by
both buyers and sellers.1 W h e n
taxes are levied on economic
activity, the quantity of the
activity will decline. As a
result, taxes impose a
cost on society over and
above the revenue collected.
The effects of tax cuts
are not so well recognized.
Under a system of progressive
taxation, where tax rates increase as incomes increase, a proportional reduction in tax rates (an equal
'See Robert Keleher, "Supply-Side Effects of Fiscal Policy:
Some Preliminary Hypotheses" (June 1979) and "Supply-Side
Tax Policy: Reviewing the Evidence" (April 1981) Economic
Review, Federal Reserve Bank of Atlanta
FEDERAL RESERVE BANK OF ATLANTA
percentage reduction in all brackets) will provide a
greater incentive for persons in upper income brackets to expand their taxable
income than those w i t h lesser
incomes. As the result, after a
proportional rate reduction,
the tax base will grow more
rapidly in the upper income
brackets than the lower brackets. The effect will be t o shift
the tax burden toward higher
income groups.
Thus, contrary to the current
wisdom, w e believe that a roughly
proportional tax reduction—such
as the program recently enacted by Congress—will shift the tax burden toward highincome taxpayers. In this article, w e outline
the theoretical case for this surprising proposition, and then investigate the historical record. Our analysis focuses on the 1964 tax
cut, the tax cuts of the 1920s, and the 1932
tax increase.
19
Table 1: The Impact of a 20 percent Proportional Tax Rate Reduction on the After-Tax Income of Taxpayers for Selected Marginal
Tax Rates.
Initial Marginal
Tax Rate for
Income Bracket
Marginal Tax
Rate After a
20% Rate
Reduction
Percent of Each
Additional Dollar
Earned that the
taxpayer is
permitted to keep:
(1)
(2)
Prior
to
Reduction
(3)
10
20
40
50
60
80
90
8
16
32
40
48
64
72
90
80
60
50
40
20
10
Tax Rates
and Taxable Income
Changes in marginal tax rates influence both
the supply of labor and capital and the efficiency
with which these resources are transformed into
desired goods. When tax rates change, individuals
have an incentive to make several adjustments
that will alter their taxable incomes. For example,
rising marginal tax rates encourage persons t o
substitute (a) leisure for work, 2 (b) tax deductible
expenditures for taxable expenditures, (c) tax
shelter investments for investments that generate
taxable income, (d) unreported income derived
from the underground economy for taxable
reported income, and (e) current consumption
for savings (taxable future income). Each of these
substitutions will reduce the individual's tax base
and thereby reduce his or her tax liability. In
short, higher tax rates reduce the personal reward
derived from productive activities, while making
tax avoidance more attractive.
Of course, tax avoidance is costly. Many tax
avoidance projects are profitable t o taxpayers
only when they face exceedingly high marginal
tax rates. Such projects typically have low returns
'Individuals may also substitute "leisure intensive" jobs for less enjoyable
but more remunerative "work intensive" jobs. Since the desirable nonmonetary elements of a job are not taxable, high marginal tax rates reduce
the incentive of workers to undertake various sacrifices (e.g., inflexible work
schedule, out-of-town travel, job pressure, and hectic work environment) in
order to earn additional taxable income.
20
Percent Increase
in the taxpayer's
Return from an
Additional Dollar of Income
Subsequent
to
Reduction
(4)
(5)
92
84
68
60
52
36
28
2.2
5.0
13.3
20.0
30.0
80.0
180.0
t o society but are "subsidized" for the individual
by a reduced tax liability. Lower marginal tax
rates will reduce people's incentive t o engage in
tax avoidance, by permitting them to capture a
larger proportion of the income derived from
work, investment, saving and other activities that
generate goods and services in exchange for
taxable income. Therefore, if supply-side incentive
effects are important, taxable income will expand in
response t o reduced tax rates.
The Incentive Effects of a
Proportional Tax Rate Reduction
A proportional tax reduction is one for which
the percentage reduction in tax rates is equal.
Table 1 illustrates the pattern of a 20 percent
proportional rate reduction, where the 10 percent
marginal rate is reduced to 8 percent. Similarly,
the 50 percent marginal tax rate is reduced t o 40
percent, down 20 percent from the initial rate.
From an incentive standpoint, the after-tax
income (or take-home pay) that taxpayers are
permitted to keep if they expand their earnings is
critical. The larger the share of additional earnings
that the taxpayer can capture, the greater the
incentive t o work, save, and invest t o generate
additional taxable income. The impact of a
proportional tax reduction on the after-tax "marginal income" of taxpayers will vary substantially
across income groupings.
For example, after a 20 percent rate reduction,
a low income taxpayer w h o had a 10 percent
marginal tax rate finds himself permitted to keep
M A R C H 1982, E C O N O M I C REVIEW
92 cents of each dollar earned compared to 90
cents prior t o the tax cut. The tax cut increases
the after-tax income per additional dollar earned
by 2.2 percent for this taxpayer. In contrast, taxpayers in the 50 percent marginal rate bracket
will experience a 20 percent increase in after-tax
earnings (from 50 cents to 60 cents per additional
dollar earned) as the result of the proportional
tax cut.
tax rates) by the largest amount. Thus, as tax rates
rise proportionally, high-income taxpayers will
have stronger incentive t o shift t o tax avoidance
than those with lesser incomes. Since the tax
increase will exert a larger negative impact on the
size of the tax base in higher income brackets, as
tax rates rise, the share of tax revenue collected
from high income taxpayers will decline.
The increase in a taxpayer's after-tax income
per dollar of marginal earnings is directly related
t o his initial marginal tax rate. For a progressive
tax system, this means that high-income taxpayers
will experience the largest increase in after-tax
earnings (per dollar of additional income) for a
proportional rate reduction. Therefore, the incentive effects of the rate reduction will exert a
greater impact upon the taxable income of highincome recipients than on low-income taxpayers.3
That is what standard economic theory tells us
should happen when tax rates are increased or
reduced proportionally. N o w let's examine the
historical record of such tax changes and see
what actually did happen.
An interesting corollary follows from this pattern
of incentive effects. Because the incentive effects
are greatest for wealthier taxpayers, they will be
more likely than those with lower incomes to
shift resources from various forms of tax avoidance
into the generation of taxable income. Predictably,
taxable income will grow most rapidly in higher
income brackets. The rapid growth of taxable
income in high-income brackets implies that the
tax revenue derived from high-income taxpayers
will decline by a smaller amount (or rise by a
larger amount) than for taxpayers in lower income
brackets. 4 Therefore, a proportional reduction in
tax rates will shift the tax burden (as measured by
the share of tax revenue derived from each
group) toward high-income taxpayers.
The analysis is perfectly symmetrical. A proportional increase in tax rates will reduce the aftertax earnings per dollar of additional income for
those with high incomes (and high initial marginal
3
Some economists have argued that since a tax cut also increases the level
of income for the taxpayer, the "income effect" may encourage individuals
to reduce their work effort, at least partially offsetting the impact of the
substitution effect While this is true for the individual, it is highly questionable
when applied to the aggregate labor market. When considering the income
effect for the economy as a whole, we must also look at how the change in
tax rates (and revenues) affects the availability of goods supplied through
the public sector. For a detailed analysis of this issue, see J. Gwartney and
R. Stroup, "Labor Supply and Taxation: A Correction of the Record,"
unpublished paper available from the Center for Political Economy and
National Resources, Montana State University.
"Tax revenues are equal to (a) the applicable tax rate(s) multiplied by (b) the
size of the tax base. When the rates are reduced proportionally, tax
revenues will vary directly with changes in the "taxable income base." Since
the positive incentive effects of a proportional tax reduction are greatest in
the upper income brackets, the rate reduction will cause both taxable
income and the revenues collected in the upper brackets to expand relative
to the lower brackets.
FEDERAL RESERVE BANK OF ATLANTA
The Distributional Impact
of the 1964 Tax Cut
The 1964 tax cut has often been referred to as
an "across the board" proportional tax cut. While
the rate reductions were accompanied by an
expansion in the number of income brackets, the
tax cut was approximately proportional. For the
lowest income bracket, the 20 percent rate of
1963 was reduced an average of 22.5 percent.At
the top of the income scale, the maximum
marginal rate was reduced 23 percent In between,
marginal rates for most income brackets were
reduced by 17 percent, to 21 percent.
Given the pattern of these incentive effects, our
theory indicates that the most rapid income
gains should be registered by upper-income
taxpayers.
The historical data support this theory. Measured in constant 1963 dollars, the total income
derived from tax returns w i t h an adjusted gross
income (AGI) of less than $10,000 rose only 1
percent. In contrast, taxpayers with adjusted
gross incomes between $10,000 and $50,000
registered approximately a 33 percent increase
in aggregate income. Moving up t o still higher
income brackets, where our analysis indicates
the most significant incentive effects of the rate
reductions, still larger gains in AGI are found.
Adjusted gross income derived from returns with
an AGI of $50,000 to $100,000 rose by 39.1
percent. For income brackets above $100,000,
constant dollar gains in AGI registered between
52.1 percent and 71.6 percent over the two-year
period.
21
Table 2: The Growth Rate of Adjusted Gross Income and Tax Revenues for Tax Returns Reporting Income of $50,000 or more
Prior to and Subsequent to the 1964 Tax Cut (Data Are Measured in Constant 1963 Dollars)
Adjusted Gross Income
(Billions of Constant
1963 Dollars)
Tax Revenue Collected
(Billions of Constant
1963 Dollars)
Returns with AGI
Greater than $50,000
Returns with AGI
Greater than $50,000
Year
1960
1961
1962
1963
1964
1965
1966
Averaqe
Growth Rate
(percent):
1961-63
1964-66
$11.99
13.75
13.62
14.60
17.67
21.33
23.65
$4.54
5.21
5.04
5.38
6.08
7.20
7.97
7.0
17.5
6.1
14.1
Average Tax
Rate
37.9
37.9
37.0
36.8
34.4
33.8
33.7
—
Source: Internal Revenue Service, Statistics of Income: Individual Income Tax Returns (Annual).
-4
Chart 1 sheds additional light on the impact of
the 1964 reductions on returns reporting an
adjusted gross income of $50,000 or more during
the 1959-66 period. Between 1959 and 1963,
the number of these high-income returns ranged
from 125,000 t o 162,000. Only a slight upward
trend was observable. After the rate reduction,
the number of high-income returns grew rapidly,
reaching 272,000 by 1966. Clearly, the tax cut
was accompanied by an acceleration in the
number of returns with an AGI of $50,000 or
more.
The Growth of Tax Returns Reporting
Income of $ 5 0 , 0 0 0 or More, 1959-66
After Tax Cuts
—
Before Tax Cuts
.
Table 2 presents data on the growth of adjusted
gross income and tax revenues collected from
returns with an AG I of $50,000 or more for the
1960-66 period. The adjusted gross income
derived from these high-income taxpayers grew
at a 7 percent annual rate during the three years
immediately prior to the tax cut. The picture was
quite different for 1964-66. During the three
years subsequent to the rate reductions, the AG I
(measured in constant 1963 dollars) derived
from returns reporting incomes of $50,000 or
more, rose at an annual rate of 17.5 percent!
The growth rate of tax revenues derived from
these taxpayers followed a similar pattern. For
the 1961-63 period, the real tax revenue collected
from returns with an income of $50,000 or more
rose at an annual rate of 6.1 percent. In the three
years after the tax cut, tax revenues collected
from these taxpayers grew at an annual rate of
14.1 percent. Even though the average (and
more importantly the marginal) tax rate of taxpayers with incomes of $50,000 or more declined,
the constant dollar growth rate of revenues
collected in this category rose substantially.
Are Other Factors at Work?
•59
'60
'61
'62
'63
'64
'65
'66
Source: Internal Revenue Service, Statistics of Income: Individual
Income Tax Returns (Annual)
22
W i t h the passage of time, many factors other
than the incentive effects of lower marginal tax
rates may contribute to income growth. As real
MARCH 1982, E C O N O M I C REVIEW
Table 3: The Share of Tax Revenue Collected from Various Percentile Groupings Ranked According to Adjusted Gross Income
Prior to and Subsequent to the 1964 Reduction in Tax Rates
Percentile of All Returns
(Ranked from Lowest to
Highest Income)
Tax Revenues Collected from Group
(in billions of 1963 dollars)"
1963
Bottom 50 percent
50 to 75 percentile
75 to 95 percentile
Top 5 percent
Total
a
$ 5.01
10.02
16.00
17.17
$48.20
Percent of Personal Income
Taxes Collected from Group*
Percent
Change
1965
$ 4.55
9.61
15.41
18.49
$48.06
9.2
4.1
3.7
7.7
0.3
1963
1965
10.4
9.5
20.8
20.0
33.2
35.6
32.1
38.5
100.0
100.0
These estimates were derived via interpolation.
Source: Internal Revenue Service, Statistics of Income: Individual Income Tax Returns (1963 and 1965).
income grows, individuals will move to higher
income brackets. Even for a period as short as
two years, the shifting of individuals across income
brackets for reasons unrelated t o the tax cut
could reduce the reliability of data based on
income brackets. For example, if such growth
reduced the total number of taxpayers in lower
income brackets and shifted them into higher
ones, that alone could explain the expansion of
higher income groups' gross income.
To correct for such shifts across income classes,
we analyzed the change in net income by percentile groupings. For the bottom 50 percent of
returns, measured in constant dollars, the adjusted
gross income rose from $70.75 billion in 1963 to
$77.66 billion in 1965, an increase of 9.8 percent
Other groupings registered even more rapid
gains. Returns with incomes in the 50 t o 75
percentile and 75 to 95 percentile range experienced growth rates in adjusted gross income of
13.1 percent and 12.4 percent respectively. Just
as our analysis w o u l d expect, the largest income
gains were registered by the top 5 percent
grouping—the category where the decline in
marginal tax rates increased after-tax earnings by
the largest amount. The adjusted gross income
(measured in constant 1963 dollars) in this top 5
percent category, representing returns with an
income of more than $13,667 in 1963 and
$15,000 in 1965, rose from $76.24 billion in
1963 to $88.86 billion in 1965, an increase of
FEDERAL RESERVE BANK OF ATLANTA
16.6 percent. Thus, the growth rate of adjusted
gross income in the top 5 percent group was
nearly 70 percent greater (16.6 percent compared
to 9.8 percent) than the growth for the lowest 50
percentile, where incentive effects of the tax cut
would have been weakest.
Given this pattern of income, Table 3 illustrates
the change in tax collections according to percentile groupings. For the bottom 50 percentile, tax
revenues (measured in 1963 dollars) declined
from $5.01 billion in 1963 t o $4.55 billion in
1965, a reduction of 9.2 percent. Clearly, the
evidence does not support the backward bending Laffer Curve for this grouping. 5
Revenue collections from returns in the 50 t o
75 percentile and 75 to 95 percentile also
declined, although by a smaller percent than for
the lowest income grouping. In all three of these
income categories, the negative impact of the rate
reductions on tax revenues overshadowed the
positive impact of income growth on tax revenues.
Therefore, tax revenues collected in these income
categories in 1965 were lower than in 1963.
