Economic impacts of commercial applications of biotechnology in field-crop production

UNIVERSITY OF

ILLINOIS LIBRARY

AT URBANA-CHAMPAIGN

AGRICULTURE

NOnCE Hetum or renew all Ubrwy MalerWsl The Minimum F«« «or

each Lost Book is $50.00.

The person charging this material is responsible for

its return to the library from which it was withdrawn

on or before the Latest Date stamped below.

Theft, mutilation, and underlining of book, are reasons for diseip«-

nary action and may result in dismissal from the University.

To renew call Telephone Center, 333-8400

UNIVERSITY OF ILLINOIS LIBRARY AT URBANA-CHAMPAIGN

L161— O-1096

AU6 1 2 1992

«WCULTURE LIWAKY

r^.^?^^^

HUNG**

Economic Impacts of
Commercial Applications
of Biotechnology in
Field-Crop Production

Wojciech ]. Florkowski
and Lowell D. Hill

Bulletin 799

University of Illinois at Urbana-Champaign

College of Agriculture

Agricultural Experiment Station

^. N • ^V'-\^>" -:• •'' -^..
'^^\-\H'!'1-^V-^^'-

.- --. -

AUTHORS: Wojciech J. Florkowski is an assistant professor of agricultural economics at
the University of Georgia and Georgia Experiment Station, Griffin, Georgia. Lowell D.
Hill is the L.J. Norton Professor of Agricultural Marketing in the Department of
Agricultural Economics at the University of Illinois at Urbana-Champaign, Urbana,
Illinois.

Editor: Mary E. Theis
Designer: Krista Sunderland

The Illinois Agricultural Experiment Station provides equal opportunities in programs and employment.

U.S. scientific leaders in the development
of agricultural biotechnology can contrib-
ute to the economic growth of many coun-
tries by commercializing plant cultivars with
improved genetic characteristics. In the face
of limited resources and the high cost of
research, the undesirable effects of past
adoptions of technology have made the
public and research community more aware
of the importance of evaluating economic
impacts before adopting a technology. Lack
of information about the potential effects
of biotechnology has led to a discussion of
its environmental hazards (Brill), social is-
sues (like those described by Buttel), and
patenting of altered cultivars (Schmid). Bio-
technology has also been the focus of
several agricultural economists (Butler;
Sundquist et al; Lu; Harl; Hill et al; Offut
and Kuchler; Kalter and Tauer; Hueth and
Just).

The U.S. Department of Agriculture (USDA)
and the Office of Technology Assessment
of the U.S. Congress have conducted stud-
ies on the impacts of agricultural biotech-
nology as well as other modern technolo-
gies. But most of these studies have avoided
the issues of distributional welfare effects
as the technologies have been commer-
cially adopted.

Analysts of economic and social conse-
quences have often relied on qualitative
evaluation without quantifying the results.
Any attempt to quantify something so com-
plex as the welfare effects of adopting a
technology must incorporate numerous
simplifying assumptions and be limited to
partial equilibrium analysis. Useful insights
into relationships among the important
variables can be obtained by quantifying
at least first-order consequences of change.

With the use of static analysis, this man-
uscript provides estimates of the welfare
changes associated with commercialization
of twelve alternative plant biotechnologies.
Welfare is measured by consumer and pro-
ducer surpluses. The analysis gives infor-
mation about the potential regional real-
location of agricultural land and about
welfare gains to consumers. The models
used in the analysis demonstrate long-

term, aggregate effects of each new tech-
nology, assuming full adoption. Impacts of
each technology are examined indepen-
dently of the other eleven; no simultaneous
adoption of two or more technologies was
allowed in the model.

Description of Selected
Alternative Biotechnologies

Symbiotic Changes. This biotechnological
alternative focuses on improving the ability
of plants to obtain nitrogen from the soil.
Nitrogen fixation technology would enable
corn and other plants to fix nitrogen on
their roots much as the soybean plant does
now. This symbiotic technology would be
of great importance to farmers because
nitrogen fertilizer represents a significant
cost and because the lack of supply in
many countries prohibits use of fertilizer
at the optimum levels.

New Rhizobia Strains. Genetic changes in
rhizobia, bacterial species capable of fixing
nitrogen, are also receiving considerable
attention. The symbiotic relationship be-
tween rhizobia and legume crops is rec-
ognized as having significant economic im-
portance in agriculture and the cost of food
production. Developing new strains of rhi-
zobia that will be effective on crops other
than legumes could extend this beneficial
relationship to other crops. These rhizobia
may also be altered to increase the amount
of nitrogen fixed in the symbiotic relation-
ship currently found in most legume crops.

Altered Protein Content. Biotechnology can
alter the chemical composition of grains.
Protein content and quality, in terms of
amino acid balance, are of special impor-
tance. In countries suffering from protein
deficiency, increased protein would directly
improve nutritional levels. Higher protein
grains could also reduce the cost of pro-
duction for livestock where supplemental
protein is now required.

New Resistant Varieties. Besides directly de-
stroying the plant, pests can indirectly cause
plant loss by creating an environment con-
ducive to other diseases. Resistant varieties

of plants can be developed by directing
genetic changes. The southern corn leaf
blight is one example of how genetic dif-
ferences have altered the impact of a dis-
ease. Developing corn plants resistant to
aspergillus flavis is also a high research
priority in many countries.

Frost Tolerance. Frost damage to crops fre-
quently lowers the yield or totally destroys
crops in many parts of the world. Some
plants are highly susceptible, others quite
tolerant to cold temperatures including
frost. Genetic manipulation could increase
the resistance of important crops to the
danger of occasional frost.

Herbicide Tolerance. Increased herbicide tol-
erance would increase crop yields and the
value of a crop where treatment for pests,
diseases, or weeds harms the protected
crop. Genetic changes could neutralize the
impact of herbicide residues in the soil.
Significant commercial progress has been
made on this technology in recent years.

Heat Tolerance. Heat stress significantly
lowers the yields of many crops. In some
cases, it precludes the cultivation of some
feed and food cereals in regions where
they are needed. The global warming trend
may also increase the interest in heat tol-
erance for many crops now grown in tem-
perate regions. Developing varieties that
are resistant to heat is one of the objectives
of genetic manipulation.

Plant Growth Regulators (PGRs). These reg-
ulators stimulate or retard plant growth.
Some growth regulators are now in com-
mercial operation. For example, wheat may
be sprayed to control plant growth and
prevent lodging. Future developments in
biotechnology may widen the array of uses
and plants for which PGRs are effective.

