Group Title: Working paper - International Agricultural Trade and Policy Center. University of Florida ; WPTC 05-03
Title: Induced innovations and foreign workers in U.S. agriculture
Full Citation
Permanent Link:
 Material Information
Title: Induced innovations and foreign workers in U.S. agriculture
Series Title: Working paper - International Agricultural Trade and Policy Center. University of Florida ; WPTC 05-03
Physical Description: Book
Language: English
Creator: Napasintuwong, Orachos
Emerson, Robert D.
Publisher: International Agricultural Trade and Policy Center. University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: March 2005
 Record Information
Bibliographic ID: UF00089800
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.


This item has the following downloads:

wp050003 ( PDF )

Full Text

WPTC 05-03

I '-ional Agricultural Trade and Policy Center

Orachos Napasintuwong & Robert D. Emerson

WPTC 05-03 March 2005





Institute of Food and Agricultural Sciences



The International Agricultural Trade and Policy Center (IATPC) was established in 1990
in the Institute of Food and Agriculture Sciences (IFAS) at the University of Florida
(UF). The mission of the Center is to conduct a multi-disciplinary research, education and
outreach program with a major focus on issues that influence competitiveness of specialty
crop agriculture in support of consumers, industry, resource owners and policy makers.
The Center facilitates collaborative research, education and outreach programs across
colleges of the university, with other universities and with state, national and
international organizations. The Center's objectives are to:

* Serve as the University-wide focal point for research on international trade,
domestic and foreign legal and policy issues influencing specialty crop agriculture.
* Support initiatives that enable a better understanding of state, U.S. and international
policy issues impacting the competitiveness of specialty crops locally, nationally,
and internationally.
* Serve as a nation-wide resource for research on public policy issues concerning
specialty crops.
* Disseminate research results to, and interact with, policymakers; research, business,
industry, and resource groups; and state, federal, and international agencies to
facilitate the policy debate on specialty crop issues.


Orachos Napasintuwong
Food and Resource Economics Department
PO Box 110240
University of Florida
Gainesville, FL 32611

Robert D. Emerson
Food and Resource Economics Department
PO Box 110240
University of Florida
Gainesville, FL 32611


A cost function approach of induced innovation is used to measure the
biases in U.S. agricultural technology between 1948-1994. The results show
significant labor-saving, capital-using technical change. Focusing on the impact
of migration policy on labor-saving technology, a simulation of different rates of
labor-saving technical change is conducted. The simulation shows decreases in
elasticity of labor demand and demand quantity, and an increase in wage rate as
technology becomes more labor-saving.

JEL codes: Q160, J430, 0300
Keywords: Induced Innovation, Biased Technical Change, Foreign Labor

Selected Paper for presentation at the American Agricultural Economics
Association Annual Meetings, Long Beach, California, July 28-31, 2002

Copyright 2002 by Orachos Napasintuwong and Robert D. Emerson. All rights
reserved. Readers may make verbatim copies of this document for non-
commercial purposes by any means, provided that this copyright notice appears
on all such copies.



Throughout much of U.S. agricultural history, labor has been the relatively scarce

factor of production. Hayami and Ruttan's theory of induced innovation suggests that an

increase in relative factor prices will induce technical development to save the factor that

becomes relatively more expensive. A change in relative demand for factors, however, is

a result of both changes in factor prices and changes in technology. As labor becomes

relatively more scarce, the wage rate also becomes relatively higher, thus inducing the

development of labor-saving technology. This paper addresses the relation between

immigration regimes and the structure of technology in U.S. agriculture.

Several studies have shown that U.S. technology is biased toward labor-saving, as

the induced innovation theory would suggest. Binswanger (1974) showed that

technology was labor-neutral from 1912 to 1944, but it was biased toward strong labor-

saving technical change between 1944 and 1968. He estimated that two-thirds of the

decrease in labor share was explained by biased technical change, and only one-third by

the price substitution effect. From 1948 to 1968, both labor and machinery prices have

accelerated, but the technology was machinery-using, suggesting machinery-using

technical change. Antle (1984) also showed technical biases against land and toward

machinery between 1910 and 1946. However, he found that technical change was labor-

saving, capital-using, and chemical-using from 1947 to 1978. A study by Shumway and

Alexander (1988) also suggested non-neutral technical change in most regions of the

United States.