5
The backward bending portion of a Laffer Curve results when lower tax
rates cause income to rise by an amount sufficient to induce an increase in
tax revenues While this was not true for the bottom 95 percent of taxpayers,
the data do indicate that it may have been present for the top five percent.
Marginal tax rates for this group of high income recipients declined from the
30-91 percent range in 1963 to the 25-70 percent range in 1965.
Nonetheless, as Table 3 shows, the tax revenue collected from them was
7.7 percent greater in 1965 than for 1963.
23
The picture for the top 5 percentile of returns
was quite different. In this category, measured in
constant 1963 dollars, federal tax revenue collections from personal income rose from $17.17
billion in 1963 t o $18.49 billion in 1965, an
increase of 7.7 percent. While tax collections
declined between 1963 and 1965 for all other
groupings, the top 5 percent of returns registered
a healthy 7.7 percent expansion. Tax revenue
grew because the rapid expansion in taxable
income more than offset the loss of tax revenue
from the rate reductions.
Just as our theory predicts, the roughly proportional tax rate reduction of 1964 shifted the tax
burden t o high income groupings. Since the
growth of income was more rapid for high than
low income taxpayers, the proportional reduction
in rates increased the share of tax revenue
collected from more prosperous recipients. As
Table 3 illustrates, in 1965 the 5 percent of
returns with the highest incomes paid 38.5
percent of the income taxes, compared t o only
35.6 percent in 1963, prior t o the rate reduction.
Simultaneously, the proportion of income tax
revenue derived from the bottom 50 percent fell
from 10.9 percent in 1963 to 9.5 percent in
1965.
The Tax Cuts of the 1920s
In addition to the 1964 tax cut, there have
been t w o other instances in this century of major
peacetime tax rate changes in the United States.
During the 1920s a series of tax reductions
substantially altered the tax structure. In 1932,
the largest peacetime tax rate increase in U. S.
history was imposed. W e will briefly analyze the
nature of these t w o changes and investigate their
impact on the growth of income (and tax revenue)
across income brackets.
Three major tax reductions—engineered by
Secretary of the Treasury Andrew Mellon—were
enacted during the 1920s—in 1922, 1924 and
1926. Like the supply-siders of today, Mellon
argued that the high marginal tax rates imposed
during World War I retarded economic growth,
while encouraging the flight of income into areas
exempt from taxation. 6 Opponents of the tax
reductions argued that high tax rates were necessary t o generate revenue that would permit a
reduction in the huge debt incurred during
World War I. Believing there was a roughly
proportional relationship between tax rates and
24
"... the tax base
proved highly responsive
to changes in the
incentive structure
during the 1920s."
tax revenues, many congressmen concerned
about the size of the debt were reluctant to
reduce the high marginal surtax rates that had
been imposed during the war.
There was also widespread disagreement about
the distributional impact of the tax reduction.
Predictably, Mellon argued that lower rates would
reduce the incidence of tax avoidance and lead
to an expansion in the tax base. He believed that
lower rates would increase the tax revenue
collected from the rich. Critics of the tax cuts
argued that they were a boon t o the rich. In his
summary of the Revenue Act of 1926 published
by the prestigious American Economic Review,
Professor Roy Blakey of the University of Minnesota concludes, " I n a word, taxes on the rich,
especially on the very rich, have been greatly
reduced. In order to secure political support for
this big reduction (in marginal tax rates) on large
incomes and estates, smaller tax cuts were handed
out all down the line." 7
Despite the opposition, tax cut advocates
carried the day. W h e n considered as a package,
t w o characteristics of the tax reductions during
the 1920s stand out. First, the marginal tax rates
in the upper income brackets were slashed.
"In his book Taxation: The People's Business published in 1924,
Secretary Mellon wrote:
"The existing system of taxation was formed to meet wartime conditions.... The vital defect in our present system is that the burden is
borne by wealth in the making, not by capital already in existence. We
place a tax on energy and initiative; and at the same time provide a
refuge in the form of tax-exempt securities, into which wealth that has
been accumulated or inherited can retire and defy the tax collector."
'Roy Blakey, "The Revenue Act of 1926," American Economic Review
(September 1926), p. 401.
MARCH 1982, E C O N O M I C REVIEW
Table 4: The Tax Revenue Collected According to Income Groupings Prior to and Subsequent to the Series of Reductions in Tax
Rates Instituted During the 1920s.
Net Income Grouping
Less than $10,000
$10,000 to $25,000
$25,000 to $50,000
$50,000 to $100,000
Over $100,000
Total
Tax Revenue Collected from Income Grouping
(In millions of constant 1929 dollars)
Percent
1921
1926
Change
$155.1
121.8
108.3
111.1
194.0
$690.2
$ 32.5
70.3
109.4
136.6
361.5
$710.2
-79.3
-43.2
+ 1.0
+ 23.0
+ 86.3
+ 2.9
Percent of Tax Revenue
Collected from Grouping
1921
1926
22.5
17.6
15.7
16.1
28.1
100.0
4.6
9.9
15.4
19.2
50.9
100.0
Source: Internal Revenue Service, Statistics of Income (Annual).
While the marginal rates ran t o 73 percent in
1921, the top rate was cut to 2 5 percent by 1926.
Clearly, the progressivity of the personal income
tax was substantially reduced during the fiveyear period. Second, the personal income tax
liability was eliminated or virtually eliminated for
low-income recipients during this same period.
As the personal exemption allowance and the
minimum income for filing were increased, millions
of taxpayers were removed from the tax rolls.
The number of returns filed declined from 6.66
million in 1921 to 4.14 million in 1926. 8
The economy's performance duringthe 192126 period was quite impressive. Price stability
accompanied a rapid growth in real output. The
tax cuts granted the largest increases in after-tax
take home pay t o those in the highest income
brackets. 9 At the bottom of the income scale, the
take-home pay of a married couple earning
between $5,000 and $10,000 rose from the 8991 percent range in 1921 to the 95-97 percent
range in 1926. In contrast, the take-home pay on
marginal earnings for a married couple w i t h a net
income of $100,000 or more rose from the 27 to
40 percent range in 1921 t o 75 percent in 1926.
Clearly, the tax rate reductions during the 1920s
indicate that the incentive t o earn additional
"The number of taxable returns declined from 3.59 million in 1921 to 2.47
million in 1926.
9
Remember, after-tax take-home pay per marginal dollar of earnings is the
major determinant of one's incentive to earn additional taxable income. Of
course, the percent of one's take-home pay per marginal dollar of earnings
is equal to 100 percent minus one's marginal tax rate.
FEDERAL RESERVE BANK OF ATLANTA
taxable income was enhanced most for people in
the upper income brackets.
H o w was the growth of income across income
groupings affected by the tax cut? Measured in
dollars of constant purchasing power, the net
income of taxpayers in the $5,000-$ 10,000 bracket rose 63.1 percent between 1921 and 1926, a
healthy increase of more than 10 percent annually.10 However, those in higher income brackets
registered even more rapid gains. For the $10,000$25,000 grouping, net income reported for tax
purposes rose by 82.9 percent during the fiveyear interval. Persons with incomes between
$25,000 and $100,000 registered gains of more
than 100 percent during the period. At the t o p of
the scale, where marginal tax rates were reduced
dramatically from 73 percent to 25 percent, the
reported net income of returns with incomes of
$100,000 or more rose 421.9 percent (a compound
annual rate of 39 percent) in just five years! Just
as Secretary Mellon had predicted, high income
individuals apparently adjusted their affairs, substituting away from the various forms of tax
shelters and leisure toward the earning of taxable
income as their marginal rates were reduced
substantially.
Table 4 presents evidence on the impact of the
1920s rate reductions and other tax changes on
distribution of the tax burden. In aggregate, real
,0
Since the filing status and exemption allowances were also changed, it is
not
possible to make income comparisons between 1921 and 1926 for
brackets below $5000.
25
tax revenues were 2.9 percent higher in 1926
than in 1921. However, tax revenue collected
from income groupings of less than $25,000
actually declined. The tax revenue derived from
incomes of $10,000 or less fell from $155.1
million in 1921 t o $32.5 million in 1926, a
decline of 79 percent. In contrast, collections
from taxpayers reporting a net income of $50,000
or more rose from $305.1 million in 1921 t o
$498.1, an increase of 63.2 percent.
Clearly, even though the progressivity of the
personal income tax was reduced substantially,
the tax changes of the 1920s shifted the burden
toward high-income recipients. As the last t w o
columns of Table 4 show, only 4.6 percent of the
personal income tax revenues collected in 1926
came from returns with a net income of $10,000
or less, compared t o 22.5 percent in 1921. In
contrast, in 1926, fully 50.9 percent of the total
income tax revenue came from returns with an
income of $100,000 or more, compared t o only
28.1 percent collected from this same grouping
in 1921.
These findings illustrate that the tax base
proved highly responsive t o changes in the
incentive structure during the 1920s. Like our
analysis of the 1964 rate reductions, as a result of
the strong response of high-income taxpayers,
the tax cuts of the 1920s actually shifted the tax
burden to the higher income brackets even
though the rate reductions were greatest in this
area.
The Tax Increase of 1932
The view that government should at least
balance its budget was an accepted orthodoxy in
the 1930s. Throughout the 1920s the budget
had run a surplus. Substantial progress was made
toward retiring the World War I debt. Even
during the recession year of 1930, the federal
budget ran a small surplus. As the recession
worsened during 1931, however, a budget deficit
was incurred. As the budget situation continued
to deteriorate in 1932, President Herbert Hoover,
assisted by the newly elected Democratic majority
in the House of Representatives, moved to
reinstitute the high marginal tax rates of the early
1920s. For taxpayers with less than $10,000 net
income, rates were increased from the 1.5 to 5
percent range in 1931 to 4 to 9 percent in 1932.
26
The progressivity of the system was increased
sharply. A person with $50,000 of taxable income
confronted a 30 percent marginal rate in 1932,
compared t o 18 percent in 1931. For $100,000 of
taxable income, the rate jumped from 25 percent
to 56 percent. While the 25 percent rate constituted a ceiling in 1931, the effective marginal
rates ran up to 63 percent in 1932.
What impact did these rate increases have on
taxable income? In aggregate, the net income
reported to the IRS fell by 4.7 percent duringthe
first year the tax rate increases were imposed.
The percentage decline across income groups
was highly uneven. The largest decline in reported
income was registered at the very top of the
income pyramid. Measured in constant 1931
dollars, the reported net income from returns
with incomes in excess of $300,000 fell a whopping 49.1 percent in a single year. Several other
income classes above $10,000 registered doubledigit reductions in reported net income.
In contrast, the net income derived from
returns with less than $10,000 of income (less
than $8,985 in 1932 in order t o adjust for the
deflation), actually increased slightly, by 2.8
percent. Just as the 1920s data indicates that a
reduction in marginal tax rates rapidly increased
the reported net income from high income
returns, the 1932 data suggest a rapid decline in
the reported net income in upper brackets as the
marginal rates increased sharply.
Table 5 illustrates the impact of the 1932 tax
increase upon tax revenues according to income
classes. Overall, measured in constant dollars,
tax revenues rose by 49.2 percent. However, the
increase in tax revenue was far greater in the
lower income groupings—particularly the under
$10,000 category. Tax revenues collected from
returns w i t h a net income of $25,000 or less rose
from $51.6 million in 1931 t o $134.0 million in
1932, an increase of 160 percent in a single year.
While the marginal rates on incomes above
$25,000 rose sharply in 1932, the revenue increases
derived from the high rates were much more
modest. Measured in constant dollar terms, the
total tax revenue collected from returns reporting
more than $25,000 of net income rose from
$194.6 million in 1931 t o $233.4 million in 1932,
a growth of 19.9 percent.
The big increases in marginal tax rates levied
against large incomes did increase revenue collections from these classes. However, the growth
MARCH 1982, E C O N O M I C REVIEW
Table 5: The Tax Revenues Collected According to Income Groupings (Measured in Constant 1931 Dollars) Prior to and
Subsequent to the Tax Rate Increase of 1932
Net Income Grouping
(constant 1931 dollars)«
Less than $25,000
$25,000 to $50,000
$50,000 to $100,000
$100,000 to $300,000
Over $300,000
Total
1931
Tax Revenues Collected from Grouping
(in millions of constant 1931 dollars)13
1932
Change
$ 51.6
40.1
44.8
51.9
57.8
$246.1
$134.0
48.1
51.7
66.2
67.4
$367.3
+
+
+
+
+
+
159.7
20.0
15.7
27.6
16.6
49.2
Share of Tax
Revenues Contributed
IMI
1932
% 21.0
16.3
18.2
21.1
23.5
100.0
% 36.5
13.0
14.1
18.0
18.4
100.0
"The parallel constant dollar brackets for 1932 are: (a) less than $8,985, (b) $8,985 to $22,462, (c) $22,462 to $44,923, (d) $44,923 to $89,846 and (e) over $89,846.
b
The 1932 tax revenue data were inflated to adjust for the decline in prices during the period.
Source: Internal Revenue Service, Statistics of Income (Annual)
of tax revenues was much more modest than for
the lower income categories. In the $50,000$100,000 brackets, where the marginal tax rates
approximately doubled, tax revenues expanded
by only 15.7 percent. Similar revenue growth
was derived from incomes above $100,000,
where marginal tax rates more than doubled.
Thus, for high incomes, the large increases in tax
rates did not increase tax revenues by very much.
Implications of the Analysis
Our analysis indicates that the tax base responds
to changes in marginal tax rates. Economic theory
indicates that a reduction in marginal tax rates
will increase people's incentive to earn more
taxable income. Further, this incentive will be
higher for higher income groups.
Like the 1964 rate reductions, the 1981 tax cut
is roughly an across-the-board proportional reduction in tax rates. By the time it is fully effective in
1984, the tax rates on money income will be
between 20 and 25 percent lower than the
parallel marginal rates of 1980. 11 Nonetheless, it
has often been criticized as a "welfare for the
" O f course, these data refer to income measured in money terms. Since the
indexing of the rate structure is not scheduled to begin until 1985, inflation
will continue to push persons with a constant real income into high marginal
tax brackets. Our work indicates than on average, once account is taken for
inflation, the marginal tax rates of a person with a 1980 income of less than
$50,000 will be virtually unchanged if earnings and the annual inflation rate
during 1981-84 averages between 10 percent and 12 percent. Therefore, it
is not obvious that significant supply-side incentive effects will result from
the Reagan tax cut.
FEDERAL RESERVE BANK OF ATLANTA
rich" approach. Even highly placed administration
officials have implied that the tax cut's major
objective was t o reduce the taxes of high-income
recipients. Therefore, it is widely perceived that
the Reagan tax cut will shift the tax burden away
from high income taxpayers and toward those
with lesser income.
The historical evidence indicates that this
criticism is misplaced. Previous changes in tax
rates suggest that the tax base will increase most
in those income brackets where the rate change
induces the largest increase in take-home pay
per dollar of marginal income. For the Reagan
plan, as for proportional rate reductions in general,
this means that the positive incentive effects will
be greatest in the upper income brackets. Predictably, a more rapid expansion of the tax base in
the upper income brackets will lead to an increase
in the share of income tax revenue collected
from high-income groups.