Ice-Retarding Bacteria. Crop production can
be influenced not only by genetic changes
in the plants but also by genetic changes
in the microorganisms associated with the
plants. Ice-retarding bacteria, one such ex-
ample, has been chosen for analysis in this
study because it has been developed to the
point of experimental application. Presence
of these microorganisms prevent damage
to crops from low temperatures of sur-
rounding air.

The Model

Assuming that the agricultural sector op-
erates under conditions of perfect compe-
tition, by maximizing the sum of consumer
and producer surpluses incorporated in an
objective function, we can estimate changes
in welfare after the use of selected agri-
cultural biotechnologies. This procedure
was developed by Samuelson and made
operational by Takayama and Judge
through application of quadratic program-
ming (QP). As an alternative to QP, Duloy
and Norton suggested a linear-program-
ming (LP) algorithm with grid linearization.
This method allows an analysis of both
separable and nonseparable demand func-
tions and has been recommended for ag-
ricultural sector analysis (McCarl and
Spreen). Taylor et al. applied this grid
linearization to separable demand func-
tions. In case of nonseparable demand
functions, the price is expressed as a func-
tion of parameters of substitutes' demand
functions. An integral incorporated into the
objective function of the model used to
estimate welfare and distribution effects of
technology adoption for a two-commodity,
two-market case is:

= /Q,c-(flc.
\

bcst)

bce)

/ (bscd + bm) (b« + bx) \
\1 - ((&„, + &„) (bcd + bct))J

where subscript c refers to one commodity,
for example, corn; subscript s refers to
another commodity, for example, sorghum;
d indicates the domestic market, e the
export market; Q, is the total quantity of
the domestic and export market; P is the
price of each commodity. The equation
becomes a part of an integral measuring
the area under the demand function for
commodity c, with Q,s as an argument. A
similar derivation procedure was followed
in specifying the objective function used
in this study (Florkowski).

Future characteristics of agricultural bio-
technologies, yield, and the use of fertilizer
and pesticides, must be incorporated into
a model in order to provide a reliable
solution. With information about the di-
rection of changes in input use from an
international survey (Florkowski and Hill),
we assumed, after consulting agronomists,
that the size of increase or decrease in a
specific input use would amount to 10
percent of its cost. The international survey
also provided estimates of expected yield
changes (Table 1). Twelve of the twenty
technologies included in the survey were
included in the analytical model. The prin-
cipal criterion for selection was the prob-
ability of rapid commercial adoption as-
signed by survey respondents. Other
selection criteria included possible future
changes in input use induced by biotech-
nologies, expected yield changes, goals of
plant-breeding programs, and the availa-
bility of the necessary data for specifying
a model. This study considered only the
impact of genetic improvement through
biotechnology and the minimum changes
in input use. Other factors have been omit-
ted, such as machinery that decreases soil
compaction and increased yields from
higher concentrations of atmospheric car-
bon dioxide or improved management.

The formulated benchmark model used
statistics published in the final version of
the 1982 Federal Enterprise Data System
(FEDS) Budgets on the cost and quantities
of inputs applied per acre for nine row
crops: barley, corn, cotton, oats, peanuts,
rice, sorghum, soybeans, and wheat. In

order to arrive at the cost of production at
the regional level, the production costs
reported for states were weighted by the
state's share in acreage planted of a crop
in a given region in 1982.

The costs used in the model, which are
variable cost categories reported by budget
data and fixed costs, include the cost of
machinery, tractors, and general farm over-
head. The cost of share rent was excluded
because it represents a part of the calcu-
lated consumer and producer surpluses.

This study uses Soil Conservation Service
(SCS) data on yield adjustment for different
land classes compiled in 1974 (USDA, 1975)
and used in studies by Nicol and by Taylor
and Frohberg. Data on the amount of land
available in different quality classes were
obtained from the USDA Natural Resource
Inventory. Yields in the model were cal-
culated as simple averages of yields re-
ported by the SCS and adjusted by land
class.

The averages were further adjusted for
genetic yield improvement between 1974
and 1982 in order to make the yield data
correspond to 1982 cost estimates. The size
of the yield increase from genetic gain was
obtained through field tests (Miller and
Kebede; Meredith and Bridge) or through
interviews with experts (Hymowitz; Lam-
bert). In the case of cotton and peanuts,
no field test data were available, so we
adjusted yields by calculating a percentage
increase using the difference between
weighted average yields for the early 1970s
and 1980s. We calculated weights as a
percentage share of the total regional har-
vested acreage for each state. The yield of
rice in the benchmark model remained
unchanged because the comparison of av-
erages between the two periods did not
show significant differences. We assumed
that the yield of oats and barley increased
by 0.3 percent annually because of genetic
improvement. For crops not explicitly in-
cluded in the model, an estimate of total
land area allocated to these crops was
withdrawn from the total land available in
the model for crop production.

Table 1. Expected Percentage Changes in Yield from Application of Selected Biotechnologies

Technology

Corn

Rice

Sorghum Soybeans Wheat

percent

Symbiotic'

-4

-3

-8

7

-2

New rhizobia strainsb

4

1

-2

12

0

Altered protein0

-3

-5

-8

1

-4

Virus-resistant11

8

9

9

7

9

Bacteria-resistantd

8

6

5

10

6

Fungus-resistantd

10

12

7

12

7

Insect-resistant5

10

10

10

8

11

Frost-tolerant'

6

9

8

7

8

Herbicide-tolerant8

4

9

5

10

6

Heat-tolerant"

5

5

7

12

8

Plant growth regulator'

12

13

3

12

5

Ice-retarding bacteria'

5

6

6

6

4

'Biotechnology altering a plant in order to induce a symbiosis between a plant and nitrogen-fixing bacteria.

bGenetically altered rhizobia strains that through symbiotic association with a plant increase the amount of nitrogen

available to a plant.

'Biotechnology increasing the content of digestible protein in kernels.

dVirus-resistant, bacteria-resistant, and fungus-resistant plants developed through biotechnology are plants that are

resistant to economically important diseases caused by viruses, bacteria, or fungi.

Tlants resistant to insect damage.

'Plants tolerating below freezing temperatures through their internal mechanism.

8Plants tolerating a high level of herbicide spray.

hPlants that have a high tolerance for extreme temperatures during a growing season.

'Plants responding to plant growth regulators applied during a growing season by increased yields.

'Genetically altered bacteria that when sprayed on frost-sensitive plants delay ice crystal formation, preventing frost

damage.