Empirical evidence of machine-using, labor-saving technical change in

agriculture follows from the many developments in farm mechanization. During the past

century several machines have been invented to substitute for human labor such as

tractors, threshers, and reapers. There have also been significant labor-saving

technologies in dairy, poultry and swine production. More recently, the development of

farm mechanization has focused on mechanical harvesters for perishable crops. Even

though the U.S. has become more advanced in mechanical technology, there remain a

number of potential mechanical technologies that have not been fully developed or

adopted such as mechanical citrus harvesting in Florida. With the most difficult

applications remaining, the technological challenges are both difficult and expensive.

Moreover, the development of new technology is relatively more expensive when an

inexpensive foreign labor supply remains readily available.

The labor market experience for the past two to three decades has been an

increasing flow of illegal labor into the U.S. with significant unauthorized employment in

agriculture. The National Agricultural Workers Survey (NAWS) during 1997-1998

shows that 81 percent of farm workers are foreign born, and 52 percent of total farm

workers are unauthorized (USDL). Given a downward sloping demand and upward

sloping supply for agricultural labor, the augmentation of labor supply through

unauthorized foreign workers results in a rightward shift in the supply of labor at a wage

rate lower than would occur in the absence of supply augmentation. From the

perspective of induced innovation models, the incentive for new labor-saving

technologies is reduced from what it would be in the absence of international labor


The first objective of this study is to estimate the biases in technical change in

U.S. agriculture over the period 1948 to 1994. During this period, there are two major

changes in migration policy pertaining to agricultural labor. The early part of the period

includes the Bracero program established in 1942 and ending in 1964. This was a

bilateral agreement between the U.S. and Mexico that allowed Mexican workers to enter

the U.S. for agricultural work. The termination of the Bracero program was followed by

the rapid adoption of the mechanical tomato harvester (Martin; Sarig, et al.). The second

legislative change was the Immigration Reform and Control Act (IRCA) of 1986. The

legislative intent of IRCA was to reduce the flow of illegal immigrants. One component

of IRCA was the Special Agricultural Worker (SAW) program legalizing 1.2 million

unauthorized workers in 1987-88 who claimed to have done at least 90 days of farm work

in 1985-86. There is little evidence, however, that IRCA reduced the flow of illegal

workers; as noted above, the 1997-98 NAWS data indicate that 52 percent of the farm

workers are unauthorized.

The end of the Bracero program and the passage of IRCA during the study period

were intended to reduce the supply of foreign farm workers. Had the program changes

truly had their intended effects, the induced innovations model would suggest that the

changes should induce the development of additional farm mechanization and other

technologies that could substitute for the potentially more expensive labor. At about the

same time, however, public institutions, particularly the U.S. Department of Agriculture

and Land Grant universities, in the early 1980s were reducing their emphasis on

mechanization research under the presumption that it favored large-scale agriculture and

displaced labor.

A second objective in this paper is to provide implications for migration policies

from alternative technological biases. Assuming that technical change biases are a result

of induced innovations, simulations of different technical change scenarios provide a

basis to evaluate alternative labor supply scenarios. Previous research has shown

agricultural labor supply to be highly elastic (Emerson and Roka), and the implication of

Gisser and Davila is the same. Farm labor supply is assumed to be perfectly elastic for

the purposes of this paper.

There is a number of studies of technical change in agriculture have been done

(Antle; Binswanger; Shumway and Alexander; Weaver; Griliches; Hayami and Ruttan;

Kislev and Peterson; Huffman and Evenson). One difference is that the present paper

builds on the previous work using the quality-adjusted data published by USDA (Ball, et

al., 1997, 1999). A second difference is that the present paper is focused on the relation

between labor migration and technological change in agriculture for the 1948-1994



We adopt the dual cost function model of biased technological change by using a

time variable to represent technology in the translog cost function. Although a time

variable may leave much to be desired as an explanation of technological change, that is

not our objective. As Chambers argues, time is a very economical variable for

representing technological change (p. 204). The estimates from parameters of the

translog cost function provide the estimates for elasticities of factor demand and

elasticities of factor substitution. In addition, the time variable provides estimates of bias

in technological change.