Therefore, the 1981-84 rate reductions seem
likely to increase the share of tax revenue derived
from high-income recipients compared with those
derived from taxpayers with lesser income. Far
from shifting the tax burden toward the poor, the
Reagan program will shift the tax burden toward
the rich.
—James Gwartney
and Richard Stroup
27
tives on
the
v g e ^
, c \ e
s
'
o n
O
b U
. „ ^ e
n^s
.vjefs
O
f
'
^
„
„
«
f e d
a
(
s
<
>
v,v,cate
p e s t
f r e e s "
B a n *
SV
S
Lumbering at Top Speed:
The Check Collection System,
1952 - 1979
Banks now have more incentive than ever to collect checks quickly. But the
historical evidence casts doubt on whether increased local clearing and
enhanced Fed and correspondent collection services will result in significantly
faster check collection.
Like some kind of awkward creature in a race
for survival, the U.S. check collection system has
evolved t o keep pace in its changing environment. Yet the speed of collection has remained
essentially unchanged for the last 27 years.
Today, the check collection system is facing
pressures that seem to presage significant change.
The pricing of Federal Reserve check collection
services, which began in 1981, and high labor
and transportation costs are raising the cost of
check collection, while high interest rates are
making it more important t o collect checks
quickly. In response to these pressures, banks
in many areas are creating or reestablishing
local clearing arrangements, and correspondent banks are upgrading their check collection
services.
FEDERAL RESERVE BANK OF ATLANTA
It seems reasonable to expect a substantially
streamlined check collection system to emerge
from these pressures. However, history suggests
otherwise. The period between 1952 and 1979
also saw considerable social, structural, and technological change, but the banking industry's
responses t o these changes produced no significant improvement in check collection speed.
1952
On a typical day in 1952 the nation's banks of
first deposit received about 28 million checks
from their customers. About 20 percent of
those items were drawn on the banks in which
they were first deposited ("on-us" checks).
The remaining 80 percent were "on-others"
checks.
29
Almost 40 percent of the 22 million onothers checks processed for collection on day
2 were drawn on banks in or near the same
locality as the first receiving bank. These items
were collected from the local banks either
directly or through local or regional clearing
arrangements. Checks drawn on non-local banks
were collected through correspondents and
Federal Reserve Banks. Correspondent banks
received 28.4 percent of all on-others items
sent for collection by banks of first deposit, and
the Fed received 32.1 percent. Thus the Fed
had about a 3.7 percent market share advantage
over the correspondent banking system.
Table 1
Percent of Bank-of-First-Deposit
Items Paid Per Day in 1952
1967
During the late 1950s and throughout the
1960s the banking industry focused on automating check processing operations to alleviate
the pressure generated by continued check
volume growth. The concept of electronic funds
transfer (EFT) was a natural outgrowth of banks'
automation efforts. A Bank Administration Institute
(BAI) study for an EFT system design resulted in
Number paid
(millions)
Percent Paid
1
2
3
4
5
5.835
15.431
5.304
1.116
.128
21.0
55.4
19.1
4.0
0.5
6
.012
7
.004
27.830
*
100.0
Weighted average speed of collection 2.1 days
Speed of Collection
Some checks were processed by only one or
t w o financial institutions while others were
processed by as many as seven banks. The
percentage of items that were paid on each day
and the weighted average number of days each
check remained in the check collection system
are shown in Table 1. Because it is based on a
model that represents only the essential features
of a real-world situation, Table 1 "tends to
overstate the speed of payment and understate
the volume of items in process of collection...
In other words, the outline shows checks being
collected about 25 percent faster than they
actually are collected." 1 This suggests the average
speed may have been as long as 2.6 days. The
estimated average time a check remained in
the process of collection in 1952 was 2.3 days,
which is approximately at the mid-point of the
2.1-2.6 day range.2
Day
*Less than 0.05 percent
an updated model of the check collection system.
On a typical day in 1967, about 74 million
items were received by banks of first deposit,
nearly triple the number received by banks in
1952. 3 A larger population and the increased
financial activity of consumers contributed to
the larger check volume.
A second change revealed by the 1967 model
was that a larger proportion of bank-of-firstdeposit items were on-us checks that were
paid on the same day they were deposited. In
1952 about 1 in 5 bank-of-first deposit items
were on-us checks; in 1967 the ratio was nearly
3 in 10. This change may have been due t o the
concentration and branch expansion that took
place during 1952-1967. In 1967 there were
347 fewer banks and 12,185 more bank offices
than in 1952. 4
Third, the proportion of on-others items
collected from local banks either directly or
through local clearing arrangements dropped
almost 6 percentage points from the 39.5
percent calculated from the 1952 model to
3
'Joint Committee onCheck Collection System, Study of Check Collection
System (Washington, D. C., 1954), p. 133, cited as the Wurts Report.
lbid., p. 22.
2
30
ln the late 1960s the Bank Administration Institute began an EFT study that
was the basis of the BAI project report titled An Electronic Network for
Interbank Payment Communications: A Design Study. Data about check
flows in the United States were collected for that study, and later they were
used as the foundation of another BAI report. The Check Collection
System: A Quantitative Description.
'Federal Deposit Insurance Corporation. Annual Report of the Federal
Deposit Insurance Corporation for the Year Ended D e c e m b e r 3 1 , 1 9 5 2
(Washington, D.C, 1953), pp. 84 and 85; and Annual Report of the Federal
Deposit Insurance Corporation: 1967 (Washington, D C. 1968), pp. 156
and 157.
MARCH 1982, E C O N O M I C REVIEW
33.6 percent in 1967. A considerable portion
of this change may be related t o a 27 percent
decrease in the number of local clearinghouses
in operation between 1952 and 1967. 5 However, it is not clear whether the demise of many
clearinghouses was the result of a decline in
the percentage of local items in the mix of
bank-of-first-deposit checks in some locations
or if local clearinghouses were displaced by
other check collection intermediaries without
regard for the local and non-local components
in the item mix.
A final difference was in the number of days
necessary for all checks received in a single
day's deposits to completely clear the system.
The 1952 model indicated that a total of seven
days was required, but the 1967 model showed
the elapsed time to be only five days. Since
both models appeared to assume overnight
transportation, 6 this difference suggests that
some of the circuitous check routing that took
place in 1952 had been eliminated. The
improvement may have been related to the
virtual disappearance of non-par checks by
1967. Some of the circuitous routing in 1952
was caused by banks' effort t o avoid exchange
charges on non-par checks.
Second, the average number of institutions
that processed each check was essentially the
same. While the BAI reported that on average
each check written was processed by 2.6 institutions, it apparently did not consider the
difference t o be significant. The BAI report
noted, "Two theoretical patterns were used by
the 1952 'Study of the Check Collection System'
to arrive at a similar figure of 2.4; one pattern
developed 2.6, and the other 2.2." 7
1979
A third model was developed in 1979 as part
of a study conducted by the Federal Reserve
Bank of Atlanta for the American Bankers Association, the Bank Administration Institute, and
the Federal Reserve System. That model showed
an increase in check volume but stability in the
average number of institutions that processed
each check. The study stated, " O n average,
each of the 32 billion commercial bank checks
written in 1979 was processed by 2.4 institutions. 8 The study also noted, "The iDest condition' estimate of the average speed of collection
was 1.9 days," a miniscule improvement over the
model developed in 1952. 9
Model Similarities: 1952 and 1967
Check Disposition Patterns
The 1952 and 1967 check flow models
displayed t w o striking similarities. First, the
relationship between correspondents' and
Federal Reserve Banks' market share of onothers checks received from banks of first
deposit was virtually unchanged. Both correspondents and Federal Reserve Banks gained
market share as the number of local clearing
arrangements declined; the Fed gained 3.1
percentage points; and correspondents gained
2.8 percentage points. This relatively equitable
distribution of items formerly sent to local
clearings seems to support the hypothesis that
clearinghouse closings were largely the result
of social, structural and technological developments, rather than of a concerted effort t o
improve the check collection system.
5
Rand McNally & Company, Rand McNally Bankers Directory (First 1953
Edition) Chicago, IL, 1953, pp. 298-305; and Rand McNally International
Bankers Directory (First 1968 Edition) Chicago, II., 1968, pp. 62-65.
6
The assumptions underlying the BAI model were not clearly specified in the
report or in the technical appendices, which are contained in the earlier
document, An Electronic Network for Interbank Payment Communications. The appendices only suggest that the Wurts Report model was
used in developing the BAI's model.
FEDERAL RESERVE BANK OF ATLANTA
The nation's banks received 126 million
bank-of-first-deposit items on a typical day in
1979. Of these, 36 million items (29 percent)
were drawn on the banks in which they were
first deposited and were paid on the day they
were received. The 89 million on-others items
received by banks of first deposit were sent t o
the three categories of check collection intermediaries in the following proportions:
Federal Reserve Banks
Local clearinghouses
Correspondents and
other banks
41.2 percent
22.1 percent
36.7 percent
'Linda M. Fenner and Robert H. Long, The Check Collection System: A
Quantitative Description. (Park Ridge, Illinois: Bank Administration Institute,
1970), p. 18. This study will be cited as the BAI Study.
"Federal Reserve Bankof Atlanta. A Quantitative Description of the Check
Collection System (American Bankers Association and Bank Administration
Institute: Washington, D C, 1981), p. 5. This study will be cited as the Atlanta
Fed Check Study.
"Ibid. p. 4.'°Rand McNally & Company. Rand McNally International Bankers
Directory (First 1968 Edition), Chicago, II., 1968, pp. 62-65; and Rand
McNally International Bankers Directory. (First 1980 Edition), Chicago, IL
1980, pp. 50-54.
31
The percentage of on-us checks received as
bank-of-first-deposit items in 1979 was very
similar t o what it was in 1967, and the change in
the distribution of on-others checks was consistent with the 1967 pattern. That is, as the
number of local clearinghouses fell (from 237
in 1967 t o 146 in 1979 10 ) the share of bank-offirst-deposit items collected through local clearing arrangements fell from 33.6 t o 22.1 percent.
Analysis of the 1979 model showed that correspondent banks picked up 5.5 of the 11.5
percentage points lost by local clearinghouses,
and the Federal Reserve System picked up 6.0
percentage points. This consistency with the
market share shift patterns of 1967 suggests
that commercial banks, acting in their role as
banks of first deposit, did not initiate the
changes to improve the check collection system
but only reacted to changes in their external
environment.
10
Rand McNally & Company. Rand McNally International Bankers Directory
(First 1968 Edition), Chicago, II., 1968, pp. 62-65; and Rand McNally
International Bankers Directory. (First 1980 Edition), Chicago, II., 1980,
pp. 50-54.
32
Impact of RCPCs
A change in availability schedules at the Fed,
not a major shift in check collection patterns,
appears t o have been responsible for most of
the reduction in the best-condition average
speed of check collection between 1967 and
1979. The change in availability was a part of
the Fed's Regional Check Processing Center
(RCPC) program implemented in the early
1970s.
The RCPC concept evolved out of the recommendations presented in the Wurts Report in
1954. In some regions of the country the Fed
established regional check processing facilities
that were remote from a Fed head office or
branch and designated the geographical area
surrounding these facilities as RCPC areas. In
other instances, RCPC areas were established,
but check processing was actually performed
at a Fed head office or branch. The RCPC
program incorporated three major changes in
the Fed's check collection policies. First, immediate credit (same day availability) was granted
on items drawn on RCPC banks. Second, the Fed
MARCH 1982, E C O N O M I C REVIEW
Table 2
Impact of RCPCs on Weighted Average
Speed of Check Collection
Percent of Average Day's Bank-of-FirstDeposit Items Paid Over 5-Day Check Flow
Period
Day
AFCS* Model
1
2
28.9%
51.1
18.0
1.8
0.2
3
4
5
100.0%
Weighted Average
Speed of
Collection
1.9
Extrapolation from
AFCS Model**
28.9%
36.8
28.3
6.1
0.6
100.0%
2.2
•Atlanta Fed Check Study
"Extrapolation based on assumption that check disposition patterns
remained constant but availavility of RCPC items was 2 days.
required RCPC banks t o make immediate payment in a form readily available t o the presenting
bank (i.e., reserve account balances at the Fed)
for the cash letters it delivered to them. Finally
nonmember RCPC banks were permitted limited
access to Fed check collection services.
The impact that RCPCs had on the average
speed of check collection under optimal conditions is shown in Table 2. The first column shows
the percentage of items that were paid on each
day of the 5-day flow period depicted in the
1979 check flow model. The second column
shows the percentages that w o u l d have been
paid on each day if items drawn on RCPC banks
had received deferred credit (next-day availability),
as they did under the Fed's pre-RCPC policies.
Table 2 shows that the RCPC program resulted in
about 0.3-day improvement in weighted average
speed of collection.
Conclusion
The check collection system has shown a
surprising degree of stability over the past 30
years. There has been a gradual reduction in
the number of local clearinghouses in operation
and, hence, in the proportion of items collected
via local clearings. However, the volume displaced by local clearinghouse closings has
been redistributed t o correspondents and
Federal Reserve Banks in nearly equal proportions. The difference between the proportion
of on-others items sent by banks of first deposit
FEDERAL RESERVE BANK OF ATLANTA
to the Fed and the proportion sent to correspondents increased less than 1 percentage point
between 1952 and 1979. Neither intermediary
has gained a substantial market advantage over
the other.
Furthermore, on average each check written
has been processed by 2.2 - 2.6 financial
institutions throughout the period. In 1952
and 1967, the average speed of collection was
equal to the average number of processors per
item. The only development that appears t o
have had a noticeable impact on the average
speed of collection is the Fed's RCPC program,
which changed the availability schedule for
items drawn on over 9,000 RCPC banks.
In view of the historical evidence, it seems
reasonable to question whether present efforts
t o establish local clearing arrangements and
enhance Fed and correspondent check collection
services will result in significant improvements
in check collection speed. In fact, one can
question whether the speed at which the
system operates can be substantially improved
(as long as the system relies solely on a paper
document). Certainly the industry can work t o
reduce the number of rejects and exception
items and improve the rate at which return items
are handled. However, the best possible speed
of collection under ideal processing and transportation conditions can only be improved to 1.7
days if the 30/70 percent split between on-us
and on-other items in bank-of-first-deposit checks
remains steady and if all on-others items are paid
on day 2. The potential of the check collection
system is limited by the need t o transport checks
physically and to batch process them.
Because of the limitations inherent in the
check collection system, banks must carefully
evaluate their responses to the changes that
are occurring today. While the check collection
system must continue to function to serve the
needs of payors and payees, and while banks
will certainly seek to profit from the changes, in
the long run investments in the check collection
system may only maintain the status quo in a
system that is reaching its maximum potential.
A delicate balance must be sought between
resources committed to maintaining the check
collection system and those invested in developing and promoting alternatives t o the check.
—Donald L. Koch,
Veronica M. Bennett,
and Paul F. Metzker
33
A benchmark descriptive study of the check collection
system. Contains detailed information not found in any
other source.