Individual crops, crop mixes (McCarl), and
rotations used in the benchmark model
were based on USDA agricultural statistics
for the last 5 to 10 years, on the graphic
summary of the location of crop production
from the latest U.S. census of agriculture,
on FEDS Budgets, and on personal inter-
views with agricultural experts from a
number of states. Crop mixes were fixed
for each region, and the share of total acres
allocated to each crop in any given mix
was based on the historical data provided
by various sources. The acreage of each
crop within the crop mix was constrained

by that crop's share of total acreage and
by the total acreage of cropland available.
Individual crops not included in a crop mix
were constrained only by the available
acreage of cropland in the region.

Benchmark Model Solution
and Validation

A benchmark model for ten regions of the
United States including nonirrigated and
irrigated land is presented below.

MAX s = 2 D. - 2 2 2 c^* -22

2

j

MAX S = Maximize the sum of the consumer and producer surpluses, subject to the
following constraints:

1. Land constraint

2 Akmi < 14 — total nonirrigated land; for all k, m, j
/
2 Aki < Lf — total irrigated land; for all k, i, j

i

2. Commodity balance

222 YL/.^ +222 n.^ - 2 Qs ^ o for aii «

m / t I > * p

3. Constraints on steps for area under the demand function for each commodity

2 ZE < 1.0 for all H

p

4. Demand-supply balance

2 Q"n ^ Tn for all n

P

5. Constraints on acreage allocated to each crop
2 A'mln < 2 ak Lkm for all k, j, n; 0 < a < \

m m

2 Al,n < 2 *>" L? for all k, j, n; 0 < b < I

i i

where

a is a coefficient allocating a proportion of the total available nonirrigated

land class to production of a crop under a given technology;
b is a coefficient allocating a proportion of the total available irrigated land

class to production of a crop under a given technology;
i subscript denoting the quality class of irrigated land;
;' subscript denoting the cropping pattern, including rotational schemes and

mix of crops in the region;

k subscript denoting the geographical production region;
m subscript denoting the quality class of nonirrigated land;
n subscript denoting a commodity;
p segment of the demand schedule;
A acres of crops produced;
C cost of production per acre;
D area under the demand curve;
L total acres of cropland;
Q quantity of a commodity represented by the area under a segment of the

demand curve, D;
T total production of a crop;

Y yield of each crop under the given technology, crop mix, and land class;
Z activity representing the pth segment on the demand schedule;
Alj acres of crop production under cropping pattern ;' on irrigated land class

i in region k;
Akin acres of crop production under cropping pattern ;' producing commodity

n on irrigated land class i in region k;
Akmj acres of crop production under cropping pattern ; on nonirrigated land

class m in region k;
Akmin acres of crop production under cropping pattern / producing commodity

n on nonirrigated land class m in region k;
C* cost per acre of producing crops under cropping pattern ;' on irrigated land

class i in region k;
Ckmj cost per acre of producing crops under cropping pattern / on nonirrigated

land class m in region k;

Dn area under the demand curve for commodity n;
Lf total acres of irrigated cropland of class f in region k;
Lkm total acres of nonirrigated cropland of class m in region k;
Qpn quantity of commodity n corresponding to D£

Tn total production of commodity n in 1982

YJ,n yield per acre of commodity n produced under production cropping pattern

j on irrigated land class i in the region k;
Ykmjn yield per acre of commodity n produced under production cropping pattern

/ on nonirrigated land class m in region k;
Z£ activity in the model drawing an amount equal to the area under the

demand curve at the pth step on the function. This amount is drawn from

the demand-supply balance for commodity n.

All of the above parameters and variables
are assumed to be non-negative. Ranges
on the production of crops included among
the constraints of the benchmark model
were used in order to arrive at a solution
that would be comparable to land alloca-
tion for actual acreage use in 1982.

The percentage absolute deviation (PAD)
was used as a criterion for model evaluation
(Norton and Schiefer). The PAD value of
6.74 percent for the benchmark model
indicates that model acreage allocation dif-
fered by 6.74 percent from the actual
acreage allocation of 1982.

Comparisons between the 1982 actual yield
and yields in the benchmark solution were
used to further validate the benchmark
model (Table 2). Yields for irrigated and
nonirrigated fields were combined to obtain
average yields for each region and crop.
Areas where the benchmark model yield
did not deviate by more than 10 percent
included the Northeast, Delta, Appala-
chian, and Pacific regions. Yield for the
Lake States, the Southeast, and Northern
Plains had a deviation larger than 10 per-
cent. These larger deviations included the
yield of oats in the Lake States, soybean
yield in the Southeast, and sorghum yield
in the Northern Plains. The yield of wheat
and oats deviated more than 10 percent
from the actual average yield in the solu-
tion for the Corn Belt. Similarly, the yield
of oats and corn in the Mountain region
deviated more than 10 percent from the
1982 level. The solution obtained for the
Southern Plains showed deviation above
10 percent in the cases of corn, sorghum,
and cotton. Some of the discrepancies be-
tween average yields and actual yields were

the result of inadequate data on crop yield
on irrigated land by land class. Comparison
of calculated yields with long-term average
yields revealed differences with regard to
the same crop — differences that, on oc-
casion, were larger than the 1982 averages.

Prices obtained from the model solution
reflected the market equilibrium deter-
mined by the model. The differences be-
tween the actual 10-year average prices
and estimated prices were substantial (Ta-
ble 3). Among the reasons for those dif-
ferences were the omission of some eco-
nomically important crops, inaccurate
estimates of long-term, own-price and
cross-price demand elasticities (Table 4),
imperfections in real markets, and forces
distorting equilibrium prices. As an ex-
ample of market imperfections, rice is traded
on a thin market (Rastegari-Henneberry).
Accuracy in estimating price elasticity is
diminished by the fact that the prices of
some commodities, such as barley, cotton,
or soybeans, are influenced by their dual
usage or the demand for a joint product.

Given the objective of the study, that is,
the evaluation of a potential change in
economic welfare due to the commercial
application of an agricultural biotechnol-
ogy under long-term equilibrium condi-
tions, the set of prices generated by the
model was considered satisfactory evidence
that the model's specifications were correct.

Results

Impact of Biotechnologies. Solutions of the
model provided estimates of land allocation
after the application of biotechnologies (Ta-
ble 5). The commercial planting of cultivars

Ol

u

O

ID CM vO

ID

o

2

d

CO i-i i-i

d

OO

CO 0 0

OO

o

Tj*

CM CO I-H

SO

tN

*•

"*

SO

_^

C!

3

C

tN so CO

OS p vO

OO Os OS

ID ID fN

O

ra

Cl_

1-H Tjl ID

OO OO' 1-H

SO i-i os'

SO tN tN

tN

Ol

^^

CM tN OS

CO ^ O

O OO OS

—f 1-H OS

tN

PH

tn

1-H OS

OO OS I-H

SO so

O O

O

C

CO CM

CM CM

1— 1 1-H

3

O

a.