Deviations from the estimated technology bias in both direction and magnitude

form the basis for simulations of alternative technology biases. One simulation provides

estimates for changes in labor demand when factor prices are assumed to remain

constant. The second simulation with alternative technology biases provides estimates of

equivalent wage rate changes while factor shares and other factor prices are assumed to

remain constant. The wage rate changes are assumed to follow from alternative

immigration policies raising or lowering the wage rate that agricultural employers face.

The counter-factual is again based on the induced innovations theory: the observed

technology biases result from the relative factor prices. Coincident with the implied

wage change is a new equilibrium quantity of labor calculated from the elasticity of

demand for labor, given a perfectly elastic labor supply.


We adopt the basic structure of Binswanger's translog cost function model; this

section draws heavily from his paper. The model assumes a single aggregate agricultural

output. Constant returns to scale is assumed: the level of output does not affect the

relative use of inputs. The production of the aggregate agricultural product (Y) requires n

variable inputs X = (Xi, X2,..., Xn). The vector of input prices is W = (W1, W2,..., Wn).

A state of technology influences a cost of production. Using time as a representative for

technological knowledge, production cost is a function of input prices and the technology

level. The translog cost function C = f(Y,W1,..., Wn, t) can be written as

In C = v + In y + v, In + y, In W, In W +v, In t + (In t)2 + In W, Int (1)
2 1 1I

Since a cost function must be homogeneous of degree one in prices, this implies for a

translog cost function that

ZV, =1; y, =o; Zy1, =O.

In addition, a symmetry restriction is also assumed to hold.

Yij Y= ji for all i, j; i j

Utilizing Shephard's Lemma, aC/aWi = Xi, the first derivative of a translog cost function

generates a share equation.

n C X S i = ,...,n (2)
SIn W1 C

S1 = = Vn + Wy, In Wj + colnt i =l,...,n (3)
a InW, l 1J

AS, = Aln Wj + Alnt i =,...,n (4)

For a discrete time period, a change in factor share is a result of changes in factor prices
and a change in technology. The direction of biased technical change is measured by a
change in factor share, keeping relative factor prices constant. In a many-factor case,
technical change is biased toward factor i-saving, neutral, or i-using if the share of factor
i in total cost decreases, stays constant, or increases.

i saving
dS 1<
B relative factor pnce = --- 0 neutral (5)
at S, >
i -u sin g

Thus, changes in factor shares as a result of only changes in technology, ASi* can be

estimated from

AS* = o,Aln t i ,...,n (6)

The sign of o% determines the bias in technical change, and co, can be interpreted as a

constant rate of the bias of factor i over the estimation period.

The parameter estimates of share equations allow us to calculate Allen partial

elasticities of factor substitution (cy) and price elasticities of factor demand (rlij).

J = + 1 for all i, j; i j (7)

(, = 2 11 + S 2 S,) for all i (8)

= i+ S for all i, j; i j (9)

n + S, -1 for all i (10)


The estimates of biased technical change for each factor are obtained from the

share equations. We assume that there are four variable inputs: capital, labor, land, and

intermediate inputs. Since there are two major changes in migration policy during the

study period, the time variable is separated into three periods to capture the potentially

different structure of technical change within each period. The first period is from 1948

to 1964 (Bracero period), the second period is from 1964 to 1986, and the last period is

from 1986 to 1994 (post-IRCA).

Since the covariance matrix of the disturbances of all four share equations is

singular, a system of three share equations is estimated using iterated seemingly unrelated

regression (SUR). In this estimation, the share equation for real estate is dropped, and

independent variables include relative prices of factor inputs to price of land, and a time


SK =v +7, In WK + 7 In WL + 7 In W + .lnt + T2 lnt + K3 Tnt + c (11)

S, = v, + YLK In WK + YL In WL + Y In + Llnt + (L2.:Tlnt + coTlnt + CL (12)

S, = v+ + 7 In WL + L, n W+ ln + C2T2 zlnt + o1 T3lnt + C (13)
S, = + In + ^ In -- + 7ni In -- + o)illnt + Tlnt + 2313!^ + e, (13)

The variables K, L, I, and A represent capital, labor, intermediate inputs, and land,

respectively. T2 and T3 represent dummy variables for period 1964-1986 and period

1986 to 1994, respectively. The Ei's are the stochastic disturbances having zero

expectation and nonzero contemporaneous covariances. A system of share equations

requires that the summation of the four factor shares equals to one. As a result, in

addition to homogeneity and symmetry constraints, the following additional

constraint, io, = 0, is assumed forj = 1,2,3.


