Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter.
Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 99
Productivity in Services
JOHN W. KENDRICK
The chief factor behind secular increases in productivity the ratio of
output to inputs in real terms—is innovation in the technology and organi-
zation of production. Other factors, such as cyclical changes in the rate of
utilization of capacity, are also important, as discussed later. However, in
the long run, it is technological advance that is primarily responsible for
reducing real costs per unit of output (the opposite side of the productivity
coin). This is obviously so in the case of process innovations or new and
improved producers' goods. Even the introduction and diffusion of new
consumer goods have a positive influence on productivity through the learning
curve effect.
In the first section of this chapter, estimates of rates of change in pro-
ductivity in the goods and services sectors of the U.S. business economy by
major industry groups, 1948-1986, are presented. This furnishes the basis
for analysis of causal factors behind productivity in the second section by
using two of the three major approaches available. The first is growth ac-
counting by which the chief forces promoting productivity growth in the total
economy are identified and evaluated. The second is the use of multiple
regression analysis to relate relative changes in productivity by industry to
differences in levels or rates of change in causal variables.
There is a third approach at the microlevel which uses case studies of
individual firms and technologies. This approach is well represented in the
companion book to this volume (Guile and Quinn, 1988~. Additionally, the
American Productivity Center in Houston has published more than 50 case
studies of productivity in various companies covering innovation, human
99
OCR for page 100
100
JOHN W. KENDRICK
resources management, and other factors.) There have also been interfirm
and interplant comparisons of productivity levels and rates of change that
have generally confirmed the findings of macroeconomic analyses.
The third and final section of this chapter presents a menu of policy options
for promoting productivity growth. These are generally policies that would
stimulate productivity across the board, but some are of particular relevance
to the services sector as pointed out in the concluding comments.
OUTPUT, INPUT, AND PRODUCTIVITY DEVELOPMENTS, 1948-1986
Sector Comparisons
The marked increase in the share of the services sector (defined in terms
of the criterion of tangibility), both in persons engaged and in real gross
product between 1948 and 1986, is shown in Table 1. Persons engaged
include proprietors and full-time equivalent employees. Real gross product
is the value added by the various industry groups in constant 1982 dollars,
as estimated by the Bureau of Economic Analysis in the U.S. Department
of Commerce.
The proportion of all persons engaged in the U.S. economy working in
services, private and public, rose from 54 percent in 1948 to more than 72
percent in 1986. This continues a trend that has been evident since at least
1870, when 23 percent of the labor force was engaged in services. The share
of real gross product originating in services has also increased, but to a
somewhat lesser degree, as shown in Table 1. This implies that real product
per person engaged rose less rapidly in services than in goods production.
An alternative picture of sectorial shares is provided by the breakdown of
final goods and services entering the gross national product (GNP), shown
in Table 2. The proportion of tangible goods declined from 48 percent in
1948 to approximately 43 percent in 1973 and 1986. The share of final
services grew from 39.5 percent in 1948 to 46.6 percent in 1986. The share
of structures remained close to 12.5 percent in 1948 and 1973, declining to
10.4 percent in 1986. This sectoring gives somewhat different results from
the industry sectoring because it is based on final products, and some of the
services industries add value to tangible goods. However, Table 2 confirms
the finding that services constitute an increasing proportion of GNP.
Returning to the broad sectorial aggregations of industries, Table 3(A)
shows that real gross product in services grew more than 1 percent faster
per year, on the average, than in goods production between 1948 and 1986-
3.7 compared with 2.6 percent. The slowdown of growth after 1973 was
less pronounced in services than in goods.
Table 3(B) shows the relative performance of total factor productivity
(TFP). As a measure of the net savings in labor plus capital inputs per unit
OCR for page 101
PRODUCTIVITY IN SERVICES
TABLE 1 Real Gross Domestic Product and Persons Engaged in U.S.
Domestic Economy by Sector and Industry Group
101
1948
1973 1986
A. Persons engaged
Total (thousands) 58,301 83,299 104,645
Percentage distribution
Agnculture 11.4 3.8 2.8
Mining 1.8 0.8 0.8
Construction 5.7 5.8 5.8
Manufactunng 27.4 23.9 18.0
Total Goods 46.2 34.2 27.4
Transportation 5.2 3.4 3.1
Communications 1.3 1.3 1.1
Public utilities 0.9 0.9 0.9
Trade 18.3 19.5 21.2
Finance, insurance, real estate 3.2 5.0 6.3
Services 13.2 17.9 23.7
Total Private Services 42.1 48.0 56.4
Government 11.7 17.8 16.2
B. Real gross domestic product
Billions of 1982 dollars 1,114.9 2,726.6 3,694.5
Pecentage distribution
Agnculture 5.5 2.6 2.7
Mining 6.5 4.9 3.2
Construction 8.1 6.2 4.6
Manufactunng 21.4 22.8 22.0
Total Goods 41.5 36.5 32.5
Transportation 6.9 4.3 3.5
Communications 0.8 1.9 2.6
Public utilities 1.2 2.7 2.8
Trade 14.5 16.0 17.4
Finance, insurance, real estate 9.7 13.5 13.9
Services 11.6 12.5 15.3
Total Private Services 44.6 50.9 56.5
Government 13.9 12.6 11.0
SOURCE: Bureau of Economic Analysis, U.S. Department of Commerce.
of output over time, TFP indicates changes in productive efficiency. Over
the long term, increases in TFP result primarily from innovations in the
technology of production, which are usually incorporated in capital goods.