•

C

0

OO SO tN

CM OO CO

p

OO O tN.

vO p ID

£

•**' CM' ID

OO CO OS

06

06 d co'

vO Os O

o

ID Csl OS

I-H CM OS

rf

V^Q (NJ ^\

OS sO OS

U

tN tN.

tN tN

SO

tN tN

CM •*

1-H

13

O)

SO CM I-H

^^ ON LO

00 ^H

CM OS tN

OO I-H SO

O CM tN.

IN. CO ^

CO so SO

•^ CM i-H

tN TJH O

OS rH ^f

CO CO i-i

so d

0 0 r-H

OS <* O

OO ID CM

OO I-H ID

tN tN O

IN. ID vO

^J< OS O

£

CO •<* O

1-H

CO CO O

1-H

CO ^f

1-H

CO rji I-H

CO CO Os

CM CO Tf

CO CO O

1-H

ID ID OS

ID ••* OS

(A

C

(0

CM

SO tN 0

ID OS I-H

O OO CM

1— 1 CO tN

CM ID CO

Tf CM Tt"

CM OO CO

o

IN

•* tN 1-H

tN OS OS

t ID ID

SO OS OO

ID tN OS

CO SO I-H

O I-H ID

jf

CM

CM CM I-H

1-H

(N CM 0

1-H

CO CO 0

1-H

CO CO 0

CM CM O

1-H

CM CM OS

CO CO O

1-H

6

3

<u

OS

tN. IN.

00

ID

tN

TJ< CM •*

so co oq

O

&C

^

csi

CO OO

SO

^

CO

•^ i— i O

O ID SO

IN!

i_,

C

SO O

IN.

Tf

ID

SO tN 1-H

Tf CO OO

tN

O

en

T-H

1-H

a.

V

on

•« co

SO

"*

CO O O

o •* ts

O

CM SO OS

1-H rj< IN.

•<* CS) 1-H

CM

13

2 o

tN

ID

d IN i-!

i-H OO CM

Q

tN TjH CSJ

so' ID OO

ID CO ^

d

O

11

ID

ID

so so I-H

1-H

so so rt

IN.

CO ^1* OO

ID ID Os

ID SO I-H

1-H

IN

H

ID O CM

"* ID

Tf CM OS

OS •**< ID

oq os I-H

OO

oo co ^

Tf SO »-H

ID O CO

o

tN. CSI i-H

OS ~~t"

ON ID ID

odd

CM' I-H I-H'

fn)

•<*< co d

OS SO tN

VO cs| CM

U

OS so O

tN OO

OS O O

1-H 0 OS

CS| 1— 1 OS

tN

O vO so

OS 00 OO

CM ^ I-H

JH

tN CO OS

p ID

OO

OS 1-H CO

SO vO 00

so oq ID

ID

)_,

SO I-H CO

d so

CO

IN! 06 d

CM

d ••*' IN!

d ID oo

1-H

•

(O

pa

ID O O

1-H i-H

ID O

ID

ID ID O

1-H

"

ID ID O

1-H

sO SO 0

1-H

SO

'aJ

I

(0

^ 1*

ll.

73

73

73.

1 ^

1 L

73

OJ -
•*-• 4-*

73

O) -

-W 0-J

la.

cs

OO

73

13

1-1 §

y ui oi

^ W PH

HI

-H 13 (-

111

< W PH

ill

y 05 ju

< W PH

ill

x c/i o>

< W PH

ill
<&&

JJ '« a3
< pq PH

III

2 <Q JU

"^ W PH

III

H crt 0*

<! W PH

Ill

£

c

1

8.

n

C

%

'f3

T3

2

0)

E

PH

<£

£

•1

Ol

<s

1

1

Southeast

Appalach

5
C/)

J

"oJ

CO

1
U

3
D

Southern

Northern

Mountair

U

<d

'0

(2

'Estimated

Table 3. Estimated Quantities and Prices for Selected Crops

Price

Crop

Estimated

Actual

Percent*

dollars VCT bushel

Barley
Corn

4.04
5.90

2.94
3.19

137.41
184.95

Cotton

.31"

.71"

43.66

Oats

2.63

1.82

144.51

Peanut

.2659"

.2659"

100.00

Rice

.3829b

.1257b

304.61

Sorghum
Soybeans
Wheat

8.24
8.26
5.27

2.92
7.89
4.38

282.19
104.69
120.32

'Estimated price as a percent of the 10-year average price.
bDollars per pound.

Table 4. Price Demand Elasticities Used for Estimating the Benchmark Model

Demand elasticity

Crop

Domestic

Export

Barley
Corn

-.40
-.70"

-1.51
-1.31"

Cotton

-.12

-.80

Oats

-.85

NAb

Peanut

-1.60

-3.20

Rice

-.11

-1.30

Sorghum
Soybeans
Wheat

-2.20'
-.30

-.55

-2.36C
-2.80
-1.82

"Cross-price demand elasticity of corn with regard to sorghum is 0.14.

blnsufficient export volume to estimate elasticities.

'Cross-price demand elasticity of sorghum with regard to corn is 1.79.

with higher protein content, virus- and
bacteria-resistant cultivars, and heat-tol-
erant cultivars will also cause a withdrawal
of more than 20 million acres of land.
According to solutions of the model, less
land will be withdrawn with the introduc-
tion of cultivars that can establish symbiotic
relationships with nitrogen-fixing bacteria,
are resistant to insects, or are tolerant to
frost and herbicides.

In general, agricultural production will be-
come limited in areas with soils susceptible
to erosion because land capability classes
five and six will be withdrawn from pro-
duction. It also may be that in some south-

ern regions, insect-and-weed pressure may
outpace the benefits of biotechnologies that
lower pesticide use under the assumed crop
mix. The use of new cultivars generally
will increase the acreage allocated to row
crops in the Corn Belt and the Southern
Plains. In the Corn Belt, new cultivars will
lead to larger production of commodities
under consideration. But the comparative
advantage of the Corn Belt will decrease
with the introduction of bacteria -resistant
and heat-tolerant cultivars elsewhere and
the use of improved rhizobia strains. After
commercialization of any of these twelve
technologies, an increase in acreage planted

73
.£

CO

—
-
a.

_O
03

—
0

2

CO

CO

o

CO

-

0)

M

(O

-

C/9

O

to

E

I

o

= v>
.2 S.

I*

•a as
S.2

in

u

c£

§1

6

D *"" '

cn

I?

u «

^ 3
j£

cux;
a.