Data used in this study are obtained from the production accounts constructed by

Economic Research Service, USDA (Ball, et al., 1997, 1999, 2001). The aggregate U.S.

data report quantity and implicit price indices of aggregate inputs and output during

1948-1994. We select four variable input categories for this study: durable equipment,

hired labor, real estate, and other intermediate inputs. Hired labor is representative for

the labor variable in the model. Self-employed labor and unpaid family labor are

assumed to be a fixed input. Intermediate inputs include agricultural chemicals, fuels and

electricity, feed, seed, and livestock purchases, and other purchased inputs which also

include contract labor services. Since contract labor quantity is not significant at the

aggregate U.S. level, excluding it from hired labor should not have a significant impact

on the estimates. Durable equipment includes autos, trucks, tractors, and machines, and

represents the capital variable in the model. Real estate represents the land variable in the

model, including both land and buildings.

Quantity indices in the production account are constructed based on the Tornqvist

index number specification. Implicit price indices are constructed as the ratio of the

value of the input aggregate to the corresponding quantity indices, and can be interpreted

as unit values (expressed in millions of dollars) of the aggregates. A detailed discussion

of data construction can be found in Ball, et al. (1999, 1997). The data obtained from this

series are adjusted for changes in quality or characteristics of the inputs over time. As a

result, quantity and price indices can be viewed as constant quality indices. It is

important to use quality-adjusted data when analyzing induced technological change

because using unadjusted quality indices will result in a biased estimation of parameters

in the induced innovation model.

Figure 1 illustrates real implicit price indices of inputs between 1948 and 1994.

Prices of labor and capital steadily increased over time. Price of intermediate inputs

decreased throughout the period, except during the early 1970s when it rose with the oil

embargo. Real estate price volatility increased over time, with the real estate price

increasing dramatically during the mid 1970s to mid 1980s, and decreasing thereafter.

Figure 2 illustrates the expenditure share of each input during the same period.

The expenditure share of labor decreased until the mid 1980s, and increased afterward.

The expenditure share of capital was stable throughout the period. The expenditure share

of real estate increased from 1948 to 1980 with some exceptions during 1970 to 1975,

and decreased after 1984 until 1993. The intermediate input share decreased

substantially from 1948 to 1984, but increased during the remainder of the period.


The parameter estimates of share equations are summarized in table 1. Since only

three equations are estimated, the estimates of YAK, YAL, YAI, AA, and oA are derived from

the other estimates based on homogeneity, symmetry, and c restrictions. The signs of the

time parameter estimates in each share equation show the direction of technical change.

The coefficients of time are significant and negative in the labor equation, suggesting that

technology was biased toward labor-saving in all three periods, but became increasingly

more labor-saving following the Bracero program and implementation of IRCA. The

coefficient of time is positive in the capital equation for the first period, but becomes

smaller in the last two periods. Somewhat surprisingly, the technology was biased

toward capital-using through the Bracero program. However, there were important

changes elsewhere in U.S. agriculture at the time, in particular the adoption of the

mechanical cotton picker in the southern U.S. This period also experienced significant

mechanization in mid-western grain farming as well as developments in animal

agriculture. The technology remained strongly biased toward capital-using until IRCA,

although at a slower rate. The mechanical tomato harvester was the major change in farm

mechanization after the end of the Bracero period; however, its impact shows a slightly

lower rate of machine-using technical change than the prior period. After IRCA,

technology became less capital-using, perhaps due to the increasing flow of foreign

workers into the U.S. for farm work.

