J‘JJJJ” J J J r
J J'" 4- -,
JJJ'4J44J: JJJJJJ JJ'L JJJ LJ44 444”}; 4J4?
JJ J44J44 4:44 44 J_' '3“ 44144 444
JJ44 J" JJJJ4JJ4 4444 444:444144
J J JJJJJ'J JJJ ‘JJJ'JJJ'U J. JJJJJJ‘JJ JJJ'J 4J
J J JJW J

49 JJJ4;'JJ'JJ|" L
4 4 " '4 4 JJJJ'L4J'J'JJJ‘JJL JJ ",JJJ
JJJ JJ‘J JJJJJJJJJ‘ JJJJ‘JJ‘JJJ JJ:
44J4 JJJJJJJ" 44 [1‘4“ JJGJJJJJWJJJJ‘J'JJ J 44 J44
J J“ J’" 4JJ'JJJJJJJ {J4 'JJ ”J

'J' be
‘4'4J." 31'4“; JJJ
JJJJJ4J44 JJJ J4JJ'JI J ' ii
J

JJJ’

4.4.
.z:.fi:rzn.___

 

it:

 

 

J 4 J 'I J
J 4 4 4 4‘ 4' ' J b J J JJJJJJ-J -
. ‘ 4. ‘,,'Jru' JJJJHJJJJJJ JJ'J JJ.JJ'J3JJ."JJJJ‘J'JJ1,IJJJJ‘JJJ'J'JJ}
- -- J". J J“.trJ.,JJJJJ‘JJJJJJJJ'J‘J J ‘JJJ'JJJJ"J'J'
. , ':"J J ’ " J' ‘J‘
.- 4, , 4 4J,444J4I4'444J JJJ J4 24444 J4J4444 44.44344 444J44J1Jq‘4l

J I, ‘J JJ ’424.
J ..J JJJ7 I} m4 JJJJ'JJ JJJJJJ

' ..J J J JJJI'J J JJJ {J "83‘ JJ
. 4J4 JJJJ
44444 4JJJJJ JJJ44J’J J44 4444 JJJ 444P4
‘JJ'J'J'J JJ" ‘ J"

J
IJ J‘ Ju J
J" J J" ‘J
4 JJJJJJJJJ'JJJJJ4JJJJJJJJ444‘J4JJ4J JJ J4JJJJJ
" J' ‘J 'JJ ,, 444-.
'(JJJJJJ'J'JJJ 4* 44(444 4444444464414?
lJJ44J44 44 JJ JJJJ '4JJJ.JJ ,4J4J4
JJJJJ JJJ"." J J J'4J4
JJJ JJJJ .44, 4,444
JJJ , :4.

JJ ‘ J
46944} ,44 J444J I (JJ 4

JJ4
'JJJJ JJJ J ‘444444 444
JJJJJJJJJ" [4444444'4144444J-JJ4JJJJ'JIJJ444 JJ'JJJJJJ' “'14" JJJJJ 44<I|I‘J.4

   
        
          
  
    
 

J4J.IJJ" J'JJJJJJJ J"

  

'4" 'JJJ’J JJLJ'44' "' '5 JJJJJJJ J" v. WJJJJ JJJ J'J'JJJJJ" J'JJJJJJJ, ,JJ'34, ";J,JJJJJ
444 4JJ|J444 4144 $4444 fiJJ JJ 4444'1JJ44IJJ4J4 44444444444 JJ4 4'44‘4'444 4J4J44J4J4a, 44 4444 44444 444J

'J'JJ "J J 'JJJ‘J' J'" JJJ . JJJJJ JJJJJJJJ " |JJJJJJ4 "JJJ'L ' J'J' JJJJJJ" J
4, ' ' J" ,4 JJJJJJJJNJJ J M“ ”J JJJJJ JJJWJJ JJJ'J-f' JJJJJUJJJJ‘ NJ mm "M

 

 

 

 

llIIIIIIII‘IIIIIIIHHIIII IIIIIIIIIIIIIHI L

we“, 3 1293 10607 9720 .1 LIBRARY

Michigan State
University

 

This is to certify that the

thesis entitled

PAST AND FUTURE TECHNOLOGICAL CHANGE IN THE
U.S. FOREST INDUSTRIES

presented by
CHRISTOPHER D. RISBRUDT

has been accepted towards fulfillment
of the requirements for

PA D degree in PM“?

 

Major professor

Date r%y¢7 3212 /777

0-7 639

OVERDUE FINES ARE 25¢ PER DAY
PER ITEM

‘Return to book drop to remove
this checkout from your record.

 

MQU

. "@352 21:35:»

in

 

 

PAST AND FUTURE TECHNOLOGICAL CHANGE IN THE
U.S. FOREST INDUSTRIES

By
Christopher D. Risbrudt

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Forestr

1979

Abstract

PAST AND FUTURE TECHNOLOGICAL CHANGE IN THE
U.S. FOREST INDUSTRIES
by
Christopher D. Risbrudt

This study examines technological change in four
forest industries through several means and discusses
probable future changes. The industries considered were:
logging camps and contractors (SIC 2411); sawmills and
planing mills, general (SIC 2421); pulpmills (SIC 2611);
and papermills, except building paper (SIC 2621).
Qualitative indicators of technological change were
compared, consisting of time series data on consumption,
prices, capital employment, number of establishments, and
several of their ratios. Secondly, five econometric
models were applied to measure technological change
between the years 1958 and 1976. The models used were:
1) arithmetic; 2) geometric; 3) Cobb-Douglas; A) constant
elasticity of substitution; and 5) Johansen. Only the
results of the first four models are reported. Several
of the models also provide information such as capital to
labor price movements, elasticities, and economies of
scale. Finally, descriptions of past and probable future
technologies were covered. Alternative projections of one

of the indexes of technological change (geometric), based

on GNP and time, were made to 2030.
Technological change in the four forest industries
over the 19 years have been modest. Between 1958 and 1976,
the average annual increase in the geometric index was
only 1.9 percent. Results of the four models are as
follows:
Arithmetic Geometric Cobb-Douglas CES

------- average annual increase--—------
Logging camps and

contractors 3.4 3.1 3.8 3.3
Sawmills and planing

mills, general 1.8 1.8 0.9 -_-
Pulpmills 2.5 2.4 2,3 --_

Papermills, except

building paper 0.7 0.M 2.4 ---
These figures represent average annual changes in total
factor productivity, or alternatively, shifts in an
industry's production function.

Output has increased in all four of the industries:
sawmilling the least, 20 percent, and logging the most,
150 percent. Capital also increased, with logging
doubling its investment and sawmilling growing only 13
percent. Employment declined slightly in logging and
papermaking, falling by one-third in sawmilling, and
increasing 10 percent in pulping. Annual productivity
increases, the most common measure of progress, were

above or near the average for all manufacturing. Logging

productivity was far above the national average for all
manufacturing (2.7 percent), at 5 percent, sawmilling
slightly above average at 3 percent, and pulping and
papermaking only slightly below the national average at
2.6 percent. The number of establishments in each of the
industries has been fairly constant in all cases except
sawmilling, which has experienced a decline of almost 50
percent.

Regional shifts in the industries have occurred. In
logging, the South gained while the Pacific Coast lost
establishments. Sawmilling establishments declined in
all regions. In pulping and papermaking, there has been
a general shift to the South.

Research and development in the forest industries is
not great. For both lumber and paper, funds for R&D as a
percent of net sales in manufacturing companies performing
R&D is less than one percent, compared to greater than
three percent for all manufacturing. Universities,
government, and manufacturers of equipment for the
industries do perform R&D that affects the forest
industries, however.

Changes in raw materials have also probably occurred.
Reliable data are not readily available, but logs have
most likely become smaller.

An additional factor may be that the entire forest
products industry has concentrated on advancement in other

areas, such as plywood, particleboard, and fiberboard.

Technological change has accounted for about 60
percent of the increases in per employee productivity
for three of the industries. Additional capital per
employee produced the remainder. Papermaking differed,
with greater capital intensity producing 72 percent of
the growth, with technological change accounting for 28
percent.

In the future, no great deviations from past trends
are expected. New capital expenditures per employee less
than the average for all manufacturing in logging and
sawmilling (21 percent less, and 40 percent less,
respectively, than the $2300 in 1976 for all manufacturing),
should bring their labor productivities down somewhat,
although they will still probably remain above average.
Per employee expenditures in pulping are more than 5 times
the national average for manufacturing, and almost twice
as large in papermaking, yet their labor productivities
will probably not increase greatly, based on capital
requirements and probable future technologies.

Technological change trends are expected to vary
only slightly from the past. Logging, which advanced
technologically 3.1 percent per year, should continue at
about the same rate. Sawmilling and planing, which
progressed at 1.8 percent per year in the past, should
be able to improve slightly on its performance. Pulping,

at 2.h percent in the past, will probably be lower,

perhaps as much as 0.5 percent per year in the long run.
Papermaking, which increased technologically at a very
low 0.4 percent yearly, will probably not be able to

improve at a much greater rate.

ACKNOWLEDGEMENTS

The impetus for this study developed during my work
with Dr. Robert S. Manthy on his book, ”Natural Resource
Commodities - A Century of Statistics.” The trends
presented in that book led me to investigate the role
technological change plays in the extraction of natural
resources from the environment over time. To Dr. Manthy
go my thanks for whetting my interest in this subject.

The U.S. Forest Service funded my project while I
was in graduate school. The Forest Products Laboratory,
Madison, Wisconsin, graciously allowed me to complete this
dissertation while I was in their employ. I am also
grateful to Dr. Robert N. Stone, my project leader, who
provided valuable guidance, and who also insisted on
completion.

I thank Mrs. Norma Jones for her skill in typing the
rough draft.

Finally, my complete admiration goes to my wife, Sue
for so bravely attacking the horrendous task of typing
the second draft and final copy. To her, and our two
children, Christine and Jill, goes my thanks for their
love, patience, and encouragement through the difficult

years involved in this endeavor.

TABLE OF CONTENTS
Egg;
LIST OF TABLES.................................... vi
LIST OF FIGURES................................... 1X
I. INTRODUCTION................................
The problem.............................
Objective of the study..................
Framework for analysis..................

scope Of the studYOOIIOOOOO0.0.0.000...O

OWWUHH

Definitions and relationships...........

II. METHODS OF INDICATING AND MEASURING TECHNO-
LOGICAL CHANGE............................
Qualitative indicators....................

Consumption and price...................

\O\OCI)CD

Employment, capital, and output.........
Per employee data....................... 10
Utilization rates....................... 11
Establishment data...................... 12
Models.................................... 12
Arithmetic.............................. 13
Geometric............................... 16
Cobb-Douglas............................ 19
Constant elasticity of substitution..... 21
Johansen index.......................... 23
III. EVIDENCE OF PAST TRENDS..................... 26
Logging camps and contractors........... 28

iii

IV.

VI.

VII.

VIII.

Sawmills and planing mills, general.......
Pulpmills.................................
Paper mills, except building paper........
A comparison..............................
Other studies.............................
PROJECTIONS AND PROBABLE TECHNOLOGIES.........
Factors influencing change..................
Timber harvesting.........................
Sawmilling................................
Pulping...................................
Papermaking...............................
The environment for technological change....
Size class distribution...................
Region of operation.......................
Research and development..................
Efficiencies..............................
Prices....................................
Resources.................................
SUMMARY AND CONCLUSIONS.......................
APPENDIX A - GEOMETRIC INDEX UTILIZATION
RATES.......................................
APPENDIX B - ARITHMETIC INDEX WEIGHT
CALCULATIONS.................... ..... ’.......
APPENDIX C - ARITHMETIC INDEXES 0F
TECHNOLOGICAL CHANGE ...................... ..

iv

101
105
108
108
111
116
117
120
123
127

136

139

148

Page
IX. APPENDIX D - STRICT COBB-DOUGLAS PRODUCTION

FIJ-NCTIONSIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII 156
X. CITED LITERATURE.......... ....... ............. 157
XI I ADDITIONAL REFERENCES I I I I I I I I I I I I I I I I I I I I I I I I I 161+

10.

11.

12.

LIST OF TABLES

Some indicators of technological change in
logging camps and contractors................

Number of establishments and per establishment
data for logging camps and contractors.......

Geometric index of technological change in
induStrySIC24110000000000.0.000000000000000

Annual growth rates in productivity, 1958-
1967, and 1968-1976 for logging camps and
contraCtorSo00000000000000.0000...00000000000

Three-factor Cobb-Douglas function

(Y = AKbLCTd) for logging camps and

contractorSIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII

Some indicators of technological change in
sawmills and planing mills, general..........

Number of establishments and per establishment
data for sawmills and planing mills, general.

Geometric index of technological change in
induStrySIC2421.0IIIIIIIIIIIIIIIIIIIIIIIIII

Annual growth rates in productivity, 1958—
1967, and 1968—1976 for sawmills and planing

millSIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII

Three-factor Cobb-Douglas function

(Y = AKDLCTd) for sawmills and planing mills,
generalIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII

Pulpwood consumption per ton of pulp produced,
by pI‘OCeSS, 1920-19760coo-cooocuooocooooooono

Some indicators of technological change in

pulpmillSIIIIIIIIIIIIIIIIIIIIIIIIII

vi

3%
36

38

42

50
52
53

56

59

Table
13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

Number of establishments and per establishment
data for pulpmillSIIIIIIIIIIIIIIIIIIIIIIIIIII

Geometric index of technological change in
induStrySIC 2611IIIIIIIIIIIIIIIIIIIIIIIIIIII

Annual growth rates in productivity, 1958-
1967, and 1968-1976 for pulpmills. c o I o o o o o o I 0

Three-factor Cobb-Douglas function

(Y: AKbLCTd) for pulpmillSIIIIIIIIIIIIIIIIII
Some indicators of technological change in
paper mills, except building paper...........

Number of establishments and per establishment
data for paper mills, except building paper..

Geometric index of technological change in
induStrYSIC 2621.IIIIIIIII-IIIIIIIIIIIIIIIIII

Annual growth rates in productivity, 1958-
1967, and 1968-1976 in paper mills...........

Three-factor Cobb-Douglas function

(Y‘= AKbLCTd) for paper mills, except
building paperIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII

Annual increase in technological change, by
industry and method of measurement...........

Size class distribution of four forest
industries in the United States, 1972........

Establishments, employment, and value added
by region for logging camps and contractors,
1972'1958IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII

Establishments, employment, and value added
by region for sawmills and planing mills,
1972'1958..IIIIIIIIIIIIIIIIIIIIIIIIIIIIII...

Establishments, employment, and value added
by region for pulpmills, 1972, 1958..........

Establishments, employment, and value added

by region for paper mills, except building
paper, 1972, 19580.00000000000000000000000000

vii

65

71

72

73

76

76

79

109

112

Table
28.

290

30.

A1.

A2.

A3.

B1.

B2.

B3.

B4.

Cl.

02.

CB.

C4.

Funds for research and development performed
by the forest industries, 1960-1975..........

Value added and capital expenditures in the
lumber and wood products industry: Ratios

of ”most

efficient" to "least efficient"

plants and to average plant, 1967............

Wholesale price indexes of selected timber
PI‘OdUCtS, 1950-197600000000000000000000000...

Percent capacity utilized, lumber production,
1958-1976, by quarteroo00000000000000.0000...

Percent capacity utilized, woodpulp
production, 1958-1976, by quarter............

Percent capacity utilized, paper production,
1958-1976, byquarter-000000000coco-00000000o

Arithmetic

index weight calculations, industry

SIC 2411IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII

Arithmetic

index weight calculations, industry

SIC 2421IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII

Arithmetic

index weight calculations, industry

SIC 26110IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII

Arithmetic

index weight calculations, industry

SIC 2621IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII

Arithmetic
industry

Arithmetic
industry

Arithmetic
industry

Arithmetic
industry

index of technological change in
SIC 2411IIIIIIIIIIIIIIIIIIIIIIIIIIII

index of technological change in
SIC 2421IIIIIIIIIIIIIIIIIIIIIIIIIIII

index of technological change in
SIC 2611IIIIIIIIIIIIIIIIIIIIIIIIIIII

index of technological change in
SIC 2621IIIIIIIIIIIIIIIIIIIIIIIIIIII

119

121

136

137

138

140

142

144

146

148

150

152

154

Figgre

LIST OF FIGURES

1. Cubic feet per thousand board feet

International % scale: softwood sawlogs,
1952-1976.I.IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII

2. Cubic feet per thousand board feet

International % scale: hardwood sawlogs,
1952—1976IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII

3. Cubic feet per thousand board feet

4.

International & scale: sawlogs, U.S.
tOtals, 1952-197600000noouocooooooooooooooooo

Input and output rates of growth for
industrial roundwood.........................

ix

47

48

84

INTRODUCTION

Over time, changes in the technologies employed by
industry affect our need for raw materials. In the forest
industries, alteration in production processes and products
produced will influence the quantities of timber cut from
our Nation's forest lands. To prepare for our future
timber requirements, then, it is important to evaluate
technological change in the forest industries.

The problem

In general, economic theory includes as the main
factors of production land, labor, and capital. Tech—
nology has not received equal treatment by economists in
developing theory. This force which controls the manner
and combination of the other factors of production has
too often been assumed away under the characterization
”ceteris paribus.”

Economists have been becoming aware that exclusion of
technology as a factor of production is resulting in
discrepancies between theory and actual fact. The theory
has clearly become deficient for many of the applications
to which it has become necessary to extend it.

This has occured because technology is a difficult
factor to include in theory. Also, technologically

1

2

induced change is becoming more and more prevalent today.
Advances in electronics, chemistry, metallurgy, and other
sciences are producing continuing changes in our society.
In fact, change has become engineered into our system.

Business and government have become more aware of the
relationship between technology and productivity.
Increased productivity is viewed as the solution to
improving resource use, reducing inflationary pressures,
and increasing our standard of living.l- Hence, technol—
ogical change in the manufacturing sector is regarded as an
important issue.

In 1974, Congress mandated in the Resources Planning
Act that the Forest Service periodically assess the
Nation's need for renewable resources, and our ability to
meet them. One of our major resources is timber; the
United States consumed 13 billion cubic feet of timber
products in 1976.g

Timber demand is expected to rise substantially in
the future, to 27.8 billion cubic feet of roundwood

consumption in 2020.2 In assessing this future, it is

 

LGilbert P. Dempsey. 1973. Toward growth in productivity.
Forest Products Journal 23(4):12-14.

gRobert Phelps. 1977. The demand and price situation for
forest products 1976-77. Forest Service Misc. Pub. No.
1357. Washington, D.C.: U.S. Government Printing
Office.

2U.S. Forest Service. 1977. The nation's renewable
resources--an assessment, 1975. Forest Resource Rep. No.
21, Washington, D.C.: U.S. Government Printing Office.

3

imperative to consider the role changing technology will
play in determining the quantities of roundwood that will
be necessary to satisfy the demand for final wood products.
For example, due to changing technology, in the future we
can expect that more board feet of lumber will be obtained
from each cubic foot of sawlog than was possible in the
past. Improvements in pulping will extract a higher
percentage of wood fiber from a cord of roundwood.
Objective of the Study

The objective of this study is to project historical
patterns of technological change in the U.S. forest products
industries based upon an evaluation of past performance and
likely future developments. The projections will cover the
years 1990-2030. Its purpose is to build a basis for the
U.S. Forest Service to better evaluate the effects of
technological change on the Nation's wood requirements.
Framework for Analysig

Technological change can be variously measured,
evaluated, and indicated by several approaches. First,
there is the consideration of the changes in processes
in manufacturing. This is simply a descriptive process.
Second, there are the economic trends that evidence
technological change. These are the more qualitative
indicators such as time trends in prices, consumption,
employment, capital and capital per employee, output and

output per employee, etc. These trends are evaluated by

L1,

their changes through time, and by comparison with other
time series. Finally, there is the measurement of
technological change through the use of mathematical
models.

A major problem in attempting to analyze and evaluate
technological change is the fact that there exist no
suitable units in which it can be measured. The approach
most often employed by economists has been to construct an
index of technological change, which is usually based on
a model involving value added or gross output in
manufacturing, and employed capital and labor. This
method allows a quantification of the trend of technological
change, without identifying an absolute level. One year
is arbitrarily picked as the base, and an index or other
measure constructed from that point. Such an approach
will be utilized in this study; five models employing
different configurations of the variables and different
assumptions are used to measure technological change in
the forest industries. The models are the: (1) arithmetic
index, (2) geometric index, (3) Cobb-Douglas function,

(4) constant elasticity function, and (5) Johansen index.

The purpose in employing more than one model is to
compare their results, due to their differing constraints
and assumptions, and not to compare the models directly.

The differences and similarities of the models will be

5
compared only in terms of the measures of technological
change which they produce, in an effort to understand the
causes and consequences of increased productivity.

The geometric model will be used for two projections,
one based on time, the other on GNP. By far the largest
portion of the literature on technological change is devo-
ted to problems of measurement rather than explanation.E
To avoid this limitation, all three of the above approaches
will be utilized in this study.

Scope of the Study

Because of the vast number and diversity of firms and
plants comprising the forest industries, the study must be
focused on only a few aggregate industries to remain
manageable. The majority of data utilized in this study
are from the various Census of Manufactures, and the
industries included conform to the Standard Industrial
Classification scheme. They are logging camps and
contractors (SIC 2411), sawmills and planing mills,
general (SIC 2421), pulpmills (SIC 2611), and papermills,
except building paper (SIC 2621). The time frame covered
is the period 1958-1976, although some data series start
in 1950.

These industries comprise major segments of the U.S.

economy. Logging and sawmilling employed 1.3 percent of

 

ED.W. Jorgenson and Z. Griliches. 1967. The explanation

of productivity change. Review of Economic Studies

34:249-283.

6

the employees, paid 1.0 percent of the payroll, and
produced 1.1 percent of the value added in the U.S. in
1972. The importance of pulpmills and papermills is only
slightly smaller, with this industry responsible for 0.7
percent of employment, 0.9 percent of payroll, and 0.9
percent of value added in 1972.
Definitions and Relationships

Technology and technological change suffer from
many definitions and interpretations. Technology as used
in this study is defined as the inputs required to produce
a given output.5 In economic terms, technology determines
the shape of the production function. In this sense, the
state of technology is the social pool of knowledge of the
industrial arts available at a given point in time.é
It defines the complete set of possibilities society
possesses to produce goods and services. Technological
change involves the creation of a new set (which includes
the old one) of production alternatives.2

The actual technologies employed at any point in time

 

iNational Science Foundation. 1976. Technological
innovation and Federal government policy. Office of
National R&D Assessment. NSF 76-9.

éEdwin Mansfield. 1971. Technological change: An
introduction to a vital area of modern economics. New
York: W.W. Norton & Co.

ZIrwin Feller. 1972. Production isoquants and the
analysis of technological and technical change.
Quarterly Journal of Economics 86:154-161.

7
are techniques. When the method of producing a specific
output is altered, it must be defined as a change in
technique. Technical change is thus defined as a change
in production method out of the existing set of
alternatives. Strictly speaking then, the changes in
production functions examined in this study are due to
changes in technique. However, because the improvements
included in these measures are all-encompassing, including
better health of workers, improved managerial skills,
changes in the quality of inputs, and greater capital
accumulations, we shall allow these events to reside under
the broader, more general term of changes in technology.

Conversion measures can be applied to all levels of
the production process, from harvesting in the forest
stand to final use such as lumber in housing. Along
each step in the process, the raw material is transformed
into a product, which in turn becomes the raw material
for the next step.

Advances in technology can be expected to improve
utilization, the result of techniques that allow more
output from a given input. Changes may also result in
additional knowledge about techniques that allow a wider
range of factor substitution.§ Changes in technology can
also create new resources, by utilizing materials that

were not previously acceptable.

 

éNathan Rosenburg. 1976. Perspectives on Technology.
New York: Cambridge University Press.

II. METHODS OF INDICATING AND MEASURING
TECHNOLOGICAL CHANGE

There are several methods of indicating or measuring
technological change. Besides an historical examination,
economic data series can be useful in evaluating progress
in an industry. Also, econometric models exist that allow
a measurement of technological change. The basis for
these methods will be discussed in the following section
with the economic data series or qualitative indicators

first, followed by the econometric models.

Qualitative Indicators

The real price trend for the output of an industry
can partially reflect the changes in technology that have
occurred. Certainly, price is the result of the
interaction of demand and supply: however, if in the face
of expanding consumption, the real price remains constant,
then we may be sure that changes are occurring that affect
supply. These changes may be structural, political,
economic, or technological. Indications of the first
three factors will aid in evaluating possible changes in
the latter cause. Additionally, direct evaluation of new

technologies occurring in an industry may be made.

Consumption and Price

Consumption and price are the only data observable
at any point in time from the interaction of economic
demand and supply. Series of these two products of
market elements, taken together, give an indication of
those market forces.

A recently completed study of raw materials in the
United States has reported that virtually all natural
resource commodities have experienced horizontal or
declining real price trends, in spite of a fifteenfold
increase in consumption.2

With expanding populations, demand can be expected
to shift outward. If no shifts occur in an inelastic
supply, prices will rise. However, if with increasing
consumption, price does not rise, there is an indication
of a shifting supply curve. The reasons for the shifts
may then be investigated for indications of technical
change.

Employmenthapitaliand Output

Labor and capital are the two basic factors of
production in any productive process. Time series of the
quantities used by an industry give evidence of changes in

the production of output. Certainly, given an adjustable

 

2Robert Manthy. 1978. Natural resource commodities--a
century of statistics. Baltimore: John Hopkins
University Press, published for Resources for the
Future.

10
productive process, changes will be made in the amounts of
capital and labor used in response to their changing
relative prices, without any differences in the level of
technology employed. Price series for the two factors are
also necessary. However, while the adaptability of the
technological productive process employed varies between
industries, in few, if any, is it perfectly adjustable.
There are imperfections in any firm or industry, compared
to its theoretical counterpart. This fact will lessen the
importance of changing relative prices versus that of new
technology: for resources employed, adjustments in the
capital/labor ratio will be made during periods of capital
growth or replacement in the industry.

The comparison of the three time series of capital,
labor, and output will provide indications of change in
the underlying economic factors involved in the productive
process.

Per Employee Data

The partial productivity measure of output per
employee is one of the most commonly used indicators of
progress in an industry. It is one of the measures used by
the Bureau of Labor Statistics to trace changes in

technology.Lg .It is a partial measure, however, because

 

AQU.S. Department of Labor. Bureau of Labor Statistics.

1974. Technological change and manpower trends in six
industries. Bulletin 1817. Washington, D.C.: U.S.
Government Printing Office.

11
it excludes changes in the other factors of production.
Such exclusion forces the assumption that all changes
in production are the result of changes in the quality of
labor. It is not a measure of efficiency, however, for a
high labor productivity can be produced as inefficiently
as a low one.;;' Such measures need to be related to the
process of change of which they are a product. It remains
a useful measure, however, if one is mindful of its
shortcomings.

Capital per employee data provide an indication of
changes in the relative amounts of the factors of input
used. It is an indication of the capital intensity of the
industry. Taken into consideration, it allows more
realistic interpretation of output per employee data.
Epilizatipn Rates

Utilization rates give the amount of resource that
is converted into a product. For this reason, they are
also sometimes called conversion factors. Examples
would be the board feet of lumber produced per cubic feet
of sawlog, or the cords of roundwood required to produce
a ton of pulp. Direct comparison of utilization rates
through time gives a reliable and valid measure of

technological change, ceteris paribus.

 

LiW.E.G. Salter. 1969. Productivity and Technical Change.

Cambridge: The University Press. 2nd edition.

12

Eptablishment_§ata

A trend toward increasing size has tended to reduce
the number of firms in American industries. Technological
changes based on economies of scale have favored the trend
toward fewer firms, lessening competition. Conversely,
large firms may need to compete nationwide, and do not
enjoy the benefits of local market monopolies.

If a technological trend favoring economies of scale
reduces the number of establishments in an industry,
ceteris paribus, then the time trend of this number will
indicate the trend of technological change. Of course,
there are many factors that can favor or retard such a
tendency. Unregulated competition, risk taking, individual
firm research and development, and many other factors favor
growth of some individual firms at the expense of others.
Government regulation, transportation limits, resource
characteristics, and other factors can limit firm size.

The amount of capital and number of persons employed
per firm, over time, can also indicate technological
change, in addition to other economic influence.

Models

Of the mathematical models available, five were

considered earlier.£§- Because of the difficulty of

describing an absolute level of technology or techniques

 

lgThe following section is based in part on Lester B. Lave.

1970. Technological Change: Its Conception and
Measurement. Englewood Cliffs, N.J.: Prentice-Hall,
Inc.

13

in use, it is necessary to resort to describing its
effects.l2' For this reason, all models use a base year
and construct an index of measure from that reference
point.
Arithmetic

The arithmetic index is a simple and direct measure
of technological change. Kendrick used this index to
develop time trends of total factor productivity for the
national economy and its major sectors, including
manufacturing, transportation, mining, agriculture, and
government.l&

The index is based on the equation

Y0 = WOL + 10K (1)
where W0 is the average wage at time zero, L and K are

labor and capital, respectively, and i is the average

0
return to capital at time zero. Calculations for the
weights W0 and i0 are shown in Appendix B.

Implicit in national incoming accounting, equation
(1) states that the output of an industry is the sum of
the products of rent to capital plus wages to labor. The
index is constructed by evaluation of the percentage

changes in the components between periods. CO is

introduced to account for changes in efficiency.

 

llEdwin Mansfield. 1968. Industrial Research and
Technological Innovation. New York: W.W. Norton and
Co., Inc.

LiJohn W. Kendrick. 1961. Productivity Trends in the
United States. Princeton, N.J.: Princeton University
Press.

Y L K
1- _;_ ._1_
Yo-Co (woL +1on) (2)

The series for Co, 01,..., Cn is a measure of
technological change. In effect, equation (2) shows the
percentage growth in output as a weighted sum of the
percentage increases in labor and capital. The weights
have the effect of producing what the output of period 1
would have been using the productive efficiency of period
0. Since the output in period 1 is generally greater, Ci
is a measure of the increase in productivity.

However, this productivity increase, Ci, is not
composed entirely of technological change. Increases may
also result from changes in scale of output, changes in
utilization of capacity, and changes in the quality of
inputs and outputs.

The changing weights used in this study are an
attempt to compensate for the effect of changing relative
prices of the two inputs, capital and labor. For any
period, the estimate of Ci employs the weights of the
relative prices of the preceding period, i.e., for C3
the weights are the relative prices in period 2 (W2 and i2).
This method necessarily makes the assumption that the
relative prices are indicative of the marginal
productivities of the inputs. Such a condition is only
attained under perfect competition in an industry.

A second and rather important implicit assumption in

this method is that the weights, W and i, are changed only

15

by technological change. This is clearly at odds with
neo-classical theory, which holds that the equilibrium
between marginal productivities and prices will be
disturbed only through demand and price shifts. Changes
in the demand or supply of a factor will change the
price, and consumers will adjust their use of the factor
until the marginal productivity again equals the price.
Such a condition must hold for the long—run equilibrium.
In the short run, however, the assumption that the
weights are changed through technological change may not
be unreasonable. To allow for the long-run adjustments,
the weights are recalculated yearly.

An example will serve to illustrate how the
arithmetic index is calculated. Equation (3) is solved

for Co:
Yl/YO

1 . K1
“0%) " 10%;) ‘3)

To calculate the arithmetic index for logging camps

CO =

 

and contractors, for the year 1961 (03), the data are
taken from Table C1 in Appendix C, and the weights are
taken from Table B1 in Appendix B (an example of how the

weights are calculated is included in Appendix B):
455i9é425.9 I 42
Z . 4 .6
“5881(73.1I + “4119 (539.5)

1.082

‘33

16

Each yearly calculation produces a figure somewhere
near the value one: it is a year-to-year measure and is
not cumulative. The indexes presented in Appendix Tables
C1 through C4 have been converted to cumulative indexes,
however, to make them more comparable to the geometric
index which is described below.

Although they are constructed differently, the
arithmetic and geometric indexes yield almost identical
results. For this reason, and others given in the
next section, only the results of the geometric index
will be discussed for the four forest industries. The
results of the arithmetic index calculations are
presented in Appendix C.

Geometric

The geometric method constructs a cumulative index
of technological change using increases in output
unexplained by increases in capital per employee. The
necessary assumptions for this method include constant
returns to scale, a perfectly competitive economy,
capital and labor are paid their marginal products, and
that technological change is neutral (marginal rates of

substitution remain the same).

17
This method, developed by Solow.!hi makes use of the

following equation:

AA-_QQ._W _A_K.-w Ala.
A-Q kK 1L (4)

where Q, K, and L are output, capital, and labor, and Wk
and W1 are the elasticities of output with respect to
capital and labor. Gross capital is used, rather than
net of depreciation. Annual depreciation and annual
replacement are not necessarily the same, and the use of
gross estimates circumvents that accounting problem, but
makes necessary the assumption that assets are replaced
fully when they are retired.l§

The equation can also be written in a ”per labor"

form:

_£;A.— AAX _W ARE .
A - y k k (5)

where y and k are Q/L and K/L, respectively. This
equation has the straight-forward interpretation that
technological change is responsible for any changes in
output unaccounted for by variations in the amount of
capital per employee.

In this model, factor changes are combined
geometrically (weighted by elasticities of output with

respect to each factor) rather than arithmetically

 

iiRobert M. Solow. 1957. Technical change and the
aggregate production function. Review of Economics
and Statistics 39:323-333.

lQWarren Hogan. 1958. Technical progress and production
functions. Review of Economics and Statistics 40:407-

413.

18
(weighted by prices). The geometric name is derived
from the weights, which, if the function is defined
as Q=f(K,L), means the elasticity of output with respect
to labor is the partial derivative (éQ/ZL)(L/Q). Another
difference from the arithmetic model is that capital is
adjusted for the capacity utilized, as an attempt to
include the flow of capital, rather than the stock.

The capital utilization rate is calculated according
to the Wharton School Econometrics Unit, as discussed by
Phillips.£z This method adjusts for idle capacity by
charting quarterly output data and selecting peaks by
inspection. Each peak was defined as 100 percent
capacity, and a straight line from peak to peak was used
to describe the capacity utilized in the intervening
years. The average annual capacity utilized was then
multiplied by the capital series to obtain an estimate
for capital in use. The capacity utilization rates are
included in Appendix A.

An example will show how the geometric index is

calculated. Equation (6) is multiplied by A to yield:

AA "' A (9:71 "Wk iii) (6)

The index for logging camps and contractors, for

1959, is calculated from the data in Table 3. The

 

izAlmarin Phillips. 1963. An appraisal of measures of
capacity. American Economic Review 53(2):275—292.

l9
residual for that year results from the two years

between 1958 and 1959:

“’50 = WI " 5‘ 2*“ )--“16(56“253225I

Since the index (A) for 1959 was arbitrarily set at one,

 

each residual is then added to the previous year's index
to produce the cumulative result (1.000 - .050 = .950, the
index for 1959: the residual for 1960 is .077, which
yields the index .950 + .077 = 1.027).

Capital's share in income, Wk, is calculated by
subtracting the payroll from value added, and dividing
the result by value added. The weight for calculating
the residual for the change between 1958-59 comes from
data for 1958: (390.4-288.0)/390.4 = .416. This data is
from Appendix Tables B1-B4.

A similarity in the two models is the fact that the
indexes of technological change are calculated as
residuals. This practice led Abramovitz to declare
them "some sort of measure of our ignorance."l§
Cobb-Douglas

The Cobb-Douglas has been one of the most popular
aggregate production functions ever since its introduction

in 1928.£2 Its success has no doubt been

 

L8“Moses Abramovitz. 1956. Resource and output trends

in the United States since 1870. American Economic
Review 46(2):5-23.

i20.E. Ferguson. 1969. Microeconomic theory. Homewood,
Ill. Richard D. Darwin, Inc.

20
due to several factors: its ease of explanation, its
plausible form (constant return to scale, diminishing
returns to a factor), and its ease of estimation by
standard regression techniques.

The Cobb-Douglas production function is given by
the form:

Q = A KB Ll'B (7)

This function is homogeneous of degree one and has
an elasticity of substitution of unity, since the
exponents sum to one. A less restrictive form of this
function is to remove the constraint that the exponents
sum to unity:

Q = AKbLC (8)
While not strictly a Cobb-Douglas, fitting such a
function by regression will provide an estimate of the
returns to scale for the industry, as indicated by the
sum of the exponents. A sum greater than one indicates
economies of scale: less than one, diseconomies of scale.-2--g
If we assume competition in factor markets and that
entrepreneurs minimize costs, then the ratio of the
price of capital, PK, to the wage rate, w, will be the
same as the ratio of the marginal product of capital to

the marginal product of labor:g$

 

ggMurray Brown and John S. de Cani. 1962. Technological
changes in the United States, 1950-1960. Productivity
Measurement Review 29:26-39.

allbid.

21
35= 79-; (9)
w cK
From this relation we see that an increase in b relative
to c implies a labor-saving technological change, and a
decrease in b relative to c implies a labor-using
technological change.

A further modification of the classical Cobb-Douglas
function also allows an estimate of the trend in
technological change. Addition of a time variable will
indicate the rate of change:

Q = AKbLCtd (10)
This form of the Cobb-Douglas function has been used to
identify "technological epochs,” during which there was
no nonneutral technological change.g§-

Constant Elasticity of Substitution

The constant elasticity of substitution (CES)
function is more general, and it includes the Cobb-
Douglas as a special case. It was developed because the
assumption of unitary elasticity of substitution was felt
to be unduly restrictive.g2' The function is expressed as:

v = A( dK’P + (1 - d)L"p)-1/p (11a)
where V is value added, K and L are capital and labor, A

is a (neutral) efficiency parameter, p is a substitution

 

ggMurray Brown and Joel Popkin. 1962. A measure of
technological change and returns to scale. The Review
of Economics and Statistics 44:402-411.

g2K.J. Arrow, H.B. Chenery, B.S. Minhaus, and R.M. Solow.
1961. Capital-labor substitution and economic '

eificiency. Review of Economics and Statistics 43:255-
2 8.

22
parameter (1/(1 + p) = s, the elasticity of substitution),
and d is a distribution parameter.
The addition of a time parameter allows estimation
of technical change:
v = 11(1)t IdK“P + (1 - a) L‘PI ‘1/P (11b)
The CBS function was derived from an observed relationship

between wages and labor productivity:
log(%) = log a + b log W + e (12)

The coefficient b is taken as an estimate of the
elasticity of substitution.

The function is developed by showing the reverse of
the implications of the standard theory of production:
that a particular relation between value added per unit
labor and wages determines the production function.

The original method for estimating this equation
involved a piecemeal process, calculating one or two
coefficients at a time. To avoid the simultaneous
equation bias this method introduces, the entire
equation (11b), save one coefficient, was estimated at
one time using nonlinear regression. The elasticity of
substitution (and therefore p) was estimated separately,
using equation (12).

It is necessary to make the usual assumptions for
this model to hold: those of constant returns to scale
and competitive labor markets. One additional

assumption is that the prices of products and material

23
inputs do not vary systematically with the wage level. A
weak test of this assumption was performed by Arrow,
pp. gl., and the assumption was not rejected.§&' This test
was not performed for the data used in this study.

The generality and adaptability of the function has
been demonstrated by its use to measure the influence of
technological change on employment,§i and to study the
distribution of income between capital and labor.g§
Johansen Index

The reliability of the results from any mathematical
model depend greatly on the quality of the data available.
The data series principly employed in all of the above
models are value added, labor, and capital. Of these
three, the capital series is of the poorest quality.
Accounting problems involving use of historical or
replacement cost, depreciation methods, and taxation
elements make estimation of capital series difficult.

Johansen derived a model which does not use a capital

series in estimating technological change.g2 Its aim is

 

gflIbid.

g5Murray Brown and John S. de Cani. 1963. A measure of
technological employment. Review of Economics and
Statistics 45:386-394.

géMurray Brown and John S. de Cani. 1963. Technological
change and the distribution of income. International
Economic Review 4(3):289-309.

g2"Leif Johansen. 1961. A method for separating the effects
of capital accumulation and shifts in production
functions upon growth in labor productivity. Economic
Journal 71:775-782.

24
to avoid use of the poor quality data series.

The equation is:
a
log(§f)= (log w) B + e (13)

where a is labor productivity, B is capital's share in
value added, w is the relative increase in wages between
periods 1 and 2, and e is the log of the average increase
in productivity. A necessary assumption to perform the
computation by regression techniques is that e is not
correlated with B.

The model is derived from a Cobb-Douglas formulation,
on the assumptions that firms minimize costs, and that
relative shares of capital and labor are constant, with
the result that the proportional increase in total wages
paid between two periods is equal to the proportional
change in total return to capital:

W L R K

WEE-12- = {If (14)
where W is wages, L is labor, R is rate of return to
capital, and K is capital. If managers consider W and R
as given, then to minimize costs they make adjustments in

the capital to labor ratio. This leads to

K1/L1 _ R2731 _ W (15)

where w is taken to be the relative increase in wages.

25

The above mathematical models are all dependent upon
simplifying assumptions. necessary to handle an otherwise
unmanageably complex reality, or to release the analysis
from otherwise crippling lacks of information. The
differing assumptions and forms of modeling can produce
striking differences in results. Further, the models
generally produce only one type of information: the time
trend of technological advance. This time trend is
usually calculated as the "residual" increase in value
added from one period to another, unexplained by increases
in capital or labor. This residual time trend, while
useful for increasing our knowledge of the changes in an
industry, needs to be corroborated with additional
indicators of technological change to be properly

interpreted.

III. EVIDENCE OF PAST TRENDS

The trends of technological change in the U.S. forest
industries will be presented first by a brief history
of the major machinery and process changes that have
occurred, followed by the information that can be extracted
from the series on prices, production, capital, employment,
etc. Lastly, four models will be employed to measure
technological change. These various types of evidence of
technological change will be presented for four forest
industries, classified by the Census of Manufactures as
(a) logging camps and contractors (Standard Industrial
Classification 2411), (b) sawmills and planing mills,
general (SIC 2421), (c) pulpmills (SIC 2611), (d) paper-
mills, except building paper (SIC 2621).

No correction is made in this study for improvements
in the quality of labor. Improved health and education
presumably increase the productivity of workers, and this
will bias the quantitative measures of technological change
upward. For example, Denison estimated that 23 percent of
the growth rate in the United States from 1929 to 1957 was
due to education. He further projected the contribution of
education to the growth rate for 1960-1980 would be 19

percent.§§ To the extent that better education is not

 

géEdward F. Denison. 1962. The Sources of Economic Growth

in the United States and the Alternatives Before Us.
Committee for Economic Development, Supplementary Paper
No. 13.

26

27
reflected in the price of labor, the measures will over-
state technological change.

A second factor affecting productivity and hence of
concern in estimating technological change is management.
Improvements in management are presumed to have accumulated
over time. In a recent survey of chief executive and
industrial relations offices, 65 percent of the respondents
rated ”more effective management” as very important in
improving productivity. The percentages for other factors
rated as very important in influencing productivity were:
Capital investment, 27 percent: improved technology, 35
percent: and human relations, 36 percent.g2 Productivity
in the survey was defined in a broad sense of overall
efficiency, effectiveness, and performance of the organiza-
tion. The importance attached to management in the survey
is taken to be an indication of both past progress in this
area, and of potential for future improvement. However,
due to the difficulty of quantifying the effects of
management on productivity, this factor is not explicitly
considered in the quantitative measures.

A third concern is factor quality. Changes in the
quality of logs are not corrected for in the indexes:
however, some indication will be given through the use of

supplemental data.

 

gBMildred Katzell. 1975. Productivity: The measure and
the myth. American Management Association survey
report.

28
L9gginngamps and Contractors (SIC 2411)

Firms in this industry are primarily engaged in
cutting timber and in producing rough, round, hewn, or
riven primary forest or wood raw materials.29

Logging and woods operations connected with pulpmills,
sawmills, etc., and not separately reported are included in
these other industries.

Examples of technological change in this industry are
mainly of new machinery. A major productivity increase
occurred with the adoption of the power chainsaw. This
development, adopted in the 1940's and 1950's, will
probably have little effect on the indexes.

Another method of cutting developed soon after the
power saw: the hydraulic shear. This means of cutting
results in lower stumps, although it is generally limited
to smaller sized timber (usually less than 24 inches,
although one model can handle trees up to 36 inches).

One problem involved with shears is the compressing and
tearing of fibers up to 10 inches from the cut. The major
advantage of this method, of course, is its speed. Timber
up to 14 inches is cut in 3 to 6 seconds, where a manually
operated power saw takes 30 to 90 seconds.ll

Other new machines have been developed to perform a

 

lgExecutive Office of the President, Office of Management
and Budget. 1972. Standard Industrial Classification
Manual. GPO. Washington, D.C.

ZlSteve Conway. 1973. Timber cutting practices. San
Francisco: Miller Freeman Publications, Inc.

29
greater number of operations. There are feller-bunchers,
which raise productivity by reducing the amount of time
necessary to choke a full load for skidding. A further
development is the Buschcombine, which fells, limbs, bucks,
and forwards the load to a trailer. These types of
machinery find limited use in the Pacific Northwest and
Rocky Mountains. The terrain and timber size limit their
use to the South, Lake States, and Northeast.

Wheeled skidders are possibly the second-most major
development after the chain saw. These machines move with
twice the speed of crawler tractors, although their loads
are generally smaller. New models of these skidders are
being adapted to operate on rougher and wetter terrains
than their predecessors.2§

A new method of logging, which also has entailed the
development of suitable equipment, is whole-tree harvesting.
Its advantages are that there are fewer pieces to skid for
a given amount of wood, and any bucking can be done by
machine in a timber yard. Opportunities for bucking for
highest value are improved.

A related development to whole-tree harvesting is
chipping in the woods. This allows more complete
utilization of the total material in the stand.22

Development of a bark/chip separator is necessary for

 

lgForest Industries. 1967. Logging handbook No. 4. San
Francisco, Calif. Miller Freeman Publications, Inc.

33mm .

30
greater use of this method, however.

There are other indicators of technological change
in any industry, some of which are given in Table 1 for
logging camps and contractors. Comparison of yearly new
capital expenditures with the year-to-year change in
gross fixed assets reveals little similarity. This serves
to point out that the quality of the capital data series
is probably low. The capital series is poor because of
changes that are difficult to adjust for, such as tax laws,
depreciation methods, etc.

New capital expenditures rose at about 5.6 percent
yearly, on the average, over the 19 years covered, while
U.S. domestic production of industrial roundwood grew by
only 1.8 percent per year.2& A rate of capital investment
in an industry greater than the rate of its market
expansion indicates either capital replacement of labor,
technological change, or both.

The number of establishments in this industry, as
estimated by the Bureau of the Census, has remained fairly
constant, as shown in Table 2. Except for the peak year
of 1967, the number has stayed near 13,000 establishments.
The real value added by each has also increased about 85
percent. The number of employees, however, has remained the
same, except for a low in 1967. That year showed an increase

in the number of establishments, but a decrease in the

 

23Robert B. Phelps. 1977. pp. cit.

.000000 wc0pC0pm psmschm>oo .m.: ..o.a .copws0am03 .0mnom 0950mm nom02 00m .0 900m
.mo0pm0p0pm zhpmsczH .HH mS:0o> .000300095202 mo mamc00 N000 .m:m:0o 0:9 mo ammusm

.mohzom

.mpmmm0 0000M mmopm 00000 5000 0000030000!

31

0

 

0.00 0.000 0.00 0.000 0000
0.00 0.000 0.0: 0.000 0000
0.00 0.200 0.00- 0.000 0000
0.00 0.000 0.00 0.00 0000
0.00 0.000 0.000 0.00 0000
0.00 0.000 0.0 0.000 0000
0.00 0.000 0.000: 0.00 0000
0.00 0.000 0.00 0.000 0000
0.00 0.000 0.00 0.00 0000
0.00 0.000 0.00 0.00 0000
0.00 0.000 0.0- 0.000 0000
0.00 0.000 0.00 0.000 0000
0.00 0.000 0.00 0.00 0000
0.00 0.000 0.00 0.00 0000
0.00 0.00: 0.0: 0.00 0000
0.00 0.0:: 0.0 0.00 0000
0.00 0.00: 0.0 0.00 0000
0.00 0.00: 0.0: 0.00 0000
0.00 0.000 0.00 0000
000000 00 000a 00020 A0 0000 00020 00 000a 00020
mmmho0msm mpmmm< 00000 Impmmm¢ cmx0m mmonw 000590UC0mxm 000%
00¢ mmopo 00909 0 :0 mwc0zo 0000900 302

 

 

.muopo0upcoo 0:0 09800 0:00000
:0 00:0:0 0000000oczomp mo 09090000S0 msom::.0 0090B

 

32

.000000 02000000 psmeshm>ou .0.0 0.0.0 .0000000002 .00100 002000 0000: 000 .0 0000
.0009009000 0090:0GH .HH 08:0o> .009390003802 mo 050800 0000 .030800 0:0 00 50005m

.009500

.0000000 mm00 .00300005808 09 00000 0:00>M
.0000000 @000 .090000 00000 mo 0:00> 00o00l

 

0
0.00 0 0.00 00000 0000
0.00 0 0.00 00000 0000
0.00 0 0.00 00000 0000
0.00 0 0.00 00000 0000
00 .aoaev 00 .zogev 00008000000000
M0000< 0:00> 000>O0Q8m M0090Q0o mo 009852 000%

:::::::::::::::: 980820009090m 00m:::::::::::::::

 

.00000009800 080 09800 0:00000
00% 0900 9808900090000 009 0:0 09808900090900 Mo 0098zz::.m 0090B

 

33

capital and employees in each establishment. These smaller.
but more numerous operations managed to maintain the
growth in value added for the industry. however.

The geometric index (Table 3) increased at a rate of
2.32 percent compounded annually.

Real value added per employee increased from $5,445 in
1958 to $13,665 in 1976, a 151 percent change. At the same
time, real capital per employee increased by 109 percent.
The corresponding annually compounded increases are 4.96
percent and 3.95 percent. respectively. while output per
unit of capital grew at an annual compounded rate of 1.19
percent. The share of capital in income increased by
52 percent from 41.6 percent to 63.3 percent. This latter
measure is also an estimate of the elasticity of capital
with respect to output.

The last column of Table 3 is the real value added per
employee net of technological change. The increase in
capital intensity accounts for 41.2 percent of the increase
in per employee productivity. Technological change is re-
sponsible for the majority of the gain. or 58.8 percent.
This indicates that during the time period covered.
technological change has had a substantial impact on the
timber harvesting industry.

The growth in output per unit of capital was greater
for the last half of the period than for the first half.

as shown in Table 4.

34

.008090

9009009089009 mo x0089 098908o0w 099 09 0009>90 .00009980 809 00000 0:90> 900mm
.000zO9980 990 mo 809858 09 0009>90 .090000 00x99 mo 0:90> 00080 0090990

.0000o9980 990 mo 809838 09 0009>90 .08390093808 09 00000 0:90> 00909909!

.00000 0390> 09 0009>90 .9908009 000009980 990 03898 08390093808 09 00000 0:90>I

mm
0
0

 

0000 000.0 00000 000. 00000 0000
00Nm 000.9 m9N0 N90. 0Nmm9 0009
9900 000.9 0000 000. wMNN9 0009
00:0 N9m.9 0000 0N0. 0000 0009
N000 00N.9 000m 000. N900 N009
0:90 m0N.9 009m 0N0. 0090 9009
0N00 090.9 0000 000. 0000 0009
0000 000.0 0000 000. 0000 0000
0000 000.0 0000 000. 0000 0000
0000 000.9 0000 090. 0000 0000
0000 000.0 0000 000. 0000 0009
0000 000.0 0000 00. 0000 0000
0000 000.0 0000 00. 0000 0000
0000 000.0 0000 000. 0000 0000
0000 000.0 0000 000. 0000 0000
9000 0N9.9 N000 00:. 0000 9009
0000 000.0 0000 000. 0000 0000
0000 000. 0000 000. 0000 0000
0000 000.0 0000 000. 0000 0000
000O998m 008090
809 00000 9009009089009
0390> 900m 90 00089 000o998m 809 08090 000o998m 809 800»
9M0090088o0 098908000 N9099900 900m M9099900 M0000< 0:90> 900m

 

08o9o089800 080 09800 0890009

 

.0000 000 00000020 00
008090 9009009089009 mo x0089 098908o00uu.m 0990B

35

08998999 9808890>00 .m.0
.0000000000 00000000 .00 05:00>

.009990

..0.0 .0000000003 .00-o0 003000 00902 000 .0 0000

.008390093802 90 030800 N009

.030800 099 90 300839

.009300

A.Q.PCOOVII.M mHQNB

36

Table #.--Annua;7growth rates in productivityL71958-1367,
and;968-1976 for logging camps and contractors.

 

Output per unit Output per unit
Year of labor of capital
1958-1967 5.26 ----- 6:65-- '

 

This increase in capital productivity, coupled with a
decline in labor productivity, is evidence of a substitution
of capital for labor in latter period, in terms of the
rate at which each factor increased output.

Kaiser and Guttenburg calculated output per manhour for
U.S. sawmills as increasing at an average annual rate of
3.2 percent between 1954 and 1967. This was in contrast to
the 1.2 percent rate for the first half of the century.2j-
They also found the greatest increase in the South, with
3.4 percent, while the West experienced 2.9 percent, and
the North lowest at 2.3 percent.

The sum of the exponents in an unrestricted Cobb-
Douglas production function (the exponents are not forced
to sum to 1) give an indication of returns to scale for an
industry. For logging camps and contractors, the un-
restricted Cobb-Douglas function is:

log Y = 0.7948 + 1.1896 log K - 0.3688 log L R2

(.1612) (.5974)

=O.782

 

25H.F. Kaiser and Sam Guttenburg. 1970. Gains in labor
productivity by the lumber industry. Southern
Lumberman 221:15, 18.

37

The sum of the labor and capital coefficients is
0.8208; a figure less than one indicates the industry is
operating at diseconomies of scale (the numbers in
parenthesis are standard errors).

A negative exponent is unexpected in a Cobb-Douglas
equation; its literal meaning is that production could be
increased by decreasing the labor input. In a Cobb-
Douglas function, the coefficients will not turn out
"right,” i.e., both positive and less than one unless the
indexed trend of value added lies between those for capital
and labor.2é

Introduction of a time factor into the function will
give an estimate of the change in output due to time, and
hence, an indication of technological change.

Table 5 shows the time trend in the Cobb-Douglas
coefficients. Regressions were fitted on the data for the
first 10 years and also on successive periods consisting
of additions of 2 years, i.e., 10, 12, 14, 16, 18, and
finally 19 years. Unfortunately, the only coefficients
with acceptable standard errors are those for the time
variable. This series of coefficients show an increasing
trend, which would suggest that the rate of technological
change was increasing in the seventies. This is also

indicated by the geometric index where the greatest

 

2é-E.H. Phelps Brown. 1957. The meaning of the fitted
Cobb-Douglas production function. Quarterly Journal
of Economics 71:546-560.

38

Table 5.--Three-factor Cobb-Douglas function (X = AKbLCTd)
forloggingcamps and contractors.

 

Period log A b c d b + c R

1958-1967 3.3510 0.4756 -0.0603 0.1287 0.4153 0.902
(.3458) (.9313)(.7704)

1958-1969 4.8054 .4020 - .2979 .1375* .1041 .921
{-2523} (.7664)(.0647)

1958-1971 1.8395 .1143 .7806 .2219* .8949 .908
{-2301) (-5955)(-0533)

1958-1973 1.2627 .1194 .9082* .2223 1.0276 .939
(.2024) (.3483)(.0487)

1958-1975 3.0267 .0626 .5596 .2812* .6222 .870
(.3266) (.5200)(.O774)

1958-1976 2.8567 .3185 .2461 .2405* .5646 .865

(.3168) (-5273)(.O796)

 

*Significant at the 10 percent level.

increases occur in 1970 and 1974. The coefficient
corresponds to a yearly rate of increase of 3.8 percent,
which is a somewhat larger increase than the geometric
index.

The relationship of the capital and labor exponents
suggests the type of technological change that occurred. A
rise in b relative to c denotes a labor-saving technical
change, while a fall in b relative to c is evidence of a
labor-using technical change.

Except for a dip in the second period, the labor
coefficient was rising, and the capital coefficient falling,

through 1973. This indicates that technological changes

39

during this period tended to be of a labor-using type.
suggesting new technologies adopted required relatively
more labor than capital. This trend was reversed in the
last two periods. however.

The sum of the exponents for the capital and labor
variables is a measure of returns to scale for the industry.
While highly variable. returns to scale have generally been
less than one. Unfortunately, only one coefficient for one
period is significant for these two variables; hence. these
estimates of returns to scale are not reliable. The full
equation, however. is relatively good at explaining the
variation in the observed output levels. as evidenced by
the R-squares.

Unlike the Cobb-Douglas function which assumes the
elasticity of substitution to equal unity. the CES function
allows its estimation. For logging camps and contractors.
fitting equation.(flb) by least squares yields:

v = 2.6536 (1.0334)13 (3075 K'35u2 + .6925 L'3542 ) '2823

The elasticity of substitution. estimated by equation
(12) is 1.548 (the standard error is .0952). This suggests
it has been relatively easy to replace labor with capital
in the logging industry. This is verified by the trends
in labor and capital presented above.

The technological change parameter indicates that this
factor has been advancing at the rate of 3.3 percent yearly.

This is close to the rate estimated by the geometric index.

4O

Sawmills and Planing Mills, Genegal(SIC 2421)

Past technological changes in the sawmilling and
planing industry have consisted of refinements to systems
in general use before 1958. One of the major adjustments
has been to process logs in a continuous flow, requiring
little manual handling. Another has been the shift from
ponding to cold-decking of logs, with transport accomplished
with large-capacity, log-loading tractors.

The move toward cold-decking logs has also allowed
better log sorting systems. The logs are sorted first for
product (veneer, sawlogs, pulp) and for sawlogs, sorted
further by size. Sawmilling runs of logs of all one size
are then made at greater speeds, due to less need for
adjustment.

Chipping of slabs and edges has reduced waste disposal
problems and added salable chips. Debarking has also
increased, with the bark often being burned for energy or
sold for mulch.

Edger saws with thin blades and carbide-tipped teeth
have reduced saw-kerf waste by one-third.

Improvements have also been made in sorting, stacking,
and packaging of lumber. The main effort has been to
replace hand labor with automated machines.

These innovations have required major investments by
sawmill and planing mills. Capital expenditures in 1972

were $1,842 per employee, considerably higher than for all

41

manufacturing ($1,335). These investments supported a
rising output per employee hour at an average annual rate
of 2.7 percent from 1958 to 1975.22' From 1971 to 1976,
however, the annual average percent change was only 0.4
percent.l§

Other indications of technological change add insight
into the sawmilling and planing industry. The trend in
new capital expenditures, shown in Table 6, averages a
compounded increase of 3.1 percent annually. This is in
contrast to the rate of decrease in total employees, {—1.95
percent) and the growth in domestic lumber production of
only .43 percent compounded annually.22

Table 7 shows the number of establishments in this
industry, as estimated by the Bureau of the Census, has
experienced a dramatic decline, on the order of 48 percent.
Further, the contraction in the number of establishments
has been quite steady. In contrast, capital per establish-
ment has greatly expanded between 1958 and 1972, by 145
percent. Capital in the industry has not been augmented

greatly: the expansion has been due to the decline in the

 

2Y-John Duke and Clyde Huffstutler. 1977. Productivity in
sawmills increases as labor input declines
substantially. Monthly Labor Review. April.

léArthur S. Herman. 1977. Productivity reports. Monthly
Labor Review. October.

lzRobert P. Phelps. 1977. pp. cit.

42

.mowwmo wcflpcwhm pawscmm>ou .m.: ..0.Q .COHmzfiammz .wmuom mmsouu Momma OHm .H puma

 

.mowpwapdpm hhpmsocH .HH mESHo> .mothoMMscmz mo msmcmo mmmfi .mzmcmo 6:9 mo smopsm

.muhzom

.mpmmmm umxflm mmopm HMPOp EOLM cmPMHSOHmoM
H.66H H.53mfi m.: m.anm 656a
“.mma w.msmH o.HoH- m.omm mama
0.6ma o.m:mH H.53H- N.Hmn aamfl
a.mmfi m.omom :.mm o.mmm mama
6.66H m.mwom o.mo: :.mfim mama
a.mmfi m.oooH m.mfi- H.66H Hamfi
m.mnfi n.maoa 0.66- m.mefi oama
a.aafi m.Hm6H m.om H.Hmfi mmmfi
m.mafi m.HmoH n.6m 5.5:H momfi
m.oma o.mooH m.mm- m.:HH moofi
o.mmH m.omofi n.HN- :.m:H coma
o.Hom m.mmoH m.mm- m.ooH mwmfi
o.mmH m.moafi m.mo 0.3:H smmfi
:.mom m.mama :.mo H.mofi mead
m.mom m.m:mH 6.05 o.mmfi momfi
m.:Hm m.maza m.mmH- m.moH Hmmfi
6.0:m 6.Hmwfi m.mm~- H.mmfi coma
m.mmm 3.:Hmfi a.oma o.mmfi mmofi
m.m:m m.mmmfi :.mmfi mama

“zones Aw mmmfi HHHSV as mmaa HHHEV Aw mama Hafizv
mmmzoamsm mpomm< cmxwm lmpmmm< ooxflm mmouu meSFHccmmxm mam»
HH< mmoao Hmpoa H :H mmcmno Hmwflmmo zoz

 

 

Hmpocmm.4maafle mcflcmHm ccm maaflssmm

 

CH mmcmno HNOHMOHonnomP mo myomeficsH mEomll.o manna

43

.moflmmo mcfipcfium pcoecpm>ou .m.: ..o.a .cOpmcflnmmg .wmuom mmsouo p0nm2 on .H puma
.moapmprpm hhpmsccH .HH oesao> .mmMSPommzcms Mo mamao mama .msmsmo map mo sampsm

.monzom

.mpmHHon wmmfi .muSPomazcms an cocoa msam>m
.mhmaaoo wmmfi .mpmmmm nmxfim mo msam> mmopol

 

H
N.mom Hm n.0mm Hmom mmmfi
6.mnH ma n.6mH Hamofi ammfi
m.oHH ma m.mmH mmfimfi moma
m.wn ma m.¢oH mnemfi mama

“a..:onav Aa .sosev
ucmcc< msam> mmmzoagem lawywmmo mpcmeanHQMPmm use»
a H Mo hopssz

uuuuuuuuuuuuu pcosnmwanmvmm pmmnnuuuuuuunuuuu

 

Hohosowx4waafle wzwcwam,o:m mHHfiESMm
mom memo psoenmfianmpmm hog and mpcmezmflanumm wo honszz:n.m magma

 

44

number of establishments and the survivors expanding to
take their place in production. This will also be
indicated by the Cobb—Douglas estimate that the industry
has been operating under rather large diseconomies of
scale.

While the amount of capital in each establishment has
been growing rapidly. the number of employees in each has
been growing only slightly. The average mill employs five
more in 1972 than it did 14 years previously. This is less
than a one-third increase. At the same time. output per
establishment has increased by 164 percent.

Improvements in processing technology in sawmilling
and planing should generally be reflected in the amount of
raw material required to produce a given output. Figures
1 and 2 show the trend in cubic feet of sawlogs required
to produce 1.000 board feet of lumber. softwood and hard—
wood. International 1/4-inch scale.

The North was the only region showing a decline in the
raw material requirements for production of 1.000 board
feet of softwood lumber. The South had a slightly rising
trend and the Pacific Coast and Rocky Mountain regions had
level trends (Figure 1).

There are two forces which would tend to offset a
decline in raw material requirements per unit of output in
this industry. One is a decline in the quality of the logs
used. manifested by smaller diameter environmental and

economic considerations requiring use of lesser quality

45

 

 

 

200 '-
l80 '-
1'3
:3 I60 -
u \- , . .53? '
g ““KHX/
b ’40 __ ""-"'" $0077!
— NORTH
_-— PACIFIC COAST
ROCKY MOUNTAIN
120;
I 1 l I 1
I950 L960 I970 I980 1990
YEAR

Figure 1. Cubic feet per thousand board feet
International 1/4 scale: softwood
SanogS. 1952-1976.

Sources:

1952 - Forest Service. 1958. Timber Resources for
America's Future. USDA Forest Resource Report
No. 14. Washington, D.C.: U.S. Government
Printing Office.

1962 — Forest Service. 1965. Timber Trends in the
United States. USDA Forest Resource Report
No. 17. Washington, D.C.: U.S. Government
Printing Office.

1970. 1976 - Unpublished Forest Service data.

46

materials that would otherwise be left in the woods. The
other is the development of new technologies that allow
utilization of smaller material for sawlogs. An example
would be the chipping headrig, which can turn very small
logs into studs. and the slabs and edgings into chips.
Increased use of such technologies would result in an
upward trend in the raw material requirements per unit of
output.

Figure 2 shows the raw material requirements to
produce 1.000 board feet of hardwood lumber. For this
resource. the trends have been much more sharply upward.
relative to those for softwoods. The trend for all
regions except the North turned upward sharply in 1962.
and the latter also turned upward in 1970. These upward
trends mean that it now takes more cubic feet of hardwood
sawlogs to produce a 1.000 board feet of lumber than it
did in the past. (The datum point for the Pacific Coast.
1976 is considered a poor estimate. In addition. the
amount of hardwood production in this region is very small.
3 percent of the national total in 1976.)

These increasing trends for hardwood are probably the
result of a decline in the quality of the resource (more
defects. sweep. crook. knots. etc.) and smaller sizes. not
offset by yield-increasing advances in technology.

The national trends in this area are shown in Figure 3.
Overall. the trend for softwoods has been slowly upward. and

that for hardwoods more sharply upward since 1962. Advances

47

CUBIC FEET

 

 

 

220 -
/
200 r ’/
//
100 — .
[60 4
[4o _ ----- SOUTH
— NORTH
— -— PACIFIC coasr
Igor 11001” yawn/1v
l l l 1 1
1950 19 60 1970 1990 1990,
YEAR

Figure 2. Cubic feet per thousand board feet
International 1/4 scale: hardwood
sawlogs 1952-1976.

Sources:

1952 - Forest Service. 1958. Timber Resources for
America's Future. USDA Forest Resource Report
No. 14. Washington. D.C.: U.S. Government
Printing Office.

1962 - Forest Service. 1965. Timber Trends in the
United States. USDA Forest Resource Report
No. 17. Washington, D.C.: U.S. Government
Printing Office.

1970. 1976 — Unpublished Forest Service data.

48

I80 "'

l60 - ___/'

f

-

fit

(3
l

----- 90F TWOODS
—— I'IA RDWOODS

IZOJ—
I; l l l l J

I950 I9 60 I970 I980 I990
YEAR

CUBIC FEET

 

Figure 3. Cubic feet per thousand board feet
International 1/4 scale: sawlogs,
U.S. totals. 1952-1976.

Source:

Figures 1 and 2, weighted by regional sawlog
removals, 1952, 1962, 1970. and 1976.

49
in sawmilling and planing technologies may have slowed
what would have otherwise been a more rapidly rising
trend.

The geometric index shows additional details. such
as that the output per employee increased 75 Percent over
the period studied (Table 8). an average annual compounded
increase of 1.7 percent.

The last column in Table 8 gives the trend in per
employee productivity due only to capital increases.

This series is net of technological change. The capital
deepening in this industry is responsible for only 35.2
percent of the increase in per employee productivity. The
major portion of the increase. 64.8 percent is due to
technological improvements. Real capital per employee
peaked in 1972. just before the housing recession. It
then fell by 23 percent in just 2 years. Had the housing
recession not occurred. there would have presumably

been no interruption in the rate of capital investment
per employee. Investment per employee has since
apparently resumed its normal upward trend of about 2.9
percent per year.

The share of capital in income also peaked in 1973.
just one year after the investment per employee high.
However, the overall trend in this estimate of output

elasticity with respect to capital has been upward.

50

.00000 0SH0> 09 c000>00 .0000000 0000oHQE0 000 03:08 00300003505 09 00000 0300>

.0mc0co H0OHwoaocno0p mo x0020 owhp0§o0w 059 an c000>00 .00000950 009 ©0000 0SH0> H00ml

.0000oamem 000 mo 009530 09 c000>00 .090000 00000 mo 0300> mmouw 00000009

.000000980 000 mo 0098:: 09 c000>00 .00:po0h:£0e 09 00000 0SH0> 0000000Qm

 

0
0000 000.0 00000 000. 0000 0000
0000 000.0 0000 000. 0000 0000
0000 000.0 0000 000. 0000 0000
0000 000.0 00000 000. 0000 0000
0000 000.0 00000 000. 0000 0000
0000 000.0 0000 000. 0000 0000
0000 000.0 0000 000. 0000 0000
0000 000.0 0000 000. 0000 0000
0000 000.0 0000 000. 0000 0000
0000 000.0 0000 000. 0000 0000
0000 000.0 0000 000. 0000 0000
0000 000.0 0000 000. 0000 0000
0000 000.0 0000 000. 0000 0000
0000 000.0 0000 000. 0000 0000
0000 000.0 0000 000. 0000 0000
0000 000.0 0000 000. 0000 0000
0000 000.0 0000 000. 0000 0000
0000 000.0 0000 000. 0000 0000
0000 000.0 0000 000. 0000 0000
00hoamem 0ws0so 0000
000 0000< nmoaocno0a
0SHO> H00m mo N0UCH 00moamsm 00m 000nm 00zoamem 00m 0000
wompo000oo 000908000 NH0PHQ0U H00m MH0PHQ0U Mc000< 0sa0> 000m

 

H000£0o .0000: wCHC0Hm 0:0 0000830m

 

0NON on 00903020 :0 00:050 0000000092000 mo x0090 00000800ouu.m 00909

51

.000000 wcflwcHhm pc0ec00>oo .m.:
.0000000000 00000000 .00 00000>

..o.m .copmswzm03

.00-00 000000 00002 000 .0 0000

.0003900HSC05 mo 050C0o N000 .msmc0o 0:9 mo 30003m

.0ohzom

0.0.00000--.0 00000

52

Table 9 shows some rather surprising results of
applying the partial measures of capital and labor
productivity to the sawmilling and planing industry.
Productivity shows a tremendous decline in the last half
of the period; in fact the capital measure becomes negative.
This decline in capital productivity is possibly a
reflection of added equipment for pollution control.

Table 9.--Annual owth rates_;n productivityL 1958—1262,
and 19E8-l9261for sawmills and planing mills.

 

Output per unit Output per unit
Years of labor of capital
Pct. Pct.
1958-1967 4.73 1.72
1968-1976 .30 -1.50

 

The unrestricted Cobb-Douglas function for this
industry is given by:

log Y = 8.5275 + .4990 log K - .1841 log L R2 = .678

(.1079) (.0896)

The sum of the coefficients equals .3149; this is much
less than one, and indicates the industry has been operating
under diseconomies of scale.

The addition of a time variable into the unrestricted
Cobb—Douglas yields an indication of technological change.
Table 10 shows the time trend of the fitted parameters.

The standard error of the labor coefficient is always
larger than the coefficient; hence, it cannot be considered

statistically significant. However, the series would

53

Table 10.--Three:£actor Cobb-Douglas function (Y=AKchTd)
for sawmills and_planing mills, general.

 

 

1958—1967 3.5381 0.4968* -o.0135 0.0645* 0.4833 0.923
(.0961) (.1793) (.0257)

1958-1969 3.5587 .4209* .0840 .0712* .5049 .834
(.1190) (.2023) (.0328)

1958-1971 3.6386 .4289* .0589 .0686* .4878 .844
(.1075) (.1750) (.0295)

1958-1973 3.6170 .4456* .0408 .0662* .4864 .934
(.0549) (.1492) (.0259)

1958-1975 2.7746 .4856* .1433 .0632 .6289 .734
(.1088) (.2508) (.0437)

1958-1976 2.9787 .4648* .1350 .0588 .5998 .712
(.1084) (.2534) (.0440)

 

*Significant at the 10 percent level.

indicate labor was becoming more important in the
productive process.

The A term, changes in which represent a change in
neutral technology, remains relatively stable until the
last few years, when it declines. This relates well to the
c term, in which changes indicate changes in nonneutral
technology. This parameter rises considerable in the last
2 years, relative to the other subperiods. The d parameter,
an indicator of the rate of change in neutral technology,
remains about the same throughout the period. The result
of this factor yields only about a 0.9 percent yearly
increase in neutral technological change. This estimate
is much lower than the geometric index; however, the

coefficients are not statistically acceptable in the last

54
periods, although they are in the first four.

The sums of the labor and capital coefficients are
all less than one, meaning the industry has been
operating under diseconomies of scale. The figures have
been generally approaching unity, however, which suggests
that the forces causing the diseconomies may be lessening.
Changes in the size and quality of sawlogs may also be
involved.

The CES equation for sawmills and planing mills,
estimated by nonlinear regression, is:

v = 1.1375 (.9989)t (.8360 K'1314 + .164 L'131“) 7'6104

The equation yields the result that the output (real
value added) of this industry can be estimated fairly
closely by using just capital data. The elasticity of
substitution of capital for labor is calculated as 1.1513
(standard error = .0747), which suggests it has not been
as easy to substitute capital for labor in this industry
as it has been for some of the other forest industries.

Ferguson found that for eleven cases in the lumber
industry (SIC 26), the elasticity was greater than zero
for eight, between 0 and 1 for two, and greater than one
for one case. His data consisted of the Census years

1947, 1954, and 1958.39

 

E9—C.E. Ferguson. 1963. Cross-section production functions

and the elasticity of substitution in American
manufacturing industry. Review of Economics and
Statistics 45:305-313.

55

The time parameter, the measure of technological
change, signals no advance. With the figure lying so
close to one, however, it is difficult for the model to
differentiate the small difference. Hence, it is viewed
as not being very different from the other measures.
PulpmillsLSIC 261;)

This industry is defined as establishments engaged
in manufacturing pulp from wood or from other materials.
Included are logging camps operated by pulpmills but not
separately reported.

One indication of technological change in this
industry would be the use of fewer cords of wood per ton
of pulp produced. Table 11 presents the trends in cords
of wood consumed per ton of pulp produced for the various
pulping processes.

In aggregate, pulping shows little progress in the
reduction of cords of wood required to produce a ton of
pulp. There was virtually no change between 1920 and
1970. Since then, there has been about a 6 percent
improvement.

The reason for this lack of improvement in pulping
lies primarily with the sulfate process, which has been
steadily increasing its share in woodpulp production to
69 percent of the total. No stable improvement in terms
of output per unit input has been made in this process

since 1940. Instead, changes in this process have been

.Unmon0mmmm U20 00009 Mo mOHPwflpMPm .0000 .mPSPHPmSH 0000a £000005<

.0 u qmm 900900 mo0zommm 000>0om pmm0om
.0.0.0.0 0000 :0 000000 000000 0:0 00 00002 00000663 .0000 .00>00002 .0 00>00

.wmohsom

.=00006= 00 00000600:
.00000s0\00000s0 00 00000600m

 

56

0
00.0 00.0 00.0 00.0 00.0 0000
00. 00.0 00.0 00.0 00.0 0000
00. 00.0 00.0 00.0 00.0 0000
00. 00.0 00.0 00.0 00.0 0000
00.0 00.0 00.0 00.0 00.0 0000
M 00.0 00.0 0 00.0 0000
M 00.0 00.0 0 00.0 0000
M 00.0 00.0 0 00.0 0000
M 00.0 00.0 0 00.0 0000
00000
0002623090 mPGMHsm mPHMHSW HNHoQO USN WCH>HommHQ HMPOB 90.00;.

 

wmu0oo~
.0000-6000 .0008600 00
060056000 QHSQ 90 now 009 nowpmasmcoo cooamasmul.00 00909

57

was

.mwcflcmmnom .Umcoamxm\cmpMMQ0mmo

.mmmhp coozmazm ©0000ommmcz

.000080500800 .cooBUSSOMw mmozaocHw

 

00.
mm.
00.
00.0

“00:90

00.
00.
00.0
00.0
00.0

00000000\0000000000

00.

No.0
no.0
00.0
30.0

00008050080m

0000
0000
0000
0000
0000
0000
0000
0000
0000

 

0.0.00000--.00 00000

58
in improvements in the quality of the paper produced
from the woodpulp. This has entailed generally more
bleaching and refining. These changes have been to the
detriment of the output of pulp per cord of wood. and
have offset any yield-increasing improvements.

The slight improvement in the aggregate yield has been
due to the introduction of new. and growth of older.
higher yielding processes. such as semichemical. and de-
fibrated/exploded pulps. These are not the type of pulps
used for high-quality. bleached and coated papers.
however. and they have not replaced the major process of
sulfate pulping.

Some other indicators of technological change are
presented in Table 12. New capital expenditures show the
decline in the industry actually started about 1970. and
this indicator didn't pick up again until 1975. Total
gross fixed assets also show the slump the industry
experienced in 1972-1974. Overall. employment in the
pulpmill industry has remained fairly stable, with the
exception of the 1972-1974 period.

Table 13 shows the number of establishments in this
industry has remained constant except for a drop in the
early 1960's. This decline was offset by the increases in
capital. employees. and output per establishment. Given
the tremendous investment required per establishment. a

sudden change in the number of establishments in the

.000000 m000000m 0:08:00>ow .m.: ..o.a .copw00nm0z .mmuom mmso0o 0000: 00m .0 p000
.00000000pm 00003000 .00 05:0o> .000300003002 mo 030000 0000 .mswcmo 0:0 mo 3000sm

.0003om

.000000 00000 00000 00000 5000 00000300000

59

 

0.00 0.0000 0.000 0.000 0000
0.00 0.0000 0.000 0.000 0000
0.00 0.000 0.00 0.00 0000
0.00 0.000 0.00 0.00 0000
0.00 0.000 0.000- 0.00 0000
0.00 0.0000 0.00- 0.00 0000
0.00 0.0000 0.000 0.00 0000
0.00 0.0000 0.00 0.000 0000
0.00 0.0000 0.00 0.00 0000
0.00 0.0000 0.00 0.000 0000
0.00 0.0000 0.000 0.000 0000
0.00 0.000 0.00 0.000 0000
0.00 0.000 0.00 0.00 0000
0.00 0.000 0.00 0.00 0000
0.00 0.000 0.000 0.00 0000
0.00 0.000 0.00- 0.00 0000
0.00 0.000 0.00- 0.00 0000
0.00 0.000 0.00 0.00 0000
0.00 0.000 0.00 0000
000000 00 0000 00020 00 0000 00020 00 0000 00020

000000mem 00000< 00x00 l00000< 00000 00000 00030000000m 000»

00¢ 00000 00009 0 c0 0wc0no 0000000 302

 

 

000050030 :0 000000
00o0wo0oczomp mo 0000000000 0Eomuu.m0 0000B

 

60

.000000 00000000 0:08:00>0o
.0.0 ..0.0 .0000000003 .00-00 000000 00002 000 .0 0000 .0000000000
00003000 .00 0E300> .000300003002 00 030000 N000 .030000 000 00 30003m

.00030m

.0000000 0000 .00300003008 an 00000 0300>I

.0000000 @000 .000000 00x00 00 0300> 0000mm

 

0
00.0 000 00.00 00 0000
00.0 000 00.00 00 0000
00.0 000 00.00 00 0000
00.0 000 00.00 00 0000
00 00020 000000000 00 00020 00000000000000 0000
M00000 000m> 00000000 00 000002

uuuuuuuuuuuuuuu 0006500000000 000nnuuuuuuunuuuuu

 

000060030 000 0000 0008500000000
000 000 00008300000000 00 00QE3ZII.M0 0000B

 

61

industry would not be expected. In addition, the Census
does not report separately pulpmills associated with
paper mills. For example, the American Paper Institute
reports 279 wood pulpmills in 1972.&l

The geometric index of technological change for pulp-
mills (Table 1#) advances at the compounded rate of 1.3
percent per year. Pulpmills have the highest real value
added per employee of the four industries studied. This
real output per man has grown from $13,803 to $22,313 in
19 years, an annually compounded rate of 2.6 percent. In
contrast, the corrected real value added per employee (net
of technological change), rose only 1.2 percent compounded
annually. Of the productivity increase per employee, 42
percent is due to capital deepening, while the remainder,
58.0 percent, is accounted for by technological change.

This is also the most highly capitalized (in per
employee terms) of the four industries. In 1976, real
capital per employee in pulpmills was roughly 7.5 times
higher than either logging camps and contractors or sawmills
and planing mills, and about twice as high as paper mills.
This investment per man grew at an annually compounded rate
of 3 percent over the period studied. Both of these

indicators had peaked in earlier years; real output per

employee in 1974, and real capital per employee in 1971.

 

EJ'utlmerican Paper Institute. 1977. Statistics of paper and
paperboard.

.wwcmno HmOHmoHoczoop mo xoozw owhpoEoom map hp coofl>ao .ommOHQEo pom Gonna ozam> Hmmml
.momhoameo Ham mo hopes: hp omcfl>flc .mpommm cmxfim mo osam> mmonw ompmammom

.umcow osam> an cmcw>wn .Haomzma mooAOHQEo Ham mSCwE whopomgscme an umoom osam>t

.momhoaQEo Ham mo #0953: an oocw>wc .onzpommscme hp cocoa osam> copmamoow

H

 

62

 

 

manna som.fi omaou mom. mammm ommfi
mooofi Hos.fi mommo mom. mmmmm mama
comma oHo.H momma Ham. mooom emmfi
ommmfi mm:.H omooo amo. mommm mama
ammofi :N:.H Heoos omo. oooom mama
mmeoa mom.fi momfio :mo. aammm Humfi
ooomfi om:.fi Hmoom ooo. mfifiom ammo
momma omm.H Hmmfim ooo. ommsm momfi
ommmfi oom.fi mmouo :mo. ofifiem oomfi
omHuH mmm.fi mfiomo :mo. mammm Roma
Hmoofi Ho:.H oomno Hoo. ooomm oomH
memoa moa.fi moomo oao. ofioem oomfi
momma mme.fl oemuo moo. NHBNN somfi
momma oo:.H emooo mmo. ofiofim momfi
onoH mom. ooomo mmo. omHoH moms
mfiomfi ooH.H ammo: moo. momma Homfi
Hommfi ofio.fi oHHoe moo. momma coma
Hmmefi moo.H oooo: mHo. ofioofi momfi
momma ooo.H mooo: oHo. momma oomfi
oohoamem
pom oocc< owsmno
osam> doom fiscawoaozzooe mo mmzoamem Mom onmnm oozoamem pom pom»
MUmoPomhcAOU NmfiGH OwHPmEOmO WHNPHQNU Hamm MHmpHQmo Mfimcc¢ mSHw> mem
maafiemasm

 

HHoN OHm aupmsccw CH mwcm:o awoflmOHocnooP mo xoccw owupmaomunu.:fi manwe

63

.m.D
hnpmsczH

.oowmmo mcwpcfium psoECQm>ow
..o.a .copmcfigmmz .om-om manage beams on .H ppmm .moapmflpapm

.HH mSDHo> .mmthommzcms Mo msmcoo Numfi

.msmzoo ozp mo sawmsm

.mohsom

A.o.pcoov--.:fi magma

6h

The share of capital in income trend roughly follows
the trend of real capital per employee. This would be
expected, unless there are significant changes in the price
ratio between capital and labor and in the elasticity of
substitution.

Table 15 shows a dramatic reversal in the two
partial productivity measures presented. Both measures
are negative in the last period because of a decline in
output in the industry (5.? percent), while capital
increased 16 percent and labor increased 2 percent.

The unrestricted Cobb-Douglas function for the
pulpmill industry, 1958-1976, is:

Y = 1.4n5 + 1.1015 log K - .1208 log L R2 = .752

(.1616) (.2772)
The sum of the coefficients is very nearly one,
suggesting that the industry has been operating under
neither economies nor diseconomies of scale.

The addition of time into the equation results in
an indicator for technological change. Table 16 shows
the results of fitting such an unrestricted Cobb-
Douglas function for various periods. Unfortunately,
the standard errors are usually large relative to the
coefficients; thus, their reliability is in question.
However, the R-squares are high; while none of the
individual coefficients may be reliable, taken as a
whole, the equations do explain major portions of the

variations observed in the data.

65

Table 15.—-Annualfgrowth ratesin productivity, 195§e1967,
and1968-1976,§or pulpmills.

 

Output per unit

Output per unit

 

 

 

Year of labor of capital
1958-1967 5.11 percent 0.82 percent
1968-1976 -.86 percent -2.28 percent
Table 16.--Three:factor Cobb-Douglas functionflY = AKELCTd)

for pulpmills.
Period log A b c d + c R2
1958—1967 -4.2449 0.2678 2.9150* 0 1470 3 1828 0.904
(.5347) (1 3248) ( 0940)

1958-1969 -4.1390 .3582 2.6545* .1410 0127 .920
(.4112) (1 0502) (.0821)

1958-1971 -4.4670 .3118 2.8909 1435* .2027 .932
(.2828) ( 7392) (.0702)

1958-1973 .0874 .5220 6536* .1746 .1756 .859
(.3639) ( 3842) (.0918)

1958-1975 .0140 .5916* 5144 1615* .1060 .834
(.3166) ( 3685) ( 0763)

1958-1976 .5517 .5566 .4081 1464* .9647 .795
(.3425) ( .3946) ( 0823)

 

*Significant at the 10 percent level.

66

Changes in log A represent change in neutral tech-
nology. This term in the function fitted to the pulpmill
industry data remains at a low level until 1973, when it
takes a tremendous jump (the figures are in logs). The
occurrence is reversed for the labor coefficient, changes
in which reflect changes in nonneutral technological
change. The level of nonneutral technological change
dropped dramatically after 1971. In contrast, the rate of
neutral technological change (d) increased after 1973, but
only slightly; the estimate for the entire 19-year period
is equivalent to an annual increase of 2.3 percent. This
estimate is much higher than the geometric index for
pulpmills.

Up to 1971 the equations show large economies of
scale (b + c) for the industry. However, there was little
entry by new firms. The industry is an oligopoly and
maintains price control in wood buying. Competition
for stumpage is lessening this practice, however.&;

High capital requirements and other barriers probably
prevented other firms from entering the industry to take
advantage of the economies of scale.

Between 1971 and 1972, the industry experienced a large

drop in capital, on the order of 29 percent, while labor

 

fig"Sam Guttenberg. 1970. Economics of southern pine

pulpwood pricing. Forest Products Journal 20(4):15—18.

67

decreased by 27 percent. and value added fell by only
16 percent (Appendix Table C3). The years 1971 and 1972
were poor for the industry. and firms had to adjust.
Output in 1976 had not yet returned to its 1970 level.
Because of these changes. after 1971. the large
economies of scale dropped closer to unity. The fitted
CES function for pulpmills is:
V = .3567 (.9958)t (.99K.414o + .01 L.4140) 2.4155
Estimated by equation (10), the elasticity of substitution
is 1.7065 (standard error = .2325). This is an indication
that it has been easy for the pulp industry to replace
labor with capital. verified by the near doubling of
capital while the number of employees has remained about
the same, Appendix 03.

The technological change parameter is less than one.
suggesting there has been negative progress in the
pulping industry. Given the results of the other measures.
this can be discounted. The confidence interval
estimated for this parameter places the upper limit at
about 2.7 percent. and the estimates of the other models
fall into this range. With the rate so close to one. it
is difficult for the model to statistically differentiate

the small deviation.

Recently Lothner constructed an index of technology

68
for the various pulping processes.£l The index also
indicates the Minnesota and Wisconsin industries'
ability to use hardwoods in pulping. Lothner's applied
technology index rises about 25 percent from 1949 to
1969. His index is based on a set theory derivation as
proposed by Scott, which is an alternate method of
estimating technological change.fl
Paper MillsI Except Building Pape£_(§;0 2621)

There are basically three types of papermaking
machines in use today. The oldest type is the cylinder
paper machine, which although in many mills, is
gradually being phased out. While the cylinder machine
has the advantage of being able to build papers of I
greater thicknesses, it is a relatively slow process.

A much faster machine is the Fourdrinier, which can
be run at speeds greater than 2,000 ft./min., and hence
produce more tons per day. In this machine the slurry'
is drained through a moving belt, sometimes using vacuum
to increase the amount of water removed.

The third machine, the Yankee machine, differs from
the Fourdrinier only in the drying section. This type

consists of a very large (up to 15 ft. in diameter)

 

ElDavid C. Lothner. 1974. The Minnesota and Wisconsin
Pulpwood Markets: An Econometric Study of Past
Changes and the Future Outlook for Forest Resource
Planning. Ph.D. dissertation. University of
Minnesota.

EEJ.R. Scott, Jr. 1964. The measurement of technology.
Journal of Farm Economics 46(3):657-661.

69
single drum for drying, rather than many small drying
drums of similar diameters.

The paper formation process can be arbitrarily
broken down into five segments: stock preparation, web
formation, wet pressing, drying, and finishing.£5
Technical improvements have occurred in all these
segments, but the majority have been in the class of "fine
tuning.”

The trend in papermaking has clearly been toward
larger and faster papermaking machines. Twenty years
ago, a ”big” machine had a width of 200 inches and a
lineal speed of about 1,000 feet per minute. Today,
many machines have widths twice as great and speeds in
excess of 2,000 feet per minute, with some (producing
lighter weight papers) with speeds of up to 5,000 feet
per minute.

Due mostly to the large capital investment involved,
new techniques and innovations have been accepted slowly
and cautiously. One change that has been accepted is the
switch from brass or bronze wire forming belts to plastic
belts. These are longer lasting than the metal belts,
which have a useful life of 7 to 21 days, are easier to

install, and experience less downtime, thus producing

 

EiJohn G. Strange. 1977. The Paper Industry: A clinical
study. Appleton, Wis.: Graphic Communications
Center, Inc.

70

more tons of paper.

Some qualitative indicators of technological change.
such as new capital expenditures. are shown in Table 17.
Yearly capital expenditures increase by 105 percent from
1958 to 1976. The average yearly increase is 6.9 percent
of total gross fixed assets.

The number of establishments and per establishment
data for this industry are shown in Table 18. The
number of paper mills has been fairly constant from 1958
to 1972. The amount of capital and output per
establishment. however. have grown by 5.2 percent and
2.7 percent compounded annually. The number of employees.
after increasing in 1963, fell to about the same level in
1972 as it was in 1958.

Table 19 shows that the geometric index increased
only 12 percent. virtually all of it in the last 5 years.
Real value added per employee increased by over

$7,000, or by 62 percent. while capital per employee
increased by 105 percent. These increases are reflected
in the value added per unit capital. which fell from .603
in 1958 to .476 in 1976 on a per—employee, real-dollar
basis. The corrected real value added per employee
increased by only 1.96 percent. compounded annually.
Increased capital provided 71.6 percent of the increase in
productivity. by far the largest proportion among the

four industries studied. Technological change accounted

.mowwmo wswpzflnm pamscpo>ou .m.D ..o.n .copwcflnmms .omuom m930hc Memos UHm .H psmm
.moapmprpm hupmsccH .HH mesao> .mmp:pommscm2 mo mamsoo mmmfi .msmsoo oSP mo sampsm

.mohsom

.mpommm coxwm mmopw Howey Sony Umpdasoamoa

71

 

n.5mfi o.Homo H.65H m.omo oamfi
m.omH o.muoo H.mmm- H.m:: mama
:.omH m.:omo o.omm- n.5m: sums
N.Hmfi o.mmoo m.mfi «.mem mums
m.mmH m.Hmoo :.mo :.Hom mama
o.mmH a.oooo «.ms- m.mmm Hmmfi
m.omH o.oooo H.m- «.mom oamfi
m.H:H H.HHoo m.om o.Ho: moms
n.6mfi m.mmoo e.Hsm 2.:H: oomfi
0.0:H :.Homo m.mu o.oHo momfi
o.omH o.oofio c.8oo o.Hmo oomfl
o.mma :.mHo: o.use o.Ho: moms
o.mmH m.HmH: m.mom s.mam somfi
o.mmH o.momm m.amfi o.mmm moms
m.mmH :.mosm H.Nmm m.oom momfi
:.mmH m.HH:m m.om o.omm Homfi
m.omfi o.ommm n.5mfi o.mmm coma
o.:mH o.mmHn m.oos 6.60m moms
m.HmH m.mnom o.mom oomfi
Asooev no oomfi Haasv Ao oomfi Hausa no oomfl Hanso

moozoamem mpmwm< coxwm Impomm< coxflm mmOAU mmHSpHGCmme paw»
HH< mmopm Havoe H ca owcmno Hopwmmo Bmz

 

 

.umgmmxwzfloawsn pmooxm .mHHHS Momma
Cw omcmno awomeHocnoop mo whomefiUCM osomuu.mfi manna

 

72

.ooflmmo mswpcflnm psoespo>ow
.m.p ..o.a .:64wcfl:mmz .om-om mmsopo pawns on .H puma .moapmapMpm
zhpmsccH .HH ossao> .moHSPommscmz %o mSmCmo mumfi .msmcoo ocv mo swopsm

.mopsom

.mnmaaon whoa .ou3966932m5 an poops osam>l
.mnmaaop mmmfi .mpommm coxwm mo mzam> mmopow

 

H
oo.o mum HH.oH mam mama
:o.o mom oo.:H son momfi
no.o mam mH.NH omm moms
om.: Hum no.5 son oomfi
Aw Haflsv Aw HHHEV mpcoesmHHQMpmm poo»
Mpocc< osam> momzoamem Manpwmmo mo 969832

uuuuuuuuuuuuuu pcoesmHHDMPmm pomsunnnnnnunuunn

 

 

.pomwm msfloaflsn pawoxm .mfiaws pomwmlpom wpmc
psosnmeQMPmm non cam mvzoenmwanmpmo mo popezz::.mH mapme

73

.owcmzo Havomoaocsoop Mo xocco convoEOow on» an omofl>fio .oonHQEo pom cocoa 63Hm> Hooml

.momhoaQEo Ham mo hopes: on cocfl>wu .mpommm coxfim mo 63Hm> mmopm povaooQ
.noocm osam> an omuw>fln .Haouhmm momzoamEo Ham mscfle manpommssme an coopm osam>
.mowhoHQEo Ham mo hopes: >9.poow>wo .onzpommscme hp poops osam> oopmammol

N
H

 

emmofl
mooofl
mamsfi
omomfl
oooufi
oomsfi
oommfi
mHHmH
oofiafi
ooooH
osmofl
oooofi
omoofi
oomsfl
oomzfi
eoeofi
Hmmmfi
moomfi
msmfifi

«NH.H
omo.a
oaH.H
ooH.H
mum.
omm.
mflm.
ozm.
mmm.
on.
osm.
mmm.
5mm.
0mm.
Hmm.
omo.H
mmo.H
Hoo.fi
ooo.H

omoo:
Hmooo
ooomm
NNHN:
momma
mood:
omomo
omooo
oaoom
oomoo
swoon
ofimao
ozoflo
mofloo
omomm
oomom
moamm
ooomm
Hoomfi

omo.
ooo.
mfio.
ooo.
ooo.
ems.
moo.
ofio.
mmo.
smo.
moo.
zoo.
ooo.
umo.
ado.
Hoo.
moo.
moo.
omo.

oaoofi
comma
mzoom
oosmfi
moasfi
mzmofi
Hmoofi
omfiofi
oHooH
oomoH
oooofi
camafi
oooefi
momma
mmmofi
soaofi
omsofi
ooaofi
mamafi

ommfi
ommfi
enmfi
mums
mums
Hmmfi
osmfi
mom“
oomfi
mood
oomH
oomfi
somfi
ooofl
momfi
Homo
coma
mood
oomfi

:

oozoamem
pom omoo<
osam> Homm
oopoopnoo

mwzmco
Havowoaosnome mo
xoUGH oflpmeOow

oozoamem pom
Mampflmmo doom

N

ohmsm

Hmpammo

H

oohoamsm pom
cocc< msam> Homm

 

hogan mzfipaflsm pmooxm .maafls Mommm

 

«mom UHm hmvmsccfl no mwsmno Hooomoaosnoop mo sons“ couposoowna.mfi mamas

74

.ooommo wzapsHpm psoscno>ow

.m.p ..o.n .:6pmzflgmms .om-om 663626 neon: on .fi puma .monpmapMPm

hnpmzocH .HH mezao>

.mohspommscmz 9o msmcmu NmmH .msmCoo may mo smopzm

.oopsom

A.o.pcooo--.mfi manna

75
for only 28.4 percent of the increase.

The trend in capital's share in income is only weakly
upward at the rate of 3 percent over the 19-year period.
There has been little change in the elasticity of output
with respect to capital.

As shown in Table 20, the growth in output per unit
labor slowed in the second decade covered in this study,
while the output per unit of capital reversed itself
from negative to positive growth. Such a reversal occurred
because of the decline in the amount of capital in the
industry after a doubling in the first decade. At the
same time, total labor employed in paper mills remained
fairly constant. These trends mean that the industry may
have been capitalizing at a rate faster than its markets
were growing.

The unrestricted Cobb-Douglas function for paper mills,
1958-1976, is fitted as:

Y = 5.4382 + .5033 log K - .4151 log L R2 = .771

(.0687) (.4411)

The sum of the exponents is very nearly zero.
suggesting there are large diseconomies of scale of
operation in this industry.

Addition of a time variable allows estimation of
the shift of the production function through time, and

hence the estimation of technological change.

76

Table 20.--Annua;_growth rates
and 1968-1976L in

in productivity, 1958-1967,
paper mills.

 

Output per unit Output per unit
Year of labor of capital
1958-1967 2.65 -315 '
1968-1976 2.09 1.68

 

Table 21 shows that, like the pulpmill industry,

major changes occurred between 1971 and 1973. Like its
companion industry, the changes occurred in neutral
technology (log A) which took a major jump, and in
nonneutral technological change (0) which experienced a
major decline.

Table 21.--Three-factop Cobb-Douglas function (Y: AKbLCTd)
for paper mills, except building paper.

 

Period log A b c d b + c R2

1958-1967 -0.2832 0.2275 1.202 * 0.0413 1.4300 0.943
(.1671) (.4566) (.0504)

1958-1969 - .1639 .2165 1.1958* .0439 1.4123 .965
(.1432) (.3504) (.0436)

1958-1971 - .2315 .2073 1.22 3* .0426 1.4326 .966
(.1301) (.29 6) (.0403)

1958-1973 3.0543 .3157 .3739 .0399 .6896 .854
(.2916) (.5819) (.0920)

1958-1975 4.1553 .1358 .4314 .1086 .5672 .792
(.2991) (.7285) (.0899)

1958-1976 3.9259 -.0058 .7040 .1501* .6982 .806
(.0066) (.4319)

(.0196)

 

* - -
Significant at the 10 percent level.

77

However. the time coefficient. d. which indicates the
rate of neutral technological change. begins to increase
in 1975. The final value. that for the entire 19 years.
yields an annual rate of increase of 2.4 percent. This
is too high when compared with either the arithmetic or
geometric indexes. A more reasonable rate is about 0.7
percent per year. which would obtain from the coefficients
of the periods ending in 1973 and before.

Economies of scale in the industry change from 1.4
to about half this value after 1971. This timing
follows the other changes in the series of equations.

Like most of the other industries. few of the
coefficients are statistically significant. Nevertheless.
the R squares are sufficiently high; meaning that while
few of the individual coefficients are reliable. the
equations on the whole explain the variation observed in
the data fairly well.

The CES equation for paper mills is:

. . 1 4. 02
)t (.0047 K 2221 + .9953 L 222 ) 5 7

V = 1.0000 (1.0198

The elasticity of substitution used in estimating
this equation was 1.2855 (standard error = .1224). Such
a level of elasticity suggests it has been possible to
replace labor with capital with relative ease.

The technological change parameter places this factor

of growth at about 2 percent per year. While a slow

78

rate, it is still about five times the geometric index
rate.
A Comparison

Table 22 summarizes the percent of technological
change for each of the industries, estimated by each
model. There is close agreement between the arithmetic
and geometric measures. For this reason, only the results
of the geometric index were discussed, as stated earlier.
The geometric model was chosen over the arithmetic
because the assumptions required for the former were
judged less restrictive. In particular, the assumption
for the arithmetic that prices are changed only in the
short run by technology shifts is a difficult one to make,
since there have been no studies performed that would
indicate this. In addition, weighting by the elasticities
of labor and capital with respect to output, as in the
geometric model, provides additional information on
the industries through the estimation of those

elasticities.

79

Table 22.--Annual increase in fitechnological fchange,
by industry and method of measurement.

 

 

------------------ Model-----—------------—--

Industry Arithmeticl Geometric; Cobb-Douglas CES
"""""""'22::I:III:IIIII§;;;;;;IIIIIIII2:222:22:
Logging 3.4 3.1 3.8 3.3
Sawmilling &

Planing 1 .8 1 .8 0.9 3-
Pulping 2.5 2.4 2.3 .2.
Papermaking 0.7 0.4 2.4 2.0
1

—Linear regression trend.
gLess than zero.

For the projections that follow in the next chapter,
the geometric index is used for several reasons, rather
than the arithmetic (for the reasons given above), the
Cobb-Douglas, or the constant elasticity of substitution.
Both the Cobb-Douglas and the CES models are fitted
through regression techniques. Hence, both are limited
by the number of data points in excess of the number of
parameters estimated, i.e., the degrees of freedom.

There are only 19 data points available, and with three
or more parameters estimated, the degrees of freedom
become somewhat lower than desired. The geometric index
does not suffer from this limitation, and so is judged
the most suitable for the projections and overall use.
Other studies

The four industries included in this study have also

been evaluated as aggregates. The Bureau of Census

80

classifies the two industries, logging camps and contrac-
tors. and sawmills and planing mills, general, into the
two-digit industries are also included in this classi-
cation.Zié Pulpmills and paper mills, except building
paper, are included in.SIC 26, Paper and Allied Productsfi-Z

Robinson constructed a geometric index of technolog-
ical change for the lumber and wood products industry
(SIC 24). He found that the level of technology had
advanced at an average rate of 1.75 percent per year
bewteen 1949 and 1970.E§ Using the translog function,
another method of calculating technological change,
Gollop and Jorgenson found the average annual rate of

growth to be 1.77 percent for 1960—1966, and 1.02 percent

 

EéThe 1972 classification, in addition to these two

industries, includes the following in SIC 24A:
Hardwood Dimension and Flooring (SIC 2426), and
Special Product Sawmills. n.e.c. (SIC 2429). Other
subgroups are Millwork, PlyWood, and Structural Wood
Members. n.e.c. (SIC 24B), Wooden Containers and
Miscelleneous Wood Products (SIC 24C), and Wood
Buildings and Mobile Homes (SIC 24D).

EZSIC 26A, in addition to the two above-named, includes
the four-digit industries Paperboard Mills (SIC 2631).
and Building Paper and Building Board Mills (SIC 2661).
Other subgroups are Converted Paper and Paperboard
Products, except Containers and Boxes (SIC 26B), and
Paperboard Containers and Boxes (SIC 26C).

EQV.L. Robinson. 1975. An estimate of technological
progress in the lumber and wood products industry.
Forest Science 22(2):149-154.

81

for 1966-1973 in the lumber and wood products (except
furniture) industry. The average annual rate of growth
for the paper and allied products industry (SIC 26), for
the same two periods are .0124 percent and .0094 percent,
respectively.£2' Massell used a geometric index to estimate
the average percentage rate of technical change to be 3.77
for lumber and wood products, and 2.34 for pulp, paper
and products for the period 1946-1957.59

A Canadian study employing the geometric index of
technological change found a 50 percent increase in that
country's pulp and paper products industry, with only a
8 percent increase in the wood products industry.i; The
period covered was the years 1940 to 1960. The figures
correspond to average annual rates of change of 2.4
percent and 0.4 percent, respectively.

A study of American manufacturing estimated the
partial elasticity of substitution of capital for labor

to be 2.54 for lumber and wood products, and 0.37 for

 

£2F.M. Gollop and D.W. Jorgenson. 1977. U.S. productivity
growth by industry 1947-1973. Univ. of Wisconsin--
Madison, Social Systems Research Institute Workshop
Series No. 7712.

59B.F. Massell. 1961. A dissagregated view of technical
change. Journal of Political Economy 69(6):547-557.

E-1-'-G.H. Manning and G. Thornburn. 1971. Capital deepening
and technological change: The Canadian pulp and paper
industry 1940-1960. Canadian Journal of Forest
Research 1:159-166.

82

pulp, paper, and allied products.2§ The former is higher
than either of the partial elasticities estimated for
the logging or sawmilling industries in this study, and
the latter figure is much lower than the elasticities
of substitution estimated for pulpmilling and papermaking.

For comparison with the general economy, Massell
estimated the rate of technological change (with a
geometric model) to be 2.54 percent per year in United
States manufacturing from 1919 to 1955.51

Schmookler included papermaking as one of the
industries in his study on inventive activity and economic
growth.52 While the years covered by his study (1837-1957)
do not overlap with those covered by this study, the
information he presents is of interest. The data on the
annual number of patents show that the inventive activity
in papermaking peaked during the late 1920's and early
1930's. There were 898 patents in 1931 versus 653 in
1957-

 

2gD.B. Humphrey and J.R. Moroney. 1975. Substitution
among capital, labor, and natural resource products in
American manufacturing. Journal of Political Economy

83(1):57-82.

52B.F. Massell. 1960. Capital formation and technolog-
ical change in United States manufacturing. Review
of Economics and Statistics 42:182-188.

igJacob Schmookler. 1966. Invention and Economic
Growth. Cambridge: Harvard University Press.

83

In a second study covering the same time period,
Schmookler traced the number of patents in a number of
specialized categories;5 For woodsawing machines, the
apparent inventive activity peaked in the 1870's and
1880's.

A final illustration of technological change in the
forest industries is contained in Figure 4. This figure
is a comparison of the trends in inputs and outputs for
the entire forest products industry (SIC 24 and 26). In
1950, 115 cubic feet of industrial roundwood were
required to produce one ton of product. By 1976, only 93
cubic feet were required, a reduction of 19 percent. A
factor that may be important involved in producing this
reduction is a changing product mix. The proportion of
woodpulp has increased from 18.3 percent to 38.6 percent,
while lumber has declined from 57.3 percent in 1950, to
32.8 percent in 1976 (measured in tons). Another factor
is the increased use of mill residue for pulp chips.
Figure 4 is unadjusted for these changes. In this way,
however, it reflects the overall progress in providing for

wood consumption with less raw material.

 

isJacob Schmookler. 1972. Patents. Invention, and
Economic Change. Cambridge: Harvard University Press.

Figure 4.

Source:

84

Input and output rates of growth for
industrial roundwood.

Robert B. Phelps. 1977. The demand and
price situation for forest products 1976-77.
USDA Forest Service Miscellaneous Pub. No.
1357. Washington, D.C.: U.S. Government
Printing Office.

DOMES TIC PRODUCT/0N INDEX //950-’IO0I

85

 

I65

I60

I55

\ \
U.
a 6

\
Q
Q

\
8

\ \
i“. 8

\ \
M
£7. Q

\
\
0

I05

mo

 

j

 

 

 

YEAR

__ PRODUCTS FROM _
INDUS TRIAL ROUNDWOOD
-- I 7' 0N5 I . _
- '-1
‘\
_ I
r / \ 1
I \ l
—. It ~ / ‘ ‘ T
I ‘ I
_. l ‘ L
Q
I I |
I \ n,
1— ] I“
K— INDUS TRIAL
- 1’ Roumawooo -
n I ICU F 7' I
I \ \ I
f \ \
1 1 1 1 1 1 '
I952 I956 I960 I964 I968 I972 I976

IV. PROJECTIONS AND PROBABLE TECHNOLOGIES

Any static equilibrium projection model must have
an exogenous variable to provide the ”driving force,” to
produce change. For the purposes of this study there
appear to be two suitable exogenous variables: gross
national product and time. Projections of the
geometric index of technological change based on GNP and
time are presented in this chapter, for each of the four
industries included in this study. These are followed by
a discussion of products and processes the industries may
adopt in the future.

In the projections, the industries maintain their
respective positions with respect to the rate of
technological change: i.e., logging camps and contractors
is the most rapidly advancing of the four industries.
while paper mills are the least rapid. These relationships
hold for both the time and GNP projections.

The rates of increase in the geometric index, for the

two projection methods are given below.

Industry Annual rate Percent
of increase of GNP
log (billions)

Logging camps

and contractors .03079 0.817904

(SIC 2411)
Sawmills and planing

mills, general .01837 .540279

(SIC 2421)

86

8?

 

Industry Annual rate Percent
of increase of GNP
log(billions)
Pulpmills (SIC 2611) .02375 .703027
Papermills, except
building aper .00424 .0807496
(SIC 2621)

Factors Influencing Change

There are many factors that affect the rate of
technological progress. These can be divided into two
broad areas. First, there are changes in the rewards and
benefits from particular kinds of technological advance.
These are the demand factors that stimulate or retard
efforts to achieve advances. Second, there are changes
and differences in the stock of materials and components,
and in knowledge about them and processes. These factors
constitute the supply side for technological advance.5é
Technological change is in many respects simply another
commodity produced by the economic system, and subject to
economic forces.

The projections for each of the industries are
dependent upon certain assumptions. Given the past rapid
growth in capital for each of the industries (doubling,
or nearly so, for all but sawmilling), future
technological advance will depend on the availability

of investment funds.

 

fiéR.R. Nelson, M.J. Peck, and E.D. Kalachek. 1967.
Technology, Economic Growth, and Public Policy.
Washington, D.C.: The Brookings Institution.

88

A second factor influencing technological change is
the structure of the industries. In terms of number of
firms in the industry. the sawmilling industry has been
the only one that has seen significant changes. It is
probable that the rate of decline in the number of firms
is this industry will slow and perhaps stop in the future.
Fewer firms are likely to mean economies of scale and the
possibility of increased profitability and hence a source
of capital for increased technological change. It is
then possible that the rate of future technological
change will be greater than that of the past. The
evaluation of available technologies reveals that such
opportunities are extant.

A third factor involved in technological change is
price trends. The trends for both inputs (raw materials
especially energy. capital. and labor) and outputs (logs.
lumber. pulp. and paper) will be important. A rising
price trend for one or more inputs should stimulate new
technologies for reducing the amount required. or for
allowing substitution of a cheaper input.

A fourth influence on the rate of technological
change is government policy. Tax policy. such as
investment credit and depreciation. play a role in the
amount of new capital a firm or industry is willing to

invest. A second area where government policy is

89

important is in the amount of research and development
government is willing to fund, in both Federal research
organizations and in universities. A third area where
government actions will specifically influence technologi-
cal change in the forest products industries is in
Federal timber sales. Changes in timber supply security
would alter the performance of affected firms.iz

One final influence important to consider (although
the list could be greatly expanded), is the rate of
technological advance in the rest of the economy. There
are two aspects here: one is the rate of advance in
competing industries, the other is the development of
technologies that can be adapted for uses in the forest
industries. There is evidence that the rate of techno-
logical advance in the U.S. economy is declining, or at
least the average rates of social return on progress-
generating activities is declining.§§’52 This general
decline will surely influence the rate of technological
change in the forest industries, and could indicate

that the several projections, since the are based on

 

iZWilliam R. Bentley. 1970. Technological change in the
forest industries--a problem analysis. The University
of Wisconsin Forestry Research Notes, No. 151.

i-8-Michael Boretsky. 1975. Trends in U.S. technology: A
political economist's view. American Scientist 63(1):
70- 20

52William Fellner. 1970. Trends in the activities
generatin technological progress. American Economic
Review 60 1):1-29.

90
past performances. are too high.

Rapid changes that lower prices. improve quality. or
add entirely new products in competing industries will
increase pressure on the forest industries to adopt
equally innovative changes or lose their markets. In
the past. the lumber industry has not been particularly
successful in preserving and expanding its market. New
technologies evaluated in this study offer some hope that
this trend can be reversed.

Often. technologies that develop in other industries
are adopted by the forest industries. An example in the
future will be cutting of lumber by laser. Other
adoptions are not so straightforward. but are equally
dependent on advances in other scientific fields. For
example. the development of plastic webs for papermaking
was dependent on advances in the plastics industry.

As covered earlier in this study. the forest
industries have achieved only modest gains in technological
advances and manufacturing productivity. An analysis of
some of the factors affecting technological change in
the industries is covered in another section of this
chapter.

Because of the low past productivities. most
probably the result of low rates of adoption of new
technologies. opportunities for improvement are
considered to be large. since new knowledge of production

has been accumulating. Many of these opportunities exist

91

in the areas of marketing. institutional arrangements.
management improvement. and employee training. While
these factors can play a large role in increasing pro-
ductivity. they are not the concern in this chapter
(although past changes of these types no doubt played a
role in the trends found in the calculated indexes). The
focus of this chapter will be on the technical
improvements in timber harvesting and processing that
have been developed. Some of these are already in use.
but have not had widespread adoption. Others have yet
to be tried by industry. but appear promising.
Timberpfigpvesting

The projections for technological change in logging

camps and contractors. based on the geometric index. are

below:
Yea; GNP Based Time Based
1990 1.883 1.957
2000 2.112 2.264
2010 2.353 2-572
2020 2.557 2.880
2030 2.770 3.188

If future technologies are adopted at the same rate
as those in the past. other things being equal. then the
geometric index of technology will be approximately 3.2
times its 1958 level in 2030. In contrast. if the
adoption of new technologies depends on the growth of
the U.S. economy. then the index for 50 years in the

future will be somewhat lower. at 2.8.

92

In either case, the level of employed technology in
harvesting timber will be roughly three times as great
50 years in the future as it was 20 years ago. The
following discussion covers some of the developments
judged probable to produce the projected levels of
technology.

The process of cutting standing trees and moving
them to a mill has shown a clear trend toward mechanization.
This trend will certainly continue in the future. With
few exceptions, one general principle has held for
logging in the past - the object has been to remove the
sound, clear bole of the preferred wood species.
Recently. however, this general principle has begun to
give way to complete tree and full tree harvesting,
usually involving chipping in the woods.é9' Full tree
harvesting involves taking the entire above-ground
portion of the tree, while complete tree harvesting also
includes the stump and a portion of the root system.
There are several machines or machine systems now in use
that utilize these harvesting methods.§l- These systems can

reduce per cord costs by about half. while increasing

output per man by more than seven times, compared to

 

éQJ.L. Keays. 1975. Forest harvesting of the future.

Western Forest Products Laboratory. Unnumbered report.

éiJ.R. Erickson. 1968. Mechanization in the timber-
producing industry. Forest Products Journal 18(7):
21-27 I

93

conventional systems.ég A more recent analysis of whole
tree chipping estimated a $3 savings per cord over chips
from debarked roundwood.é2-

A major emphasis in the development of new timber
harvesting techniques will be on reducing the wood
residues left in the forest after logging operations. To
a large degree, the reduction in residues will depend
upon the prices of chips containing bark and foliage.
These prices will in turn depend on the development of
separating methods, or new pulping processes that can
digest the bark and leaves. The increasing possibility
of using wood for fuel may also play a major role in
reducing forest residues.é&- Research in the area of bark
and chip separation is continuing.é5

New machinery will evolve the fastest in the

pulpwood and chip harvesting areas, rather than in the

 

égK.K. Neilson. 1967. The present state, problems, and
outlook of mechanized tree processing in Eastern
Canada. Pulp and Paper Canada 67:WR 297-WR 301.

ézFrank E. Biltonen, J.R. Erickson, and J.R. Mattson.
1974. A preliminary economic analysis of whole-tree
chipping and bark removal. Forest Products Journal

24(3) 34'5““‘7 -

é-LE’T.H. Ellis. 1975. The role of wood residue in the
national energy picture. ip_Proceedings of the
International Meeting of the Forest Products Research
Society on "Wood Residue as an Energy Source,” Denver,
Colo.

éiLogging research progress report, No. 45. 1974. Pulp
and Paper Research Institute of Canada, Pointe Clare,
Quebec.

94
sawlog and veneer harvesting areas, because of movement
toward continuous flow harvesting techniques.

Utilization efficiency during harvesting is expected
to allow the minimum tree removed to be 6 inches DBH
with a 4-inch top, for second growth timber in the West.
Currently (1976). the minimum tree removed in the West
is 9 inches DBH with a 6-inch top. In the East, the
minimum will drop from 9 inches DBH and a 7-inch top to
9-inch DBH with O-inch top.-6-é This will increase the
amount of material removed per acre. This increased
harvesting utilization is estimated to possibly reduce
logging residues by 1.4 billion cubic feet.éz-

If past trends in real value added per employee and
capital share continue, then to reach a geometric index
of 3 (in 2030). it will be necessary for real capital per
employee to be over $17,200 in 1958 dollars. Considering
the past pattern of investment, this level should not be
difficult to attain. While logging is projected to
continue to be the most progressive technologically
(relative to its own 1958 level), the possible improve-
ments cited above are judged sufficient for the

industry to meet the time series projection levels.

 

ééR.L. Porterfield. 1977. Utilization efficiency

during harvesting--a survey of current and prospective
status. Forest Products Journal 27(12):17-20.

ézL.E. Lassen and Dwight Hair. 1970. Potential gains
in wood supplies throu h improved technology.
Journal of Forestry 68%7):404-407.

95
Sawmilling
The projected geometric indexes of technological

change for sawmills and planing mills, general, are:

 

Year GNP Based Time Based
1990 1.717 1.720
2000 1.868 1.904
2010 2.027 2.087
2020 2.162 2.271
2030 2.302 2.455

A range of technological change projections is
provided by the two bases of GNP and time. If the rate
of increase will be the same as it has been in the past,
then the geometric index of technology will be roughly
two and one-half times its 1958 level in 2030 for
sawmilling. Alternatively, if adoption of new tech-
nologies depends upon future economic activity, then
the rate will be somewhat lower than in the past, and
the geometric index will reach only 2.3 times the 1958
level by the year 2030. Either way, progress in the
sawmilling and planing industry will continue.

Promising technologies that will contribute to future
progress are covered in the remainder of this section.

The process of cutting solid lumber from logs in the
United States has evolved slowly since the first sawmill
was built in Maine in 1624. Today, however. there are a
host of new products and processes that are in development
or beginning to be commercially accepted. The trend is

strong toward producing wood as an ”engineered”

96
material, i.e., with prespecified qualities and properties.

These new developments can be broken into three
main groups: Those sawing processes that convert more
of the log into solid lumber: those that control the
quality of lumber: and those that produce new products
similar to or that can be called lumber.

Improvements in the sawing of logs into lumber
include high-strain headsaws with narrow kerf, more
accurate set works, and computer-controlled or assisted
sawing decisions. By simply using currently available
technologies, lumber recovery factors can be increased
by over 27 percent.é§

There are now about a dozen sawmills using the
computer-controlled sawing operation called ”Best
Opening Face (BOF),” and approximately an additional
fifty are using some type of less sophisticated computer
control.é2- The BOF sawing can increase yields on an
average in excess of 20 percent over conventional

methods.29 The number of mills using some degree of

 

é§-H.C. Mason & Associates, 1973. Study of softwood sawlog

conversion efficiency and the timber supply problems.
Report to the USDA Forest Service, Forest Products
Laboratory. Madison, Wis.

é2Hiram Hallock. 1977. Precision-quality and value.
Expo '77 logging-sawmilling seminar. ed. by Keith
Judkins. Southern Forest Products Association.

ZQ-Hiram Hallock and David W. Lewis. 1971. Increasing
softwood dimension yield from small logs--best
opgning face. USDA Forest Service Research Paper FPL
1 .

97

computer control can be expected to increase. Systems
to also control ripping and crosscutting are now under
development.Z$

Technological advance can create resources out of
otherwise useless material. An example of this is the
shaping-lathe headrig which shows promise of being able
to economically convert small. low-grade hardwoods into
marketable products.zg This machine can convert small logs
into solid lumber products plus flakes for board.zz-

More exotic methods of cutting wood other than by
saw are being investigated. One method showing promise
involved lasers for cutting both solid wood and wood-
based products. The advantage of such a method is the

very thin kerf produced.2&' One company is now using a

 

ZiAbigail Stern and Kent McDonald. 1978. Computer
optimization of cutting yield from multiple-ripped
boards. USDA Forest Service Research Paper FPL 318
(in press).

ngeter Koch. 1976. Key to utilization of hardwoods on
pine sites: the shaping-lathe headrig. Forest
Industries 103(11):48-51.

22Peter Koch. 1975. Shaping-lathe headrig will convert
small hardwoods into pallet cants plus flakes for
structural exterior flakeboard. ip Proceedings of
the Ninth Particleboard Symposium. Washington State
University, Pullman, Wash.

Z)iCurtis C. Peters and Conrad M. Banas. 1977. Cutting
wood and wood-base products with a multikilowatt
laser. Forest Products Journal 27(11):41-45.

98
laser for cutting puzzles and blocks in toy manufacture.zj'

Another area for improvement is in sawing methods.
Research has shown that'some sawing methods are superior
to others for given log sizes.-Z§-"ZZ Adoption of a
differing method may require log sorting prior to
breakdown, but this can also prove profitable, if there
is opportunity for the conversion into more than one
product.

Improvements can be expected throughout the saw-
milling process. Research into new methods of drying
have yielded faster curing of green lumber. Microwave
kilns can dry large pieces of Douglas-fir and hemlock in

only 5 to 10 hours with minimum degrade.Z§

 

25Gordon R. Connor, Sr. 1977. The central hardwoods
response. ip Resource Availability and the Hardwood
Forest Products Industry. W.L. Hoover and H.A. Holt,
eds. Department of Forestry and Natural Resources.
Purdue University.

ZéHiram Hallock, Abigail R. Stern, and David W. Lewis.
1976. Is there a "best" sawing method? USDA Forest
Service Research Paper FPL 280, 12 pp.

ZZo.w. Bousquet, and 1.3. Flann. 1975. Hardwood sawmill
productivity for live and around sawing. Forest
Products Journal 25(7):32-37.

Z§L. Admiral Barnes, R.L. Pike, and V.N.P. Mathur. 1976.
Continuous system for the drying of lumber with

miciowave energy. Forest Products Journal 26(5):
31- 2.

99

There are several methods for maintaining the quality
of lumber produced by a mill. One of these, of which there
are already about fifteen machines in use, is high-speed
machine stress rating (MSR). These machines grade lumber
on the basis of its stiffness, at speeds of up to 1,000
feet per minute.22

Another quality-control process locates specific
defects in lumber using ultrasound. This is also a
computer-controlled system: it reduces waste made by
inaccurate sawing decisions resulting in lower grade.§2-

The third area of technological advance lies in the
area of new lumber products. These include press-lam and
EGAR. Press-lam is dimension lumber from parallel-grain,
rotary-cut. thick veneer laminates. The product yield
from 12- to 18-inch-diameter logs averaged 60 percent.§l’

A new process of producing solid lumber is by edge

gluing and ripping (EGAR). In this process, logs are

live sawn, the unedged flitches are dried, ripped to the

 

22W.L. Galligan, D.V. Snodgrass, and G.W. Crow. 1977.
Machine stress rating: practical concerns for
lumber producers. USDA Forest Service General
Technical Report FPL 7.

‘QQKent McDonald. 1978. Lumber defect detection by
ultrasonics. USDA Forest Service Research Paper FPL
311.

QLFPL Press-Lam Research Team. 1972. FPL press—lam
process: fast, efficient conversion of logs into
strugtural products. Forest Products Journal 22(11):
11-1 0

100

largest usable width, and edge-glued into panels up to
48 inches wide. Lumber of any feasible width can then
be ripped from the panel. A product recovery of about
10 percent over conventional systems is produced by this
method.§g

Although many new innovations have been researched
and developed since World War II, the sawmilling industry
has been slow to accept them, and will probably remain
slow, although with some improvement, into the future.
Reasons cited for this situation are:

--A shortage of skilled implementation engineers
who can analyze a mill operation to determine the
feasibility of a new application.

--Communication with mill managers, and getting
their cooperation, is difficult.

--Training operators is costly and difficult.

-—There is a shortage of skilled maintenance crews.§2-
Without remedies to correct these problems, acceptance of

new techniques and products will continue to be sluggish.

 

ggK.C. Compton, H. Hallock, C. Gerhards, R. Jokerst. 1977.

Yield and strength of softwood dimension lumber pro-
duced by EGAR system. USDA Forest Service Research
Paper 293.

§3w. Bennett. 1978. Sawmill technology outruns industry's
skill at using it. Forest Industries 105(1):28—29.

101

Should past trends in real value added per employee
and capital's share in income continue. the sawmilling
and planing industry can attain the projected geometric
index of technology of 2.4 in 2030 with an investment per
employee of slightly more than $19,900 (1958 dollars).
This means less than a doubling of the 1976 level of real
capital per employee. Since the industry achieved a
near doubling in the nineteen years covered in this study.
considering this past trend. the required investment
should be easily attained. and probably surpassed.

Adoption of new technologies has been slow in the
past. which is probably a strongly contibuting factor to
the decline in the number of establishments recorded for
this industry. This decline. however. suggests that the
least progressive firms have been "weeded out" of the
industry. As a result. the industry is composed of
fewer. but larger and more progressive firms. These
points. plus the promising technologies discussed above.
lead to the judgement that this industry should be able
to surpass the geometric projections. possibly to an
index level of 3.5 to 4.

Pulping
The projections for technological change in

pulpmills. based on the geometric index. are:

102

 

 

Year GNP Based Timepgased
1990 1.859 1.860
2000 2.056 2.098
2010 2.263 2.335
2020 2.439 2.573
2030 2.621 2.811

Two alternative projections are provided by the
different bases of GNP and time. Should the pulping
industry continue to progress as it has in the past. its
geometric index of technology will increase to 2.8 by
2030. On the other hand. if future progress in pulping
is linked to growth in the economy. then the adoption of
new technologies will be lower. This industry has been
the second-most technologically progressive of the four
industries. Following is a discussion of some of the
probable technologies that will support future progress.

Current pulping methods are relatively old. being
built from discoveries first practiced over a century
ago. Even so. there are few new pulping processes under
development which show promise of becoming important in
the years ahead. Most research today is on aspects of
"fine tuning" existing processes.

In spite of its disadvantages of odor. high costs.
and high pollutant loading. the kraft process has expanded
its share of pulp production. Its advantages of
versatility. energy generation. and pulp strength will
ensure that it will continue to be the major pulping
process into the future. The kraft process may see

competition from some other methods. however.

103

An old pulping method that may see a return to
greater usage is the soda process. The addition of
oxygen for pulping and bleaching has renewed soda as a
viable alternative to kraft, due to its reduced pollution
loading.§&’§i Research indicates higher hardwood pulping
yields than that for kraft. Further improvements may be
expected.§é

Thermomechanical pulping (TMP) may expand the
fastest of all the methods in the future. Its advantages
are improved pulp strength and adaptability for potential
chemical treatment.§z The minimum acceptable size of a
TMP mill is only about one-third that of a kraft,
allowing future plants to be built more cheaply and in
areas without the large wood supplies required for kraft.

Several new pulping systems are being developed.§§
Holopulping, a selective delignification, three—stage

process, will retain all cellulose, hemicelluloses,

 

gitAnonymous. 1976. .Where's pulping headed? A review

of state of the art. Pulp & Paper 60(9):?8-80, 89.

£3iAJ-I. Nissan. ed. 1973. Future technical needs and
trends in the paper industry. Special Technical
Association Publication No. 10. TAPPI.

§éAnonymous. 1978. Funded research plan. The Institute
of Paper Chemistry. Unnumbered report.

ézJohn G. Strange. pp. cit.

§§J. Rauch, ed. 1976. Kline guide to the pulp and
paper industry. Charles H. Kline & Co., Inc.

Fairfield. N.J..

104

and other polysaccharides of wood. The process will
yield between 65 and 80 percent compared to 45 to 50
percent for kraft. Nonsulfur pulping in Canada is
expected to occur before 1990. although sulfur-based
processes will still dominate.-8-2

Another pulping method. hydrorefining. will have
enormous impact on the industry. if it is fully
developed and put in commercial operation. This
method is envisioned as producing yield of up to 90
percent. by retaining almost all lignin through
hydrogenation.

Other advances in this industry will involve
improved bleaching with oxygen or ozone. and increased
use of computers for process control.

Based on the assumption of continued past trends of
real value added per employee and capital's share in
income. the pulping industry will have to invest over
$125,000 (1958 dollars) per employee to reach the projected
geometric index level of 2.8 in 2030. This is a very
high level of investment. and it is doubtful that the
industry can attain it. On this assumption. it is

suggested that the rate of technological progress in the

§2K.M. Jege and K.M. Thompson. 1975. The Canadian
pulp and paper industry--threats and opportunities.
1980-1990. Unnumbered report. Pulp and Paper
Research Institute of Canada.

105

pulping industry in the future will be less than that of
the past. Because of the tremendous investment required,
changes will probably be restricted to continued
refinements of existing and in place production methods,
rather than to additions of totally new ones. Hence,
it is judged that the future level of technology in this
industry will probably reach only 2.5 on the geometric
index.
Papermaking

Papermaking is a very ancient process: in the
United States, it is a mature industry and relatively
little technological progress can be expected. The
projections of geometric technological change for

papermills. except building paper, are:

 

Year GNP Based Time Based
1990 1.058 1.093
2000 1.081 1.135
2010 1.104 1.178
2020 1.124 1.220
2030 1.145 1.262

The two projections of the geometric index of
technological change for papermaking are significantly
lower than for any of the other three forest industries.
With either assumption, of the same progress as in
the past through time, or of progress linked to the
growth of the U.S. economy, the projections are not of
very large increases in the level of technology in the

papermaking industry. The methods of making paper are

106

very old. and there possibly is not much improvement that
remains to be made. in terms of efficiency per unit of
capital equipment and labor. There are some new develop-
ments that may become important in the industry. however.
and they are discussed in the next few paragraphs.

The two basic papermaking machines are the
Fourdrinier and the cylinder. and their basic principles
of operation have remained unchanged for over a century.
A new sheet-forming machine was commercialized in the mid
'60's. the twin-wire former. By draining the sheet from
both sides. the method is more rapid. with better formation
and uniformity. and improved physical properties. It also
eliminates two-sidedness. These advantages will lead
to an expansion of this type of paper forming in the
future.

The U.S. papermaking industry had 1,210 Fourdriniers.
536 cylinder machines, 6 combination units. and 8 twin-
wire formers in 1975.29

One problem with papermaking is the large amounts
of water it requires. The furnish (fiber—containing
slurry) typically consists of over 99 percent water and
less than 1 percent wood fibers. Research is underway to

develop processes using higher consistency forming and

 

EQJ. Rauch. ed. pp. cit.

107
also closed-loop systems.

Dry forming is a papermaking method without water.
Its use has been predicted to form 2 percent of Canada's
paper by 1990.2;

Other improvements in papermaking will occur in the
drying sections and in process control, using computers.
These will consist of adjustments to existing systems,
however, rather than radically new technologies. Emphasis
for some time into the future will be on pollution control
and reducing energy requirements.

Should past trends in real value added and capital's
share in income continue as in the past, the industry will
only have to invest $33,000 (1958 dollars) in real capital
per employee to reach the projected 1.2 geometric index in
the year 2030. The actual level of investment is
already past this figure. The reason the geometric index
of technological change in this industry has not
increased much is because increases in capital per
employee have not produced proportionately large increases
in output per employee. Thus, the industry has been
increasing its investment per man, doubling it in the
nineteen years covered, but output per man has gone up
only 62 percent. Similar conditions can be expected to
remain true in the future. with increasing investment, but

output per man lagging behind. Hence, the judgement is

 

2-1-K.M. Jege and K.M. Thompson. pp, cit.

108
that the projections of little technological change will

occur in this industry are accurate.

The Environment for Technological Change

 

Technological change does not occur in isolation.
Its formation and rate of change depend on many factors--
social. economic, institutional, and political.

Size class distribution

One of these factors affecting technological change
is the size of the firm. Table 23 shows the size class
distributions for the four industries included in this
study. It is evident that the two industries dealing
with solid wood tend to be small units in terms of
employment. These units are also low in capital per
employee, as shown earlier.

The pulp and paper industries, in contrast, tend to
be much larger. Nearly half of the pulpmills have 50
or more employees. while three-fourths of the papermills
have that number.

All of these forest products industries are mature,
in the sense of having been in production for many years.
Mature large industries tend to reduce employment
through increased capital spending and thereby improving
labor productivity. It is also suggested that large
corporations view innovation largely in terms of cost

reduction and increased labor productivity for existing,

109

.oowmmo msopsfinm pacesno>ow .m.D ..o.n .sopwswzmmz .omuom masono acnms on .H puma
.mofipmwpmwm znpmzcsH .HH ossao> .monsvoMM:CMS mo msmcoo Numfi .mSmsoo can we scopsm

.mopsom

.Pmmwgma ozp spas wcficcfim6m*

 

+oom.m
0H OH H H ma:.muooo.H
:m :H m : mmmnoom
as no 2H m mmsuoom
mam oom so on asm-ooH
on ms: ooH HHH mmuoo
saa.fi Hem omo Hoo m:-om
mm:.m oan mom.fi HmH.H mfiuoH
ooo.o mao.fi «66.: osH.N 61o
amo.m oom.: wmm.mH omfi.a 21H
moonss: oneness
* mpzos * magma
o>wpdassso unmfianmpmm o>fipmassso unmoHQMPmm mwmzoamsm
Hmsm UHm «Ham OHm

 

.INNM

mmpmpm 6696:: on» an menopause“ pmmaoo 666% mo coapaaaapmae mmeao 666m--.om wanes

 

110

.oOHmmo msHchpm pcosspm>oo .m.= ..o.Q .copmzHgmms .omnom mmsomo ponms UHm .H Pawn
.mOHPmemvm zhpmscsH .HH oESHo> .monspommscmz mo msmcoo mmmH .msmCoo may mo swoszm

.oOHSOm

.pmomHMH esp 29H; mcHQQHwom*

 

H H
mm mm H H
no mo 5 o
HnH on pH oH
HmN om mm M
:0m mm mm 0
omm Ho on a
mom :H on m
Nmm MH mm H
mdm um 00 HN
muonssc mucosa:
* masos * muses
6>HpeHasau -gmHHnmpmm m>HpmHasao -gmHHnmpmm
HNQN UHm HHWN UHm

+ooo.m
mme.m-ooo.H
mmm-ooo
663-com
m:m-ooH
mm-oo
meuom
6H16H
61o

momHOHmEm

 

A.G.P£oov11.mm mHDMB

111

in place processes.2g' Such a view tends to hold technical
change to ”fine tuning” of technologies already in use
rather than adoption of radically different methods.
These facts help explain why the rate of technological
change is so low in the pulp and paper industries. The
logging and sawmilling industries, being smaller and less
capital intensive, can adopt new technologies more rapidly.
Regign of opgpation

The region of the United States an establishment
operates in may also play a role in affecting the adoption
of new technologies. Tables 24 through 27 present
regional data on the number of establishments, employment,
and value added for the four industries in 1972 and 1958.

These data show that the number of logging operations
has grown in the southern and mountain regions to the
detriment of the other regions. A comparison of the
percentage figures for establishments versus employment
and value added, however, reveal that operations in the
South tend to be small, while those in the Pacific region
are larger than average.

All regions experienced a decline in the number
of sawmills and planing mills between 1958 and 1972.
In percentage terms, the South had the greatest decline

in employment, but its share of value added remained the

 

2gCommerce Technical Advisory Board. 1976. The role of
new technical enterprises in the U.S. economy. A
report to the Secretary of Commerce, 15 pp.

112

.moHHHo moHpeHaa
Psmsspo>oo .m.= ..o.m .zoprHnmmz .mm op om masons Homo: cam hpmsssm Hohmcou .H whom
.moHpmHPMPm hupmsqu .HH ossHo> .monspommscms mo msmcoo mmmH .msmcoo map mo sompsm

.mOHHHo meraHaa pomsaam>oo .m.o ..o.o .eopmeHgmos .om-om maaoao aons on .H poem
.mOHPmvapm hypmscsH .HH masHo> .mmthommscmz Ho mzmzmo NumH .mSmCoo map Ho smopsm

.mohsom

 

o.ooH o.ooH o.aoo o.oqu o.ooH o.ooH o.Ha o.oo o.ooH o.ooH ooo.mH oom.oH H6969

 

o.oo o.oo o.on o.omo o.oo o.oo o.oo o.oo o.sm H.om Hoo.o ooo.m OHHHoma
o.o s.o o.om s.aa o.o 3.5 H.o o.o o.o o.o moo ooo :Hopcaos

o.oo o.oo a.oo o.ooo o.oo o.oo o.oo H.om m.oo o.oo oHs.o ooo.a opaom
Hoppcmo
o.o H.o H.6H o.oo o.s o.o o.o o.o a.oH o.o oao.H HNH.H apaoz
H.o o.o o.Ho ~.ss o.oH o.s o.a o.o s.oH o.o Hos.H oHH.H pmmmopaoz
Ho msoHHHHso Hooo.Ho
--Ham-MW...:wmdflimm--mm:wm-mmwwwmm1mm---mm ............
pcooamm Honesz accouom HmnESZ pcoonmm Hopssz
ocean 65Hw> acmSHOHmsm mpcmeanHnmumm GOHMmm

 

 

.wme amme .muovomhpcoo cam mnemo
wchwomhom :onmH hm pound osmm> paw .QGoEHOHQEo .mPCmsanHnmumm11.:m oHQme

 

113

.mOHHHo maHpaHaa
pcoscpo>ow .m.D ..o.Q .covwchmmz .mmnom mmsopo acnms and hpmsssm HMHoCoU .H whom
.moHpmempm hhpmsccH .HH oESHc> .mohsvommscms mo mzmcoo mmmH .msmcoo map mo smohsm

.moHHHo mszaHaa pcmscam>oo .m.o ..o.o .copmeHammz .om-om maaoao aoHes on .H atom
.moHpmHPMPm aupmsccH .HH mssHo> .mop390mmscm2 Ho mszmo NmmH .mSmcou map mo swonsm

.mopsom

 

o.ooH o.ooH a.omH.H o.ooo.m o.ooH o.ooH s.oom o.ooH o.ooH o.ooH ooo.oH Hao.o Hopes

 

w.o: :.m: m.mmm m.mm:.H o.Hm m.mm H.om d.mo m.HH :.HH «mm.H mHm onHomm
o.o o.oH o.oHH o.oom o.s o.oH o.oH o.oH o.o H.o mos oHo cHopcaos
o.oo H.mo o.ooo H.omo o.oo o.Ho o.omH o.oo o.oo o.oo Hom.o oom.o opaom

HmHPCmo

H.o o.o o.oo H.moH o.o o.o o.oH “.6 o.oH o.oH moH.m ooo.H spaoz
H.o H.o N.ma o.oHH o.o o.o o.oH o.o o.oH o.oH moo.~ amH.H 6666:9662

Aw msoHHHHsv Hooo.Hv
as as 2%-..“me..1mm“..-MAWJWWMWMWWN-mm---wm:-mm .........
Psmopom ponesz psoommm Hopssz psoonom nonesz
coopm o:Hm> psoEHOHmmm . mpcosgmHHQMPmm COHmmm

 

 

.onH .mmmH 6:? wcHsquocm
mHHHEBmm pom soHMoH an 66666 msHm> ocmx4wcoeron86 .mvsmsanHnmpmm11.om oHpme

 

 

 

114

.mOHHHo maHchaa
pamssso>oc .m.= ..o.n .sopmchmmz .mN op om masono news: was zpmsssm HmHmCou .H whom
.moHpmemHm znpmscsH .HH ossHo> .mmsspowmscwz mo mamcou mmmH .mSmcoo one mo sompsm

.moHHoo maHchaa Hsoseao>oo .m.o ..o.o .copmsHomez .om-om maaoao tones on .H puma
.mOHPmHPMPm zhpmsccH .HH mESHo> .moH390mmssmz Ho mausoo NmmH .mszoo map mo swohsm

 

 

.mopsom
o.ooH o.ooH o.ooH o.ooo o.ooH o.ooH «.1: o.oH o.ooH o.ooH oo oo H369
o.oo o.oo o.oo o.ooH 6.5m o.oo o.o o.o o.oo o.om oH.A.mH QHHHoma

II II II II I... II GHMPCSOS
o.oo o.ao o.oo o.ooH o.oo o.oo o.o o.o o.oo a.oo oH om opaom
HMHpcmo
o.o o.o o.o o.oH o.o o.s a. o.o o.oH m.Hm HH oH opaoz
o.oH - o.oo 1- N.Hm 11 o.o 1- o.oo 11 oH -1 Homeopaoz
Ho mgoHHHHso Hooo.Ho
...... 1.1mm--Mm---mm:mm-Mm-memm:mmm-wm-www-mommmww---1-..---
Pfloohwm .HmDESZ Pfimohmm .HmQEdZ Pfimohmm HGQESZ
cocoa msHm> PQmEonmsm mpcmesmHHnmpmm SOHWmm

 

 

.oomH .NmoH
.mHHHsmHsm Mom SOHmou an cocoa 03Hm> and psoSHOHmam14mpcmszmHHQMHmm11.om oHnma

115

.mOHpmHHMPm hnpmsccH .HH oESHo> .moHSHommscms mo mamcoo mmmH .msmcoo asp mo smopsm
.oOHmmo mchcHnm pzoesho>ou .m.=

.mOHHmempm sapwooeH .HH asaHo>

..o.o .eopwcHommz .om-om masoao genes on .H Hana
.mopspommscms mo msmsoo mmmH .mSmcoo esp Ho smousm

 

 

 

 

 

 

.oopsom
o.ooH o.ooH a.mooH o.ooom o.ooH o.ooH m.HoH o.omH o.ooH o.ooH ooo moo Hopes
5.6 on OHHHoom
o.oH H.NH o.ooH o.Hoo o.a o.o o.oH a.mH o.a o.o mm H :Hmpsaos
o.oo o.oo a.ooo o.oao H.om o.oo oJoo H.oo o.HH o.oH oo oo opaom

HMHPCmo
o.om o.om o.oo: N.Haa H.sm o.oo s.oo o.am o.om o.om oo oo opaoz
a.oo o.Ho o.ooo o.oHo o.oo o.oo s.mo o.oo o.oo o.oo 26H moH pmooapaoz

Ha msoHHHHEV Hooo Hv
--Hamm--maimmflwmw---mom--m2..m--mmm1mm:mm:mm:wmMm ...........
PGOOHmnH 900.832 PSO 09mm” .HQQESZ Pfimohmm .HmQESZ
venom 63Hm> pzmsonmem mPQoEQmHHQMPmm QOHwom
.oomo .Nama .amamarmaHeHaaa 66164.

.mHHHS Honda 90% coHMon an cocoa 63Hm> cam .psoeamHQSo .mpcmsanHQMHmm:-.mm oHnme

116
same. This would indicate this region made the greatest
gain in technological advance.

With so many establishments failing, only the most
economically viable survived. This industry, then, has
experienced strong pressure to accept new, more productive
technologies.

Both pulpmills and paper mills have been very stable
in terms of the number of establishments (the small number
of pulpmills relative to paper mills is due to the fact
that they are often not reported separately in the Census
of Manufactures). Pulpmills have been relatively more
successful in reducing their employment than paper mills.

While the data are incomplete, due to disclosure
rules, it is apparent the South has gained in the number
of establishments. The new plants in the South would
tend to employ the latest methods. and therefore this
region should be slightly more technologically advanced
than the others.

Research and development

Research and development plays a major role in

 

technological advance. The great majority of studies on
the subject indicate that the rate of return from R&D is

very high, usually ranging well above 20 percent.22

 

22Edwin Mansfield. 1972. Contribution of R&D to

efiggomic growth in the United States. Science 175:477

117
Yet, Table 28 shows the forest industries to be poor
performers in this area. Funding for R&D in the forest
industries as a percent of net sales for companies
performing R&D is significantly below the average for all
industries. This dismal record is sometimes defended on
the basis that much of the research and development in
the forest industries is done by the equipment
manufacturers supplying the industry. While this is true,
it also holds for many other industries. Further, the
figures on the bottom of Table 28 are only for those
companies performing R&D: it is probably justified to
assume most companies engaged in logging and sawmilling
perform no R&D. Therefore, if funding for R&D were
calculated as a percent of total net sales for the
industry, the figures would be even lower.

The above discussion is only for nongovernmentally
performed R&D. In these industries, significant efforts
are made by government and universities in research and
development.

Efficiencies

The minor emphasis placed on technological advance
in the forest industries is shown by another symptom
other than low R&D funding. This symptom is also shown
by the ratios in Table 29. The wide discrepancies
between the most efficient plants and the least efficient

plants show the potential for improvement in the forest

118

Table 28.--Funds_fog_reseapch and development
performed by the:ipre§:,industries,
1260519750

 

1960 1965 1970 1975

Lumber, wood products,

and furniture 13 13 48 68*
Paper and allied

products 56 76 178 253
Total, all

industries 10,509 14,197 18,062 23,540

COMPANY FINANCED

Lumber, wood products,
and furniture 11 NA 48 NA

Paper and allied
products 56 76 NA NA

FUNDS FOR R&D AS A PERCENT OF NET SALES
IN MANUFACTURING COMPANIES PERFORMING R&D

Lumber 0.6 0.4 0.5 0.4
Paper 0.7 0.8 0.8 0.7
Average, all industries 4.3 4.3 3.7 3.1

 

*29 of which was for furniture
Sources:

National Science Foundation. Research and development in
industry, 1975. Survey of Science Resources Series, NSF
77-324. Washington, D.C.

National Science Foundation. Basic research, applied
research, and development in industry, 1965. Survey of
Science Resources Series, NSF 67-12. Washington, D.C.

National Science Foundation, Research and development in
industry, 1960. Survey of Science Resources Series, NSF
63-7. Washington, D.C.

119

.mOHmmo
wsHchHm psoECHo>ou .m.= ..o.n .cOpmspnmmz .mHmH :HpoHHzm .moHHpmsch me :H mosmpp
Hmsomcms cam owcmso HooHMOHoccooe .ommH .mOHpmeMpm Honda mo smopsm ..H.Q.m.:

 

.oossom
m. H o. H a. H H5 mHHHs 38126
one mHHHezmm
:.H o.N m.H m.m mHopoMHpcoo 6:6
mmst wchmoH
HeeHeeac
pcMHm mp:6Hm pcmHm mpCMHm
mwmpo>m epCoHonmo pmon: owMHo>m epCmHoHpmm pmdoHe
op op op op
:psoHonmo pmoze epsoHoHMHm pmoze :psoHOHHmo pmoze epCoHoHHHo pmose Hopomm
hupmsosH
sacs-cue
ooHOHAEm Hon mehos GOHpospopm
moHSpHUCono HapHQmo H69 @6666 63Hm>

 

 

.mmmH .pcmHQ mmmpo>m 0p cam meMHm
apcoHOHmmo pmNoH: op apCoHOHmmo pmoE: mo mOHpMm dHHpmsosH
mpozcohm @003 was HoQESH map :H monszozmQNm HmpHmmo ccm cocoa osHm>11.mN mHnme

 

 

120

industries. Moreover, the differences between the two
measures (value added and capital expenditures) show
that while the average plant does not lag by a large
degree in capital expenditures per employee, there is
a greater difference in value added per production
worker man-hour. This implies that the less efficient
plants lag in the utilization of their capital,
probably by investing in outdated technologies, or
poorly managing that which they possess.
Prices

The real price movements of the products manufactured
by an industry are both a reflection of past technological
change and a factor influencing further change. A lag
in productivity relative to growth in demand should,
ceteris paribus, result in an increase in real price,
relative to other goods. If total productivity rises
faster than demand, then the real price should decline.2&

As shown in Table 30, the wholesale price index of
lumber and wood products has risen relative to that of
materials and components for construction, although the
relationship has been quite variable. Woodpulp and paper

are compared relative to the all commodities index. Prior

 

2l-J'-V.W. Ruttan and J.C. Callahan. 1962. Resource inputs
and output growth: Comparisons between agriculture
and forestry. Forest Science 8(1):68-82.

121

 

 

 

H.oo o.ooH o.oo o.ooH s.oHH o.HoH H.oHH o.ooH o.oHH mooH
o.oo o.ooH o.so o.ooH o.HHH s.sHH o.ooH o.oHH o.ooH oooH
o.ooH o.ooH o.ooH o.ooH o.ooH o.ooH o.ooH o.ooH o.ooH sooH
a.sm o.so H.ooH o.ooH o.HoH H.ooH s.HoH o.ooH o.oo oooH
o.so o.oo o.ooH H.ooH s.am o.oo p.66 o.oo o.oo oooH
o.oo o.oo o.ooH H.oo o.so o.oo o.ooH o.oo o.oo sooH
o.oo H.oo o.oo o.oo o.oo N.Ho o.oo o.oo o.oo oooH
o.oo o.oo o.ooH H.oo o.oo o.oo «.56 o.Ho o.oo mooH
o.oo o.oo o.ooH o.oo o.oo o.so o.oo o.Ho o.oo HooH
H.Bo a.mo s.aoH m.moH H.oo o.oo o.oo o.oo o.oo oooH
o.oo o.Ho o.ooH o.ooH o.oo o.oo o.ooH o.oo o.oo mooH
o.oo s.oo s.ooH o.ooH o.oo o.oo o.oo o.oo o.oo oooH
o.so o.oo o.aoH a.ooH a.oo o.oo o.oo o.oo o.oo sooH
o.oo o.so o.oHH o.oo o.ooH o.oo o.ooH o.oo o.oo oooH
o.oo o.oo o.ooH s.oo o.ooH o.oo o.ooH H.so o.oo oooH
o.oo o.oo o.ooH o.oo o.ooH o.oo o.ooH o.oo o.oo sooH
o.Ho H.oo s.ooH o.oo o.ooH o.oo o.oHH o.oo H.oo oooH
o.oo H.oa s.ooH o.oo H.ooH o.Ho o.oHH o.oo a.oo NooH
o.oo o.os o.ooH o.oo m.HHH a.oo o.oHH m.so o.oo HooH
o.oo o.ao o.oo o.Ho o.oHH o.oo o.oHH o.oo o.ss oooH
|0>Huumamm HMSPO¢ |m>HPmem HMSPO< lm>fivmflmm Hm3v0< lm>wvmflmm HGSPO< COMPOHHHPmCOO
m N H H Mom mpsoGOQEoo Homw
Hmmmm mHsmuooz Honezq mpospopm 6cm mHmHHopmz
coo? cam HmQESH
HooH n sooHo

 

.wNmHIQWMH .mpozuomm1HoQEHp popooflom Ho moxoocH mOHHm mammoHo2311.om oHnme

.oOHHmo wchcHHm pCoEQHm>ou
.pCoonopm map mo pHomom oHeosoom .mumH .pCmchmHm one

.HHe .mm .uaHeoa .m Hamper

.m.o ..o.o .sopmsHsmsz

122

 

 

.moohsom
.xoch moHpHooEEoonHHm map 0p o>HpmHomM
.xocsH SOHposhpmcoo Hop mpcozomEoo cam mHmHHmpme onp op mo.>HpsHom.H
o.oo o.ooH o.ooH o.oom o.ooH o.oom o.ooH o.oom o.ooH oaoH
o.oo o.oaH o.ooH o.oom H.ooH o.ooH o.ooH o.oaH o.osH osoH
o.oo o.ooH o.ooH o.sHm o.ooH H.Bom o.oHH o.ooH o.HoH saoH
H.oo o.HNH o.oo o.ooH H.ooH o.oom o.omH m.asH a.ooH oaoH
o.so o.oHH o.oo o.HHH o.omH o.ooH o.oHH o.ooH o.ooH NsoH
o.ooH H.oHH o.oo o.mHH o.oHH o.ooH H.ooH o.amH H.oHH HsoH
o.ooH o.HHH o.oo o.ooH o.HoH a.oHH o.HoH H.oHH o.oHH osoH
lo>HpMHom Hm5p0< lo>HpMHmm HMSpo< lo>HpMHom Hm3p0¢ Io>HpMHom Hm5p0< QOHposnmeOO

N . m .
Hommm

H . H How mpCoQOQSOO How»

one mHmHHopmz

 

mHsmoooz Hopesq mposoohm

coo; ccm HoDESH

 

A.o.psooV-.om oHnme

123

to the last 3 years, woodpulp had remained at about the
same level as the all-commodities index. The sudden
increase in the last few years is possibly due to the
necessity of raising prices to cover the costs of added
polution control equipment.

Softwood lumber shows a more variable index than the
other commodities because of swings in the housing
market.

Competing materials have generally shown less of a
rise in their price indexes than lumber. Aluminum
siding, concrete products, building brick, and gypsum
products have all declined relative to the materials and
components for construction index. This difference in
price behavior has no doubt led to the level consumption
of lumber, while the Nation's population has been expanding.
Resources

The characteristics of the raw material base an
industry utilizes also determine the direction technological
change may take. However. resources cannot be defined
without references to the level of technology. Technolog-
ical knowledge has been defined as ”information which
improves man's capacity to control and to manipulate the
natural environment in the fulfillment of human goals, and

to make that environment more responsive to human needs."25

 

giNathan Rosenburg. 1972. Technology and American
Economic Growth. New York: Harper & Row.

124

It is then technological knowledge which determines
which materials in the environment the forest industries
can utilize to satisfy the needs of our society.

Data in the area of raw material quality are
virtually nonexistent for stumpage, sawlogs, or pulpwood.
The general concensus is that, overall, the quality of
sawlogs, at least as reflected by sawlog size, has
declined. In contrast, there is no good reason to
believe that pulpwood quality has changed.

Technological changes have redefined resources for
the forest industries, however. Semichemical pulping has
allowed the utilization of small, low-grade hardwoods for
pulp. The chipping headrig has also allowed use of small
material. This fact has led Irland to declare that ”The
major role of technological development in United States
forest industry over this century has been one of
resource-expanding change.”2é

The changing resource base in the forest industries
is the result of two forces: One is the expansion to the
physical limits of traditional resources, such as softwood

sawlogs, limited by the allowable cut policies of the

 

géLloyd C. Irland. 1973. Resource endowment, technology,
and trade: The case of U.S. timber resources.
Unpublished paper presented at meeting of the Southern
Economic Association and Southern Forest Economics
Workers, Houston, Texas.

125

U.S. Forest Service; the second is the realization of the
opportunity represented by huge amounts of harvesting
residues. Utilization of these residues was viewed both
as an untapped raw material resource and as a response to
rising pressures by the public and government to lessen
impacts on the environment.22

The difficulties of prediction are numerous, the
last not being that of defining an invention or innovation.
Should the high-strain bandsaw be classified as a separate
innovation from an ordinary bandsaw? Or should both
be included in a general class of headrigs? Prediction
must be based on counting, which cannot be done without
definition. Lack of consistent definition, because of
evolving techniques and equipment, makes prediction
difficult.2§

Future technological change is expected to produce
more output from a given amount of raw material.
Projections by the U.S. Forest Service place softwood
lumber yields 15 percent higher in 2000, based on 1970
yields. The increased yield for hardwoods is projected to
be 5 percent. Both of the projections are based on the

relative price of lumber rising at 1.5 percent per year.

 

22Richard L. Porterfield. 1977. 92. cit.

2&5. Colum Gilfillan. 1952. The prediction of technical
change. Review of Economics and Statistics 34:368-385.

126

Pulp yields are also expected to increase by about 7
percent over their 30-year projection period, based on

relative prices rising 0.5 percent per year.22

 

22U.S. Forest Service. 1973. The outlook for timber in
the United States. USDA Forest Service, Forest
Resource Report No. 20. Washington, D.C. U.S.
Government Printing Office.

SUMMARY AND CONCLUSIONS

Technological change in the four forest industries of
logging, sawmilling, pulping, and papermaking has been
modest. Between 1958 and 1976, the average annual increase
in the geometric index of technological change for the four
industries was only 1.4 percent. Such an indicator of
technological change is based on the changes in production
unaccounted for by concurrent changes in labor and capital
in the industry. Its interpretation is that of an index of
shifts in the production function, or alternatively, that
of a measure of total factor productivity. It may also be
considered as an indicator of progressiveness, inasmuch as
the index of technological change measures the success of
the industry in producing extra output, in excess of
relative changes in capital and labor. The index may also
be considered an indicator of the adaptiveness or
adoptiveness of an industry in utilizing the technological
opportunities available.

A descriptive evaluation of possible future technology
for the four industries reveals that there are both
improved versions of currently employed processes and
totally or radically new technologies available. The range
and magnitude of new technologies varies between the indus-
tries. with the greatest opportunities apparently in the
logging and sawmilling industries, and less in papermaking.
Capital requirements for new technologies also varies

127

128
considerably between the industries, with less investment
required for new machines in logging and sawmilling than in
pulping or papermaking.

One of the four econometric models measuring technol-
ogical change also allowed estimation of the elasticity of
substitution between labor and capital. The constant
elasticity of substitution function placed the elasticities
for all the industries above one, ranging from 1.2 for saw-
mills and planing mills, to a high of 1.7 for pulpmills.
Estimates for all in the elastic range is reasonable
because all have increased their levels of capital, while
increasing output usually without increasing labor.
Apparently the price of capital relative to that of labor
has also declined, evidenced by the proxy price weights
calculated for the arithmetic index (Appendix Tables B1
through B4).

The geometric technological index weights changes in
labor and capital by their respective elasticities of
output. An estimate for the elasticity of output with
respect to capital is capital's share in income, which has
been increasing for all of the industries. This parameter
of the production function has increased only slightly in
pulping and papermaking, but has increased fairly substan-
tially in sawmilling and even more so in logging. An
increasing elasticity of output with respect to capital
means that it is becoming relatively easier to increase

output through the addition of capital than of labor (the

129

two elasticities were assumed to sum to one in the model).
The declining relative price of capital (evidenced from the
arithmetic model) and the additions to each industry's
capital reflect this trend.

Several qualitative indicators of technological
change also show varying rates of progress, although not in
such an exact fashion. All of the industries have
increased their output. The real value of output in
logging camps and contractors grew the most (150.3 percent),
while in sawmills and planing mills, general. it grew the
least (20.6 percent). Real gross assets also increased,
and the same industries were first and last in this
ranking. Table 31 summarizes most of the qualitative
indicators considered.

Employment over the nineteen year period declined in
three of the industries, with sawmilling decreasing by
over 30 percent. The sole industry to increase employment
was pulping which climbed slightly more than 10 percent.
Per employee productivity, the most commonly used
measure of progress in manufacturing, grew almost
uniformly in three of the industries, with increases
between 60 and 75 percent for the nineteen years. Labor
productivity in logging increased twice as much as the
others, increasing by over 150 percent. Annual growth
rates in labor productivity for pulping and papermaking
ranged slightly below the national average for

manufacturing of 2.7 percent, with sawmilling slightly

130

Table 31.--Percent change in some qualitative indipatopg
of technological change in the forest
industries, 1958-1976f

 

 

Indicator Logging Sawmilling Pulping Papermaking
Percent

Real output 150.3 20.6 78.7 57.4
Real capital 100.0 13.1 90.3 96.5
Employment ‘003 '31 02 10.6 -209
Output/unit

labor 151.0 75.3 61.6 62.1
Output/unit

capital 25.2 606 -601 '1909
Annual growth

in labor

productivity 5.0 3.0 2.6 2.6
Number of

establishments 3.4 -48.4 1.7 -1.4
Capital/employee 108.9 71.1 67.4 105.3
New capital

expenditures/

employee 89.4 160.7 300.7 111.0
Source:

Bureau of the Census. 1972 Census of Manufactures. Volume
II. Industry Statistics. Part 1, SIC Major Groups 20-26.
Washington, D.C.

131

above and logging well above the national average of
growth in labor productivity.£99

The number of establishments in each of the industries
in this group has remained fairly stable over the years,
with the exception of sawmilling. In this manufacturing
activity, the number of establishments as recorded by the
Bureau of the Census has declined by almost one half. With
such a reduction, it is probable that the least efficient
and progressive plants have been the most likely to cease
production. All in the group have achieved output growth
by increasing in size (in capital terms), and not generally
by establishment numbers.

Regional shifts have occurred, however. Some
movement in logging establishments has been from the
Pacific Coast to the South. Sawmilling establishments
declined in all regions. In pulping and papermaking, there
has also been a general shift to the South.

Research and development in the forest industries is
not great. For both lumber and paper. funds for R&D as a
percent of net sales in manufacturing companies performing
R&D is less than one percent, compared to greater than

three percent for all manufacturing. Universities,

government, and manufacturers of equipment for the

 

lQQ—LE. Henneberger. 1978. Productivity growth below

average in the household furniture industry. Monthly
Labor Review 101(11):23-29.

132
industries do perform R&D that affects the forest
industries, however.

Changes in raw materials have also probably occurred.
Reliable data are not readily available, but logs have most
likely become smaller.

An additional factor may be that the entire forest
products industry has concentrated on advancement in other
areas, such as plywood, particleboard, and fiberboard.

Technological change has accounted for about 60
percent of the increases in per employee productivity for
three of the industries, while additional capital per
employee produced the remainder. Papermaking differed,
with greater capital intensity producing 72 percent of
the growth, with technological change accounting for 28
percent.

Capital per employee increased in all of the
industries, however this factor change did not produce
equal results among the four industries. Logging
augmented their investment per man the most, and also
succeeded in leading the group in most of the qualitative
measures of technological change, especially growth in
labor productivity and the change in real output. Yet
in papermaking, an increase in capital per employee of
almost the same proportions resulted in low growth in
labor productivity and the geometric index of technological
change. Growth in the ratio which measures the rate of

investment per man was highest in pulping. This fact

133
coupled with the others, suggests that it has been
easier, and less expensive in terms of capital, to produce
technological change in logging than in any of the other
three industries in the group.

The two industries producing solid wood products,
logging and sawmilling, improved their output per unit of
capital ratios, while the two fiber industries, pulping
and papermaking. suffered declines in their output per
unit capital input ratios. In general, growth in an
industry's capital investments will lead to expansions
in its productivity. All of the industries have expanded
their expenditures per employee, as shown in Table 32.

Three of the forest industries in this group were
ahead of the national average in new capital investment
per employee in 1958, and in the cases of the two fiber-
based industries, much ahead. Sawmilling was only
slightly behind in 1958, but had fallen proportionately
further behind by 1976. By 1976, logging camps and
contractors had also fallen behind in new capital invested
per man. In contrast, pulping and papermaking were still
well ahead of the national average, and for the former,
more than five times as much was invested per employee as
in the average U.S. manufacturing industry. Yet,
technological change (measured by the geometric index) was
not spectacular, and growth in per employee productivity was

slightly below the national average.

134

Table 32.--New capitaigexpenditures per employee ig four
forest industries and all manufacturingL
1958 and 1976.

 

 

Industry New Capital Expenditures
Per Employee
1958 1926
Logging camps and
contractors 954 1,807
Sawmills and planing mills,
general 526 1,371
Pulpmills 3,028 12,134
Papermills, except
building paper 1,971 4,158
All manufacturing 620 2,300
Source:

Bureau of the Census. 1972 Census of Manufactures. Volume
II, Industry Statistics. Part 1, 810 Major Groups 20-26.
Washington, D.C.: U.S. Government Printing Office.
Bureau of the Census. Various Annual Survey of Manufactures.
Washington, D.C., U.S. Government Printing Office.

In the future, trends will probably not vary much
from what they were in the past, with a few exceptions.
New capital expenditures per employee of less than the
national average should bring the growth in labor
productivity down somewhat in logging and sawmilling.
However, evidence examined above suggest that it is easier
to produce technological change (shifting the production
function) in these two industries than in pulping and

papermaking, at least in terms of investment per man.

While there is no hard and fast evidence yet to suggest it,

135

the decline in the number of sawmills may slow in the
future. Distances economical to transport logs will
tend to limit the area from which the average mill can
draw raw materials and hence its size. Improved processing
technologies can compensate for the trend to smaller logs.

While new capital investment per man has been high
in pulping and papermaking, advances in technological
change have not been as great per dollar. Alternatively.
it has been difficult to increase total factor productivity
in these two industries. In terms of the geometric index,
pulping will probably not be able to retain its past rate
of 2.4 percent annually. Should past trends in capital
share and real output per employee continue, capital
requirements will probably be too great to maintain
pulpings past rate of technological improvement. If the
industry continues to invest as it has in the past,
productivity could fall even lower. Technological change
in papermaking will probably remain low, as it has in the

past.

APPENDICES

APPENDIX A

GEOMETRIC INDEX UTILIZATION RATES

Table A1.--Percent capacityputilizedI lumber productigp,
1958—1976, by qparter.

 

 

-------------- Quarter----—--—-----

Year I II III IV Average
1958 81.0 92.1 100.0 97.7 92.7
1959 96.1 100.0 100.0 95.2 97.8
1960 91.8 100.0 95.1 81.8 92.2
1961 79.9 93.4 93.8 87.3 88.6
1962 83.8 99.2 100.0 93.5 94.1
1963 89.9 98.2 100.0 95.0 95.8
1964 93.8 98.0 100.0 89.4 95.3
1965 88.8 95.8 100.0 94.7 94.8
1966 92.8 100.0 94.8 84.5 93.0
1967 87.1 93.6 92.0 89.2 90.5
1968 89.7 100.0 98.9 93.6 95.6
1969 95.2 100.0 96.1 92.9 96.1
1970 89.9 93.1 91.9 86.0 90.2
1971 91.3 100.0 96.6 91.5 94.9
1972 93.0 100.0 98.8 9 .0 96.7
1973 97.1 100.0 100.0 9 .6 97.9
1974 91.0 100.0 88.9 70.3 87.6
1975 69.3 87.7 90.7 85.6 83.3
1976 92.1 96.1 100.0 97.6 96.5
Note: The above capacity utilization figures were used as

a proxy for logging camps and contractors.

137

Table A2.--Percent capacity utilized, woodpuipppoduction,
i958-1976y_by_quartep.

 

-------------- Quarter------—--------
Year I II III IV Average
1958 99.0 97.1 100.0 100.0 99.0
1959 100.0 100.0 96.3 95.6 98.0
1960 100.0 96.7 92.5 91.9 95.3
1961 94.0 100.0 95.4 100.0 97.4
1962 100.0 100.0 94.4 94.8 97.3
1963 98.8 100.0 97.6 100.0 99.1
1964 100.0 100.0 94.8 95.4 97.6
1965 96.8 96.5 95.1 95.7 96.0
1966 97.3 100.0 97.3 98.4 98.3
1967 99.5 100.0 93.2 94.1 96.7
1968 100.0 100.0 94.4 93.4 97.0
1969 93.9 100.0 96.9 98.9 97.4
1970 99.3 100.0 95.3 96.5 97.8
1971 98.9 100.0 96.6 100.0 98.9
1972 99.0 100.0 95.7 96.9 97.9
1973 99.0 100.0 97.1 97.2 98.3
1974 97.3 100.0 95.2 93.1 96.4
1975 83.2 76.6 81.9 89.3 82.3
1976 97.1 100.0 93.2 94.8 96.3

 

138

Table A3. --Percent capacity utilized, paper production,

1958-1976, by quarter.

 

Year

1958
1959
1960

1961
1962
1963
1964
1965

1966
1967
1968
1969
1970

1971
1972
1973
1974
1975

1976

-------------- Quarter------------

I

94.8
100.0
100.0

97.5
100.0
100.0
100.0
100.0

100.0
100.0
100.0
100.0
100.0

100.0
100.0
100.0
98.8
82.8

99.2

II

94.9

100.0
100.0

100.0
100.
100.
100.
100.

\O
(I)
OOH OHOO\O 0000

III

93.4
94.2

91.8

94.6
93.0
95.0
95. 0
95 O

97.1
93.3
94.3
95. 7
921

94.3
95-7
95.1
95 2
85 5

93.7

IV

100.
96.
94.

100.
96.
100.
98.
97.

99.

(I) Old-FOO O\N\OU-F oomouo \OCDO

Average

95.8
97.8
96.7

\0
0\
N {TWP-‘04? FWCDCDH NWCD

 

APPENDIX B

ARITHMETIC INDEX WEIGHT CALCULATIONS

139

moHd.

 

mmmw. 1 H u panoz HmpHmmo
ammo. n
Am.oHH\Ho.oom-o.moso~+Hs.ooH\o.oomo 1
. mcmOHh. OWN -1 PSWHQS .HOQMHH

.Hm oHnme
Eopm coxmp .mmmH How .mHopomecoo ocm mmEmO wsHmwOH pom mHmmem cm mH BOHom
.prchs acncH - H 1 HcoHcs HeHHaso

wamch HmpHQmfi\AHHonmmuomopm msz>V~+mxoccH pamshonsokHHOHzmmv u pcwHoB Honoq
xcczH psoEHonEo\HHOHamm .

.mMHSEHom mcHBOHHom map wQHms oopMHSOHmo who: xcoCH oHpcSQpHHm ccp Hop mpanoz one

140

.ooHummmH pm ooxoch .HcQESC .mooHOHmEm HH¢I

.ooHnmmmH pm coxooQH .mpommm ccme Ho 63Hm> mmoawm
.mHMHHoo mo mCoHHHHE .mmoonmEm HHm .HHoHHmmI
.mHmHHop mo mQOHHHHE .cHSpommscmE an cocoa osHm>m

H

 

moom. moms. p.66 o.omo m.moa o.mmHm omoH
oHom. room. a.oo o.oo: H.ooo H.oHsH osoH
oooo. sooo. o.ooH o.Hoo m.oao o.oooH smaH
oomo. osmo. o.oHH o.moo o.moo o.mooH omoH
omHm. ammo. o.HHH o.oom o.omo o.ooHH msoH
ammo. Hmso. o.oo o.omm o.oo: o.moo HmoH
ammo. oomo. H.ooH m.oom m.ooo o.omo omoH
ammo. msoo. o.moH m.oom o.oo: m.ooo mooH
oooo. ooso. o.mo m.soH o.omo o.Hos oooH
omao. memo. o.oo m.oaH o.omo H.ooo nooH
oHHs. oooo. H.HoH m.HoH o.oHo o.ooo oooH
ooos. oHoo. o.moH o.ooH o.Hoo a.oHo oooH
moon. Homo. o.mo o.ooH o.mom o.ooo sooH
Homo. oHHo. o.moH m.ooH o.Hom o.omo oooH
omoo. oooo. m.oo s.omH o.msm o.oo: mooH
ooHo. HHoo. o.oo o.mHH m.msm o.oo: HooH
oHHs. Homo. o.moH a.mHH o.mom o.oo: oooH
ooHo. sooo. m.ooH m.oHH o.oom o.moo oooH
ooHo. osoo. o.ooH o.ooH o.omm o.omo oooH
pszoz pcmHmz AxocQHv Axochv IHHOHzmm Mucous osz> Homw
HcpHaco acnmg macnnH chHHaco m

 

 

.HHom on HapnsosH .nschcHacHuc HruHcs UHcosH cHecsspHs<-.Hm cans

141

.oOHmmo wchcHHm pcoecpc>oo .m.:

..o.a .copmcHnmmz .mcHSpcmmssmE mo zc>usm Hmzcc< mSOHHm> .msmcco map Ho smmpsm

.oonmo msHpsHHm pCmECHm>ou .m.:
.nchnHHcpm appusocH .HH csch>

..o.n .:opmcH:mm3

.om-om mascao HcHsz on .H puma

.mcHSpompzsms Ho machu NumH .mSmcco cap mo smopsm

.mcHsom

A.U.pcoov11.Hm mHQmB

.ooHuwan pm poxoccH .Hmnesn .moomoHQEm HH<|

.ooHuwmaH pm ooxmch .mpommm coxHH Ho msz> mmopwm

.mHmHHoc mo mCOHHHHe .mooHOHQEm HHm .HHonmml
.mHMHHoc mo mcoHHHHE .mHzpommssme an cocoa msz>m

 

142

oomm. ooms. o.oo o.mmm m.aooH s.ooso omoH
osom. Hmoa. o.oo o.mHm o.mmsH o.Hoom omoH
somo. oomo. o.om o.ooH m.oosH H.oooo smoH
osoo. omHo. o.oa o.ooH a.oooH a.osao omoH
oomo. oomo. o.mo o.ooH m.mmmH o.ooom mmoH
oooo. oooo. o.oo o.ooH m.ooHH o.oon HsoH
Hoom. oon. m.om o.omH o.mooH m.momH osoH
moon. Homo. o.ms o.omH o.HHoH o.omom oooH
moon. HoHo. o.Hm o.sHH o.oso m.sHoH oooH
ammo. oaoo. o.os o.HHH o.omo o.oooH mooH
sHoo. oooo. o.om o.oHH o.Hao a.oooH oooH
mono. aooo. o.mo H.ooH o.omo m.oosH oooH
omoo. smoo. o.oo o.oHH m.moo o.oHoH sooH
omoo. ammo. o.oo H.ooH o.HHo H.omoH oooH
oHoo. ooHo. o.oo m.om o.oma m.omoH mooH
ssoo. oooo. H.mo o.oo o.mom o.HomH HomH
omoo. ammo. o.mo o.moH m.Hoo o.mmoH oooH
msoo. omoo. o.moH o.oHH o.ooo s.oooH mooH
moon. asso. o.ooH o.ooH o.omm o.ooHH oooH
pamHoz panoz Axochv AxmcsHv IHHonmm Icocpm 65Hm> Ham»
HspHano acncH macncq anpHaco m H

 

 

.Hmom on appososH .uschcHschc Hnuch acosH chcscpHa<-.mm cans

143

.moHpmo wchCHHm psoECHo>oo .m.D

“.o.Q .copwsHsmmz .moHSpommscms mo ho>shm stcc¢ msoHHm> .mszco onp mo scopsm

.oOHmmo msHpsHHm pzoECHo>ou .m.=
.nchanch aapnsosH .HH csch>

..o.a .:OpwsH:mmz

.om-om mascao ncHss on .H puma

.moHSpcmmssmz Ho mamcoo NBQH .msmcoo map mo scopsm

.oopzom

H.o.psocV-1.mm cHnss

144

.ooHnmmmH pm coxcccH .Honss: .mooonmEo HH<I

.ooHummmH pm ocxmccH .mpommm pome mo 63Hm> mmopwm

.mHMHHoc mo mCoHHHHe .momonQEc HHm .HHopzmml

.mHMHHoc mo mQOHHHHE .oHSpommssms an cocoa 63Hm>m

H

 

oars. mmoo. o.oHH o.omo o.oom o.omo omoH
oooo. HHoo. o.mo o.mmm o.HoH o.Hmo omoH
omoo. osHs. o.oo o.omm m.ooH o.oms omoH
oHos. mooo. H.os o.oom m.ooH m.moo omoH
anon. ommo. o.oa m.ooH m.omH o.oom mmoH
soon. oooo. H.moH o.oom o.ooH m.moo HmoH
ooos. mono. o.ooH o.oom m.ooH o.oo: omoH
ooso. sooo. m.ooH o.oom o.moH o.mom oooH
omoo. oaHo. o.ooH m.moH o.ooH m.Hom oooH
woos. moHo. o.ooH o.ooH o.omH m.ooo mooH
oooo. moss. o.mHH o.omH o.omH o.oom oooH
mooo. HoHs. m.ooH o.ooH o.mHH H.osm oooH
osoo. smHo. m.moH H.ooH o.oHH o.Hoo sooH
ammo. oomo. o.ooH o.ooH o.moH o.oom mooH
sass. oooo. o.mo o.HoH o.oo o.ooH mooH
oooo. moor. o.oo m.moH o.mo m.ooH HooH
Homo. mooo. H.moH o.oHH o.oo m.Hom oooH
mooo. moHo. H.moH o.mHH H.mo o.omm oooH
soHo. oHoo. o.ooH o.ooH o.os o.ooH oooH
psmHoz pancz AxonsHv AxonsHv (HHopzmm lomoom msHm> Hmow
chHaco acncH macncH chHHaco m H

 

 

.HHom on HapmsocH 1nscHHcHschc HouHc; xcosH cHecscHHLo1-.om cHnes

145

.ooHHmo wsHpsHHm pccacnm>oc .m.:

..o.Q .copmcHnmmz .moHSpomHScmz mo ac>psm Hmzzc< mSOHHm> .msmcmu mzp mo swopzm

.ooHHHo mchsHpm psoECHo>om .m.:
.nchancpm aapnsosH .HH csch>

..o.o .scpwsHancz .om-om mascao acHss on .H Hana
.mmHSpcmmssmz mo mzmzmu mmmH .msmcoo 62p mo smmhsm

.copsom

H.o.pscco-1.om cHnee

146

.ooHnmmmH pm umxoccH .Hmnezc .momHOHQeo HH¢I
.ooHummmH pm omxmccH .mpmmmm omme mo csHm> mmopo
.mHmHHoo mo mGOHHHHE .momHOHmsm HHm .HHopzmml

.mHMHHoo mo mcoHHHHS .opspommscms an ocoom msHm>m

H

 

mmom. moon. H.mm m.ooo m.osoH o.mmm: omoH
mmom. omom. H.mm m.moo o.oooH m.omoo osoH
ammo. mmso. o.oo o.mmm m.oooH o.oom: smoH
mHom. oon. o.oo m.mHo o.mooH o.oHoo osoH
mmom. oars. o.oo o.oom o.oosH m.ooom mmoH
omom. omom. o.ooH o.oom o.oooH o.oHom HaoH
ooom. osHa. o.ooH o.mmm m.oooH m.ooom osoH
mHHo. oooo. H.ooH m.oom o.oomH o.omom oooH
ommo. oaso. H.ooH m.osm o.ooHH o.mssm oooH
moon. oooo. o.ooH o.omm m.omHH m.ooom mooH
mmsm. osmo. o.moH o.on m.oooH m.ooom oooH
omom. soHo. o.HoH o.ooH o.oom m.ooom oooH
omoo. mmoo. m.oo o.ooH o.mHo o.omoH somH
ooHs. msoo. o.oo o.ooH o.mmm m.mooH mooH
moor. mooo. o.ooH o.moH o.oom m.oooH mooH
ooos. oooo. o.ooH o.moH o.oHo m.oooH HooH
omHo. oooo. H.ooH o.mmH o.omo o.msoH oooH
oooo. mooo. m.moH o.oHH m.oma o.omoH mooH
soHo. moms. o.ooH o.ooH o.Hoa m.mooH oooH
panoz pzmHoz Axocch Axocch IHHouzmm looccm msz> Hmcw
HmHHaeo acpcg macnsH chpHaco m H

 

 

.Hmom on sponsosH 1cscheHschc Hschs scosH cHHcscHHuo-1.sm cane

147

.mchmo msHchHm pSmECHo>oo .m.:

..o.Q .copmcHnmmz .moHSpommscmz Ho zm>Hsm Hmscc< mSOHHm> .msmcmo map Ho scopsm

.oonmo wsHchHm pscsspo>ou .m.:
.cernHHcem snpmaosH .HH csch>

..o.o .scpmsHsmcg .om1om nasceo ucflss on .H puma
.mothommszmz Ho mSmCoo mmmH .msmcoo 85p mo sompsm

.ocpsom

A.G.Pcoov11.:m mHQmB

APPENDIX C

ARITHMETIC INDEXES OF TECHNOLOGICAL CHANGE

148

.pcpmHHmc .mpspommzcma an poops osHm>
.copmHHoc .Hmnesc .mooHOHmEo HH<I

w
m

 

.copMHmoc .mpommm Umme Ho osHm> mmopwa
ooo.H o.mmm o.Hm o.mmm omoH
HHo.H o.omo m.mo m.osm omoH
mHo.H o.omo o.om o.omo ssoH
ooo.H o.omo o.oo o.mom osoH
oHo.H o.oom o.oo o.oom mmoH
omo.H o.omo o.oo o.omo HsoH
oms.H o.oma H.oa H.moo osoH
Hmm.H o.omo o.os o.omo oooH
mHo.H o.oso o.oo o.omo oooH
omm.H o.mso o.om H.moo aooH
oom.H o.ooo o.ms o.omo oooH
oom.H H.mmo m.om o.ooo oooH
omH.H o.omo H.os H.moo sooH
moH.H o.oHo H.mm o.mmo mooH
moH.H o.moo o.oo o.Hos mooH
moH.H o.oo: o.Ha o.mso HooH
amo.H o.omo H.os o.omo oooH
omo. o.mHs o.om o.mos oooH
ooo.H o.omo m.Hm m.ooo oooH
Ho nscHHHHso Hmosccsccpo Ho mscHHHHso
scosH OHHcsapHpo Nocooo cch> Hecm Mscncq HHmpHaco Harm ace»

 

moOHHm pcmpmcoo

 

HHsm on HornsosH sH cosmos Hcchchscccp Ho acocH chcerpHno1-.Ho cans

149

.mcHHHo wchsHHm psmscpc>ou .m.:
.nerancpm snpnsosH .HH csch>

.mOHmmo mchcHHm pzoEQHo>ow .m.:
..o.n .:opw:H£mmz .mopzpommscmz Ho >m>Hzm Hmssc< msoHHm>

.msmCoo mzp Ho swopsm

..o.o .scpwchncg .om1om mascso Home: on .H puma

.moHSpommzzmz mo msmcoo NmmH

.mzmcoo ozp mo smmpsm

.couzom

A.©.pCooV11.Ho mHQme

150

.cmpMHmoo .oHSpommzcms an cocoa mSHm>N

 

.uopmHHop .Hmness .mcchonEo HH<M
.copmHHoc .mpommm come mo msz> mmoHUM
soo.H o.HooH H.ooH H.mooH omoH
mom.H o.momH m.ooH o.msoH omoH
ooo.H m.HosH o.ooH o.oooH omoH
ooo.H o.oooH m.omH m.ooom mmmH
Hso.H m.HooH o.ooH m.ooom msoH
ooo.H o.oooH m.ooH o.oooH HmoH
Hmm.H H.mooH o.omH o.mHoH osoH
ooo.H o.osoH H.BmH o.HooH oooH
mHo.H o.oooH o.omH o.HooH oooH
ooo.H o.oooH o.ooH o.oooH mooH
ooo.H o.HooH o.ooH o.oooH oooH
omm.H o.oooH o.Hom o.mooH oomH
mmo.H o.HosH o.ooH o.oomH oooH
oom.H o.oooH o.mom m.mHoH oooH
mom.H m.oooH o.oom o.oooH mooH
moH.H H.HomH o.on m.omsH HooH
ooo.H m.oomH o.oom o.HooH oooH
moH.H o.mooH m.mom o.oHoH oooH
ooo.H o.ooHH o.oom m.oooH oooH
acosH cHHcsapHso Ho nacHHHHso Hmosmmsccev Ho nscHHHHso
commpom wcccod osHm> Hmcm Mponmg MHmpHQmu Hmom use»

 

moOHHm paupmcoo

 

Hmsm on HupnsoaH 6H mucosa Hcchchsnccp He scosH chcsapHao-.mo cans

151

.moHHHo wchcHHm pscECHo>oo .m.:

..0.Q .copwcHnmms .moHSpommssmz Ho >o>Hzm HosCC¢ mSOHHm> .mSmCoo onp mo swopsm

.oOHHmo wchcHHm pcmsspo>ow .m.:
.mOHpmeMpm hppmzocH .HH mesHo>

H.0.Q .CoprHnmmz

.om1om nascao schs on .H puma

.moHSpommzcmz mo mSmCoo NumH .msmsco cap Ho smcpsm

Hoopsom

A.u.p:oov .No mHQmB

152

.copmHmcc .opzpompszms an @6666 csHm>N

.cmpMHmcc .Hmnesc .momHOHmsc HH<I
.oopMHmco .mpcmmm oome mo msHm> mmopo

N

H

 

o.oom
o.mmm
m.mom
o.oom
m.mom
o.mmm
o.oo:
H.mso
o.Hso
o.oom
m.omo
o.oom
o.moo
o.omo
m.HHm
H.oom
o.mom
o.omm
o.ooH

m.oH
H.oH
o.mH
o.oH
o.oH
o.oH
o.oH
o.oH

COUNVMOChrquDC%vL#
33:? (“WWW-3' “min
«1 Pi

HHHHHHHHH

o.oomH
m.oHoH
o.oom
o.ooo
o.Hoo
o.moHH
m.mHmH
H.HooH
m.osoH
o.oooH
o.oooH
o.oHo
o.omo
o.oom
o.omo

omoH
osmH
osmH
osoH
msoH
HmoH
osoH
oooH
oooH
sooH
oooH
oooH
sooH
oooH
mooH
HooH
oooH
oooH
oooH

acosH cHrcaapHao

Ho nscHHHHso
wocooo cch> Hscm

HmUGMmsogev

N

Honmq

Ho mscHHHHsv
HHcHHaco Hccm

 

mooHHm pCMpmcoo

 

HHom on HapusosH cusses HccHuchscccp Ho acocH cHecsawHao11.oo cane

153

.ooHHHo mchcHHm pcoschc>oo .m.:

..o.a .CopmcHnmmz .moHSpcmmszmz mo zo>Hsm Hmzcc< mSOHHm> .mSmsmo map Ho smcpsm

.ooHpHo msHpQHHm psoEQHm>ou .m.:
.mcHHnHHeHm sponsosH .HH cech>

..o.n .:opwsH:mmg

.om-om naacpo Home: on .H puma

.mopzpommscmz mo mamcoo NumH .msmcoo ocp mo smohsm

Hoopsom

H.o.psccV1-.oo cHose

154

.cmpMHmoo .mHSpoMHscma an cocoa csz>N
.copmHmoc .Honesc .mocmonEc HHom

.ompMHHoc .mpmmmm omxHH Ho 03Hm> mmopw

H

 

ooH.H m.omom o.mmH o.Homo ommH
Hmo.H m.ooom o.omH o.mmoo omoH
oom.H m.ooom o.ooH m.oooo omoH
ooH.H o.moom m.HoH o.mooo omoH
ooo.H o.oomm o.omH o.Hmoo mmoH
Hoo. o.mmom o.moH o.oooo HaoH
moo. o.omHm o.ooH o.oooo osoH
ooo.H o.oomm o.HoH H.HHoo oooH
mam. o.oomm o.ooH o.mooo oooH
soo. m.mon o.ooH o.Homo sooH
ooo.H o.mon o.ooH o.ooHo oooH
smo. o.HooH o.moH o.mHoH oooH
Ham. m.oooH o.omH o.HaHs oooH
moo. o.oooH o.omH o.moom oooH
ooo. o.HooH o.moH o.ooao moaH
ooo.H o.HmoH o.moH o.HHoo HooH
moo.H o.oooH m.ooH o.omoo oooH
ooo.H o.mHoH o.ooH o.omHo oooH
ooo.H m.mHoH o.HoH m.maom oooH
xeosH ercesHHao Ho nscHHHHso Huosenscsso Ho nscHHHHzo
Umwflflmm NU®©©< mSHm> H.mmm MHOQMHH anmpHmnmmv H.mmm Ham»

 

mooHHm pQMpmcoo

 

Hmom on aapnsosH sH cmccnc HccHuchsaecH He accsH cHHescpHno-1.so cane

155

.oonmo msHpsHHm pCoECHc>ou .m.:

..o.a .:opwcH£mm3 .moHSpomHscmz mo zo>psm Hmsccs mSOHHm> .mscho opp Ho smopsm

.oOHmmo wchcHHm psoECHc>ow .m.:
.nchnHHch sapusosH .HH csch>

..o.o .scpmsHanez .om1om nascso achs on .H Hana
.mcHSpomHscmz Ho mamcoo NmmH .msmCoo mzp Ho smopsm

Hoopsom

A.U.pcocv11.:o oHQmB

APPENDIX D

STRICT COBB-DOUGLAS PRODUCTION FUNCTIONS

156

Strict Cobb-Douglas production functions for the forest
industries:

Logging camps and contractors

16g Y = -1.1619 + 1.1868 log K - .1868 log L R2=.771
(-1567)

Sawmills and planing mills, general

log Y = 3.4068 + .8991 log K + .1009 log L R2=.634
(.0835)

Pulpmills

-1.5087 + 1.1046 log K -.1046 log L R2=.741

log Y
(.1536)

Paper mills, except building paper

2
log Y = 4.6416 + .5037 log K + .4962 log L R =.769
(.0752)

C ITED LITERATURE

CITED LITERATURE

Abramovitz, Moses. 1956. Resource and output trends
in the United States since 1870. American Economic
Review 46(2):5-23.

American Paper Institute. 1977. Statistics of paper
and paperboard.

Anonymous. 1976. Where's pulping headed? A review
of state of the art. Pulp & Paper 60(9):?8-80, 89.

. 1978. Funded research plan. The Institute
of Paper Chemistry. Unnumbered report.

 

Arrow, K.J., H.B. Chenery, B.S. Minhaus, and R.M. Solow.
1961. Capital-labor substitution and economic efficiency.
Review of Economics and Statistics 43:225-248.

Barnes, L. Admiral, R.L. Pike, and V.N.P. Mathur. 1976.
Continuous system for the drying of lumber with
microwave energy. Forest Products Journal 26(5):31-42.

Bennett, W. 1978. Sawmill technology outruns industry's
skill at using it. Forest Industries 105(1):28-29.

Bentley, William R. 1970. Technological change in the
forest industries--a problem analysis. The University
of Wisconsin Forestry Research Notes, No. 151.

Biltonen, Frank E., J.R. Erickson, and J.R. Mattson. 1974.
A preliminary economic analysis of whole-tree chipping
and bark removal. Forest Products Journal 24(3):45-47.

Boretsky. Michael. 1975. Trends in U.S. technology: A
political economist's view. American Scientist 3(1):
70-820

Bousquet, D.W. and I.B. Flann. 1975. Hardwood sawmill
productivity for live and around sawing. Forest Products
Journal 25(7):32-37.

Brown, E.H. Phelps. 1957. The meaning of the fitted
Cobb-Douglas production function. Quarterly Journal of
Economics 71:546-560.

Brown, Murray and Joel Popkin. 1962. A measure of
technological change and returns to scale. Review of
Economics and Statistics 44:402-411.

157

158

Brown, Murray and John S. de Cani. 1962. Technological
changes in the United States, 1950-1960. Productivity
Measurement Review 29:26-39.

, and . 1963. A measure of technological
employment. Review of Economics and Statistics 45:386-
39 -

, and . 1963. Technological change and the
distribution of income. International Economic Review

4(3):289-309.

Bureau of Labor Statistics. 1974. Technological change
and manpower trends in six industries. U.S. Dept. of
Labor. Washington, D.C. U.S. Government Printing Office.

 

Commerce Technical Advisory Board. 1976. The role of
new technical enterprises in the U.S. economy. A
report to the Secretary of Commerce.

Compton, K.C., H. Hallock, C. Gerhards. and R. Jokerst.
1977. Yield and strength of softwood dimension lumber
produced by EGAR system. USDA Forest Service Research
Paper 293.

Connor, Gordon R., Sr. 1977. The central hardwoods
response-V. ip Resource Availability and the Hardwood
Forest Products Industry. W.L. Hoover and H.A. Holts,
eds. Department of Forestry and Natural Resources,
Purdue University.

C0nway, Steve. 1973. Timber cutting practices. Miller
Freeman Publications. Inc., San Francisco.

Dempsey, Gilbert P. 1973. Toward growth in productivity.
Forest Products Journal 23(4):12-14.

Denison, Edward F. 1962. The Sources of Economic Growth
in the United States and the Alternatives Before Us.
Committee for Economic Development, Supplementary
Paper No. 13.

Duke, John and Clyde Huffstutler. 1977. Productivity
in sawmills increases as labor input declines
substantially. Monthly Labor Review. April.

Ellis, T.H. 1975. The role of wood residue in the
national energy picture. ip Proceedings of the
International Meeting of the Forest Products Research

Society on ”Wood Residue as an Energy Source,” Denver,
Colo.

159

Erickson, J.R. 1968. Mechanization in the timber-
producing industry. Forest Products Journal 18(7):21-27.

Executive Office of the President, Office of Management
and Budget. 1972. Standard Industrial Classification
Manual. Washington, D.C.: U.S. Government Printing
Office.

Feller, Irwin. 1972. Production isoquants and the
analysis of technological and technical change.
Quarterly Journal of Economics 86:154-161.

Ferguson, C.E. 1963. Cross-section production functions
and the elasticity of substitution in American
manufacturing industry. Review of Economics and
Statistics 45:305-313.

. 1969. Microeconomic theory. Homewood, Ill.:
Richard D. Darwin, Inc.

 

Forest Industries. 1967. Logging Handbook No. 4. San
Francisco, Calif.: Miller Freeman Publications, Inc.

FPL Press-Lam Research Team. 1972. FPL press-lam process:
Fast, efficient conversion of logs into structural
products. Forest Products Journal 22(11):11-18.

Galligan, W.L., D.V. Snodgrass, and G.W. Crow. 1977.
Machine stress rating: Practical concerns for lumber
producers. USDA Forest Service General Technical
Report FPL 7.

Gilfillan. S. Colum. 1952. The prediction of technical
change. Review of Economics and Statistics 34:368-385.

Gollop, F.M., and D.W. Jorgenson. 1977. U.S.
productivity growth by industry, 1947-1973. University
of Wisconsin-Madison, Social Systems Research Institute
Workshop Series no. 7712.

Guttenberg, Sam. 1970. Economics of southern pine
pulpwood pricing. Forest Products Journal 20(4):15-18.

H.C. Mason & Associates. 1973. Study of softwood sawlog
conversion efficiency and the timber supply problems.
Report to the USDA Forest Service, Forest Products
Laboratory, Madison, Wis.

Hallock, Hiram. 1977. Precision-quality and value. Expo
'77 Logging-Sawmilling Seminar. Ed. by H. Keith Judkins.
Southern Forest Products Association.

160

Hallock, Hiram, Abigail R. Stern, and David W. Lewis.
1976. Is there a ”best" sawing method? USDA Forest
Service Research Paper FPL 280.

Henneberger. J.E. 1978. Productivity growth below
average in the household furniture industry. Monthly
Labor Review 101(11):23-29.

Herman, Arthur S. 1977. Productivity reports. Monthly
Labor Review. 100(10):57-60.

Hogan, Warren P. 1958. Technical progress and
production functions. Review of Economics and
Statistics 40:407-413.

Humphrey, D.B., and J.R. Moroney. 1975. Substitution
among capital, labor, and natural resource products
in American manufacturing.Journal of Political
Economy 83(1):57-82.

Irland, Lloyd G. 1973. Resource endowment, technology,
and trade: The case of U.S. timber resources.
Unpublished paper presented at meeting of the Southern
Economic Association and Southern Forest Economics
Workers, Houston, Texas.

Jege, K.M. and K.M. Thompson. 1975. The Canadian pulp
and paper industry--threats and opportunities. 1980-
1990. Unnumbered report, Pulp and Paper Research
Institute of Canada.

Johansen. Leif. 1961. A method for separating the
effects of capital accumulation and shifts in
production functions upon growth in labor productivity.
Economic Journal 71:775-782.

Jorgenson, D.W., and Z. Griliches. 1967. The explanation
of productivity change. Review of Economic Studies

34:249-283.

Katzell, Mildred E. 1975. Productivity: The measure
and the myth. American Management Association, survey
report.

Keays, J.L. 1975. Forest harvesting of the future.
Western Forest Products Laboratory. Unnumbered report.

Kendrick, J.W. 1961. Productivity trends in the
United States. Princeton, N.J.: Princeton University
Press.

161

Koch, Peter. 1975. Shaping-lathe headrig will convert
small hardwoods into pallet cants plus flakes for
structural exterior flakeboard. ip Proceedings of the
Ninth Particleboard Symposium, Washington State
University, Pullman, Wash.

Lassen, L.E. and Dwight Hair. 1970. Potential gains in
wood supplies through improved technology. Journal
of Forestry 68(7):404-407.

Lave, Lester B. 1970. Technological change: Its
conception and measurement. Englewood Cliffs, N.J.:
Prentice-Hall, Inc.

Lothner. D.C. 1974. The Minnesota and Wisconsin
Pulpwood Markets: An Econometric Study of Past Changes
and the Future Outlook for Forest Resource Planning.
Ph.D. dissertation. University of Minnesota.

Manning. G.H., and F. Thornburg. 1971. Capital deepening
and technological change: The Canadian pulp and paper
industry 1940-1960. Canadian Journal of Forest
Research:159-166.

Mansfield, Edwin. 1972. Contribution of R&D to

economic growth in the United States. Science 175:
“77-4860

. 1971. Technological change: An introduction to a
vital area of modern economics. New York: W.W. Norton
& Co.

 

Manthy, Robert M. 1978. Natural resource commodities--
a century of statistics. Baltimore: John Hopkins
University Press, published for Resources for the
Future.

Massell, B.F. 1960. Capital formation and technological
change in United States manufacturing. Review of
Economics and Statistics 42:182-188.

. 1961. A dissagregated view of technical change.
Journal of Political Economy 69(6):547-557.

McDonald, Kent. 1978. Lumber defect detection by
ultrasonics. USDA Forest Service Research Paper FPL
311.

National Science Foundation. 1976. Technological
innovation and federal government policy. Office of
National R&D Assessment. NSF 76-9.

162

. Research and development in industry, 1975.
Survey of Science Resources Series, NSF 77-324.
Washington, D.C.

 

. Basic research, applied research, and development
in industry, 1965. Survey of Science Resources Series,
NSF 67-12. Washington, D.C.

. Research and development in industry, 1960. Survey
of Science Resources Series, NSF 63-7. Washington, D.C.

Neilson. K.K. 1967. The present state, problems, and
outlook of mechanized tree processing in Eastern
Canada. Pulp and Paper Canada 67:WR 297-WR 301.

Nelson, R.R., M.J. Peck, and E.D. Kalachek. 1967.
Technology, Economic Growth, and Public Policy
Washington, D.C.: The Brookings Institution.

Nissan, A.H., ed. 1973. Future technical needs and
trends in the paper industry. Special Technical
Association Publication No. 10, TAPPI.

Peters, Curtis C. and Conrad M. Banas. 1977. Cutting
wood and wood-base products with a multikilowatt laser.
Forest Products Journal 27(11):41-45.

Phelps, Robert B. 1977. The demand and price situation
for forest products 1976-77. USDA, Forest Service
Misc. Pub. No. 1357. Washington, D.C.: U.S.
Government Printing Office.

Phillips, Almarin. 1963. An appraisal of measures of
capacity. American Economic Review 53(2):275-292.

Porterfield. R.L. 1977. Utilization efficiency during
harvesting--a survey of current and prospective status.
Forest Products Journal 27(12):17-20.

Pulp and Paper Research Institute of Canada. Logging
research progress report No. 45. 1974. Pointe
Claire. Quebec.

Rauch, J., ed. 1976. Kline guide to the pulp and paper
industry. Charles H. Kline & Co., Inc. Fairfield,
N.J.

Robinson, V.L. 1975. An estimate of technological
progress in the lumber and wood-products industry.
Forest Science 22(2):149-154.

163

Rosenburg, Nathan. 1972. Technology and American
Economic Growth. New York: Harper and Row.

Ruttan, V.W. and J.C. Callahan. 1962. Resource inputs
and output growth: Comparisons between agriculture
and forestry. Forest Science 8(1):68-82.

Salter, W.E.G. 1969. Productivity and Technical Change.
Cambridge: The University Press. Second edition.

Schmookler, Jacob. 1966. Invention and Economic
Growth. Cambridge: Harvard University Press.

. 1972. Patents, Invention, and Economic Change.
Cambridge: Harvard University Press.

 

Scott, J.R., Jr. 1964. The measurement of technology.
Journal of Farm Economics 46(3):657-661.

Solow. Robert M. 1957. Technical change and the
aggregate production function. Review of Economics
and Statistics 39:323-333.

Stern, Abigail and Kent MCDonald. 1978. Computer
optimization of cutting yield from multiple-ripped
boards. USDA Forest Service Research Paper FPL 318
(in press).

The President. 1978. Economic Report of the President.
Washington, D.C.: U.S. Government Printing Office.

U.S. Forest Service. 1973. The outlook for timber in
the United States. USDA Forest Resource Report No. 20.
Washington, D.C. U.S. Government Printing Office.

. 1977. The Nation's renewable resources--an
assessment, 1975. USDA Forest Resource Rep. No. 21,
Washington, D.C.: U.S. Government Printing Office.

 

. 1958. Timber Resources for America's Future.
USDA Forest Resource Report No. 14. Washington, D.C.:
U.S. Government Printing Office.

 

. 1965. Timber Trends in the United States. USDA
Forest Resource Report No. 17. Washington, D.C.: U.S.
Government Printing Office.

ADDITIONAL REFERENCES

ADDITIONAL REFERENCES

Arrow, K.J. 1962. The economic implications of learning
by doing. Review of Economic Studies 29:155-173.

Boorstin, D.J. 1978. The Republic of Technology.
New York: Harper & Row.

Bright, James R. 1970. Evaluating signals of
technological change. Harvard Business Review, Jan.-
Feb.

Champernowne, D.G. 1954. The production function and
the theory of capital: A comment. Review of Economic
Studies 21(2):112-135.

, and R.F. Kahn. 1954. The value of invested
capital. Review of Economic Studies 21(2):107-111.

Christensen, L.R. 1975. Concepts and measurement of
agricultural productivity. University of Wisconsin,
Social Systems Research Institute reprint series no.

294.

Coates, Vary T. 1972. Technology and public policy:
The process of technology assessment in the Federal
government. Program of Policy Studies in Science and
Technology. Washington, D.C.: The George Washington
University.

Cox, L.A. 1974. Transfer of science and technology in
2 (

:Ecfiessful innovation. Forest Products Journal 9):
-80

Diamond, P.A. 1965. Technological change and the
measurement of capital and output. Review of Economic
Studies 32:289-298.

Domar, E.D. 1961. On the measurement of technological
change. Economic Journal 71(284):709-729.

. 1963. Total productivity and the quality of
capital. Journal of Political Economy 61(6):586-588.

Fabricant, Solomon. 1965. Measurement of teChnological

change. Seminar on Manpower Policy and Program. U.S.
Department of Labor.

164

165

Ferguson, C.E. 1968. Neoclassical theory of technical
progress and relative factor shares. Southern Economic
Journal 34(4):490-504.

Fisher, Franklin M. 1969. The existence of aggregate
production functions. Econometrica 37(4):553-577.

Gold, Bella. 1975. Technological Change: Economics,
Management and Environment. New York: Pergamon
Press Inc.

Greenburg, Leon. 1973. A Practical Guide to Productivity
Measurement. Washington, D.C. The Bureau of National
Affairs, Inc.

Griliches. Zvi. 1957. Hybrid corn: An exploration in
the economics of technological change. Econometrica
25(4)1501-522.

Griliches. Zvi. and Dale W. Jorgenson. 1966. Sources of
measured productivity change: Capital input. American
Economic Review, Papers and Proceedings 56(2):50-61.

Heertje, Arnold. 1973. Economics and Technical Change.
London: Weidenfeld and Nicholson.

Jaszi, George. 1962. An improved way of measuring
quality change. Review of Economics and Statistics.

43:332-335.

Kaiser, H.F., Jr. 1971. Productivity gains in forest

piodgcts industries. Forest Products Journal 21(5):
1'1 0

Kamien, M.I., and N.L. Schwartz. 1968. Optimal ”induced"
technical change. Econometrica 36(1):1-17.

Kennedy, C. and A.P. Thirlwall. 1972. Surveys in applied
economics: Technical progress. Economic Journal 82:11-
72.

Lanford, H.W. 1972. Technological Forecasting
Methodologies: A Synthesis. American Management
Association.

Linstone, H.A., and D. Sahal, ed. 1976. Technological
Substitution: Forecasting Techniques and Applications.
New York: Elsevier.

Mansfield, Edwin. 1961. Technical change and the rate of
imitation. Econometrica 29(4):?41-766.

Mason, H.C. 1971. Wood industry technology: What's new
now, what's to come. Forest Industries 98(11):22-24.

166

Massell. B.F. 1962. Another small problem in the
analysis of growth. Review of Economics and Statistics

43:330-332.

Nadiri, M.I. 1970. Some approaches to the theory and
measurement of total factor productivity: A survey.
Journal of Economic Literature 8:1137-1177.

Peterson. W., and Y. Hayami. 1977. Technical change in
agriculture. ip_Lee R. Martin, ed. A Survey of
Agricultural Economics Literature. Minneapolis:
University of Minnesota Press.

Resek, Robert W. 1963. Neutrality of technical progress.
Review of Economics and Statistics 44:55-63.

Robinson, Joan. 1954. The production function and the
theory of capital. Review of Economic Studies 21(2):
81-10 a

Samuelson, P.A. 1962. Parable and realism in capital
theory: The surrogate production function. Review of
Economic Studies 29:193-206.

Sandoe, M., and M. Wayman. 1977. Productivity of capital
and labor in the Canadian forest products industry, 1965
to 1972. Canadian Journal of Forest Research 7:85-93.

Schultze, Charles L. 1963. Uses of capacity measures for
short-run economic analysis. American Economic Review,
Papers and Proceedings 53(2):293-308.

Shell. Karl. 1966. Toward a theory of inventive activity
and capital accumulation. American Economic Review,
Papers and Proceedings 56(2):62-68.

Smith, V.K. 1974. Technical Change, Relative Prices,
and Environmental Resource Evaluation. Baltimore: John
Hopkins University Press, published for Resources for
the Future.

Solow. R.M. 1956. The production function and the theory
of capital. Review of Economic Studies 23(2):101-108.

Utterback, James M. 1974. Innovation in industry and the
diffusion of technology. Science 183:620-626.

Williams, B.R. 1967. Technology, Investment, and Growth.
London: Chapman and Hall, Ltd.