3 "

O

tn

s

Technolo

QOOO

I i— i CM CM CM <— i I I CM I CM i— i

I I I I I I II

OOOOOOOOOOCMO

O

in

I

LO O^ t*"** ^^

<N in iO ON

i i « i

CO ^^ *"^ CM CO ^1^ OO CM *"^

i— ii— i in in i— i i CM txm
III II I II

^ + +|+CM+1+ + +,+

cs

*_ o mO'-iooooooooN'-1

ocoincoo i* oooroin

VO CO

1 7

IIII

ooinoo-^<oooooooin+
+ 1+7^ + + +!

t-xooinomooinO'— i
I ^ i CM + i i + ox

oo

i— i o fMvo+i in >— i m i

I I

III

*-H ^H to CN 10 m co *~< '•^ '•^ m 10
ooirjmrMiofNOOOir>tn

CS|fN»—t\JD I »— I i— » CN CS CN| r-n r-<

I I I I I I I I I I I

*-" ^ I I

, «

OO

in

I

o

.2 "C
c 5o "C

(0 Ol rO

in the Southern Plains can be expected.
Solutions obtained for the Southern Plains
should be evaluated cautiously because of
specification problems related to irrigated
acreage.

Total Welfare Changes. The process of pro-
duction and adjustment to a new market
equilibrium leads to a change in the welfare
of producers and consumers. Predicted
changes were based on several simplifying
assumptions. In particular, the quantities
demanded by domestic and export buyers
were expected to increase by no more than
50 percent above 1982 consumption. The
domestic demand was assumed to be ine-
lastic. Shifts in crop acreages among re-
gions, changing crop mixes, and altered
input use changed the production and price
of each crop and determined the size of
consumer and producer surpluses. Under
these assumptions, an increased supply
lowers the price and increases consumer
surplus. But the producer surplus can
change in either direction because of the
interaction of production costs and input
use (Tayler et al.).

The change in total welfare, measured as
the sum of producer and consumer sur-
pluses, differed with the technology ap-
plied (Table 6); but the average gain in the
total surplus for all technologies in the
model amounted to $13.4 billion. The total
surplus — including both producer and
consumer surpluses — will be the largest
following commercialization of cultivars
that contain higher amounts of protein,
that are resistant to diseases caused by
viruses and bacteria, or that react to PGRs.

According to the solutions of the model,
total surplus was increased with the adop-
tion of any of the technologies. Gains from
seven of the technologies — symbiotic,
fungus-resistant, insect-resistant, herbicide-
tolerant, and frost-tolerant cultivars as well
as ice-retarding bacteria and new rhizobia
strains — were below the average of $13.4
billion. The other five technologies resulted
in above-average increases in total surplus
relative to the benchmark solution
(Table 6).

Distributional Effects Among Sectors. Al-
though adoption of all technologies pro-
duced an increase in total welfare, the
distribution of welfare among producers
and consumers differed with the nature of
the new technology. Solutions of the twelve
models indicated the largest gain to con-
sumers was from the development and use
of cultivars with altered protein content
(Table 6). The consumer surplus increased
by $51.1 billion. Producer surplus de-
creased by $835 million after the commer-
cialization of cultivars with altered protein
content, but the decrease was less than
that for any other technology.

Another technology that resulted in large
welfare gains to consumers was the de-
velopment of virus-resistant cultivars. The
gain of $35.2 billion in consumer surplus
offset the decrease of $4.8 billion in pro-
ducer surplus — the second largest of any
biotechnology examined. The application
of bacteria-resistant cultivars resulted in an
increase in consumer surplus of $23.0 bil-
lion and a decrease of $3.8 billion in pro-
ducer surplus.

Disease-resistant cultivars could bring some
of the largest gains in consumer surplus.
They also may cause some of the most
significant decreases in producer surplus,
but developing cultivars resistant to bac-
teria and viruses is not an easy task. The
numbers of both viral and bacterial dis-
eases are large, and absolute success is
unlikely. Nevertheless, results of this study,
stress the economic importance of disease-
resistant cultivars.

The technology that resulted in the largest
decrease of producer surplus, $5.0 billion,
was the application of plant growth reg-
ulators. The increase in consumer sur-
plus — a substantial increase, though not
the largest one — amounted to $19.3 bil-
lion. The use of cultivars with a symbiotic
mechanism for fixing nitrogen caused a
medium-sized decrease in producer sur-
plus, compared with the effects of other
technologies. At the same time, the increase
in consumer surplus with this biotechnol-
ogy was one of the smallest ($4.7 billion).

10

*g

.SB-

a> S

c

J/j

<u

^ t*x

00 CO

1-1 Os

^ co

m os

OO so

SO

QJ Q;

c

Tf CO

CM

i— i 1—1

1—1

in

b s

a.

C 3

i— i en

C

O

u

en

)H

Jj

¥ en

"3 ^

ro ^f

SO IX

IX 00

^ CO

•* oo

in I-H

in

6 3

o

rH 0

OS OS

CM Os

CM OO

Os OO

OO Os

CM

3 "n

^-oo

in IN.

sO Os

oo in

in os

CO Os

O Os

in

C £

0 en

go

fN lO

»-H r-l

Os CM

in ^

o tx

CO I-H

CM i-H
CM CM

CM SO

i—l r— 1

tx CO

CM 1-1

CM

U

e

'"en

_3

CD 3

«.

C/J

en en

(u

<u

sO tx

CM O

OO so

o in

in oo

o in

oo

03

s >-

s^

1 1

i v

1 1

V '

1 1

V '

SO

*5>

o

ai 3

a.

1

1

1

1

o

Q-g

•

C

u

03

Lri

Oi en

CO 1-1

O"\ ^*

IX OO

"* OO

m rj<

in so

in

^~

y

SO

CM CM

CM "*

in oo

IX CM

oo tx

1-1 O

Os

•^

•o "F^I

o

CM CM

CM CM

CM so

CO Os

O so

sO

O

8 s

Os

so in

OO •*

in so

•* so

sO ^f

T^ SO

in

c

_0

£ <"

"*

•^ •*

•* Tjl

en

-a

o

0

g

0.

o.

sd

as

*iC

— "en

.5

u

QJ

.S 3

"o

C

£

el) "On

en B
to 3

o

oo tx
so in

SO CM

tx co
in tx

O CO

SO 00
CM O

O CO
CO ^
Os IX

1—1 Os
Os O

CM Os
CO CM
CM IX

CM

in

CO

<u

8

T3

O
g

•a
a>

Jl

(f>

K

•2

.«

1-1 0

in co

Os tx

Os i— '
i— t

1-1 in

•** CO

i— i

CO

1—1

c
•o

u

. g

"S

43

••-.

to

C £

g

u

O v

09

CD

•£ c

C

£

-. ^

a>
C9

«

60

on

° 01

CO

09

"B-

oo

SO

so in

CO CM

in 1-1

CM ••*

T^ SO

00 00

OO i-l
Os 1-1

Os CM
IX so

0 IX

O OS

o

CM

'S,

j3

a 60
g £

—3

3

OO

in 1-1

Os CM

O 00

tx so

IX Os

I-H m

CM

^

en

SO

OO 1-1

tx tx

SO CO

SO 00

OO 1-1

1-1 O

0

*m

^ 1

t-

"<3

in

in so

O OO

tx so

SO so

in so

tx so

IX

O

£ I

V3

s

O

£. »

•S 01

^B

•o

2 ']?