Throughout the whole period, technical change was biased toward intermediate

inputs-using. The change in technology implies an increasing use of intermediate inputs

that include chemicals, fuel, electricity, contract labor, feed and seeds, and others. This

result is also consistent with Binswanger's and Antle's findings. Huffman also shows

that there is an increasing amount of contract labor after IRCA, particularly in California

and Florida. The technology was biased toward land-saving in all three periods in

differing degrees.

The estimates from table 1 are converted to the elasticity estimates evaluated at

the means in table 2. All own price elasticities of demand have the right sign. Except for

real estate, all own elasticities of substitution are significant. The elasticity of

substitution between capital and labor is inelastic and positive, but not significant. The

aggregate U.S. data during the study period suggest that there are limitations for

substituting capital for labor. Since the mechanization during the past fifty years has

been focused on developing mechanical harvesters, the difficulties of replacing labor may

result from uneven ripening, and physical damage during mechanical harvesting process.

The elasticities of substitution between intermediate inputs and capital, intermediate

inputs and labor, and real estate and capital are positive and inelastic. This suggests that

they are weak substitutes. The elasticity of substitution between labor and real estate,

however, is not significant.

Table 1. Restricted Estimates of the Coefficients of the Translog Cost Function and t-Ratios1
Input Price
Factor Share Capital Labor Intermediate Input Real Estate ln(t) T2*ln(t) T3*ln(t) Intercept
0.0644 -0.0044 -0.0476 -0.0124 0.0152 -0.0030 -0.0100 0.1669
(7.7074) (-1.0761) (-6.8211) (-2.7804) (5.2055) (-2.2027) (-7.6116) (8.9304)
0.0441 -0.0168 -0.0228 -0.0116 -0.0038 -0.0041 0.1977
(10.4013) (-3.8409) (-9.4105) (-6.8936) (-4.7570) (-4.6864) (14.8628)
Intermediate 0.1818 -0.1175 0.0411 0.0042 0.0106 0.1730
Inputs (17.0298) (-15.4536) (5.8221) (1.4991) (3.5687) (4.9404)

l E e 0.1526 -0.0446 -0.0343 0.0035 0.4680
(21.8633) (-7.1608) (4.7132) (1.3360) (19.3337)

1 Restrictions imposed:

IV, = 1; 1 r" = 0; 1 r" = 0, Y u, 0"= 0.

Table2. Estimates of Allen Partial Elasticities of Substitution and of Own Price Elasticities
of Factor Demand and t-Ratios
Capital Labor Intermediate Inputs Real Estate
Elasticities of Substitution
Capital -2.8149 0.5672 0.2979 0.4574
(-4.3620) (1.4104) (2.8945) (2.3439)
Labor -4.6732 0.6852 -0.2616
(-8.9039) (8.3611) (-1.9514)
Intermediate Inputs -0.1667 0.0188
(-5.5304) (0.2958)
Real Estate -0.1975
Elasticities of Demand -0.3203 -0.4198 -0.0992 -0.0397
(-4.3621) (-8.9039) (-5.5304) (-1.1438)

Simulation of biased technical change

During 1948-1994, there were significant changes in farm mechanization, notably

the cotton picker and tomato harvester. Although the results show that the technology is

biased toward labor-saving and machine-using, the availability of foreign labor supply

may have slowed the development of labor-saving machine-using technology; for

instance, the development of fruits and vegetable mechanical harvesters for fresh

markets. The simulations conducted here assume alternative rates of technical bias.

Our first simulation provides estimates of the elasticity of labor demand if the

technology had been at alternative rates of labor-saving. The simulation assumes fixed

input prices, but allows factor shares to change. Table 3 summarizes the results from this


Table 3. Estimates of Labor Share and Elasticity of Labor Demand and t-Ratios at different Level
of Labor-Saving Technical Change
Rate of Labor-Saving Technical Change
0.00 0.005 0.01 0.015 0.02 0.025
Labor Share 0.1323 0.1178 0.1032 0.0887 0.0741 0.0596
(25.1249) (23.3619) (19.5988) (16.8357) (14.0727) (11.3096)
Elasticity of Labor -0.5348 -0.5082 -0.4700 -0.4146 -0.3316 -0.2010
Demand (-18.0561) (-15.4326) (-12.5748) (-9.4759) (-6.1685) (-2.8016)
% Change in Labor
40.83 24.63 8.42 -2.92 -23.98 -40.18