Organizational changes, including those associated with increasing scale, are
als~important. Between 1948 and 1986, the average annual percentage rate
of increase in TFP in services was 1.4 percent, compared with 2.1 percent
in goods production. The slowdown in goods productivity from 1973 to 1979
OCR for page 102
102
TABLE 2 Distribution of Real Gross National Product by Major
Category of Final Product
1948
Billions of 1982 Dollars
JOHN W. KENDRICK
1973
1986
Total GNP 1,109 2,744 3,713
Goods 532 1,175 1,595
Services 438 1,219 1,731
Structure 139 350 387
Percentage Distribution
Total GNP 100.0 100.0 100.0
Goods 48.0 42.8 43.0
Services 39.5 44.4 46.6
Structures 12.5 12.8 10.4
SOURCE: Bureau of Economic Analysis, U.S. Department of Commerce.
was greater than in services. However, services productivity growth fell a
bit further from 1979 to 1986, whereas goods productivity accelerated smartly
due mainly to comebacks in manufacturing and farming.
The reconciliation item between rates of change in TFP and in the more
conventional real product per labor hour ("labor productivity") is the rate
of substitution of capital for labor. This is indicated by the rates of change
in the capital/labor ratios shown in Table 3(C) when they are weighted
(multiplied) by the capital share of gross product, which averages about one-
third. The capital/labor ratio increased much more in goods than in services
between 1948 and 1986 2.3 percent a year, on average, compared with
1.3 percent. There was a slowing in the 1973- 1979 subperiod but a comeback
after 1979 in both sectors. The slower growth of capital (primarily structures
and equipment) in services has important policy implications, which are
discussed later.
Labor productivity increased at a 2.2 percent average annual rate in the
U.S. business economy between 1948 and 1986, compared with 1.7 percent
for TFP, the difference reflecting the growth of capital goods per worker.
The increase in goods productivity averaged 2.7 percent annually, and in
services 1.9 percent. As in the case of TFP, there was a sharp deceleration
in labor productivity from 1973 to 1979, with a comeback in the goods sector
from 1979 to 1986, in contrast to continued slow growth in the services
sector as a whole.
An important caveat must be attached to the output and productivity es-
timates based on real gross domestic product (GDP) by industry as estimated
by the Department of Commerce. As documented elsewhere (Kendrick,
1986a), about half of the base-period value added in the finance and private
services industry groups has, in effect, been extrapolated by employment or
OCR for page 103
PRODUCTIVITY IN SERVICES
TABLE 3 Output and Productivity Ratios in U.S. Business Economy by
Major Sector and Industry Group (average annual percentage rates of
change 1948-1986, by subperiod)
103
1948-1985 1948-1973 1973-1979
1979- 1986
A. Real Gross Product in 1982 Dollars
Business economy 3.2 3.6 2.5 2.5
Goods production 2.6 3.2 1.3 1.6
Farming 1.4 0.4 1.5 5.0
Manufacturing 3.3 3.9 1.9 2.2
Mining 1.3 2.5 - 0.4 - 1.4
Construction 1.7 2.6 0.6 - 0.4
Services production 3.7 4.0 3.4 3.1
Transportation 1.4 1.7 2.7 - 0.9
Communications 6.4 7.1 6.2 4.0
Public utilities 5.6 7.2 1.8 3.2
Trade 3.7 4.1 2.7 3.4
Finance and insurance 4.1 4.3 3.4 3.9
Real estate 3.7 4.2 4.3 1.4
Services 4.1 4.0 4.4 4.3
B. Total Factor Productivity
Business economy 1.7 2.3 0.3 0.8
Goods production 2.1 2.7 0.1 1.8
Farming 3.6 3.2 1.8 6.7
Manufacturing 2.1 2.3 0.6 2.5
Mining 0.7 2.7 - 6.0 - 0.5
Construction - 0.3 0.6 - 2. 2 - 1 .9
Services production 1.4 2.0 ().5 -0.3
Transportation 1.6 2.3 1.5 - 0.5
Communications 3.8 4.9 2.4 1.3
Public utilities 3.3 4.9 - 0.6 0.9
Trade 2.0 2.5 0.4 1.4
Finance and insurance 0.4 1.1 - 0.7 - 1.0
Real estate 0.7 1.3 1.4 - 2.0
Services 0.6 0.7 0.4 0.3
C. Capital/Labor Ratios
Business economy 1.9 2.0 1.2 1.7
Goods production 2.3 2.4 1.9 2.2
Farming 4.2 5.2 2.6 2.2
Manufacturing 2.8 2.6 3.2 3.2
Mining 4.1 5.1 - 3.3 7.1
Construction 1.3 2.3 1.5 - 2.3
Services production 1.3 1.5 0.6 1.2
Transportation 0.2 0.2 - 0.1 0.3
Communications 5.0 5.2 4.0 5.1
Public utilities 2.5 3.3 1.5 0.4
Trade 3.0 3.1 2.4 3.2
Finance and insurance 2.9 1.9 4.0 6.3
Real estate 0.5 0.7 -1.7 1.5
Services 1.0 1.8 0.5 - 0.7
continued
OCR for page 104
104
TABLE 3 Continued
JOHN W. KENDRICK
1948-1985 1948-1973 1973-1979 1979-1986
D. Real Product per Labor Hour
Business economy 2.2 2.8 0.6 1.4
Goods production 2.7 3.2 0.5 2.5
Farming 5.0 4.7 3.1 7.9
Manufacturing 2.7 2.8 1.4 3.5
Mining 1.8 4. 1 - 6.9 - 1.7
Construction - 0.2 0.8 - 2.1 - 2.1
Services production 1.9 2.5 0.7 0.7
Transportation 1.7 2.4 1.5 - 0.4
Communications 5.4 6.0 4.3 4.1
Public utilities 4.4 6.3 0.3 1.2
Trade 2.4 2.9 0.8 2.0
Finance and insurance 0.9 1.4 -0.1 0.2
Real estate 1.2 2.0 - 0.2 - 0.6
Services 0.9 1.3 0.2 0.1
SOURCE: American Productivity Center.
hours data. Thus, real gross product in those portions of the finance and
services groups, as in government, contains no allowance for productivity
advance. If productivity grew as much in those industries as in the ones for
which independent output estimates were used, productivity in the services
sector as a whole would have grown by 0.3 percentage point more over the
period as a whole, narrowing the gap with the goods sector as measured.
There may, however, be some downward bias in estimates of real product
and productivity in the goods sector, particularly in contract construction.
So the conclusion still holds that productivity growth in the services sector
is significantly below that in goods.
Industry Differences
A glance at Table 1 shows the great differences among the industry groups
and their relative importance in terms of persons engaged or gross product
originating. This is also true, of course, of the detailed industry components
of the groupings.
Within the services sector, the groups differ in the proportions in which
they render services to business and to consumers, just as the goods industries
produce both final and intermediate products, as detailed in input-output
tables. The regulated groups—transportation, electric and gas utilities, and
communications (TUC) are highly capital intensive, while the other groups
are generally not so. The other services groups are also quite fragmented
into many small establishments. They tend to have higher labor turnover,
OCR for page 105
PRODUCTIVITY lN SERVICES
105
less well-educated workers (except for professional services), and a higher
rate of business failures. All of these characteristics have a bearing on pro-
ductivity and technological innovation.
With respect to productivity growth rates, shown in Table 3, communi-
cations and public utilities groups were among the highest in the economy.
Transportation and trade, whose output is measured in terms of the physical
volume of goods distributed, had productivity growth rates close to the
economy average. It was in the finance and insurance, real estate, and services
(FIRES) groups that average rates were well below average, even after
allowance was made for the probable downward bias in the output estimates.
Rates of change in labor productivity for a wider range of services in-
dustries, prepared by the Bureau of Labor Statistics (BLS), are presented in
the chapter by Mark. The finer the industry detail, the greater is the dispersion
in rates of change. BLS estimates suggest somewhat better performance of
productivity in services than those based on Bureau of Economic Analysis
(BEA) real product numbers, which supports the contention that the latter
are subject to some downward bias.
Causal Factors Behind Productivity Change
The wide dispersion in rates of productivity change among industries
provides a convenient handle for analysis of causal factors. The differential
productivity growth rates can be regressed against differences in levels or
rates of growth of independent variables believed to have a causal relation-
ship, to identify those that are significant. Results of recent studies have been
reported by Kendrick (1986b). Variables that had a significantly large positive
association with industry rates of productivity growth in the United States
since 1948 are the following:
· ratios of research and development outlays to sales, direct and indirect,
· changes in R&D ratios,
· rates of growth of real fixed capital stocks per worker,
· average education per worker,
· proportion of nonproduction workers and of females in total industry
employment,
· variability in layoff rates, and
· economies of scale.
Variables that were negatively associated with productivity changes in-
clude:
· the amplitude of cyclical changes in output,
· average hours worked per week and changes in average hours,
· the percentage of workers in an industry belonging to unions,
· days lost because of strikes,
OCR for page 106
106
JOHN W. KENDRICK
· changes in concentration ratios, and
· changes in the female proportion of the work force.
Since some of the explanatory ratios were correlated (e.g., average education
of workers and R&D ratios), only one or the other variable shows a significant
(t value) in any given multiple regression.
Interindustry analyses help explain the differences in productivity growth
between the goods and services sectors, particularly the FIRES group of
services (Kendrick, 1987~. Stronger and less cyclical output growth was
favorable for most services industries. However, these factors were out-
weighed by others that were less favorable. In particular, the capital/labor
ratio grew less rapidly in services; little if any formal R&D was performed
directly, and less benefit was derived from the R&D of suppliers of capital
and intermediate products. Hours are longer and the length of the standard
work week and year has declined less in most services industries than in
goods production. The proportions of females and youth, who have less
average experience, have generally increased more. Further, concentration
is far lower in most services industries than in manufacturing, and the average
education of employees is lower. The same is probably true of training,
although data are hard to come by concerning this variable.
These findings are generally consistent with analyses of interfirm produc-
tivity differences, and with studies of international differences in levels and
rates of change in productivity of industries and establishments. A recent
study in the latter category also found that exposure to international trade,
both exports and imports, and competition from domestic subsidiaries of
foreign companies have a positive influence on productivity growth (Davies
and Caves, 19871.
At the national level, regression analysis is not a useful analytical tool
because of multicollinearity among the time series representing causal forces.