3

s.

£ §

«

I

m u
U S

c

** , ,

13

en

w

* 'm

CO

ffi

O

^

i °

P

(A

3 **"

C

3

yi T3

i

o
o

CO

35
•2

Technology

Benchmark

Symbiotic
New rhizobia strai

Altered protein
Virus-resistant

Bacteria-resistant
Fungus-resistant

Insect-resistant
Frost-tolerant

Herbicide-tolerant
Heat-tolerant

3 ^
OJO u
0) (0

11
00 *

c V
E £

bo

V-i

<u

B-

3

0

C

i

bPercentage change in
'Average was calculate

Among biotechnologies for stress tolerance,
development of frost-tolerant cultivars,
herbicide-tolerant cultivars, and ice-retard-
ing bacteria caused relatively small de-
creases in producer surplus: $2.4 billion
for frost-tolerant cultivars, $2.7 billion for
herbicide-tolerant cultivars, and $2.5 bil-
lion for the development of ice-retarding
bacteria. The gain in consumer surplus for
the frost-tolerant technology was $14.2
billion; for herbicide-tolerant cultivars, it
was $4.6 billion; and for the technology of
ice-retarding bacteria, it was $6.2 billion.

Solutions were influenced by the meth-
odological framework, the assumption
about future demand, and the regional crop
mix. Results indicated the existence of mul-
tiple optimal solutions, a natural occurrence
in a competitive environment (Paris).
Therefore, any future changes in agricul-
tural policy, economic conditions, or tech-
nological development could alter the im-
pact of commercial biotechnology used in
agricultural production.

Distributional Effects Among Regions. Spatial
distribution of aggregate income will vary
among technologies. The withdrawal of
large portions of acreage from production
in the Delta and the Southeast can be
expected to decrease the total farm revenue
in those regions (Table 3). A similar situ-
ation will occur, to a lesser degree, in three
other regions: the Appalachians, the Moun-
tains, and the Northern Plains.

The production of major crops is likely to
remain concentrated in the Lake States and
the Corn Belt. As a result, a larger portion
of the aggregate income will be received
by mid western producers. The effects of
income concentration in the Corn Belt and
the Lake States are strengthened by the
cropping pattern that consists primarily of
corn and soybeans, and to some extent,
wheat. The model reflected the domination
of the commodity markets by corn, soy-
beans, and wheat. Depending on govern-
ment programs, land may be removed from
production in the Delta and Southeast, or
alternative crops may be introduced.

Shifts in the spatial distribution of income

will generate a second wave of effects.
Because new technologies will be neutral
with respect to economies of size, benefits
from their application will occur in pro-
portion to acreage planted with new cul-
tivars. New technologies may accelerate
the trend toward larger farms. Also, if
information about new technologies is not
made equally available to all farmers, early
adopters, who often are large farm oper-
ators, will be among the first to identify
and use the opportunity for increasing their
income.

Environmental Impacts. Application of all
new technologies, except for new rhizobia
strains and symbiotic nitrogen fixation, ac-
cording to results of the international sur-
vey would lead to increased use of nitro-
gen, phosphate, and potash fertilizers.
Plants can only use a portion of the fer-
tilizer applied at any given time because
their nutrient requirements are limited, be-
cause their root zone is finite, and because
moisture often cannot be controlled. There-
fore, increased use of fertilizers, particu-
larly nitrogen, increases the content of un-
desired chemical substances in the soil.
Leaching of nitrogen is particularly harmful
because it causes water pollution and leads
to additional costs related to upgrading
water quality and maintaining drainage.
Increased use of fertilizers as a result of
some applications of biotechnology may
not be welcomed by environmentalists,
despite increased commodity supply and
lower prices.

Application of cultivars resistant to viruses,
bacteria, fungi, and insects would lower
pesticide use. A decrease in the use of
pesticides would slow down the develop-
ment of mutant insects. It would also help
to eliminate some fears of harmful pesticide
residue in agricultural commodities.

Changes in plants will cause researchers
to focus on manipulating specific, well-
characterized genes (Brill). It seems unlikely
that an addition of several genes to a plant
could create a weed. In the opinion of
experts, weeds require a large number of
genetic traits in order to maintain their
character. If any negative characteristics do

12

occur, breeders can recognize them in a
plant; and "because the genetic alteration
in a recombinant plant is well-controlled,
the likelihood of a problem is far less than
[it is] in standard breeding practices" (Brill),
which mix specific and uncharacterized
genes of different plants. In addition, the
safety of developing biotechnologies and
their application has been assessed (Fiskel
and Covello) and is regulated.

Conclusion

Model solutions suggest a decrease in total
acreage used for the production of nine
crops subject to the analysis following the
introduction of biotechnology. Irrigated and
nonirrigated land withdrawn from pro-
duction is located in the Delta and South-
east and, to a smaller extent, in the Ap-
palachian, Mountain, and Northern Plains
regions. The affected regions represent a
range of different climates and growing
conditions that offers a potential for de-
veloping specialized agricultural produc-
tion, which could potentially neutralize the
negative effects on farm income.

A decrease in agricultural activity will slow
the degradation of the environment. Re-
planting the withdrawn land with peren-
nial or cover crops would lower soil ero-
sion. The technologies presented in this
paper that would cause the largest relo-
cation of crops and prove beneficial from
the standpoint of soil protection are the
use of PGRs, heat-tolerant cultivars, bac-
teria- and virus-resistant plants, and cul-
tivars with altered protein content.

The four technologies most beneficial to
society, as measured by the change in total
surplus are cultivars with altered protein
content, virus- and bacteria-resistant cul-
tivars, and cultivars responding to PGRs.
This ranking was largely influenced by the
size of consumer surplus, which was the
highest for these technologies. All biotech-
nologies negatively affected producer sur-
plus — the smallest effect being that from
commercialization of cultivars with altered
protein content, and the largest being the
effect of widespread use of PGRs. Under

the assumption of no change in demand,
a larger volume of commodities causes
lower gross income in the aggregate as a
result of a decrease in prices. In the cost
data used in this model, the new technol-
ogies did not sufficiently reduce the cost
of production to compensate for lower
prices.

The introduction of new technologies de-
creases aggregate farm income, as meas-
ured by producer surplus. But aggregate
income of the agricultural sector in each
region will be affected differently. A larger
portion of total farm income will go to
producers in the Midwest. Individual farm
income may decrease or increase, depend-
ing on market price and skillful application
of the new technologies. The reduction in
farm income shown by the models is the
direct result of increased supply under the
assumed price elasticities. The negative ef-
fects on the producer sector can be alle-
viated by expanding demand, finding new
uses, and controlling supply through gov-
ernment action; by transferring income from
consumers, processors, and other groups
that benefit from lower crop prices; and
by lowering costs of production.

The impact of biotechnology as presented
here illustrates a polar case of a long-term
full adoption of twelve separate technol-
ogies applied to a limited number of field
crops. The information about potential fu-
ture land allocation and welfare changes
contributes to the constantly expanding
pool of knowledge concerning predictions
of the impact of agricultural technology.
Specifically, this study indicated to agri-
cultural research administrators the per-
ceived probabilities of developing different
biotechnologies and economic impact of
their commercialization. Allocation of re-
search funds may be determined not only
by the short-term success in developing a
technology but also by its long-term wel-
fare effects. Welfare effects, in turn, may
not be limited to the easily quantifiable
changes in total surplus. These may also
include the technology on quality and sus-
tainability of natural resources, such as
unpolluted water or uneroded soil. Some

13

of the biotechnologies considered in this
study will lower pesticide use and with-
draw land from agricultural production.

Policymakers may use the information from
this study to formulate policy goals that
would make the necessary adjustment eas-
ier and to fully explore benefits offered by
the use of biotechnology in crop produc-
tion. For example, programs for alternative
land use or economic programs that sustain
rural community growth may be needed
as agriculture diminishes in importance.

For farm groups and checkoff programs,
the results of this study suggest paying
more attention to the demand for agricul-
tural crops. Traditional food, feed, and fiber
use of grains, oil crops, and cotton could
be augmented by industrial uses of crops.
Industrial use of agricultural crops would
change the demand structure and create
new markets. Checkoff funds applied to-
ward research on new uses of commodities
and on feasibility studies of new markets
can make biotechnology work to the benefit
of farmers.

14

References

Brill, W.J. "Safety Issues in Perspective" in
Biotechnology in Agricultural Chemistry. Eds.
H.M. LeBaron, R.O. Mumma, R.C. Honey-
cutt, and J.H. Duesing. Washington, D.C.:
American Chemical Society, 1987.

Butler, L.J. "Current Status of Emerging
Agricultural Biotechnologies." Unpub-
lished paper presented at the American
Association of Agricultural Economics An-
nual Meeting, August 1983.

Buttel, F.H. "Biotechnology and Genetic
Information: Implications for Rural People
and the Institutions That Serve Them." The
Rural Sociologist 5(2):68-78.

Duloy, J.H., and R.D. Norton. "CHAC. A
Programming Model of Mexican Agricul-
ture" in Multi-Level Planning: Case Studies
in Mexico. Eds. L. Goreux and A. Manne,
pp. 291-337. Amsterdam: North Holland
Publishing Co., 1973.

Fiksel, J., and V.T. Covello, eds. Biotech-
nology Risk Assessment. Issues and Methods
for Environmental Introductions. New York:
Pergamon Press, 1986.

Florkowski, W.J. "Economic Impact of
Commercial Application of Biotechnology
on Selected Crops." Dissertation, Univer-
sity of Illinois at Urbana-Champaign, 1986.

Harl, N.E. "Economic and Environmental
Effects of Developments in Biotechnology."
Unpublished paper given at the Western
Agricultural Economics Association An-
nual Meeting, San Diego, July 9, 1984.

Hill, L.D., J. Stockdale, H.F. Breimyer, and
G.E. Klonglan. Economic and Social Conse- .
quences of Biological Nitrogen Fixation in
Corn Production. AE-4608. Urbana-Cham-
paign: University of Illinois Department of
Agricultural Economics and Illinois Agri-
cultural Experiment Station, 1986.

Hueth, O.L., and R.E. Just. "Policy Impli-
cations of Agricultural Biotechnology."
American Journal of Agricultural Economics
69(1987):426-431.

Hymowitz, T. Personal communication.
May 12, 1986.

Kalter, R.J., and L.W. Tauer. "Potential Eco-
nomic Impacts of Agricultural Biotechnol-
ogy." American Journal of Agricultural Eco-
nomics 69(1987):420-425.

Lambert, R.J. Personal communication. May
12, 1986.

Lu, Yao-chi, ed. Emerging Technologies in
Agricultural Production. Washington, D.C.:
USDA Cooperative State Research Service,
1983.

McCarl, B.A. "Cropping Activities in Ag-
ricultural Sector Models: A Methodological
Proposal." American Journal of Agricultural
Economics 64(1982):768-772.

Meredith, Jr., W.R., and R.R. Bridge. "Ge-
netic Contributions to Yield Changes in
Upland Cotton." Genetic Contributions to
Yield Gains of Five Major Crop Plants. Pro-
ceedings of a Symposium Sponsored by
Division C-l of the Crop Science Society
of America, 2 December 1981 in Atlanta,
Georgia. CSSA Special Publication No. 7.
Ed. W.R. Fehr. Madison, Wisconsin: Crop
Science Society of America, 1984.

Miller, F.R., and Y. Kebede. "Genetic Con-
tributions to Yield Gains in Sorghum, 1950
to 1980." Genetic Contributions to Yield
Gains of Five Major Crop Plants. Proceedings
of a Symposium Sponsored by Division C-

1 of the Crop Science Society of America,

2 December 1981 in Atlanta, Georgia. CSSA
Special Publication No. 7. Madison, Wis-
consin: Crop Science Society of America,
1984.

Nicol, K.J. "A Modeling Approach to the
Economic and Regional Impacts of Sedi-
ment Loss Control." Dissertation, Iowa State
University, 1974.

Norton, R.D., and G.W. Schiefer. "Agri-
cultural Sector Programming Models: A
Review." European Review of Agricultural
Economics 7(1980):229-265.

Rastegari-Henneberry, S. The World Rice
Market. Giannini Foundation Information
Series, No. 85-2. Davis: Giannini Foun-
dation of Agricultural Economics/Univer-
sity of California Division of Agricultural
Sciences, 1985

15

Safety in Biotechnology Working Party of
the European Federation of Biotechnology.
"Safe Biotechnology. General Considera-
tions." Applied Microbiology and Biotech-
nology 21(1985): 1-6.

Samuelson, P.A. "Spatial Price Equilibrium
and Linear Programming." American Eco-
nomics Review 42(1):283-303.

Schmid, A.A. "Biotechnology, Plant Variety
Protection, and Changing Property Insti-
tution in Agriculture." Unpublished paper
presented at the American Association for
the Advancement of Science Annual Meet-
ing, New York, May 27, 1984.

Sundquist, W.B., K.M. Menz, and C.F. Neu-
meyer. A Technology Assessment of Com-
mercial Corn Production in the United States.
Station Bulletin 546-1982. St. Paul: Min-
nesota Agricultural Experiment Station,
1982.

Takayama, Takashi, and G.G. Judge. Spatial
and Temporal Price and Allocation Models.
Amsterdam: North-Holland Publishing Co.,
1976.

Taylor, C.R., and K.K. Frohberg. "The Wel-
fare Effects of Erosion Controls, Banning
Pesticides, and Limiting Fertilizer Appli-

cation in the Corn Belt." American Journal
of Agricultural Economics 59(1977):25-26.

Taylor, C.R., PJ. van Blokland, E.R. Swan-
son, and K.K. Frohberg. Two NationalSpatial-
Equilibrium Models of Crop Production. AERR
147, February 1977. Urbana-Champaign:
University of Illinois Department of Agri-
cultural Economics and Illinois Agricultural
Experiment Station, 1977.

U.S. Congress. Office of Technology As-
sessment. Technology, Public Policy, and the
Changing Structure of American Agriculture.
March 1986. Washington, D.C.: Office of
Technology Assessment, 1986.

U.S. Department of Agriculture. FEDS
Budget 1982. September 1984. Stillwater,
Oklahoma: USDA Economic Research Ser-
vice — Oklahoma State University, 1984.

U.S. Department of Agriculture Soil Con-
servation Service. Basic Statistics — 1977
National Resources Inventory. Statistical Bul-
letin Number 686. December 1982. Ames:
Iowa State University Statistical Labora-
tory, 1982.

U.S. Department of Agriculture Soil Con-
servation Service. Unpublished data on soil
loss and yield adjustment for land resource
areas and land capability units. 1975.

16

1o

OJ

O 0

o en m oo

o o m

oo in o o

ovoommooooooooo

0 0

0

m o o o o vo o

.c

Ol f-H T— t

*— <

eNi

^^ VO *m? * ^O — LO

£

in 3 ^

t— <

s

CO

Ol

O 0

o o o in

•<* in •*

o o o o

OOOOOOOOOOOOO«N

O 0

o

0 O O O O O O

o

C/5

CM

00

E

Ov

tn

OL,

3

o o

o o o o

o o o

o o o o

oinoo\o\ooinoomooo

o o

o

rx. vf> o o o so o

£

£P

^-HCNr-i^-irsifN^en en

CM

£

|H

o

D_

U

en

.6-

"0

2

Ol

ex

Ol

o

O 0

0 0 O O

o o o

CN O O O

oooooooooooooo

0 0

o

0 O O O O O O

1

(O

5

\o

0 tx O

01

2

t— 1 1— 1

TJ

1

IS
>,

j;

<o

3

D.

8

o o

O tx O es

O 0 0

o o o o

oinoooooooocNOOO

o o

0

0 0 0 0 0 O O

s

00

s

(0

" *— <

1

o

<n

CO

o

"u

01

T3

(A

—

o cr>

O O O O

o in o\

o o o o

OOOOOOOOOOOOOOO

0 0

0

in o o o o o o

1

JO

RJ

0

i-i CN

t

01

*o

1

4-*

C

c

^

Ol

">

03

^i

1

—

0

U

o o

o in «N ts

O O O

o in o o

oinvominMJOinooenooo

m r-l T— (

0 0

o

o o in o o CN o

'€

•o

e

o.

O

o

S

£

O en

0 0 en en

00 0 OS

0000

OCNrl.eSO^txOOOOOO^

0 0

0

0 en o 0 ^ CN 0

£

-*e

o

OO IX

CN (N en en

^ LT> ^

in

i— i CN i— i

00 0

CN CN CN

S

ro

U

r-t

E

w
Jl

u

^

|

05

V,

Z

CO

Ol

("I <"")

o o o o

oo o en

o o o o

OOOOOOOOOOOOOM5

o o

c

in o in o vo o o

oi

0)

CO

T— 1

*o

CN CN r-i en CN en

u

CQ

C

^»

en

^

|

as

•o

n

3
U

1

V)

£

1

£ <N

U

a

m

CN K>

'5

o

CO

1 E

CA tn

(A CO

£

^ tn •* •" J co C

tn

>

S2 « ••*

to

01

0

^— 1

"u

•* •£

> Ol Ol

cn 0> ^

CN

•^ <U it 09 (^ -^ {J *J ^ r^

£

u

Ol CN

•£..01 tn

X

^pi

c

T3

"o

o o

vO IX o

/S ^

? CO CO ^ • f^ C^J CO ^* ^

(N 0

CO

O p^

£

o

C

.3

£=3
en co

CO (O fO W

r4 \o 13

=3 = _- ^

1 ^ ^ «N CO ^ ^ tN tN 1 4 CN in ^

t— i cB

Cl

to »~H en en en en co

1

<

c

09

O)

01

2

CO

<x>

M

£

u

X

J-~ •-

•B

u

*^~

— ^ »-S

^s

(0

o

(0

(A
nj

PH DL, CU

CL,

CO

£

Allocation

_o

:2

Northeast

« y

(D CQ

1 1

en <

•4-*

03

•M

en

Ol

3

4-*

c 5 2
o "o3 "3
U D Q

1 J 1

O O Jx

en en Z

Northern

Mountain

c
'fa

c <y
§ -o

S fi

OS *O

"u a;

"S a

JS

« A

1.5M-3-90-United Graphics-MET

vv;?i

s^i.n^

-

'^is^S^-
&£atyi

1S1H

"$&&y$*

V

. ,.„ - .-.-,. .,. .

-s '.••'?.,"'•-; 'V-,i S'v"'; ,-.~> '"'•'. - > "V-V-. :>, :'-v-i-i • '.'-i ,;-, -v '1 ' •"> ^ i "' • •' "^ ^ ^ . i ••> '•• '"1 ft >"S'i i

s^-^

' ^v **r ^

;t\V

UNIVERSITY OF ILLINOIS-URBAN*