A QUANTITATIVE
DESCRIPTION
OF THE
CHECK
COLLECTION
SYSTEM
VOLUMES
l&ll
A publication
of comprehensive
national scope
brought to you by
the Federal Reserve Bank of Atlanta,
the American Bankers Association,
and the Bank Administration Institute
A unique study, conducted by the Federal Reserve Bank of Atlanta
Jointly sponsored by the Federal Reserve System, the American
Bankers Association and the Bank Administration Institute.
343 commercial banks and 48 Federal Reserve check processing facilities gave
detailed information concerning:
• the volume and dollar value of the checks they processed;
• the sources from which they received those checks;
• the ways they processed them;
• and, the ways they disposed of the items they processed.
RETURN THIS
RESERVATION FORM
QUANTITY CATALOG
NUMBER
064600
ro:
DPERATIONS AND
AUTOMATION DIVISION
\merican Bankers Association
1120 Connecticut Avenue, N.W.
A/ashington, D.C. 20036
TITLE
PUBLICATION
UNIT COST
(per set)
A Quantitative Description
of the Check Collection S y s t e m two volume set
$65.00
$50.00
(two or more sets)
GRAND TOTAL $
• I am sending no money now. I understand that a bill will be sent at the time the publication is
shipped. This bill will include a $3.00 per set handling and shipping charge.
Name
Bank
Street Address.
34
TOTAL
City
Telephone.
-State
_ZIP_
Regulation of Bank Capital:
An Evaluation
Shrinking profit margins in the 1980s may well make it
increasingly difficult for banks to maintain currently required
capital ratios. Regulators should consider innovative
alternatives, such as the equal marginal capital approach
or the deregulation of large institutions.
Introduction
How much capital should a commercial bank
have t o ensure its own health and the health of
the banking system? The question has been the
subject of increasingly vigorous discussion in
the industry and its regulatory agencies. 1 The
regulators impose minimum capital ratios on
banks to ensure that public confidence in
financial institutions remains high. Banks, on
the other hand, argue that they are hurt by the
requirements, since they must retain significant
portions of their earnings or raise capital in
direct competition w i t h unregulated firms.
Determining what are adequate capital levels
in commercial banking organizations is a complex and difficult problem. The debate has
centered on t w o major issues. First, what are
adequate levels of capital for each bank and
'The Federal Reserve regulates state chartered member banks, the Comptroller of the Currency regulates national banks, and the FDIC regulates
state chartered nonmember banks. Various states also actively enforce
capitai requirements on their institutions.
FEDERAL RESERVE BANK OF ATLANTA
35
bank holding company, and for the banking
industry? Second, how should achievement of
adequate capital levels be ensured? Should
regulators set ratios for the entire industry or
for large groups of similar banks? Should standards be tailored t o the individual bank? Alternatively, could private financial markets rather
than regulators be relied upon to determine
capital adequacy?
At the base of the debate between regulators
and the banking industry are the different
objectives of the t w o groups. The regulators
are concerned w i t h maintaining the soundness
of the entire banking system. Traditionally,
regulators have argued that bank failure is
"contagious." In other words, failure of one
institution could create difficulties for other
financial institutions if it causes depositors and
other creditors to lose confidence. Capital
requirements act as a form of insurance against
failure.
An individual banking organization's view of
capital, on the other hand, is strongly influenced
by the increasingly competitive financial markets in which banks operate. Commercial banks
and their parent companies compete for funds
with firms operating under less severe regulatory
constraints. Banks often argue that regulators
force them to maintain excessive levels of
capital. 2 As a result, commercial banks are
placed at a competitive disadvantage. 3
'Anthony M. Santomero and Ronald D. Watson, "Determining an Optimal
Standard for the Banking Industry," Journal of Finance, September 1977,
pp. 1267-1281.
3
lt is possible that this disadvantage is offset by other forms of regulation,
especially limitations on competition and the provision of government
supported deposits insurance.
What is Capital?
Financial analysts and accountants look o n
capital as owner's equity in the banking firm.
Conceptually, it is the book value (the value
shown in the firm's o w n accounting statements)
of the firm's assets, less the book value of its
liabilities. If market values were equal t o book
values, capital w o u l d represent the net value
after the firm's assets were liquidated and the
depositors and other creditors were repaid.
Some bank regulators have permitted banking
firms to consider subordinated notes and debentures as an additional source of capital for t h e
regulators' purposes. Recently the Federal Reserve
Board and the Office of the Comptroller of the
Currency have agreed on a definition of capital
36
These disagreements reach the heart of several issues vital to bankers. Competition between
large banks and small banks, between domestic
banks and foreign banks, between banks and
nonbank providers of financial services is significantly influenced by the capital levels that
these institutions hold. The ability of banking
organizations to expand their services and
geographic service areas is constrained by
regulators' views of what is adequate capital.
Why the Capital Issue is Important
Three recent developments in the financial
industry make it crucial for banks and their
regulators t o resolve the capital issue. First,
capital levels are falling (Table 1), a fact which is
unsettling to banks and regulators. Historically,
remember, regulators have tended to accept
the view that less capital in the industry means
greater risk exposure for the financial system. 4
Second, capital levels vary by banking firm
size (Table 2). Larger banks generally have lower
capital ratios than smaller banks. Higher capital
at smaller banking organizations may result from
several forces; regulation is important among
these. There is no d o u b t that bank regulators
require higher capital-asset ratios in smaller banks.
While this may have been a reasonable approach in
"There is no doubt that, other things being equal, reduced capital levels lead
to a higher incidence of failure within the banking industry. As capital levels
fall, the probability increases that banks will suffer losses that force them
into bankruptcy. For an excellent discussion of this, see Santomero and
Watson, op. cit. The entire industry could still be safer, however, if the
probability of capital being needed has fallen at a faster rate because of
safer portfolios or less volatility in the economy. It isquite unlikely that this is
the case, especially in recent years.
that divides capital into t w o parts on the basis of
its permanence. Those sources of funds that have
no due dates—common and perpetual preferred
stocks and convertible debt that must be converted
into stock, surplus, undivided profits, and certain
reserves—are called primary capital. Limited-life
preferred stock and subordinate notes and debentures—sources of funds w i t h due dates—are
called secondary capital if they have an original
maturity of more than seven years and are phased
out of the banking firm's capital as their maturity
approaches. Secondary capital is counted only
up t o 50 percent of primary capital.
Except where otherwise noted, the discussion of capital in the rest
of this article will be in terms of primary capital.
MARCH 1982, E C O N O M I C REVIEW
Table 2
Equity Capital to Assets
Banks in Sixth District States
1980
Table 1
Changes in Capital Ratios* 1970-1980
Banks in Sixth District States
Year
Equity to
Total Assets
(percent)
Equity to
Risk Assets
(percent)
7.7
7.7
7.4
10.8
9.9
9.6
1970
1975
1980
Asset Size
(million $)
Equity to
Total Assets
(percent)
Equity to
Risk Assets
(percent)
11.2
9.1
8.2
6.7
14.6
11.3
10.3
9.0
Source: Federal Deposit Insurance Corporation: Bank Operating Statistics 1980, 1975, 1970
Under 10.0
10.0 to 24.9
25.0 to 99.9
100 or more
'Average weighted by assets
Source: Federal Deposit Insurance Corporation: Bank Operating Statistics 1980
light of risks inherent in small banks, the small
banks argue that the policy places them at a
significant competitive disadvantage.
Third, capital requirements imposed by regulators may be restricting banking firms' growth.
As their assets grow over time, banking firms
must make additions t o their capital accounts
commensurate with their growth or else face
declining capital to asset and capital t o risk
asset ratios. To rectify this, they may sell common
or preferred stock. They may also add to equity
by retaining earnings in the firm. In practice, a
substantial portion of banking firms' earnings are
not distributed, but are retained in the bank.
The market for banking firms' stock has been
depressed for many years. Most small firms
find it difficult t o sell additional shares. Thus,
they prefer to retain earnings.
The capital
formation rate (shown in Table 3) provides a
measure of the ability of banks t o generate
capital funds internally through retaining earnings. This rate is defined as net income after
taxes, less dividends, relative t o equity capital.
The rate is not excessive for commercial banks
in the Sixth District, relative t o growth in assets.
Banks in the most rapidly growing states in the
District are least able to generate sufficient
funds internally t o maintain their asset growth.
Smaller banking firms must either have greater
earnings, or retain a greater percentage of their
earnings, in order to maintain higher capital
ratios. In competitive markets in which small
banks compete directly with larger banks, it is
unlikely that earnings rates will be higher.
FEDERAL RESERVE BANK OF ATLANTA
BANK CAPITAL RATIOS
IN SIXTH DISTRICT
Aggregate capital ratios for recent years for commercial banks in the Sixth Federal Reseive District
are presented in Tables 2 and 3. In these Tables,
equity capital is defined to exclude subordinated
notes and debentures. In Table 2, the ratio of capital to
assets is indicative of the total leverage of the banking
firm. It represents the percentage of the bank's assets
purchased with funds invested by stockholders. The
remaining assets were purchased with funds from
depositors or other creditors. The greater is the stockholder share of investment, the less likely is the bank
to be unable to meet the demands of its depositors
and other creditors.
On average, the capital asset ratio is 7.4 percent.
Therefore, 92.6 percent of assets are purchased by
debt. While this is extraordinarily high relative to most
business or manufacturing firms, many other financial
firms are leveraged to an even greater degree.
The ratio of equity capital to risk assets in Table 3
more narrowly considers the role of capital as a buffer
to protect depositors and other creditors against
losses on loans and other risky assets held by the
bank. Losses that exceed capital in magnitude will
force the bank into bankruptcy. In this event, creditors,
uninsured depositors, and the FDIC representing
insured depositors will not receive full value on their
funds held at the bank.
Risk assets are defined as total assets less risk free
assets. Risk free assets are cash, balances due from
other banks, and holdings of government securities*
The capital to risk asset ratio focuses on the value of
capital as a buffer against losses on the value of
assets.
*Of course losses may also occur resulting from a mismatch of asset and
liability maturity structures. In that event, the government security
portfolio does not play a role in the risk exposure of the bank as changes
in interest rates and liquidity needs may force sales at market values
below book value.
37
Table 3
Median Capital Formation Rates
Sixth District States and U.S.
1980
Area
Capital Formation
Rate
(percent)
Increase in
Total Assets
(percent)
8.8
8.1
9.6
11.2
10.4
8.0
8.2
6.8
13.8
5.7
14.4
8.3
7.9
9.7
Alabama
Florida
Georgia
Louisiana
Mississippi
Tennessee
United States
Source: Federal Deposit Insurance Corporation: Bank Operating Statistics 1980
Therefore, they must retain earnings at a greater
rate through reduced dividends. A lower dividend rate will tend to reduce the value of the
bank to the shareholders.
of funds is expensive compared t o non-capital
sources. The costs of generating new capital
are substantially higher than the costs of raising
funds through alternative sources. Dividends
on equity are not deductible from taxes whereas
interest on debt is. For this reason, capital may
cost twice as much (in terms of investor return)
as other debt.
In order t o determine its capital level, the
management of a banking firm must balance
the costs of capital against the benefits of
capital as a buffer against losses and a source of
funds. So long as uninsured depositors are
aware of the bank's leverage and associated
risk and require compensation for depositing
funds in riskier institutions, banking firms will
find that overall costs of funds rise after their
capital-asset ratio falls below a certain level.
The capital level at which costs of funds begin
to rise is the one which the bank should choose.
This ideal level, however, may not coincide w i t h
regulatory requirements. As a result, banks are
unhappy with the present system.
Capital From the Regulatory Perspective
Capital From the Bank's Perspective
From the banking firm's point of view, capital
has t w o functions other than satisfying regulatory requirements. First, since losses may be
charged against equity, but not against debt,
without causing the bank to fail, equity capital
serves as a buffer against losses. This role of
capital is most critical t o the uninsured depositors and other creditors. Insured depositors are
protected by the deposit insurance fund (FDIC),
and are likely t o be indifferent t o the intricacies
of the capital issue. Uninsured depositors and
other creditors do not have this protection and
must look to capital accounts as buffers t o
protect them against losses.5
The second role of capital within the banking
firm is as a source of funds, similar t o deposits,
notes, and other forms of debt. It has certain
attributes that are attractive t o banks. It has
infinite maturity and requires no fixed payment
of interest as does debt. In terms of assetliability maturity structure of the firm, it plays
an important role.6 However, capital as a source
The regulatory agencies are charged with the
responsibility of maintaining a safe and sound
commercial banking system for at least t w o
major reasons. Demand deposits are the largest
element of the nation's money supply. Failure
of a bank or its parent holding company could
weaken public confidence in the safety of their
deposits in commercial banks. If weakened confidence led bank customers t o withdraw their
deposits from other banks, a widening circle of
failures could lead to both significant decline in
the nation's money supply and t o disruption of
financial markets. 7
Ideally, regulators should set minimum capital
levels for each firm individually based on the
firm's risk of failure and the impact of its failure
on financial markets. The bank examination
process does allow for fitting capital requirements t o the characteristics of the individual
bank. This fitting, however, has been highly
judgmental and generally has ignored the social
costs of failure of the individual bank. Because
of the high costs of determining such risks for
each institution, federal regulatory agencies
5
Of course, their primary protection is the ongoing nature of the bank, the
safety of its assets and the flow of income from all the bank operations that
normally would be sufficient to offset any specific losses.
6
For an analysis of optimal capital from this perspective see John Pringle "The
Capital Decision in Commercial Banks," Journal of Finance, June 1974.
38
'Robert C. Clark, "The Soundness of Financial Intermediaries," Yale Law
Journal, Vol. 86, No. 1, November 1976.
M A R C H 1982, E C O N O M I C REVIEW
have placed banking firms into categories which
are t o some degree grouped by their relative
soundness. Target capital ratios were initially
set for banks within each group. The targets
were then adjusted to reflect specific considerations for the firm.
Until late in 1981, the Federal Reserve and
the Comptroller of the Currency applied this
approach in t w o separate ways. The first, generally adhered to by the Federal Reserve and
the FDIC, was t o set ratios of capital for similar
sizes of banking firms. W i t h i n each size class,
firms were expected to maintain target levels
of capital relative to assets, deposits, or risk
assets. For many years, the Federal Reserve
used an elaborate version of this approach in
which capital targets were based upon the
asset and liability portfolio composition. The
relative weights for required capital for each
category of assets and liabilities were somewhat
arbitrary. They were roughly based upon the
loss experience of the 1930s. 8
An alternative approach, developed by the
Comptroller of the Currency, was to group
banks into homogeneous classes based upon
size, branches and the degree of competition
in their markets. Each bank was then expected
to maintain capital ratios consistent with the
average bank within its peer group. Over time,
as banks below the average are required to
increase capital, their ratios would tend t o rise.
This approach was based upon the current
actual levels of capital within this group and, if
the entire group has too much capital, requiring
each bank to achieve the average is effectively
requiring each to have too much capital. Alternatively, the entire group may have insufficient
capital, resulting in targets being set too low for
each bank within the group. In reality, the peer
group levels were lower for the larger banks
since they are based on current levels. Therefore,
larger national banks were required to hold
less capital.
New Regulatory Policies
Late in December of 1981, each of the three
federal bank regulatory agencies made specific
policy statements about its approach to capital
adequacy. The Federal Reserve and the Comp-
8
trailer of the Currency together announced
capital guidelines for the banking organizations
they supervise; the Federal Deposit Insurance
Corporation announced a separate policy.
The four principal features of the Federal
Reserve-Comptroller plan are (1) basing capital
requirements on bank size, (2) accepting subordinate notes and limited-life preferred stock as
capital, (3) treating the largest firms individually,
and (4) continuing to consider the specific situation of each institution within the broader frame
of the guidelines.
This policy divides banking firms into three
groups, on the basis of size. The first group is
made up of 17 large multinational firms—all
bank holding companies w i t h assets of more
than $15 billion. These large firms have no
specific capital requirements; however, the supervisory agencies stated that they expected the
long-term decline in capital in these organizations
t o be reversed under the new policy. Each
institution is, however, to be subject to separate
special analysis.
The second group—those banks with assets
of $15 billion to $1 billion—called regional
banks—are subject to somewhat less strict
capital requirements than the third group,
firms with assets of less than $1 b i l l i o n community banks. W i t h i n each group of banks
three "zones" of capital were established: "adequately capitalized," "possibly undercapitalized,"
and "presumed undercapitalized." Banking firms
falling in the first zone are presumed to have
acceptable capital. Banks in this category receive
no special supervision. Firms falling in a range of
capital asset ratios below that which is acceptable
are thought to have inadequate capital and are
subject t o extensive contact with the regulator
and are required to submit capital plans that are
acceptable to the regulator. Banks with very low
capital, in zone 3, are continuously supervised.
Unlike the policy of the Federal Reserve and
the Comptroller of the Currency, the FDIC
policy does not count subordinate debt or limitedlife preferred stock as capital.This policy does not
base capital requirements on firm size, but since
nearly all banks that the FDIC supervises have
assets below $1 billion, the policy differs little
from that of the other agencies in that respect.
The FDIC policy also establishes three categories
similar to zones in the policy of the other agencies.
Its requirements are summarized in Table 4.
For a description of the ABC formula, see Ronald D. Watson, "I nsuring Some
Progress in the Bank Capital Hassle," Federal Reserve Bank of Philadelphia,
Business Review, July, August 1974, pp. 3-18.
FEDERAL RESERVE BANK OF ATLANTA
39
Table 4
Standard Capital-Asset Ratios
in Newly Announced Policies of the
Federal Regulatory Agencies
Capital Asset Ratios (percent)
Federal Reserve-Comptroller*
FDIC*'
Regional Banks
Community Banks
All Banks
Zones
Acceptable
Possibly
undercapitalized
Undercapitalized
6.5 or more*** 7.0 or more****
5.5 to 6.5
6.0 to 7.0
6.0 or more
5.0 to 6.0
Less than 5.5 Less than 6.0 Less than 5.0
"May include certain debentures and limited-life preferred stock.
"May not include debentures or limited-life preferred stock.
'"Primary capital must be greater than 5.0 percent of assets.
""Primary capital must be greater than 6.0 percent of assets.
Rough estimates of the impact of the new
capital policy of the Federal Reserve and the
Comptroller indicate that the vast majority of
regional and community banks have acceptable
capital-asset ratios and that only a small proportion of these banks have capital-asset ratios
low enough to warrant continuous supervision.
There is a slightly lower proportion of regional
banks in the acceptable zone and slightly
greater proportion of regional banks in the
unacceptable zone.
In adopting the new capital policy, the Office
of the Comptroller appears to be withdrawing
from the peer group approach that it developed
in the mid-1970s. That approach failed t o
reverse the long-term decline in capital-asset
ratios. It may, indeed, have been a casualty of
that decline since it compared each bank with
an average bank and capital in the average
bank was generally declining.
Should Capital Requirements
be Based on Bank Size?
These capital-adequacy guidelines based on
size are not above attack, however. Distinction among groups of banks, especially on the
basis of size, has been widely challenged by
the industry. Size distinctions have generally
been based on the argument that larger firms
are safer. However, the probability of failure is
complexly related to the size of the banking
firm. W i t h similar asset portfolios, larger banks
would have less risk exposure because they
40
have potentially greater diversification. Similarly,
they are less exposed to the risks of losses
brought about by sudden liquidity needs.
Although it has access to the federal funds
market and to some interest-sensitive deposits,
a smaller firm facing a reduction in deposits
may be forced t o sell assets. If interest rates are
unfavorable, this could entail losses as the
securities would be sold at a discount. Larger
firms have access to Eurodollar markets, large
certificates of deposit markets and other forms
of borrowing and therefore can avoid some of
this risk. Additionally, regulators have argued
that larger firms have been able to attract a
more sophisticated management.
Despite these factors, many larger firms behave in ways that may increase their risk. They
hold different portfolios from smaller community banks. They typically hold large commercial
and international loans not found on smaller
banks' books. Even though access to financial
markets reduces the risk of disintermediation,
the cost of large firms' liabilities is more volatile
as they must pay the market rate for a large
percentage of their funds. Thus, the net relationship between size and soundness is unclear;
there are strong, conflicting forces operating in
each direction.
In addition to the financial cost of failure,
regulators face the question of how to determine the social cost of failure. There is little
doubt that the likely social costs of failure are
higher for larger institutions. They are more
visible. They have larger proportions of uninsured
deposits and other forms of debt, thereby increasing the probability of a change in expectations
leading t o panics. Their portfolios tend t o be
similar. If one large firm is in trouble, investors
and depositors are more likely to see their
institution has a similar asset-liability structure
and may be vulnerable. Their larger size relative
to the FDIC insurance fund may change general
expectations regarding the safety of deposits. 9
Policy Options
There are several additional options for dealing with the capital problem. They range from
maintaining the status quo to letting the financial
"The Federal Deposit Insurance Fund was 11.6 billion on December 31,
1981. Each of the largest 230 banks had total assets exceeding 1.0 billion.
A failure of any one of these size firms would have a strong impact on the
FDIC reserve fund. See Federal Deposit Insurance Corporation, Annual
Report, 1980.
MARCH 1982, E C O N O M I C REVIEW
"Rapidly changing competitive forces
may well make maintenance of
size-related capital standards difficult
during the next several years."
markets regulate capital. Intermediate proposals include developing a more sophisticated
analysis of the probability and social cost of
failure and adopting a new concept of equal
marginal capital requirements for all firms.
Let's look briefly at the major alternatives.
A. Maintain the Present System
The present system provides size related
capital guidelines administered with cognizance
of each individual bank's characteristics. Its
flexibility is an advantage, particularly if regulators make advances in their systems for analyzing risk in financial institutions.
However, rapidly changing competitive forces
may well make maintenance of size related
capital standards difficult during the next several
years. Banking's stable profits and protected
markets have attracted non-depository firms
from less stable industries into markets for financial services such as transactions accounts, time
deposits, and commercial and consumer lending
where banks had held a dominant position.
Other new competitors for banks in several
product markets have arisen through the Monetary
Control Act of 1980 which expanded powers
of thrift institutions and credit unions to such
an extent that they are able to join banks as
providers of a full line of consumer financial
services. In addition to these breaks in product
and industry barriers, breaks are occurring in
geographic barriers. Nonbank institutions and
large banks are expanding their service areas
FEDERAL RESERVE BANK OF ATLANTA
across state lines, entering local markets in
many parts of the nation in competition with
local banks. These factors are likely to increase
the number of competitors in markets in which
banks offer services and, as a result, to put
downward pressure on bank profits. Lower
profits will mean fewer internally generated funds
to increase bank capital and diminished ability of
banks to go to the capital markets. Consequently,
banks will have even greater difficulties in generating sufficient new capital t o meet regulatory
requirements through retained earnings.
If earnings decline, the end result of maintaining strict size-related guidelines will be a substantial shift of activity away from the commercial
banking system. The market-regulated sector,
free from size-related capital guidelines and
from their growth restrictions, will be able to
grow at the expense of the commercial banking
sector. At the same time, in the future smaller
firms will likely lose market share t o larger ones
as the smaller banks continue t o be placed at a
competitive disadvantage and as larger firms
expand the geographic scope of their operations.
The size problem of the present approach
may be particularly critical for all banking
organizations in the Southeast. Southeastern
organizations increasingly are facing direct
competition in commercial and international
lending and deposit markets from large money
center and foreign banks and from nonbank
institutions. Most of these institutions have
much greater leverage than even the largest
regional banks. Consequently, they can operate
with smaller margins between the cost of funds
and loan revenue, and still return equal or even
higher levels of profits t o their stockholders.
With this advantage, their share of the commercial
and international markets is likely t o increase
unless size-oriented capital regulation is restructured.
B. Improve Financial Analysis
to Determine Capital Levels
Many analysts have argued that a more sophisticated system of capital determination would
eliminate many of the present system's problems. Such a system would be an extension of
the present system toward the ideal of accurately calculated requirements for each bank
based on its o w n characteristics. This approach
41
would require advances in knowledge of the
factors that determine the soundness of financial
institutions.
If the system could be improved in this way,
it would better enable the regulators to identify
institutions w i t h high probabilities of failure
and t o require more capital in them. The
competitive advantage would more certainly
go t o those firms that are in fact safer.10
Current research in this area is quite promising.
Analysts are improving their ability t o identify
banks that are likely to face difficulty in the
new environment. They focus on levels and
trends in key financial variables in the bank. 11
New developments in pricing of contingent
claims may be very useful in this regard. 12
Better financial analysis of banks would not
avoid the current system's failure t o consider
the social cost of bank failure in setting capital
requirements. However, present knowledge of
these costs is inadequate for more than a very
broad generalization about costs of failure.
One can be fairly certain that the social costs of
failure—the shakeup of public confidence in
the financial system—rise with the size of the
failing bank. Beyond that, there is no quantitative
evidence on this problem and no formula for
reflecting the social cost in capital requirements.
Considerable study would have to precede
allowing for the social cost of failure in setting
capital requirements.
C. Require Equal Marginal Capital Levels
The equal marginal capital requirement would
set the same capital adequacy levels for every
bank in relation t o additional assets. It would
accept the fact that capital levels now differ.
However in the future, all firms would be
treated equally. Each would have to add an
equal amount of capital for every dollar's growth
in total assets or risk assets.13
This approach would have two primary implications. First, it provides competitive equality at
10
An alternative approach would be to vary the deposit insurance premium
with the degree of risk.
' See Joseph H. Sinkey, Jr, "Problem and Failed Banks, Bank Examinations, and
Early Warning Systems: A Summary," in E. I. Altman and A W. Sametz,
Financial Crisis, New York: John Wiley and Sons, 1977.
"Robert Merton, "An Analytic Derivation of the Cost of Deposit Insurance and
Loan Guarantees: An Application of Modern Option Pricing Theory,"
Journal of Banking and Finance, Vol. 1,1977, pp. 3-11.
' 3 For an excellent discussion of this approach, see Thomas B. Walker,
"Regulating Capital at the Margin," Mimeo, Southeast Banking Corporation,
Miami, 1981.
1
42
"Deregulation of capital
may be a viable option, at least
for a large banking organization."
the margin. No organization would have a competitive advantage in expanding its activities
based upon its ability to leverage to a greater
degree than other firms. Second, the marginal
capital approach will move all firms toward the
same average level of capital. For those whose
existing capital ratios are higher than the marginal
ratio, capital ratios will tend to fall. Conversely,
those w i t h lower ratios w h o must add more to
the capital than they have in the past will find
their average ratios increasing.
This approach, however, fails to consider social
costs of bank failure and it requires regulators to
set a single marginal capital ratio that is applicable to
the entire industry. By arbitrarily setting equal
capital requirements for all institutions, it penalizes
the safer institutions.
D. Deregulation
The final option is t o deregulate capital
totally. This view w o u l d be consistent with
many policies that have been adopted in other
industries in recent years. The regulators w o u l d
simply rely upon private financial markets to
monitor the riskiness of each bank or its parent
holding company relative to its capital levels.
Banking firms would be free to choose the level
of capital that gives them the minimum cost of
funds.
There are three distinct problems w i t h this
approach. First, banking firms acting without
M A R C H 1982, E C O N O M I C REVIEW
regulation could not be expected to consider
the social cost of bank failure in setting their
capital levels. I n the case of small firms, this is
unlikely to be a major disadvantage t o the
public; however, large banking firms with high
visibility may have large social costs of failure.
Omission of these costs in setting their capital
under a deregulated system could result in
deficient capital from the public's perspective.
As discussed earlier, however, no other present
or proposed system formally factors in these
costs.
Second, the stocks and uninsured liabilities
of most small banking firms are not traded
actively. Therefore there will be no continual
monitor and barometer of the financial condition
of the institution. Other regulations would be
necessary in this instance.
Third, for deregulation to be a viable approach, the financial markets must penalize
firms that take on too much leverage. It is not
fully established that the market does penalize
banking organizations that expand their leverage.
Earlier studies, in fact, concluded that leverage
had no impact on stock prices. 14 Later work
based on the decade of the 1970s suggests
market prices of stock and uninsured liabilities
do change in reaction to different levels of
leverage within large publicly traded bank
holding companies. Therefore, deregulation of
capital may be a viable option, at least for large
banking organizations. 15
Conclusion
In summary, capital adequacy has often been a
bone of contention between the banking industry and its regulators. It is quite likely that
increased competition in the 1980s will shrink
profit margins and will make it increasingly
difficult to maintain current capital ratios.
Regulators should give strong consideration
t o improving the present size-related system of
determining capital adequacy. The present
approach is likely to cause increasing difficulties
as financial markets become integrated with all
firms offering similar products. At a minimum,
greater effort should be expended in developing a more objective system of financial analysis.
In addition, the regulatory agencies should
give full consideration t o innovative proposals,
such as the equal marginal capital approach or
deregulation of large institutions.
—Arnold A. Heggestad
and B. Frank King
Note: William Estes and Caroline Harless made valuable comments on an earlier draft of this article.
15
"•See, for example, Richard Pettway, "Market Tests of Capital Adequacy
ofLarge Commercial Banks," Journal of Finance, June 1976, pp. 865-875.
FEDERAL RESERVE BANK OF ATLANTA
See for example, Beishely, H. Prescott, "The Risk Perceptions of Bank
Holding Company Debtholders," Journal of Bank Research (8), Summer
1977, pp. 85-93; Herzig-Marx, Chayim,"Comparing Market and Regulatory
Assessments of Bank Regulation," Proceedings of a Conference on Bank
Structure and Competition, Federal Reserve Bank of Chicago, 1977;
Pettway, Richard H. "Potential Insolvency, Market Efficiency and Bank
Regulation of Large Commercial Banks," Journal of Financial and
Quantitative Analysis (15), March 1980, pp. 219-236.
43
Cost of Living Data:
A Guide to Sources
With average costs for transferring an executive running around $30,000,
reliable information on cost-of-living differences becomes more important A
review of some major sources —as well as some often overlooked ones—
discusses strengths, weaknesses, and methodological differences.
"How much will it cost me? Can I afford to buy a
home in a new city? What kind of life style will be
possible?"
W h e n an employee is offered a promotion or a
new j o b requiring a transfer t o another city, these
are usually the first questions asked. Even when
offered large increases in salary, many potential
transferees hesitate t o relocate given the high
cost of acquiring a new home at current interest
rate levels. To encourage valued employees to
accept a transfer or to attract a prime candidate
for a new j o b in a more expensive city, many
companies routinely offer relocation assistance
which may include paying the interest-rate difference between the employee's old and new
mortgages, covering closing costs, and guaranteeing sale price. According t o the Washingtonbased Employee Relocation Council, whose mem44
bers include corporations w i t h such relocation
incentives, the average cost of transferring an
executive within the U.S. currently runs around
$30,000. 1 Faced with the present costs of relocation, any employee or company should have the
most reliable and trustworthy information available for making decisions about whether to
accept or reject a transfer.
Although several sources of data enable regional
cost-of-living comparisons, none will provide
instant answers for all the questions related t o a
specific relocation situation, and all have weaknesses that create problems for anyone attempting to use the data. This article will point out
some often overlooked sources of information
and will discuss differences in the types of data
and collection methodologies.
'Kenneth M. Pierce, "Housing the Company Way," Time October 12,1981,
page 85.
MARCH 1982, E C O N O M I C REVIEW
Map 1 : Cost of Living Relative to Atlanta
Based on BLS "higher" budget for a four-person family
(percent deviation from Atlanta Budget)
U.S. Urban Average: +10%
Source: U.S. Bureau of Labor Statistics, Federal Reserve Bank of Atlanta calculations.
BLS Urban Family Budgets
The Urban Family Budgets estimates, issued
once a year by the Bureau of Labor Statistics
(BLS), are widely used, high quality data for
comparing cost of living differences between
major urban areas.2 The budgets specify the
costs for maintaining three standards of living—
lower, intermediate, and higher—for a hypothetical family of four living in any of 25 different
metropolitan areas.3 According to the most recent
2
The Consumer Price Index often is used for geographical cost-of-living
comparisons but was never desig ned for that purpose. See James T. Fergus,
"Cost of Living Comparisons: Oasis or Mirage?" Federal Reserve Bank of
Atlanta Economic Review, July/-August 1977, pages 92-100.
3
The Bureau of Labor Statistics defines the hypothetical family as: "a 38-yearold husband employed full time, a non-Working wife, a boy of 13, and a girl of
8". See U. S. Bureau of Labor Statistics, "Autumn 1980 Urban Family Budgets
and Comparative Indexes for Selected Urban Areas", April 22, 1981.
FEDERAL RESERVE BANK OF ATLANTA
BLS survey in autumn, 1980, the average annual
cost of the " l o w e r " family budget was $14,044,
the "intermediate" budget was $23,134, and the
"higher" budget was $34,409 per year. The
budgets are updated each autumn by applying
price changes for each metropolitan area as
reported in the Consumer Price Index.
Although an executive considering a transfer
might find this information useful, the fact that
the budgets are produced for only 25 cities
obviously limits their value. Furthermore, the
intermediate and higher budgets assume that
homeowner costs include payments for a house
purchased six years ago.4 BLS clearly states that
'According to BLS, housing costs at the lower budget level provide only for
rental housing.
45
" t h e geographic indexes do not measure cost
differences associated with moving from one
area t o another or the living costs of newly
arrived residents in a given community." 5 Therefore, a transferee w h o plans to purchase a home
in a new community at current market rates will
need t o look elsewhere for comparative costs of
housing. 6
Other problems w i t h the budgets' methodology and applicability have been well-documented in this Review and elsewhere. 7 The BLS
recognizes these problems and currently is studying a completely new approach to constructing
the budgets. 8
Atlanta is the only southeastern city now included in BLS Budgets estimates, but the inclusion
of Atlanta does enable cost-of-living comparisons
for Atlanta with other cities outside the Southeast
Map 1 demonstrates that in Autumn 1980 Atlanta
had one of the lowest living cost levels of all the
U. S. urban areas for which BLS data are available.
ACCRA Cost of Living Indicators
The American Chamber of Commerce Researchers Association (ACCRA) produces quarterly
Inter-City Cost of Living Indicators for a w i d e
sample of cities of various sizes. City chambers of
commerce participate in the ACCRA survey on a
voluntary basis, gathering their local data through
telephone, letter, or in-person pricing surveys.
Regional coordinators review the data for consistency and to ascertain whether substantial deviations from quarter to quarter are justified before
they forward the data for final processing at the
national level.
The survey covers a large sampling of cities—
237 communities participated in the Third Quarter 1981 survey—and the information reported
is updated every quarter. The Intercity Cost of
Living Indicators report shows price levels in the
form of index numbers for an all-items index and
for the various major components of the index:
grocery items, housing, utilities, transportation,
health care, and miscellaneous goods and services.
Potentially, the survey's most useful component
for the transferee is the index of housing costs.
The housing component carries the greatest
weight of all components of the all-items index.
Unlike the BLS Urban Family Budgets, the ACCRA
survey measures the cost of home ownership at
current market rates.9 ACCRA also measures
apartment rental costs and converts the home
price and rental data w i t h appropriate weights
into an index number for each participating city
and into a U. S. average.
Anyone using these figures, however, should
heed ACCRA'S warning that " t o avoid misrepresentation...users should be aware of the limitations
of the Index and should understand the kinds of
information it does provide." 1 0 Differences of
less than three index numbers between cities
should not be considered significant.
The manner in which the ACCRA survey is
conducted introduces a number of potential
problems or sources of statistical error. The
burden of reporting complete, accurate, and
timely data rests w i t h local volunteers. The user
of ACCRA data also should consider the possibility
of bias in the chamber of commerce data since
these organizations have a vested interest in
attracting jobs and businesses t o their communities.
Each ACCRA component consists of a very
small market basket—only 44 goods and services
are priced each quarter for the entire Index.
Incorrect pricing of only a few items could bias
the entire Index.
The ACCRA survey does not include taxes,
except for those incorporated in the product's
price, as in the case of tobacco, gasoline, and
liquor. Taxes can vary significantly from one city
and state t o another, and ACCRA has found no
reliable way of calculating tax estimates.
Despite its flaws, the ACCRA Intercity Cost of
Living Indicators provides a source of current
living cost data for a large sample of cities. Most
people w h o need cost-of-living comparisons for
making relocation decisions should find the
ACCRA data readily available at local chamber of
commerce offices.
9
5
U. S. Bureau of Labor Statistics, "Autumn 1980 Urban Family Budgets-,"
page 7.
6
For example, the Federal Home Loan Bank Board each month releases
information on the average terms of conventional home mortgages for
selected metropolitan areas
7
See Fergus, "Cost of Living Comparisons."
"Harold W. Watts, "Special Panel Suggests Changes in BLS Family Budget
Program," Monthly Labor Review, December 1980, pp. 3-10.
46
The specifications describe a house with 1800 square feet: 3 bedrooms, 2
baths, living-dining area, kitchen with built-in cabinetry, finished family room,
one average fireplace, utility room, and attached 2-car garage. The house is
sited on a 10,000 square foot lot with 70-foot frontage. Other detailed
specifications outline the type of construction required, with some leeway
allowed for regional practices and local codes.
10
American Chamber of Commerce Researchers Association, "Inter-City
Cost of Living Index Instruction Manual for Participating Cities (Revised
January 1979)," p. 35.
MARCH 1982, E C O N O M I C REVIEW
Map 2: Intercity Cost of Living Comparisons between Atlanta
and Other Southeastern Cities, Third Quarter 1981
( p e r c e n t d e v i a t i o n f r o m Atlanta b u d g e t )
Dyersburg
TLANTA
ou
Mie
Shreveport
Lake Charles
Jacksonville
Pensac
-14
ist Palm Beach
Boca Raton
+11
Fort Lauderdale
+8
Source: American Chamber of Commerce Researchers Association, Federal Reserve Bank of Atlanta calculations.
Given the limited coverage of BLS data, it is
natural to want t o use ACCRA data for living cost
comparisons within the Southeast. As a further
check on the reliability of the ACCRA data in
relation to the higher-quality BLS data, we compared the two measures in the 13 cities measured
by both indexes. The comparison shows that,
although actual price levels of specific cities vary
widely between the t w o sets of data, intercity
differences in living costs correlate highly. 11 Map
2 displays ACCRA data for the southeastern
" A simple correlation of data available for the same 13 cities from BLS and
ACCRA indexes shows a 0.73 correlation coeffieient or a highly significant
degree of correlation with less than 1 % chance of occurring from a random
sample.
FEDERAL RESERVE BANK OF ATLANTA
states and gives,a rough indication of living cost
levels for southeastern cities compared to Atlanta.
Atlanta's living cost is relatively high within the
region according to this measure, although cities
in south Florida show an even higher cost of
living.
Relocation Consulting Companies
To those w h o seek further information beyond
that available from BLS or ACCRA and for those
w h o are willing t o pay, another option is t o utilize
the services of a company specializing in employee
relocation or compensation consulting. These
firms provide a variety of services to their clients,
47
Table 1
Annual Expenditures for a Family of Four
Earning $40,000
City
Housing
Taxation
Trans.
Total
$7,437
$5,079
$31,356
$18,840
Atlanta, GA
16,472
8,303
4,678
29,453
Birmingham, AL
5,682
24,987
6,242
36,911
Boston, MA
33,314
21,026
7,215
5,073
Cincinnati, OH
5,300
30,812
18,514
6,998
New Orleans, LA
San Francisco, CA
35,264
3,414
6,453
44,131
(E. Bay)
36,521
24,598
5,768
6,155
Washington, DC (VA.)
5,012
19,498
7,046
31,556
Standard City, USA
LEGEND:
Costs are based on a family of 4 persons with an annual income of $40,000. All living communities used in gathering data,
represent communities where families at this income tend to reside.
The home has 7 rooms, 4 bedrooms, 2.5 baths, and 2,300 square feet of living area. The mortgage costs are premised on a 20%
downpayment, 28 year amortization at interest rates in effect on August 17,1981.
The transportation costs are based on a 1980 Chevrolet Impala driven 14,000 miles annually, and a 1977 Chevrolet Nova driven
6,000 miles per year.
Source: Runzheimer and Company, Inc., Rochester, Wisconsin
such as conducting special living cost surveys,
preparing detailed reports, or providing subscriptions to indepth cost-of-living studies. Two of
these options are described here, and several
others undoubtedly exist 1 2
Runzheimer and Company specializes in employee relocations consulting, providing companies customized reports on the expected costof-living variations between specified cities for a
given level of income. If a company plans to
transfer an employee, Runzheimer can calculate
how far that employee's salary will go toward
maintaining a given standard of living in various
cities by estimating expenditures for three major
living cost components—housing, transportation,
and taxation. Runzheimer asserts that the cost
differential for food, clothing, health care, and
other expenses varies only a little between
different cities and levels of income.
Table 1 shows the total of housing, transportation,
and taxation costs as of August 17, 1981, for
"Although we contacted several such companies, only two responded with
significant information about their methodologies and data
48
seven cities and a "Standard City, USA" for a
family of four with an annual income of $40,000.
The difference between a $40,000 income and
the total expense of the three components is the
amount the family will have left to spend on food,
clothing, health care, and other discretionary
items in those cities. Individual experience will
vary from these average figures, according t o
differences in family size, home-buying preferences, and other variables.
Through a network of primary information
sources, Runzheimer constantly updates a database on housing, transportation, and taxation
costs so that the firm can provide a report on
current living costs for almost any community in
the United States. In addition, Runzheimer does
primary research as needed in order t o calculate
estimated state, local, and federal income taxes
for any income level and any location in the U.S.
Runzheimer's strength clearly is its ability t o
generate up-to-date living cost data for almost
any specific relocation situation, enabling companies t o make better decisions and to reimburse
transferred employees fairly. Although the data
MARCH 1982, E C O N O M I C REVIEW
'
Table 2
Cost of Living Compared to Atlanta
BEFORE-TAX INCOME IN ATLANTA:
$35,000
Before tax
Income
Boston
Chicago
Cleveland
Denver
Detroit
Houston
Los Angeles
Miami
Minneapolis
New Orleans
New York
Philadelphia
Phoenix
Pittsburgh
St. Louis
San Francisco
Seattle
Stamford
Washington, D.C.
$70,000
Percent
Difference
$41,960
39,050
37,420
33,960
37,710
36,090
41,100
35,320
38,890
35,040
44,310
37,290
33,190
36,460
33,900
41,700
34,500
40,100
40,190
+ 20
+ 12
+ 7
+ 3
+ 8
+ 3
+ 17
+ 1
+ 14
0
+ 27
+ 7
- 5
+ 4
- 3
+ 19
- 1
+ 15
+ 15
Before tax
Income
$81,300
78,870
71,060
66,580
71,830
69,280
83,220
68,240
75,020
66,650
87,870
71,390
63,030
69,690
62,710
84,480
64,590
77,750
79,790
Percent
Difference
+ 16
+ 7
+ 2
- 5
+ 3
- 1
+ 19
- 3
+ 7
- 5
+ 26
+ 2
-10
0
-10
+ 21
- 8
+ 11
+ 14
Source: Associates for International Research, Inc., Cambridge, Massachusetts, October 1980.
provided do not include expenses for food,
clothing, health care, and discretionary spending,
Runzheimer's cost-of-living information is probably sufficient for a company or potential transferee t o make an informed decision.
Aspecialist in domesticand international compensation consulting, Associates for International
Research Inc. (AIRINC), provides a variety of data
and counseling services t o its corporate clients.
One such service geared toward assisting with
executive transfer decisions is the annual subscription publication entitled Executive Living Costs.
Based on data collected about executive living
patterns, the survey is designed t o provide costof-living comparisons and a measure of price
increases from year t o year for approximately 20
metropolitan areas.
Each year AIRINC staff conduct price surveys
for over 300 goods and services weighted to
represent typical purchases of executive families
at t w o income levels, $35,000 and $70,000.
These surveys are conducted at shopping areas
most frequently used in suburbs where these
families typically live. The data are tabulated and
FEDERAL RESERVE BANK OF ATLANTA
converted into indexes that reflect the cost-ofliving differences between each city.
AIRINC gives special emphasis to the effect of
taxes on living costs, and the Executive Living
Costs report provides detailed comparative information on how taxes will affect the income of a
transferee in various locations. Table 2 displays
the adjustments needed to maintain a given level
of income for a transferee from Atlanta to other
selected U. S. cities in 1980. Al Rl NC reports that
Atlanta's living costs in 1980 were 4-7 percent
below the U. S. average, with property values and
taxes, utilities, and transportation accounting for
part of Atlanta's lower expenditures. As with the
Runzheimer data, the user should be aware that
individual family experiences will vary from these
average figures.
Conclusion
Current, reliable sources of regional cost-ofliving comparisons are essential to any company
or executive in trying to reach a "go" or " n o go"
decision t o relocate. This article has pointed out
49
several reliable sources of living cost data and
assessed some of their advantages and pitfalls.
While all the sources mentioned here strive t o
report differences in living cost levels between
cities, each has a different approach and methodology. The potential user should consider the
advantages and limitations of each source before
arriving at a decision to relocate.
—Leigh Watson Healy
and William N. Cox III
BIBLIOGRAPHY
American Chamber of Commerce Researchers Association, "Inter-City Costof-Living Instruction Manual for Participating Cities," Revised January 1979,
Photocopied.
Perry, Joseph M. and ZaharaTandet, "Developing a Consumer Price Index for
a Local Area The Case of Jacksonville," Texas Business Review, March/April,
pp. 106-109.
American Chamber of Commerce Researchers Association, Inter-City Costof-Living Indicators, Third Quarter 1 9 8 1 .
Pierce, Kenneth M., "Housing the Company Way," Time, October 12,1981,
pp. 85-86.
Associates for International Research Inc., Executive Living Costs, October
1980.
"Pinpointing the Cost of Living for Executives," Dun's Review, April 1978, pp.
62-63.
Buckley, John E, "Do Area Wages Reflect Area Living Costs," Monthly Labor
Review, November 1979, pp. 24-29.
"Pinpointing the Cost of Living: The Runzheimer Plan of Living Cost
Standards," Rochester, Wisconsin: Runzheimer and Company, Inc.
Gottschalk, Earl C. Jr., "Home Buyer's $180,000 Goes Far or It Doesn't,
Depending on the Area" Wall Streét Journal, March 6, 1981, p. 1.
U. S. Dept. of Labor, Bureau of Labor Statistics, Autumn 1980 Urban Family
Budgets and Comparative Indexes for Selected Urban Areas," News Release,
April 22,1981, Washington, D. C.
Fergus, James T., "Cost of Living Comparisons: Oasis or Mirage?" Federal
Reserve Bankof Atlanta Economic Review, July/August 1977, pp. 92-100.
"Living Costs Take Biggest Bite in Northeast," Washington Post, January 8,
1979, p. 9.
Milbrandt, Gaylord, F. "Relocation Strategies: Part I," Personnel Journal,
July 1981.
U. S. Department of Labor, Bureau of Labor Statistics, "A Guide to Living
Costs: Three Budgets for an Urban Family of Four Persons, Autumn 1978,"
Spring 1979, Atlanta, Georgia.
Watts, Harold W., "Special Panel Suggests Changes in the BLS Family
Budget Program," Monthly Labor Review, December 1980, pp. 3-10.
Milbrandt, Gayloard F. "Relocation Strategies: Part II," Personnel Journal,
August 1981.
50
M A R C H 1982, E C O N O M I C REVIEW
-LE.
FINANCE
HHMmHMI
FEB
1982
Commercial Bank Deposits
Demand
NOW
Savings
Time
Credit Union Deposits
Share D r a f t s
Savings & T i m e
JAN
1982
1,099,407 1,120,920
289,176
328,111
53,775
54,615
148,279
148,931
634,166
625,036
41,672
40,734
2,683
2,645
36,405
35,580
FEB
1981
ANN.
%
CHG.
999,015
297,696
29,783
159,486
538,090
35,084
1,751
31,577
+10
- 3
+81
- 7
+18
+19
+53
+15
Savings <5c Loans
T o t a l Deposits
NOW
Savings
Time
Mortgages Outstanding
Mortgage C o m m i t m e n t s
C o m m e r c i a l Bank Deposits
Demand
NOW
Savings
Time
Credit Union Deposits
Share D r a f t s
Savings & T i m e
118,513
34,167
7,030
14,714
65,431
4,088
278
3,487
120,231
37,601
7,012
14,765
64,599
4,077
281
3,486
106,516
34,825
3,547
15,853
55,140
3,212
204
2.803
+11
- 2
+98
- 7
+19
+27
+36
+24
& Loans
T o t a l Deposits
NOW
Savings
Time
commercial Bank Deposits
Demand NOW
Savings
» Time
Ì Credit Union Deposits
Share D r a f t s
•-. Savings & T i m e
13,409
3,504
612
1,530
8,190
717
55
617
13,621
3,795
612
1,527
8,168
698
56
604
12,028
3,536
328
1,664
6,855
514
44
476
+11
- 1
+87
- 8
+19
+39
+25
+30
Savings & Loans
T o t a l Deposits
NOW
Savings
Time
C o m m e r c i a l Bank Deposits
Demand
NOW
Savings
Time
C r e d i t Union Deposits
' Share D r a f t s
Savings & T i m e
39,240
12,180
3,106
6,374
18,174
1,845
156
1,431
39,682
13,341
3,043
6,409
17,885
1,861
156
1,453
35,828
12,988
1,541
7,019
15,023
1,487
116
1,170
+10
- 6
+102
- 9
+21
+24
+34
+22
Savings & Loans
T o t a l Deposits
NOW
Savings
Time
C o m m e r c i a l Bank Deposits
Demand
NOW
Savings
Time
Credit Union Deposits
Share D r a f t s
Savings & T i m e
16,151
5,877
997
1,573
8,634
755
23
703
16,661
6,445
989
1,575
8,604
750
24
700
14,027
5,824
506
1,601
7,002
542
12
517
+15
+ 1
+97
- 2
+23
+39
+92
+36
Savings & Loans
T o t a l Deposits
NOW
Savings
Time
C o m m e r c i a l Bank Deposits
Demand
P/ NOW
21,511
6,227
941
2,380
12,493
114
8
106
21,610
6,812
937
2,381
12,244
114
8
106
18,917
5,928
465
2,456
10,521
80
4
74
+14
+ 5
+102
- 3
+19
+43
+100
+43
Savings ic Loans
T o t a l Deposits
NOW
Savings
Time
C o m m e r c i a l Bank Deposits
Demand
NOW
Savings
Time
' C r e d i t Union Deposits
Share D r a f t s
Savings ,5c T i m e
9,799
2,336
521
731
6,449
N.A.
N.A.
N.A.
9,849
2,550
515
733
6,332
N.A.
N.A.
N.A.
8,767
2,427
274
798
5,536
N.A.
N.A.
N.A.
+12
- 4
+90
- 8
+16
Savings & Loans
T o t a l Deposits
NOW
Savings
Time
commercial Bank Deposits
Demand
NOW
1
Savings
Time
> e d i t Union Deposits
Share D r a f t s
Savings & T i m e
18,402
4,044
852
2,125
11,491
657
36
630
18,808
4,659
864
2,139
11,366
654
37
623
16,949
4,122
433
2,315
10,203
589
28
566
+ 9
- 2
+97
- 8
+13
+12
+29
+11
r
Time
Credit Union Deposits
Share D r a f t s
Savings & T i m e
Notes:
Mortgages Outstanding
Mortgage C o m m i t m e n t s
Mortgages Outstanding
Mortgage C o m m i t m e n t s
Mortgages Outstanding
Mortgage C o m m i t m e n t s
Mortgages Outstanding
Mortgage C o m m i t m e n t s
Mortgages Outstanding
Morteracre C o m m i t m e n t s
Mortgages Outstanding
Mortgage C o m m i t m e n t s
Savings & Loans
T o t a l Deposits
NOW
Savings
Time
Mortgages Outstanding
Mortgage Commitments
ANN.
%
CHG.
FEB
JAN
FEB
1981
521,213
8,377
92,714
420,617
DEC
508,685
14,584
518,445
8,562
93,396
417,429
NOV
510,009
15,661
507,725
3,304
101,286
402,356
DEC
494,179
16,021
+ 3
+154
- 8
+ 5
76,566
1,372
11,766
63,471
DEC
74,487
3,246
76,020
1,398
11,824
62,928
NOV
74,606
3,505
73,670
477
12,913
60,033
DEC
71,098
3,652
+ 4
+188
- 9
+ 6
4,404
71
579
3,782
DEC
4,004
4,381
71
581
3,756
NOV
4,002
4,334
23
665
3,655
DEC
3,948
+ 2
+209
-13
+ 3
46,371
962
7,893
37,444
DEC
45,579
2,846
46,036
991
7,909
37,110
NOV
45,595
3,090
44,744
356
8,716
35,402
DEC
42,792
2.983
+ 4
+170
- 9
+ 6
9,720
143
1,183
8,430
DEC
9,368
112
9,646
140
1,187
8,360
NOV
9,441
107
9,385
39
1,346
8,004
DEC
9,315
183
+ 4
+267
-12
+ 5
7,519
83
1,216
6,238
DEC
7,114
17fi
7,469
82
1,240
6,181
NOV
7,138
is?
6,924
23
1,219
5,689
DEC
6,777
91 fi
+ 9
+261
- 0
+10
2,378
37
222
2,131
DEC
2,197
18
2,387
38
232
2,130
NOV
2,206
19
2,343
11
243
2,088
DEC
2,181
60
+ 1
+236
- 9
+ 2
6,173
75
673
5,445
DEC
6,225
44
6,101
76
675
5,389
NOV
6,224
55
5,940
25
724
5,195
DEC
6,085
70
+ 4
+200
- 7
+ 5
+ 3
- 9
+ 5
-11
+ 1
-64
+ 7
+ 5
+ 1
-39
+ 5
-1 Q
+ 1
-70
+ 2
-37
A l l deposit data are e x t r a c t e d f r o m the F e d e r a l Reserve R e p o r t o f T r a n s a c t i o n A c c o u n t s , o t h e r Deposits and V a u l t Cash (FR2900),
and are r e p o r t e d f o r t h e average of t h e week ending t h e 1st Wednesday of t h e m o n t h . This data, r e p o r t e d by i n s t i t u t i o n s w i t h
over $15 m i l l i o n i n deposits as o f December 31, 1979, represents 95% o f deposits in t h e six s t a t e a r e a . Savings and loan mortgage
data are f r o m t h e F e d e r a l H o m e L o a n Bank Board Selected Balance Sheet D a t a . The Southeast data represent t h e t o t a l of t h e six states.
Subcategories were chosen on a s e l e c t i v e basis and do not add t o t o t a l .
N.A. = fewer than four institutions reporting.
FEDERAL RESERVE BANK O F ATLANTA
51
EMPLOYMENT
DEC
1981
C i v i l i a n Labor F o r c e - thous.
T o t a l Employed - thous.
T o t a l Unemployed - thous.
U n e m p l o y m e n t R a t e - % SA
Insured U n e m p l o y m e n t - thous.
Insured Unempl. Rate M f g . A v g . Wkly. Hours
M f g . Avg. W k l y . E a r n ^ - $
NOV
1981
DEC
1980
ANN.
ANN.
%
CHG.
DEC
1981
N o n f a r m E m p l o y m e n t - thous.
Manufacturing
Construction
Trade
Government
Services
F i n . , Ins., & R e a l Est.
Trans. C o m . & Pub. U t i l .
92,015
1.9,854
4,156
21,403
16,156
18,771
5,345
5,167
NOV
1981
32,272
20,115
4,368
21,131
16,164
18,794
5,345
5,182
DEC
1980
CHG.
91,750
20,238
4,343
21,138
16,435
18,149
5,237
5,150
C i v i l i a n Labor Force - thous.
T o t a l Employed - thous.
T o t a l Unemployed - thous.
Unemployment Rate - % SA
Insured Unemployment - thous.
Insured U n e m p l . R a t e M f g . A v g . Wkly. Hours
M f g . A v g . W k l y . Earn. - $
ALSbam^
C i v i l i a n Labor Force - thous.
T o t a l Employed - thous.
T o t a l Unemployed - thous.
Unemployment R a t e - 96 SA
Insured Unemployment - thous
Insured U n e m p l . R a t e M f g . Avg. Wkly. Hours
M f g . A v g . W k l y . Earn. - $
5MÜZ3*
C i v i l i a n Labor F o r c e - thous.
T o t a l Employed - thous.
T o t a l Unemployed - thous.
Unemployment R a t e - % SA
Insured Unemployment - thous
Insured U n e m p l . Rate - %
M f g . A v g . Wkly. Hours
M f g . Avg. W k l y . Earn. - $
H H H ^ H H I H H i H
C i v i l i a n Labor F o r c e - thous.
T o t a l Employed - thous.
T o t a l Unemployed - thous.
U n e m p l o y m e n t Rate - % SA
Insured Unemployment - thous
Insured U n e m p l . R a t e - %
M f g . A v g . Wkly. Hours
M f g . A v g . W k l y . Earn. - $
C i v i l i a n Labor F o r c e - thous.
T o t a l Employed - thous.
T o t a l Unemployed - thous.
Unemployment R a t e - % SA
Insured Unemployment - thous.
Insured Unempl. Rate M f g . A v g . Wkly. Hours
C i v i l i a n Labor F o r c e - thous.
T o t a l Employed - thous.
T o t a l Unemployed - thous.
Unemployment R a t e — % SA
Insured U n e m p l o y m e n t - thous.
Insured U n e m p l . R a t e - 96
M f g . Avg. Wkly. Hours
M f g . A v g . W k l y . Earn. - $
C i v i l i a n L a b o r F o r c e - thous.
T o t a l Employed - thous.
T o t a l Unemployed - thous.
U n e m p l o y m e n t R a t e - % SA
Insured U n e m p l o y m e n t - thous
Insured Unempl. Rate - %
M f g . A v g . Wkly. Hours
M f g . A v g . W k l y . Earn. - $
Notes:
N o n f a r m E m p l o y m e n t - thous.
Manufacturing
Construction
Trade
Government
Services
Fin., Ins., & R e a l Est.
Trans. Com. & Pub. U t i l .
A H labor f o r c e data are f r o m Bureau o f Labor S t a t i s t i c s reports supplied by s t a t e agencies.
Only the unemployment r a t e data are seasonally adjusted.
The Southeast data represent the t o t a l o f t h e six states.
The annual percent change c a l c u l a t i o n is based on t h e most recent data over p r i o r year.
http://fraser.stlouisfed.org/
52
Federal Reserve Bank of St. Louis
M A R C H 1982, E C O N O M I C REVIEW
CONSTRUCTION
DEC
1981
NOV
1981
DEC
1980
ANN.
%
CHG.
150,189
149,232
148,393
+ 1
58,249
1,166.3
58,234
1,179.5
52,491
1,200.4
+ 11
- 3
31,877
29,001
32,234
- 1
25,597
25,843
26,326
- 3
8,383
195.5
8,188
194.2
7,688
183.7
+ 9
+ 6
4,919
4,825
5,530
-11
1,774
1,792
1,919
- 8
577
14.0
566
13.3
558
13.9
+ 3
+ 1
350
361
458
-24
12,299
12,598
12,847
- 4
3,732
90.8
3,614
89.5
2,928
78.5
+27
+16
1,707
1,683
2,461
-31
3,841
3,896
3,939
- 2
1,202
33.4
1,193
32.9
1,320
36.3
- 9
- 8
884
885
799
+11
3,775
3,526
3,270
+ 15
1,508
24.4
1,341
23.8
1,213
18.5
+24
+32
946
869
921
+ 3
1,343
1,406
1,561
-14
307
7.1
356
8.4
629
9.6
-51
-26
480
499
331
+45
2,565
2,625
2,789
- 8
1,056
25.7
1,117
26.3
1,040
26.9
+ 2
- 4
553
528
560
- 1
NOV
1981
DEC
1980
60,063
1,123.7
61,998
1,170.1
63,668
1,331.4
- 6
-16
R e s i d e n t i a l P e r m i t s - Thous.
Number s i n g l e - f a m i l y
Number m u l t i - f a m i l y
557.5
411.6
575.8
424.4
704.0
466.9
-21
-12
Value - $ m i l .
Number of U n i t s - Thous.
12,296
262.3
12,829
274.6
13,107
312.2
- 6
-16
R e s i d e n t i a l P e r m i t s - Thous.
Number s i n g l e - f a m i l y
Number m u l t i - f a m i l y
117.9
100.9
123.5
106.7
154.4
124.1
-24
-19
Value - $ m i l .
Number o f U n i t s - Thous.
847
21.7
864
22.3
903
25.0
- 6
-13
R e s i d e n t i a l P e r m i t s - Thous.
Number s i n g l e - f a m i l y
Number m u l t i - f a m i l y
5.4
5.5
5.8
6.0
9.2
7.4
-41
-26
Value - $ m i l .
Number of Units - Thous.
6,860
146.4
7,301
155.8
7,458
176.6
R e s i d e n t i a l P e r m i t s - Thous.
Number single-family
Number m u l t i - f a m i l y
70.4
72.9
74.6
77.9
89.1
86.2
-21
Value - $ m i l .
Number o f U n i t s - Thous.
1,755
37.0
1,819
38.3
1,820
44.4
- 4
-17
R e s i d e n t i a l P e r m i t s - Thous.
Number s i n g l e - f a m i l y
Number m u l t i - f a m i l y
21.1
8.8
21.4
8.3
26.7
8.6
-21
+ 2
Residential Contracts
Value - $ m i l .
Number o f Units - Thous.
1,321
24.5
1,316
25.2
1,136
24.0
+16
+ 2
R e s i d e n t i a l P e r m i t s - Thous.
Number s i n g l e - f a m i l y
Number m u l t i - f a m i l y
9.9
8.1
10.1
8.3
11.6
8.3
-15
- 2
Residential C o n t r a c t s
Value - $ m i l .
Number of Units - Thous.
556
12.6
551
12.6
601
14.9
- 7
-15
R e s i d e n t i a l P e r m i t s - Thous.
Number s i n g l e - f a m i l y
Number m u l t i - f a m i l y
3.5
1.7
3.6
1.8
5.1
5.1
-31
-67
956
20.1
979
20.5
1,189
27.3
-20
7.6
3.9
8.0
4.5
12.7
8.4
-40
-54
12-Month C u m u l a t i v e R a t e
Total Construction Contracts
Value - $ m i l .
Nonresidential C o n t r a c t s
Value - $ m i l .
Sq. F t . - m i l .
Nonbuilding C o n t r a c t s
Value - $ m i l .
•I
ANN.
%
DEC
1981
Residential Contracts
Value - $ m i l .
Number of U n i t s - Thous.
CHG.
—
Total Construction Contr
Value - $ m i l .
Nonresidential C o n t r a c t s
Value - $ m i l .
Sq. F t . - m i l .
Nonbuilding C o n t r a c t s
Value - $ m i l .
\MA
Total Construction Contr
Value - , $ m i l .
Nonresidential C o n t r a c t s
Value - $ m i l .
Sq. F t . - m i l .
Nonbuilding C o n t r a c t s
Value - $ m i l .
Total C
Value - $ m i l .
Nonresidential C o n t r a c t s
Value - $ m i l .
Sq. F t . - m i l .
Nonbuilding C o n t r a c t s
Value - $ m i l .
Value - $ m i l .
Nonresidential C o n t r a c t s
Value - $ m i l .
Sq. F t . - m i l .
Nonbuilding C o n t r a c t s
Value - $ m i l .
m
-15
IANA
Total C o n s t r u c t i o n C o n t r a c t s
Value - $ m i l .
Nonresidential C o n t r a c t s
Value - $ m i l .
Sq. F t . - m i l .
Nonbuilding C o n t r a c t s
Value - $ m i l .
»ISSIPPl
Total Construction Contracts
Value - $ m i l .
.Nonresidential C o n t r a c t s
Value - $ m i l .
Sq. F t . - m i l .
Nonbuilding C o n t r a c t s
Value - $ m i l .
Value - $ m i l .
Nonresidential C o n t r a c t s
Value - $ m i l .
Sq. F t . - m i l .
Nonbuilding C o n t r a c t s
Value - $ m i l .
t
Notes:
Value - $ m i l .
Number o f U n i t s - Thous.
R e s i d e n t i a l P e r m i t s - Thous.
Number s i n g l e - f a m i l y
Number m u l t i - f a m i l y
-26
C o n t r a c t s are c a l c u l a t e d f r o m the F . W. Dodge C o n s t r u c t i o n P o t e n t i a l s . P e r m i t s are c a l c u l a t e d f r o m the Bureau o f the Census,
Housing U n i t s A u t h o r i z e d By Building P e r m i t s and Public C o n t r a c t s . The Southeast data represent t h e t o t a l of the six states. The
annual percent change c a l c u l a t i o n is based on the most recent m o n t h over p r i o r y e a r .
FEDERAL RESERVE BANK O F ATLANTA
53
Ü
•
•
•
•
»
GENERAL
DEC
1981
Personal Income-$ b i l . S A A R
(Dates: 3Q, 2Q, 3Q)
R e t a i l Sales - $ m i l . - SA
Plane Passenger A r r i v a l s (thous.)
P e t r o l e u m Prod, (thous. bis.)
Consumer P r i c e Index
1967=100 (JAN.)
NOV
1981
DEC
1980
ANN.
%
CHG.
2,412.9
87.1
N.A.
8,607.6
2,340.5
87.2
N.A.
8,613.3
2,155.8
83.4
N.A.
8,541.1
+ 2
282.5
281.5
260.5
+ 8
282.1
N.A.
4,239.7
1,407.8
272.8
N.A.
3,719.3
1,412.0
249.2
N.A.
4,026.2
1,436.3
+13
N.A.
N.A.
N.A.
32.4
N.A.
105.2
59.4
31.4
N.A.
102.4
60.0
29.1
N.A.
113.5
61.0
N.A.
N.A.
N.A.
Personal Income-$ b i l . S A A R
(Dates: 3Q, 2Q, 3Q)
Taxable Sales - $ m i l . ( J A N . )
Plane Passenger A r r i v a l s (thous.)
P e t r o l e u m Prod, (thous. bis.)
Consumer P r i c e Index - M i a m i
N o v . 1977 = 100
102.4
66,806
2,109.3
90.4
JAN
155.2
98.3
66,715
1,725.5
93.0
NOV
153.6
88.8
59,194
2,182.1
116.5
JAN
137.3
+ 15
+ 13
- 3
- 22
GEORGIA
Personal Income-$ b i l . S A A R
(Dates: 3Q, 2Q, 3Q)
Taxable Sales - $ m i l .
Plane Passenger A r r i v a l s (thous.)
Petroleum Prod, (thous. bis.)
Consumer P r i c e Index - A t l a n t a
1967 = 100
48.7
N.A.
1,599.1
N.A.
DEC
282.2
47.6
N.A.
1,464.9
N.A.
OCT
281.5
43.7
N.A.
1,697.5
N.A.
DEC
258.3
+11
40.4
N.A.
255.2
1,164.0
39.1
N.A.
259.6
1,165.0
35.3
N.A.
253.8
1,160.5
+14
N.A.
N.A.
N.A.
18.3
N.A.
30.8
94.0
17.7
N.A.
30.0
94.0
16.5
N.A.
33.7
98.3
N.A.
N.A.
N.A.
39.8
N.A.
140.1
N.A.
38.8
N.A.
136.8
N.A.
35.8
N.A.
140.1
N.A.
N.A.
N.A.
N.A.
SOUTHEAST
Personal Income-$ b i l . S A A R
(Dates: 3Q, 2Q, 3Q)
Taxable Sales - $ m i l .
Plane Passenger A r r i v a l s (thous.)
P e t r o l e u m Prod, (thous. bis.)
Consumer P r i c e Index
1967=100
Personal Income-$ b i l . S A A R
(Dates: 3Q, 2Q, 3Q)
Taxable Sales - $ m i l .
Plane Passenger A r r i v a l s (thous.)
P e t r o l e u m Prod, (thous. bis.)
Consumer P r i c e Index
1967=100
Personal Income-$ b i l . S A A R
(Dates; 3Q, 2Q, 3Q)
Taxable Sales - $ m i l .
Plane Passenger A r r i v a l s (thous.)
P e t r o l e u m Prod, (thous. bis.)
Consumer P r i c e Index
1967 = 100
Personal Income-$ b i l . S A A R
(Dates: 3Q, 2Q, 3Q)
Taxable Sales - $ m i l .
Plane Passenger A r r i v a l s (thous.)
P e t r o l e u m Prod, (thous. bis.)
Consumer P r i c e Index
1967 = 100
Personal Income-$ b i l . S A A R
(Dates: 3Q, 2Q, 3Q)
Taxable Sales - $ m i l .
Plane Passenger A r r i v a l s (thous.)
P e t r o l e u m Prod, (thous. bis.)
Consumer P r i c e Index
1967 = 100
+12
+ 4
+ 5
- 2
+ 12
- 7
- 3
+13
- 6
+ 9
+ 1
+ 0
+11
- 9
- 4
+11
0
JAN
1982
JAN
1981
DEC
1981
ANN.
%
CHG.
••
Agriculture
Prices R e c ' d by Farmers
Index (1967=100)
130.0
78,942
B r o i l e r Placements (thous.)
C a l f Prices ($ per c w t . )
57.90
27.1
B r o i l e r Prices (<t per lb.)
Soybean Prices ($ per bu.)
6.05
Broiler Feed Cost ($ per ton)
211
128.0
77,961
57.80
24.6
6.00
210
145.0
79,388
69.80
30.2
7.54
237
-10
- 1
-17
-10
-20
-11
Agriculture
Prices R e c ' d by F a r m e r s
117.9
Index (1967=100)
B r o i l e r Placements (thous.)
31,337
C a l f Prices ($ per c w t . )
52.70
B r o i l e r P r i c e s (<t per lb.)
25.6
Soybean Prices ($ per bu.)
6.21
Broiler Feed C o s t ($ per ton)
207
111.6
31,078
53.78
23.5
6.09
203
129.2
31,198
66.71
28.9
7.66
234
- 9
+ 0
-21
-11
-19
-12
Agriculture
F a r m Cash Receipts - $ m i l .
(Dates: NOV, NOV)
1,886
B r o i l e r Placements (thous.)
9,684
C a l f Prices ($ per c w t . )
54.40
B r o i l e r Prices (<t per lb.)
23.5
Soybean Prices ($ per bu.)
6.21
B r o i l e r Feed C o s t ($ per ton)
230
9,691
54.20
22.5
6.12
215
1,806
10,525
62.00
28.5
7.66
230
+ 4
- 8
-12
-18
-19
0
Agriculture
Farm Cash Receipts - $ m i l .
(Dates: NOV, NOV)
3,610
B r o i l e r Placements (thous.)
1,904
C a l f Prices ($ per c w t . )
54.40
B r o i l e r Prices ($ per lb.)
25.0
Soybean Prices ($ per bu.)
6.21
B r o i l e r Feed Cost ($ per ton)
220
2,006
57.30
24.0
6.12
215
3,759
1,728
65.70
29.0
6.97
240
- 4
+10
-17
-14
-11
- 8
Agriculture
F a r m Cash Receipts - $ m i l .
(Dates: NOV, NOV)
2,913
B r o i l e r Placements (thous.)
12,344
C a l f Prices ($ per c w t . )
49.20
B r o i l e r Prices (<t per lb.)
25.5
Soybean Prices ($ per bu.)
5.99
B r o i l e r Feed Cost ($ per ton)
194
12,162
50.80
23.0
5.95
194
2,514
12,346
61.80
28.5
7.40
240
+16
- 0
-20
-11
-19
-19
Agriculture
F a r m Cash Receipts - $ m i l .
(Dates: NOV, NOV)
1,546
B r o i l e r Placements (thous.)
N.A.
C a l f Prices ($ per c w t . )
56.60
B r o i l e r Prices (<t per lb.)
28.5
Soybean Prices ($ per bu.)
6.27
B r o i l e r Feed C o s t ($ per ton)
245
N.A.
52.90
25.5
6.17
240
1,109
N.A.
66.50
30.0
7.83
250
+39
Agriculture
F a r m Cash Receipts - $ m i l .
(Dates: NOV, NOV)
2,041
B r o i l e r Placements (thous.)
6,102
50.70
C a l f Prices ($ per c w t . )
B r o i l e r Prices (4 per lb.)
29.0
Soybean Prices ($ per bu.)
6.24
B r o i l e r Feed Cost ($ per ton)
183
5,873
54.30
25.5
6.08
188
1,598
5,619
71.00
30.5
7.79
225
+28
+ 9
-29
- 5
-20
-19
Agriculture
F a r m Cash Receipts - $ m i l .
(Dates: NOV, NOV)
1,607
B r o i l e r Placements (thous.)
1,303
C a l f Prices ($ per c w t . )
51.60
B r o i l e r Prices (<t per lb.)
24.0
Soybean Prices ($ per bu.)
6.21
Broiler Feed Cost ($ per ton)
210
-
1,346
52.30
22.0
6.07
200
1,539
1,178
71.10
27.0
8.02
215
+ 4
+ 11
-27
-11
-23
- 2
-15
- 5
-20
- 2
Notes:
Personal Income data supplied by U. S. D e p a r t m e n t o f C o m m e r c e . Taxable Sales are r e p o r t e d as a 1 2 - m o n t h c u m u l a t i v e t o t a l .
Plane
Passenger A r r i v a i s are c o l l e c t e d f r o m 26 a i r p o r t s . Petroleum P r o d u c t i o n data supplied by U. S. Bureau o f Mines. Consumer P r i c e
Index data supplied by Bureau o f Labor S t a t i s t i c s . A g r i c u l t u r e data supplied by U . S. D e p a r t m e n t of A g r i c u l t u r e . Farm Cash
Receipts data are reported as c u m u l a t i v e f o r the calendar year through t h e m o n t h shown. B r o i l e r placements are an average w e e k l y
r a t e . The Southeast data represent the t o t a l o f the six states. N . A . = not a v a i l a b l e . The annual percent change c a l c u l a t i o n is based
on most recent data over prior year.
54
M A R C H 1982, E C O N O M I C REVIEW
*
Federal Reserve Bank of Atlanta
P.O. Box 1731
Atlanta, Georgia 30301
Address Correction Requested
Bulk Rate
U.S. Postage
PAID
Atlanta, Ga.
Permit 292
@FedFRASER