The simulation shows that as technical change becomes more labor-saving, the

demand for labor decreases. As a result, labor's share decreases, given constant factor

prices. Demand also becomes more inelastic as technical change becomes more labor-

saving. This suggests that non-neutral technical change for labor could rotate the labor

demand schedule. A change in labor's share due to a change in the rate of labor-saving

technology can also be converted to a change in labor demand given constant factor

prices, providing an estimate of the extent to which the demand curve shifts in or out. A

labor demand is calculated at the mean of the data, and compared to labor demand at the

actual rate of labor saving technical change1. Assuming a constant rate of technical

change for each input, it is found from the data that the average rate of labor-saving

technical change is 0.014. Thus, the simulation of a percentage change in labor demand

is compared with the labor-saving change at the rate of 0.014.

The simulation shows that if the rate of labor-saving technical change is below

0.014, the demand for labor would increase more than it had been. On the contrary, if the

rate of labor-saving technical change is greater than 0.014, labor demand would be less

AL C Awo ln(t)
1 *100 = AoLln(t)- *100 -=- *100

than it had been. A near doubling of the rate of the labor-saving technology (to 0.025)

would shift the labor demand curve to the left by 40 percent.

The second simulation assumes constant factor shares and holds all factor prices

constant, except the wage rate. This simulation provides deviations in labor price and

quantity indices from the sample means at different levels of labor-saving technical

change. Table 4 summarizes a percentage change of nominal implicit wage of hired

labor, and a percentage change of labor quantity indices at a given rate of technical


Table 4. Estimates of Percentage Change in Wage and Labor Quantity and the t-Ratios at different
Level of Labor-Saving Technical Change
Percentage Change Rate of Labor-Saving Technical Change
0.00 0.005 0.01 0.015 0.02 0.025
Wage -11.7046 -7.6974 -3.6903 0.3167 4.3250 8.3312
(-8.2267) (-5.7920) (-2.7468) (0.2166) (2.5966) (4.3298)
Labor Quantity 4.9135 3.2313 1.5491 -0.1330 -1.8152 -3.4973
(5.1631) (4.6337) (2.7696) (-0.2143) (-2.1638) (-3.1058)

Assuming induced innovations, for technology to have been neutral over the

sample period, the wage rate would have to have been about 12 percent lower, and the

equilibrium labor quantity would have been by about 5 percent higher. Should the

technology become more labor-saving, the wage rate would have to increase, and the

labor quantity would have to decrease. Given a perfectly elastic labor supply function,

labor-saving technical change would offset an upward shift in the supply curve resulting

in a higher wage rate, and a lower equilibrium quantity of labor.


Our study of biased technical change in U.S. agriculture over the 1948 to 1994

period shows that U.S. technology has become more labor-saving, capital-using,

intermediate inputs-using, and land-saving. Following the Bracero program, the

technology was biased toward more labor-saving, and after IRCA, it became even more

labor-saving. Given some evidence of more change toward mechanization, such as the

development of mechanical cotton picker in the 1950s, and of mechanical tomato

harvester after the end of the Bracero program, our result suggests technology was less

capital-using. Despite the government's attempt to reduce the flow of illegal foreign

labor into the U.S, there has still been an increasing flow of illegal foreign farm labor into

the U.S. Our study tries to answer whether relatively abundant labor supplies may have

forestalled labor-saving technology in the currently labor-intensive areas of U.S.


The simulation of different rates of labor-saving technical change suggests two

outcomes. First, if all factor prices remain constant, labor-saving technology would

result in a reduced expenditure share for labor, a more inelastic labor demand, and a

significant reduction in the equilibrium quantity of labor. An alternative scenario is

holding all factor shares constant and all factor prices constant except labor, and

evaluating the wage increase corresponding to greater labor-saving technology. From the

perspective of induced innovation theory, labor-saving technology can be induced by the

decrease in supply of labor, and the corresponding increase in wage rate. Given the fact

that a large majority of U.S. farm labor is unauthorized foreign workers, this would


suggest that an increase in the stringency of migration policy as well as law enforcement

would be necessary to induce further labor-saving technical change in U.S. agriculture.