The growth accounting approach developed by Edward F. Denison has proved
more useful in estimating at least roughly the contributions of various causal
factors to macroeconomic growth (Denison, 19854. The technique involves
quantifying variables and establishing their relative weights based on their
marginal productivities, i.e., their contributions to national income and prod-
uct. For example, the contributions of various types of investment, tangible
and intangible, are evaluated in terms of their rates of return.
Studies by Denison (1985), Maddison (1987), and Kendrick (1979) have
identified the following factors as the most important determinants of growth
of real gross business product per unit of labor input: (1) the rate of growth
of real tangible capital (primarily structures and equipment) per worker; (2)
advances in technological knowledge as applied to the processes of produc-
tion; (3) human capital (primarily education and training) per worker; (4)
volume factors: economies of scale associated with growth rates and changes
OCR for page 107
PRODUCTIVITY IN SERVICES
107
in rates of utilization of capacity associated with cyclical variations; (5)
changes in labor efficiency relative to sustainable norms; and (6) govern-
mental measures not reflected in the other proximate determinants, partic-
ularly regulations that increase real costs (inputs) but not measured outputs.
It is with respect to these categories of causal factors that policy options are
discussed in the next section.
It may be noted that the results of the growth accounting studies are
generally consistent with the findings of the interindustry analyses. The latter,
however, are able to include certain institutional factors, such as levels and
changes in concentration of firms and the degree of unionization of workers,
which differ by industry but may not exhibit pronounced trends at the national
level.
There are other causal forces that are less important at present or are not
amenable to policy measures. An example of the former is changes in the
composition of output toward industries with higher rates of return. For
example, the relative shift out of agriculture until the early 1970s added
approximately 0.3 percentage point to productivity growth, according to
Denison (1985), but has little effect at present. Also, the decline in the
quality of natural resources subtracts about 0.1 percentage point, but there
is little that can be done about it. Similarly, the decline in average experience
of the work force, associated with the youth bulge beginning in the 1960s,
and the increase in the female proportion are not susceptible to policy. In
any case, the influence of the changes in age composition of the labor force
has been mildly positive in the 1980s as the baby boom generation entered
the prime working ages.
The factors discussed above are proximate determinants of productivity
change. Underlying these are the basic values and institutions of a society.
Factors such as material aspirations, the work ethic, and the efficacy of the
profit motive change slowly. The institutional framework of the economy
continues to evolve, but it is usually far from clear what the effects of given
changes will be. Consequently, issues relating to social and economic reforms
are not discussed, but possible changes in tax laws or government regulations,
which can have fairly immediate effects on the economy, are considered.
Growth accounting has been applied to explain differences in rates of
productivity growth among countries. Denison found that between 1950 and
1962, the chief causes of the faster growth in other Organization for Economic
Cooperation and Development countries were
· economies of scale from expansions of local and international markets,
· improved allocation of resources,
· more intensive utilization of capacity, and
· increases in capital per unit of labor.
Education per worker did not increase as rapidly in any of the other countries
OCR for page 108
108
JOHN W. KENDRICK
as it did in the United States. Surprisingly, the technological gap with the
United States did not decline significantly in the European countries except
for France and possibly Italy.
A follow-up study, based on Denison's model, indicated that between
1960 and 1979 a dramatic narrowing of the technological gap did occur,
which reflected technological catch-up coupled with significant increases in
R&D spending abroad (Kendrick, 19811. This was associated with stronger
increases in capital goods per unit of labor than in the United States as a
result of higher savings rates and the opportunities for profitable investments
opened up by technological progress and expanding markets.
Growth accounting has also been used to explain changes in growth rates.
Thus, the productivity slowdown in the United States between 1973 and
1981, which is also evident in other industrialized countries to varying de-
grees, was due primarily to the following developments:
· slower growth of real capital per unit of labor, as a result of accelerated
labor force growth and a reduced rate of capital formation;
· a decline in advances of technological knowledge consequent on declines
in the ratio of R&D to GNP;
· less favorable effects of resource reallocations;
· lesser economies of scale with deceleration of output growth;
· a reduction in rates of utilization of capacity; and
· increasing real costs of compliance with government regulations.
The oil shocks of 1973 and 1979 were major factors behind some of these
negative developments.
Since 1981, a reversal of most of the depressing influences has resulted
in a modest recovery of the productivity growth rate. However, the recovery
has been strong mainly in the manufacturing and agricultural industry groups.
Productivity growth continued to be slow in the services sector where a lesser
degree of disinflation was associated with a slower growth of output than in
the goods sector through 1987.
POLICIES TO PROMOTE PRODUCTIVITY AND
TECHNOLOGICAL ADVANCE
The policy options discussed in this section are general and not industry
or technology specific. The focus is on measures that can be taken mainly
by the federal government to stimulate advances in technological knowledge
and their application to the ways and means of production in the directions
judged to be most productive by the players in the enterprise economy.
A lengthy paper "Policies to Promote Productivity Growth" (Kendrick,
1980) contained 99 policy options. A good many of the measures have since
been adopted, and others proposed. A number of measures are now suggested
OCR for page 109
PRODUCTIVITY IN SERVICES
109
under six major headings that correspond to the major causal forces behind
productivity advance: economic expansion, increasing technological knowl-
edge, savings and tangible investments, human investment, labor efficiency,
and government policy not covered in the previous points.
Volume Factors Rates of change in output and productivity show a strong
positive correlation, reflecting economies of both scale and utilization. Econ-
omies of scale are a function of the rate of growth of output, since growth
opens up opportunities for greater specialization of people, plants, and equip-
ment, and spreads overhead functions over more units. Cyclically, expansion
from lower rates of utilization of capacity toward most efficient rates boosts
productivity and vice versa. In general, innovation is easier when output is
expanding smartly than when it stagnates or declines.
The economic expansion beginning in late 1982 proceeded rapidly through
mid-1984, then slowed drastically through 1986. Rates of utilization of in-
dustrial capacity declined, and so did the rate of productivity growth, es-
pecially in services. It would be desirable if demand in the economy increased
at least as fast as the expansion of productive capacity, estimated at 3-4
percent per year. Marked cyclicality of demand also slows the growth of
capacity and productivity. The slowdown for the decade beginning in 1973
was due in part to the most severe contractions since World War II, which
took place in 1974 and 1982.
The objective of relatively steady growth of real GNP around the "natural"
rate of unemployment requires appropriate macroeconomic policies. Since
the unemployment rate in 1987 averaged 6.2 percent approximately 1 per-
centage point more than the natural rate below which wage and price inflation
tends to accelerate growth could probably proceed at about 4 percent for
a couple of years without seriously heating up inflation. Then growth should
slow somewhat, reflecting the slower growth of the labor force projected for
the l990s and the assumed leveling of unemployment at 5-5~/: percent of
the civilian labor force.
An important part of the stronger expansion should be a continuation of
the faster growth of exports than of imports in real terms that started in the
fall of 1986, due in part to the major decline in the foreign exchange value
of the dollar since early 1985. To promote continued significant declines in
the trade balance, U.S. foreign economic policy must not only seek to main-
tain an equilibrium value of the dollar; it must also encourage stronger
economic growth of our major trading partners and continue negotiations for
freer, fairer trading rules and practices on their part and ours. It should be
noted that exports of services as well as of goods have benefited from the
improved trade balance.
More generally, the expansion of the share of the services sector in the
economy has tended to reduce cyclical fluctuations. This makes the task of
OCR for page 110
110
JOlIN W. KE1!,'DRICK
macroeconomic policy easier. On the other hand, the shift to services tends
to reduce the growth of productivity and productive capacity, which under-
lines the importance of seeking measures to accelerate productivity growth
In services.
Promoting Technological Knowledge In the modern era, the fountainhead
of scientific knowledge and technological advance is research and devel-
opment. Ratios of R&D, direct and indirect, to sales within manufacturing
are significantly correlated with rates of productivity advance. The great
increase in the ratio of national R&D to GNP from a small fraction in 1919
to approximately 3 percent in the mid-1960s was a major reason for the
acceleration in productivity growth after World War I. The subsequent decline
in the ratio until the latter 1970s is cited as a reason for the productivity
slowdown after 1973, given the lags between R&D and commercial effects.
The increase in both business- and government-financed R&D since the late
1970s contributed to the pickup in productivity growth after 1982, particularly
in manufacturing.
It is important that R&D continue to increase at least in line with GNP,
and preferably somewhat faster if the supply of scientists and engineers
permits without undue inflation of salaries. Most of the increase in govern-
ment-financed R&D has been for national security purposes. As that increase
slackens, it will be desirable that financing of applied R&D for civilian
purposes be increased, particularly in areas that would benefit production of
services. Recent increases in funding of basic research and the planned
doubling of appropriation requests for the National Science Foundation (NSF)
over the next five years are welcome developments.
The passage of a 25 percent incremental R&D tax credit in 1981 helped
maintain the strong increases in business R&D. This was extended in 1986,
but at a 20 percent rate, and the extension was for a temporary, three-year
period. This may have contributed to a slowing of increases in business
R&D. It is recommended that the credit be increased to at least 25 percent,
that the definition of R&D be broadened, that increments be computed on a
stable base, and that the credit be made pe~anent because R&D programs
require long-term planning (Brown, 1984~.
Patent, copyright, trademark, and trade secret protection has been strength-
ened pursuant to recommendations in the 1985 report of the President's
Commission on Industrial Competitiveness. The protection of intellectual
property has been raised to a priority concern in the coming round of mul-
tilateral trade talks. Some people recommend lengthening the period of patent
protection from 17 to 20 years, which would help compensate for delays in
commercialization of inventions because of federal regulations. There is also
a need to strengthen the Patent and Trademark Office to improve search
capabilities and the reliability of patent grants.
OCR for page 111
PROD UCTIVI7~Y IN SERVICES
111
Serious consideration should be given to proposals in the President's Com-
petitiveness Initiative based on the recommendations of his commission (Council
on Competitiveness, 1987~. Among the initiatives are the following:
establishment of new university-based, interdisciplinary science and
technology centers to perform research in areas that contribute to U.S.
competitiveness (including services);
· creation of a technology share program involving multiyear, joint basic
and applied research with consortia of U.S. firms and universities;
· initiation of an exchange program between scientists and engineers in
.
the public and private sectors;
improved industry access to federal science and technology efforts to
increase transfers and accelerate the spin-off of defense technologies to
the private sectors;
· programs to increase scientific literacy; and
· grants to update the equipment in academic research laboratories, 20
percent of which is obsolete.
The National Cooperative Research Act of 1984 removed antitrust barriers
to joint research. More than 40 consortia have notified the Department of
Justice and the Federal Trade Commission that they intend to take advantage
of the new opportunities. It is important that companies in the services
industries be fully represented in these and related initiatives.
Savings and Investment There is a significant positive correlation between
rates of change in real capital per worker and output per worker. This is not
just a matter of the quantity of capital goods, but also of their quality, because
new plant and equipment are carriers of cost-reducing technological inno-
vations. The volume of business investment depends on expected rates of
return and on the cost of capital, particularly interest rates, which interact
with savings rates. Savings constitute ultimate constraint on the volume of
investment that can be undertaken.
A negative aspect of income taxation is that it drives a wedge between
_ ~ ~
.1~ 1 ~ . . ~ . ~ · · . . _
the returns earned by Investment and the income accruing to investors. It
also drives a wedge between market rates of interest and the interest received
by savers. Another way of expressing the negative influence of income
taxation on savings and investment is that it represents double taxation of
savings in that income is taxed and so is the income from the investments
into which savings flow.
From this point of view, the reductions in both corporate and personal
income tax rates in the tax acts of 1981-1982 and 1986 have had a positive
effect on savings and investment. On the other hand, the elimination of the
investment tax credit and the accelerated cost recovery system in the 1986
act resulted in a drop of business investment between the last quarter of 1985
OCR for page 112
112
JOHN W. KENDRICK
and the first quarter of 1987, contributing to sluggish economic growth.
Consideration should be given by Congress to restoration of an investment
tax credit.
More fundamentally, it will be desirable to further de-emphasize income
taxation, or eliminate it, and to substitute consumption-based taxes. Econ-
omists at the Brookings Institution have made a good case for a consumed
income tax (Bosworth, 19841. A value-added tax is also neutral with respect
to savings and consumption decisions, but appears to be politically unpopular.
An unfortunate aspect of the Tax Act of 1986 was the increase in capital
gains taxation beginning in 1987. Business investments and the formation
of new businesses are undertaken in part in hope of realizing capital gains
in the future. Capital gains taxation reduces that incentive. One attractive
proposal for reducing capital gains tax rates is to reduce the capital gains on
which taxes are levied to the extent that the price level (as measured by the
consumer price index) had risen between the dates of purchase and sale of
the assets.
With respect to interest rates, the most effective way to hold them down
or reduce them further at this time is through continued reductions in the
federal deficit. Government deficits absorb funds that would otherwise be
available for private investment, and competition between business and gov-
ernments in the money markets keeps interest rates higher than they would
be if budgets were balanced. Monetary policy should not be overly stimu-
lative, however, because of the inflationary potential, but it can at least be
neutral in view of the remaining slack in the economy at this time.
Human Investments Growth accounting studies show that increases in av-
erage education and training of the work force contributed about 0.7 per-
centage point to the 2.2 percent increase in labor productivity between 1948
and 1986 (Denison, 1985; Kendrick, 19791. On a cross-sectional basis, as
noted earlier, there is a significant positive correlation between rates of change
in average education and output per labor hour by industry. Along with
tangible investments and technological advances, education and training rank
among the most important forces promoting productivity advance. They are
basic to producing the scientists, engineers, and business managers respon-
sible for innovation and to preparing the labor force generally to operate
increasingly complex technology.
It is important that both public and private outlays for education and
retraining continue to increase as a ratio to GNP. The 1985 report of the
President's Commission on Industrial Competitiveness made several rec-
ommendations in the human resources area. Among these were:
.
expansion of federal training and assistance programs for displaced
workers;
OCR for page 113
PRODUCTIVITY IN SERVICES
113
greater funding of engineering education, including expansion of NSF
engineering centers;
sustained federal support for a program of basic and prototype research
in educational software through NSF and the Department of Labor; and
more effective dialogue among government, industry, labor, academia,
and other interested parties to develop consensus on educational and
other measures to increase competitiveness.
A task force headed by the Secretary of Labor concluded that government
programs to deal with displaced workers should be consolidated and funding
for retraining nearly tripled. The President's fiscal year 1988 budget provided
almost $1 billion for worker adjustment assistance, well above the previous
year. The President's Competitiveness Initiative recommends an $800 million
program under the Job Training Partnership Act to provide summer jobs,
remedial education, and skills training for disadvantaged youths.
The administration's 1988 budget requested an increase of $600 million
for an experimental loan program to allow students to borrow more money
than currently permitted. However, this would come at the expense of pro-
posed reductions in vocational training and in funding of higher education.
An older proposal is for tuition tax credits. Another is for development
and diffusion of new educational methods and technologies by the National
Institutes of Education, along with research to determine those that are most
effective in facilitating learning. Increasing emphasis should be placed on
lifelong learning. The National Academv of Fn~ine.~.rino {~6
. . . . . .
__~ ~ ^A~ i 1988) has
stressed this with respect to the need for continuing education of engineers.
Health is another form of human investment. Here the emphasis must be
on the development of new and more effective medical technolo~v and
treatments, and on the continuing education of the public in healthful life-
styles. Further increases in the productivity of health care facilities are es-
sential to bring down their relative costs.
Labor Efficiency With a given level of technological knowledge, the actual
output of individuals and organizations depends on the degree of efficiency
relative to a sustainable optimum. In the short run, the productivity of most
individuals and groups can be increased significantly with DrOner len~l~.rshin
motivation, and inputs of brain and brawn.
An, +1~ ~~` ~ ~ _~ ~ _ .1_ _ _ 1_ _ _ 1 · . · ^.
~ ~ _ _ ~ . lo,
Over one past decade there has been a major chits in management phi-
losophy and practice in many firs and other organizations from the old
"Theory X" military chain of command model to "Theory Y." involving
more participative management and the use of employee involvement (EI)
systems. Some of the EI plans, such as quality circles and joint labor-
management productivity teams, do not involve special financial incentives.
Others, such as Scanlon or Rucker plans, Improshare, productivity gain-
sharing and profit-sharing plans, do incorporate financial incentives with
. . .
OCR for page 114
114
JOlIN W. KENDRICK
rewards linked to performance. A number of surveys of EI systems, both
with and without financial incentives, indicate that managers believe they
have significantly enhanced productivity.2
Installation and maintenance of EI plans, and systems for tying pay to
performance, depend on management initiatives, of course. However, both
the Department of Labor and the Federal Mediation and Conciliation Service
have staffs that can assist in developing cooperative efforts of management
and labor to promote productivity. They can also assist in union-management
negotiations involving the moderation or elimination of restrictive work rules
or jurisdictional conflicts that impair productivity. The longer tea payoff
of EI plans comes when they elicit worker suggestions for cost-reducing
innovations in the workplace.
An executive order issued by the Office of Management and Budget in
February 1986 requires federal agencies and departments to develop pro-
ductivity measures, to set up systems for involving workers in efforts to
improve productivity, and to use productivity improvement goals in bud-
geting. These efforts are being reviewed and evaluated by the General Ac-
counting Office, which is in a position to recommend modifications to make
them more effective. The federal government is prepared to help state and
local governments in similar efforts. Getting "more bang for the tax buck,"
or spending fewer bucks for the bang, in government is a significant element
in raising national productivity.
Governmental Policies In providing the legal framework within which pri-
vate enterprises operate, government plays an important role in the produc-
tivity of the business economy. The aggregates and composition of government
expenditures and revenues also have a significant impact on the private
economy. The effects of taxation and the government surplus or deficit on
private savings and investment have been discussed. Likewise, the propor-
tions of government expenditures devoted to public investments in infra-
structure and in intangibles such as R&D, education, training, and health
affect the productivity of the private economy. Many economists think that
a separate capital budget for government would help focus attention on the
importance of public investment. Certainly, strict cost-benefit analyses should
be applied in selecting specific investment projects to be undertaken.
Of particular concern to business has been the proliferation of governmental
regulations since 1970. Because the costs of complying with regulations
increase inputs while the presumed benefits do not increase output as mea-
sured, social regulations are believed to have played a part in the productivity
slowdown. It is important that the administration persevere in its efforts to
rationalize the regulatory process, requiring early public involvement before
promulgation of new regulations, coordinating agency activities, tightening
procedural requirements for evaluative choices of proposed regulations and
OCR for page 115
TECHNOLOGY AlVD THE SERVICES SECTOR
115
reevaluation of existing regulations, and improving evaluation methods. Where
regulations are found to have negative impacts, provisions for minimizing
or eliminating such impacts should be made. Agencies should be held ac-
countable for the claimed benefits relative to costs and risks. Performance
should be reviewed regularly, and failure to achieve satisfactory results would
be cause for deregulation or substitution of alternative procedures.
Regulations should be goal oriented, rather than specifying means. Agen-
cies should try to compensate for the greater relative burdens placed on small
firms by regulations. They should study and take account of the impact of
their actions on innovation, productivity, and costs, and thus the international
competitiveness of the U. S. economy.
Reducing or eliminating economic regulations in a variety of industries
has had salutary effects, including increases in productivity (Bailey, 19861.
Technological advances have increased competition within and among in-
dustries, and it is to be hoped that the movement toward reducing or elim-
inating economic regulations will continue.
In recent years, antitrust laws have been interpreted more broadly by the
Department of Justice and the Federal Trade Commission to permit mergers
that would increase efficiency without restraining trade. This contributes to
productivity growth. However, it remains essential that mergers and collusive
activity of firms which would lead to monopolistic pricing be prohibited.
Healthy competition is a powerful force promoting innovation, productivity
advance, and reduction of real costs per unit of output. The effective func-
tioning of the U.S. economy depends on its maintenance.
CONCLUSION
During the golden quarter century (1948-1973) after World War II, pro-
ductivity trends in major portions of the services sector were significantly
below the national average. It is a particular cause for concern that since the
1973-1981 productivity slowdown which affected all industries, productivity
growth in services failed to rebound (except for trade) as it did in goods
production other than construction. Because output and productivity in fi-
nance and services industries are understated, it is important that statistical
agencies devote more attention to improving the measures. Even after al-
lowance for understatement, however, it seems clear that productivity growth
in most services groups, other than communications and trade, has been
significantly below the U.S. business sector average.
Most of the policy options discussed above would help to improve pro-
ductivity in the business economy generally, but since most services indus-
tries are much more labor intensive than the goods sector, priority should
initially be given to measures that improve the quality and efficiency of labor,
including managers. Schools of business should increasingly slant their cur-
OCR for page 116
116
JOHN W. KE!IDRICK
ricula toward the probability that their graduates will pursue careers in ser-
vices establishments, now that manufacturing accounts for only 18 percent
of total employment in the United States. Improvements in the quality of
public education and the reduction of dropout rates are of special value to
services industries. The Job Training Partnership Act and other formal train-
ing programs should stress training for services jobs. Companies in the
services sector should try to increase the amount of training they can offer
employees. More widespread adoption of employee involvement plans and
the use of incentive pay systems could have a significant effect on efficiency
as discussed earlier. In those services industries in which difficulties in
measuring work units would militate against productivity gain sharing, profit-
sharing schemes offer an attractive alternative.
With respect to technological advance, firms in the service industries
should take full advantage of the present opportunities to form research
consortia free of the threat of antitrust prosecution. New university-based
interdisciplinary science and technology centers must give full weight to the
important role of services in contributing to U.S. competitiveness in inter-
national economic relations. Given the fragmented nature of many services
industries, trade associations in these industries should move to sponsoring
more research, development, and engineering activity and should apply for
grants from the NSF and other interested federal agencies.
Economists generally hold that it it not desirable to give special investment
incentives to selected sectors or industries. These tend to distort the allocation
of investment funds based on the best judgments of financial institutions
regarding relative profit prospects. The best stimulus to investment comes
from steady, noninflationary economic expansion in line with the growth of
productive potential of the economy as discussed earlier. The decline of the
dollar since October 19, 1987 should help promote further reductions in the
negative trade balance. If additional stimulus to growth is needed, resumption
of the 10 percent investment tax credit would help promote increases in real
product and productivity across the board. Effective marketing strategies of
capital goods manufacturers involving education of managers especially of
smaller firms in applying the latest technologies, are helpful to services
industries. Beyond this, continuing efforts of the Small Business Adminis-
tration to aid small firms in meeting regulatory requirements, entering export
markets, gaining access to financing on reasonable terms, etc., will be of
special benefit to services industries.
The biggest factor that most services industries have working for them is
something unaffected by policy, i.e., generally high income elasticities and
low price elasticities of demand. This ensures the continued expansion of
the service sector with attendant economies of scale. Relatively strong ex-
pansion provides a favorable environment for managements of services firms
to meet their greatest challenge technological innovation.
OCR for page 117
TECHNOLOGY AND THE SERVICES SECTOR
NOTES
117
For publications and information, write the American Productivity Center, 123 North Post
Oak Lane, Houston, TX 77024.
2. See New York Stock Exchange (1982). A new survey was conducted in 1987 by the
American Productivity Center, Houston.
REFERENCES
Bailey, E. 1986. Price and productivity change following deregulation: The U.S. experience.
Economic Journal (March): 1- 17.
Bosworth, B. P. 1984. Tax Incentives and Economic Growth Washington D.C.: The Brookings
Institution.
Brown, K. M. 1984. The R&D Tax Credit, Issues in Tax Policy and Industrial Innovation.
Washington, D.C.: American Enterprise Institute for Public Policy Research.
Council on Competitiveness. 1987. America's Competitive Crisis: Confronting the New Real-
ity. Washington, D.C.: Council on Competitiveness (April).
Davies, S., and R. E. Caves. 1987. Britain's Productivity Gap. New York: Cambridge Uni-
versity Press.
Denison, E. F. 1985. Trends in American Economic Growth, 1929-1982. Washington D.C.:
The Brookings Institution.
Guile, B. R. and J. B. Quinn. 1988. Managing Innovation: Cases from the Services Industries.
Washington, D.C.: National Academy Press.
Kendrick, J. W. 1979. Productivity trends and the recent slowdown: Historical perspective,
causal factors, and policy options. Pp. 17-69 in Contemporary Economic Problems, Fellner,
ed. Washington, D.C.: American Enterprise Institute for Public Policy Research.
Kendrick, J. W. 1981. International comparisons of recent productivity trends. Pp. 125-170
in Essays in Contemporary Economic Problems, W. Fellner, ed. Washington: American
Enterprise Institute for Public Policy Research.
Kendrick, J.W. 1980. Policies to promote productivity growth. Pp. 45-135 in Agenda for
Business and Higher Education. Washington, D.C.: American Council on Education.
Kendrick, J. W. 1986a. Output, inputs, and productivity in the service industries. Pp. 60-89
in Statistics About Service Industries: Report of a Conference. Washington, D.C.: National
Academy Press.
Kendrick, J. W. 1986b. Differences among industries in productivity growth. Economic Policy
Studies Occasional Paper. Washington: American Enterprise Institute for Public Policy
Research (November).
Kendrick, J. W. 1987. Service sector productivity. Business Economics (April):18-24.
Maddison, A. 1987. Growth and slowdown in advanced capitalist economies. Journal of
Economic Literature 25(2):649-698.
National Academy of Engineering. 1986. Federal Actions for Improving Engineering Research
and Education. Washington, D.C.
National Academy of Engineering. 1988. Focus on the Future: A National Action Plan for
Career-Long Education for Engineers. Washington, D.C.
New York Stock Exchange. 1982. People and Productivity, A Challenge to Corporate America.
New York: New York Stock Exchange Office of Economic Research.
Representative terms from entire chapter:
services industries