Figure 1. Real Input Price Indices (GDP deflated)


Real Estate

Hired Labor

1948 1950 1952 1954 1956 1958 1960 1962 1964 1966 1968 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994

Source ProducttnAccount, ERS, USDA

Figure. Expenditure Share

Intermediate Inputs

0000000 .

Real Estate




Source Production Account. ERS. USDA


0 7000

0 5000-

o 04000-

0 3000

0 2000

0 1000 -


1948 1950 1952 1954 1956 1958 1960 1962 1964 1966 1968 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994


Antle, J. M. "The Structure of U.S. Agricultural Technology, 1910-78." American
Journal ofAgricultural Economics. 66 (November 1984): 414-421.

Ball, E., J.-C. Bureau, R. Nehring, and A. Somwaru. "Agricultural Productivity
Revisited." American Journal ofAgricultural Economics. 79 (November 1997):

Ball, V. E., F. M. Gollop, A. Kelly-Hawke, and G. P. Swinand. "Patterns of State
Productivity Growth in the U.S. Farm Sector: Linking State and Aggregate
models." American Journal ofAgricultural Economics. 81 (February 1999): 164-

Ball, E., J. Butault, and R. Nehring "U.S. Agriculture, 1960-96: A Multilateral
Comparison of Total Factor Productivity." Technical Bulletin Number 1895.
Economic Research Service, USDA, May 2001. 1895/tb 895.pdf

Binswanger, H. P. "A Cost Function Approach to the Measurement of Elasticities of
Factor Demand and Elasticities of Substitution." American Journal of
Agricultural Economics. 56 (May 1974): 377-386.

Binswanger, H. P. "The Measurement of Technical Change Biases with Many Factors of
Production." The American Economic Review 64: 6 (December 1974): 964-976.

Chambers, R. G. Applied Production Analysis: A Dual Approach. New York:
Cambridge University Press, 1988.

Emerson, R. D. and F. Roka. Income Distribution and Farm Labor Markets, in Dynamics
of Hired Farm Labor: Constraints and Community Response, ed. Jill L. Findeis.
New York: CAB International, in press.

Gisser, M. and A. Davila. Do Farm Workers Earn Less? An Analysis of the Farm Labor
Problem. American Journal ofAgricultural Economics 80:4 (November

Griliches, Z. Agriculture: Productivity and Technology. International Encyclopedia of
the Social Sciences 1 (1968):241-245.

Hayami, Y., and V. W. Ruttan. Factor Prices and Technical Change in Agricultural
Development: The United States and Japan, 1880-1960. Journal of Political
Economy 78:5(September 1970):1115-1141.

Huffman, W. E. Changes in the Labor Intensity of Agriculture: A Comparison of CA, FL
and Whole U.S., in Dynamics of Hired Farm Labor: Constraints and Community
Response, ed. Jill L. Findeis. New York: CAB International, in press.

Huffman, W. E. and R. E. Evenson. Sciencefor Agriculture: A Long-Term Perspective.
Ames: Iowa State University Press, 1993.

Kislev, Y. and W. Peterson. Induced Innovations and Farm Mechanization. American
Journal ofAgricultural Economics 63:3 (August 1981):562-565.

Martin, P. "There Is Nothing More Permanent Than Temporary Foreign Workers."
Center for Immigration Studies, Backgrounder, April 2001.

Sarig, Y., J. F. Thompson, and G. K. Brown. "Alternatives to Immigrant Labor? The
Status of Fruit and Vegetable Harvest Mechanization in the United States." Center
for Immigration Studies, Backgrounder, December 2000.

Shumway, C. R. and Alexander, W. P. "Agricultural Product Supplies and Input
Demand: Regional Comparisons." American Journal ofAgricultural Economics.
80 (February 1988): 153-161.

U.S. Department of Labor. Findings form the National Agricultural Workers Survey
(NAWS) 1997-1998. Research Report 8. OASP. Washington, D.C. March 2000.

Weaver, R. D. Multiple Input, Multiple Output Production Choices and Technology in
the U.S. Wheat Region. American Journal ofAgricultural Economics 65:1
(February 1983):45-56.

University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs