-------
.seven percent • were^ determined to be in vulnerable financial
condition; Survey observations relating to this group pf farms were
used to develop the characteristics of the Kansas wheat cattle farm
in vulnerable financial condition.
In the baseline .(no EPA actions) the vulnerable Kansas wheat cattle
farm goes out of business in 1993. The decline in net cash farm,
income experienced by the vulnerable farm under the maximum impact
case for-'""Scenario 1 causes it to go out of business one year earlier
than in the baseline. The farm does not go out of business earlier
than 1993 under any of the other scenarios.
RESULTS OF SPECIALTY CROPS IMPACT ANALYSES
The impact of .EPA actions on specialty crop producers was estimated
in a two-step process/ similar to that used -for livestock and major
field'crops.. First, commodity price changes resulting from, EPA
actions were predicted. Next, the new set of commodity prices,
along with the initial cost and yield impacts were used to determine
the impacts of EPA actions on the net returns per acre (returns to
land and farmer; provided labor) of, selected producers via income
budgeting analyses.
Results of average and maximum impact cases for four of the
specialty crops under consideration for Scenarios 1 and 3•are
presented^ below along with? a :brief introduction of 'this .crop;.,
Results of the inc.ome ^budgeting analyses for all scenarips are
contaihe'd in Appendix E- along; with: the initial .cost and yield impact,
estimates.
As this study developed, data deficiencies forced the exclusion of
caneberries and peanuts from the analysis. Data which were
available are presented in Appendix E along with those of other
specialty crops.
Applas
Apple production in the UvSl has approximately dbubled;since the
19,4'Osv-• The trend in cuJLtivars has b£en toward /higher -quality;
dessert apples. Current cultivars of major importance, are Red .
Delicious (39 percent), Golden -Delicious- (17 percent), Mclritosh (7
percent), Rome (6 percent), Granny Smith (6 percent), Jonathan (A
percent) and York (4 percent).
Apples are grown widely throughput'the U.S., with commercial
production in about 35 states. However, the-principal states (and
their approximate share of total U.S. production) are Washington (36
percent), New York (12 percent) and Michigan (10 percent).
Harvested acreage in these states is ^approximately 161,000, 62,000
and 68,000 acres respectively. According to ,1982 estimates,
Washington, has the largest number of farms with approximately 5,400,
followed by Michigan with 2/800 and New York with 2,000. .
27
-------
In recent years apple production has been most profitable in the
Washington growing areas.where slightly higher yields and higher
valued production more than offset higher per. acre production costs.
Returns have been more modest in New York and Michigan growing
areas.
SCENARIO 1
Apple producers in all three study regions (Washington, New York,
Michigan) experience similar decreases in net returns per acre under
Scenario 1 -- from $2..30 to $6.60 per acre — but these decreases
are higher on a percentage basis' .in. Michigan, .because, of the state's
lower2average returns-per-acre (Figure 10). Decreases;in: net
returns under Scenario 1 .are caused by farm worker safety
restrictions arid restrictions on the use of prganbphosphates.,
SCENARIO 3
Changes in net returns per acre for the average impact. ;case under
Scenario 3.differ substantially among production regions (Figure
il). Net returns increased 18 percent in Washington in 1990 while
during the same year net returns in New York and Michigan decreased
134 i percent and 214 :perceht respectively,. Such dramatic decreases
in net returns, may bring about- substantial structural changes, the
discussion of: which is beyond the sccipe of this study. ' The .large
differeritiai in net returns among different regions is due to
"proposed restrictions on the.use of fungicides in 1990. These
restrictions would- substantially affect New York and Michigan apple
production (e.g., 17 and 12 percent • yield reductions) but have no
production effect in Washington. 5/ ; The rise in Washington
producers', net .returns is.due to the. i. 8 .percent increase in price
above the base year.caused by the national decline in apple supply.
Potatoes
Potatoes are groWn; commercially in nearly every state. Total U.S.
production ranges :ftorn .16 to 20 million; tons, • depending on the year.
Of this production:, approximately one-third is used for. table stock
and one-half for processing. The remainder is used for seed,
livestock feed, and export.
While potatoes :are grown throughout the U.S., production is .
concentrated in several areas. The most important area is Southern
Idahp, which typically accounts for'about 25 percent ..of total
production-. Southr-central Washington is the second largest
5/ The fungicide restrictions considered :under Scenario 3 are the
~ cancellation of.all .EBDCs and chlorothalonil (see Appendix A) .
See Appendix E, Table E-2 for regional cost and yield impacts..
28
-------
Impacts on WA Apple Net Returns
impacts on NY Apple Net Returns
330r-
325
320
315
220r-
215
210
Average annual change (1987-1996):
Average Impact Case: $-2.30 (-.71)
Haxlnua Impact Case: $-3.30 (-11)
Average annual change (1987-1996):
Average Impact Case: $-4.4X1 (-H)
Haxlmm Impact Case: $-6.60 (-31)
IB87 1989 1991 1993 1995
1988 IS90 1992 1994 1996
1087 1869 1991 1993 1995
taaa leao 1992 1994
Yeai
Year
Impacts on Ml Apple Net Returns
flOr-
75
70-
65
Average
Maximum
Base
Average annual change (1907-1996):
Average Impact Case: $-3.20 (-41)
Maximum Impact Case: 1-5.60 (-71)
1987 1989 1991 1993 1995
taaa 1990 1992 1994 i996
Yeai
Figure 10. Scenario 1 regulatory impacts on apple production
-------
Impacts on WA Apple Net Returns
Impacts on NY Apple Net Returns
Average annual change (1987-1996):
Average Impact Case: $.70 (.21)
Maximum Impact Case:' $-9.90 (-31)
2SO«-
200
160
100
GO
0-
•50-
•100
•160
1987 1989 1991 1993 199S
1988 1990 1992 1894 1996
Average annual change (1987-1996):
Average impact Case: $-132.00 (-COS)
Kaxtoun Impact Case: $-163.00 -741)
1987 1889 1991 1993 1995
1988 1990 1992 1994 1998
Yaw
Year
Impacts on Ml Apple Net Returns
Avaiaga
Minimum
Base
•250-
Average annual change (1987-1996):
Average Impact Case: $-67.00 (-841)
Maximum lopact Case: $-145.00 (-1021)
1987 1989 1991 1993 1995
1988 1990 1992 1994 1996
Year
Figure 11. Scenario 3 regulatory impacts on apple production
-------
production area, followed by the Red River Valley of North Dakota
and Minnesota, and northern Maine. Together these regions account
for up to 60 percent of total U.S. production, with Washington-Idaho
harvesting approximately 437,000 acres, North Dakota-Minnesota
194,000 acres, and Maine 98,000 acres. According to 1982 estimates
of potato farm numbers, Washington-Idaho has approximately 2,400,
followed by North Dakota-Minnesota with 1,400 and Maine with 1,100.
Cultural practices vary among the major production regions. In
Idaho and Washington most of the potato acreage is irrigated and
crop yields are among the highest in the country. Acreage in the
Red River Valley and Northern Maine is primarily dryland with
appreciably lower yields and more modest contributions to farm
income from an acre of production.
SCENARIO 1
Net returns per acre in 1987 for the average impact case are
slightly lower than the baseline in all regions due to effects of
the 1984 cancellation of EDB and the 1987 suspension of dinoseb
(Figure 12). In 1990 net returns for Washington-Idaho producers
increase above the baseline by .2 percent (average impact case)
while net returns for the other regions also increase, but still
remain below the baseline. This is explained by the simultaneous
increase in the national price (.26 percent above the baseline) and
proposed 1990 groundwater regulations which do not affect the
Washington-Idaho producers.
In all three production regions the decrease in net returns is
substantially larger in the maximum impact case than in the average
impact case. Average annual net returns (1987-1996) decreased by .7
percent in Washington-Idaho, four percent in Minnesota-North Dakota,
and 8 percent in Maine under the maximum impact case. Maximum
impact estimates are considerably larger than the average for such
regulations as the dinoseb cancellation in 1987 and the groundwater
regulations in 1990 'because only a small percentage of producers are
affected.
SCENARIO 3
Results of regulatory impacts on potato producers' net returns per
acre are dominated in this scenario by the 1990 proposed restric-
tions on organophosphate use (Figure 13). Average impact estimates
in 1990 include 6.4 and 7.0 percent yield declines in Minnesota-
North Dakota and Maine respectively, while the yield decline in
Washington-Idaho was estimated at .96 percent (less organophosphates
are used in this area). Such a large decline in production results
in price increases of 1.8 percent above the base year of 1987 to its
highest level during the study period. In Washington-Idaho this
increase in price was able to offset the relatively small decline in
yield and net returns actually increased above the baseline for the
31
-------
Impacts on WA/ID Potato Net Returns
Impacts on MN/ND Potato Net Returns
610r-
605
600
595
Average •mual change (1987-1996):
Average Impact Case: «•«>£;«)
Maximum Impact Case: $-4.20 (-.71)
590 1987 i5l9 iS5i 5553 iSST
igaa 1990 1002 1994 iwe
245,-
240-
235-
230-
225
220
Average annual change (1987-1996):
Average Impact Case: $-1.90 (-.81)
Maximum Impact Case: $-9.60 (-41)
1987 1989 1991 1993 1995
1988 1990 1992 1994 1996
Year
Year
ui
K>
Impacts on ME Potato Net Returns
135
130
125
120 -
115
no
Average annual change (1987-1996):
Average Impact Case: $-1.00 (-.81)
Maximum Impact Case: $-10.00 (-81)
1087 1989 1991 1993 1995
1988 1990 1992 1994 1996
Average
Base
Year
Figure 12. Scenario 1 regulatory impacts on potato production
-------
Impacts on WA/IO Potato Net Returns
Impacts on MN/ND Potato Net Returns
700r-
650-
600
550-
500
450
Average annual change (1987-1996):
Average Impact Case: $18.00 (3X)
Maximum Inpact Case: $-54.00 (-91)
1987 1989 1091 1993 I99S
taaa 1990 IS92 1994 1996
i
255,-
240-
225
210 -
195
180
Average annual change (1987-1996):
Average Inpact Case: $-12.00 (-51)
Maximum Impact Case: $-26.00 (-US)
1987 1989 1991 1993 1995
1988 1890 1992 1994 1996
Year
u
u>
Impacts on ME Potato Net Returns
I
s
Average annual change (1987-1996):
Average Impact Case: $-13.00 (-10S)
Maximum Impact Case: $-27.00 (-211)
1987 1989 1991 1993 1995
1988 1990 1992 1994 1998
Year
Figure 13. Scenario 3 regulatory Impacts on potato production
-------
average impact case. In the other regions/ the commodity price
increase was modest in relation to the crop yield decreases/ and net
returns decreased sharply.
Maximum impact results are substantial in all production regions. A
yield reduction of eight percent was applied equally in all regions
as the result of the proposed 1990 organophosphate restrictions.
This reduction in yield when combined with other regulatory actions
resulted in an average annual decrease in net returns of nine
percent in Washington-Idaho, 11 percent in Minnesota-North Dakota,
and 21 percent in Maine during the 1987-1996 period.
Ton&tioos
Tomatoes rank second to potatoes in dollar value among all
vegetables produced in the U.S. Nearly 85 percent of total
production is used for processing/ with the remainder utilized
fresh.
California is the major tomato growing area, typically accounting
for about 75 percent of the total U.S. crop. Ninety to 95 percent
of the California crop is used for processing. Florida is the
second largest state in terms of production/ accounting for six to
eight percent of total U.S. production. Unlike California/ nearly
all Florida production is for the fresh market. California harvests
approximately 225,000 acres yearly while Florida harvests 45/000
acres. There are approximately 1600 tomato farms in California and
400 in Florida.
The value of tomatoes is much higher for the fresh market/ compared
to the processing market. Fresh market tomatoes are typically worth
approximately $500 per ton at the farm gate/ with some variance
depending on season/ location/ and quality. Tomatoes used for
processing are typically sold by producers for $70 to $80 per ton.
Yields per acre are also quite different for processed and fresh
tomatoes. Tomatoes used for processing are generally direct-seeded
(without transplanting) and have relatively higher plant populations
per acre. Tomatoes for the fresh market/ at least in Florida/ are
generally transplanted/ and the plants are staked; per acre plant
populations are much lower.
Net returns per acre of production are considerably higher for fresh
tomatoes grown in Florida than for California processing tomatoes.
While tomatoes grown in Florida for the fresh market have lower
yields and higher growing and harvesting costs, the higher price
they command more than offsets these factors. Net returns to
management and land are estimated at $1500 per acre compared to $700
per acre for California processing tomatoes.
34
-------
SCENARIO 1
The impact on net returns per acre from regulatory actions in the
tomato producing regions of California and Florida are very similar
(Figure 14). The 1988 farm worker safety regulations produce a
minimal (less than .3 percent) decline in net returns as measured by
average impacts. A more noticeable feature of impacts on tomato
producers' net returns is the difference between average and maximum
impacts. This difference is explained by the fact that some
regulatory actions (e.g., the EDB cancellation which occurred in
1984) have a significant effect on a small number of producers.
Under the maximum impact case/ the most severe declines in net
revenue occur in 1987, with reductions of 1.9 and .8 percent in
California and Florida, respectively. Even under the maximum impact
cases the decreases in average annual net.returns per acre are less
than one percent in both Florida and California.
SCENARIO 3
Maximum impacts on yields associated with the proposed 1990
restrictions on fungicides were estimated at 20 percent for both
California and Florida. (>/ Such substantial reductions of. yield
decrease net returns in California by 49 percent and in Florida by
39 percent (Figure 14). Average impacts in California affect net
returns less due to a more modest estimate for yield decline of
approximately 5 percent.
The impact estimates for tomatoes under Scenario 3 must be viewed
with some caution. Yield declines and cost increases were based on
information provided by pesticide registrants that has not been
thoroughly reviewed by EPA.
Green Peas
Green peas are a relatively minor specialty crop, with production
concentrated in the Washington-Oregon and Wisconsin-Minnesota areas.
Wisconsin leads all other states in terms of production:
Approximately 86,000 acres are harvested yearly in Wisconsin
compared to 64,000 acres in Washington. There are approximately
1,700 farms in Wisconsin and 500 in Washington/ Yields in
Washington average the highest in the nation due to more capital
intensive farming practices such as pivot irrigation. This also
accounts for the high cost of production per acre in comparison to
other states.
j>/ See Appendix E, Table E-5 for the regional cost and yield
impacts associated with the fungicide restrictions as well as
other actions affecting tomato production.
35
-------
Scenario 1
Impacts on CA Tomato Net Returns
Scenario 1
Impacts on FL Tomato Net Returns
670r-
660 -
650
1520.-
1510-
1500 -
Average annual chMige (1987-1996):
Average Inpact Case: $-1.10 -.21
Maximum Impact Case: $-5.30 J-.8X)
-J9B7 i§19 iBIi iB3 iS5
1086 1990 1992 1994 1996
Yew
149
Average annual change (1987-1996):
Average Impact Case: S.60 (CIX)
Maximum Inpact Case: $-4.50 (-.31)
1991 1993 1995
1968 1990 1992 1994 1996
Yoar
0k
Scenario 3
Impacts on CA Tomato Net Returns
Scenario 3
Impacts on FL Tomato Net Returns
Average annual change (1987-1996):
Average Inpact Case: $-6.60 (-1X)
Maximum Impact Case: $-132.00 (-20X)
1987 1989 1991 1993 IMS
1986 1990 1992 1994 1996
Yeai
Avaiage
Base
Average annual change (1987-1996):
Average Impact Case: $-210.00 (-MX)
Maximum Impact Case: $-240.00 (-16XJ
1987
1989 1991 1993 1995
1990 1992 1994 I99S
Yeai
Figure
Scenarios 1 and 3 regulatory impacts on tomato production
-------
SCENARIO 1
Average impacts on pea producers' net returns per acre in 1987
result in an initial increase of over one percent in Wisconsin
producers' net returns and a corresponding decrease of over seven
percent in Washington's net returns (Figure 15). This dichotomy
results from the 1987 cancellation of dinoseb which affects only
Washington producers. Their response is to decrease production,
which results in a commodity price increase of .53 percent over the
price in 1986. Wisconsin producers' increase in net returns
reflects this price increase. However, the price increase is not
enough to offset the costs to Washington producers from the
cancellation of dinoseb and their net returns subsequently decline.
Additional regulatory impacts (e.g., farm worker safety regulations
in 1988 and organophosphate restrictions in 1992) combine with a
declining price to decrease net returns in Wisconsin up until 1994.
SCENARIO 3
Regulatory impacts in this scenario are similar to those in Scenario
1 up until 1992 (Figure 15). A noticeable difference occurs in this
year when impact estimates of proposed organophosphate restrictions
increase sharply over those in Scenario 1. Nevertheless, impacts
are still relatively modest even under the maximum impact case when
net returns decline 2.0 and 7.8 percent in Wisconsin and Washington,
respectively, in 1992t the most severe impact year.
Canebarries
Major caneberry crops include red raspberries, black raspberries,
loganberries, boysenberries, and blackberries. Commercial cane-
berry crops are grown in the Pacific Northwest, almost exclusively
west of the Cascade mountains in the mild marine climates of Oregon,
Washington and to a lesser extent in California. Caneberry
production has been declining in recent years, due in part to urban
expansion in the principal berry regions of Oregon and Washington.
A major problem with the estimation of impacts on caneberries is the
lack of information concerning crop production. Very little
information is available regarding pesticide use and the efficacy of
pesticide alternatives. The cancellation of pesticide registrations
can have severe impacts on the industry because of the lack of
efficacious alternatives. In general, only a limited number of
pesticides are registered for use on caneberries. This is largely
because it is such a minor crop and the cost of registering a
pesticide for use outweighs the profits from modest pesticide sales.
Because of the lack of reliable data on caneberry production as well
as the caneberry market, impact estimates associated with regulatory
scenarios could not be completed.
37
-------
Scenario 1
Impacts on Wl Pea Net Returns
£
205r-
200-
195
190
Average annual change (1987-1996):
Average Impact Case: S-.40 (-.21)
Maximum Impact Case: $-.40 (-.21)
u
09
1987 1989 1991 1993 1995
I9B0 1990 1992 1994 1998
Year
Scenario 3
Impacts on Wl Pea Net Returns
205,-
200
195
190
Average annual change (1987-1996):
Average Impact Case: $.10 «.II)
Maximum Impact Case: $-1.20 (-.61)
1987 1989 1991 1993 1995
19BU 1990 1992 -1994 1996
Scenario 1
Impacts on WA Pea Net Returns
80|-
75-
70
65
Average
Maximum
Average annual change (1987-1996):
Average Impact
Maximum Impact
[I a*jc annual wianifv \_v_. »_—,-
Average Impact Case: $-3.20 (-41
- Case: $-4.00 j-SX
1987 1989 1991 1993 1995
1988 1990 1992 1994 1996
Yeai
Scenario 3
Impacts on WA Pea Net Returns
75
70-
65
Average
Maximum
Base
Average annual change (M»7-1M6I:
Average Impact Case: $-3.20 (-41
Maximum Impact Case: $-4.80 (-61)
1987 1989 1991 1993 1995
1988 1990 1992 1994 1996
Yeai Year
Figure 15. Scenarios 1 and 3 regulatory impacts on peas
-------
Peanuts
The peanut ia not actually a nut but rather a legume, more closely
related to the pea and bean. The major peanut growing areas/ are
North Carolina-Virginia, accounting for approximately 15 to 20
percent of total U.S. production, Georgia-Alabama (60 to 65 percent).
and Texas-Oklahoma (10 to 15 percent).
Overall profitability of peanut production depends heavily on the
U.S. farm program for peanuts. According to the farm program,
peanuts are classified as either 'quota' or 'additional', each
having a separate pricing system. The price support for quota
peanuts is based on the national average cost of production from the
previous year,, adjusted to reflect any increase in the average cost
of production, though restricting annual price increases to 6
percent. Quotas were assigned to farmers on the basis of historical
allotments, determined primarily on acreage allotments in place in
1981.- (Quotas in 1980 were based on an acreage allotment. Since
that time they have been defined based on production, with no regard
to acreage.) The quota support price has been $550 per ton since
1983. For purposes of this analysis, quota production was assumed
to equal 0.4 million tons at a price of $558 per ton.
Additional or nonquota peanuts may be grown by anyone. They are
used for oil and export (with some buy-back provision if quota
production is not adequate to meet domestic edible demand in a given
year). The price support for additional peanuts is set to avoid any
net cost to the Government, in effect, making the production of
additional peanuts responsive to free-market condition.
Because of unreliable cost and yield estimates associated with
various environmental regulations and the lack of critical crop
production parameters (e.g., supply elasticities), impact estimates
for the regulatory scenarios could not be completed. However,
several of the regulatory actions are expected to have significant
impacts (over 10 percent decline in yields) on peanut producers
including the suspension of toxaphene, the cancellation of certain
fungicides and use restrictions stemming from pesticides in
grpundwater regulations.
SUMMARY AND RECOMMENDATIONS
Summary results for the representative livestock and major field
crop farms in average financial condition are presented in Tables 1
and 2. Table 1 indicates the average base net cash farm income for
each producer forecasted over the 1987-1996 period and shows the
average annual change in income predicted for the same period under
Scenarios 1 and 3. Table 2 shows the average base debt to asset
ratio and predicted changes for the forecast period. As revealed in
these summary results and the preceding report, on average, major
field crop and livestock producers are not expected to experience
U.S. EPA Headquarters Library
39 Mail code 3201
1200 Pennsylvania Avenue NW
Washington DC 20460
-------
Table 1. Average Annual Effect of EPA Actions on Net Cash Farm
Income (MCFI) 1987-1996 for Farms in Average Financial
Condition (1986 *> I/
Scenario 1
IL Corn Soybean
MS Cotton Soybean
KS Wheat Cattle
Avg . Base
NCFI 1987 -
1996
35,000
58,900
11,600
Avg.
Impact
Case
-270
(-.8X)
-1,700
(-3X)
-380
(-3X)
Max.
Impact
Case*
-2,900
(-8X)
-10,700
(-18X)
-2,800
(-24X)
Scenario 3
Avg.
Impact
Case
+4,800
(+14X)
-1,300
(-2X)
+ 310
< + 3X>
Max.
Impact
Case*
-9,200
(-26X)
-14,200
(-24X)
-9,700
(-84X)
\J Average percent changes are indicated in parenthesis.
* All of the representative farms have a 90 percent chance of incurring
cost and yield impacts that are less than half of those corresponding
to the maximum impact case. The maximum impact cases, therefore,
must be viewed as very unlikely worst cases.
Table 2. Average Percentage Change in Debt to Asset Ratios' (D/A)
Caused by EPA Actions (1987-1996) for Farms in Average
Financial Condition I/
Avg. Base
D/A 1987
1996
Scenario 1
Avg. Max.
Impact Impact
Case Case*
Scenario 3
Avg. Max.
Impact Impact
Case Case*
IL Corn Soybean
MS Cotton Soybean
KS Wheat Cattle
.26
.28
.26
.6X
.3X
IX
6X
3%
-.3X
.5X
.6X
2X
6X
22X
I/ Note that increases in the debt asset ratio (appearing as a positive
percentage change in this table) represent a worsening of a farm's
financial condition.
* All of the representative farms have a 90 percent chance of incurring
cost and yield impacts that are less than half of those corresponding
to the maximum impact case. The maximum impact cases, therefore,
must be viewed as very unlikely worst cases.
40
-------
large financial impacts due to EPA actions. For the average impact
case/ average annual decreases in farm income are three percent or
less and the resulting changes in debt to asset ratios are less than
one percent. Although the average impact cases indicate that, on
average/ the losses, under these scenarios are minor, the impact on
any given producer is a function of both initial' financial and
production conditions and the extent of the initial cost and yield
impacts that are incurred. Large variations in losses incurred by
different farmers under any given set of EPA actions are possible.
Maximum impact cases were designed to set an upper bound on the
losses that each of the representative farms might incur under each
scenario. These cases indicate the income losses that would be
incurred if the representative farms were.assumed to be impacted by
all the EPA actions that could possibly affect them, and represent
unlikely worst case scenarios. Even under the extreme maximum
impact cases/ however, none of the producers in average financial
condition go out of business as a result of EPA actions.
Since the ability of farms to withstand losses is a function of
their initial financial condition/ each scenario of EPA actions was
simulated for representative farms in vulnerable financial
condition. Although the reductions in net cash farm income were
similar for vulnerable farms and farms in average financial
condition/ these income reductions resulted in larger changes in the
debt to asset ratios for vulnerable farms. .Only one of the
vulnerable farms went out of business any earlier than it otherwise
would have due to EPA actions. Under the maximum impact case for
Scenario 1, the vulnerable Kansas wheat cattle farm went out of
business in 1992, as opposed to in 1993 in the baseline.
Because of limited data availability/ the study did not forecast
changes in the financial condition of the specialty crop farms.
Instead/ it examined changes in net returns per acre (which reflect
returns to land and farmer provided labor). Summary results for the
specialty crops are provided in Table 3. The base net returns per
acre are indicated for each of the crop and regions considered/
along with the absolute and percentage changes.
As indicated in Table 3, effects on specialty crop producers are
fairly small under Scenario 1. Net returns are reduced by four
percent or less under the average impact case/ and by eight percent
or less under the maximum impact case.
Both average and maximum impact cases result in significant losses
for specialty crop producers under Scenario 3. The largest absolute
reductions in net returns per acre are incurred by tomato growers in
Florida and apple growers in New York and Michigan, with decreases
in net returns of $210/ $132/ and $67, respectively, under the
average impact case. These dramatic decreases in net returns may
bring about substantial structural changes in the production and
markets for the crops affected. Large differences in the impact of
EPA regulations on crops grown in different regions occurred because
41
-------
Table 3. Average Annual Change in Net Returns Per Acre (NR/A)
Caused by EPA Actions 1987-1996 (1986 «)
Scenario 1
Apples
HA
NY
MI
Potatoes
HA/ ID
MN/ND
ME
Tomatoes
CA
FL
^3^kO Q
reaa
HI
HA
Avg. Base Avg.
NR/A 1987 - Impact
1996 I/ Case
330 -2.30
(-0.7X)
220 -4.40
(-2X)
80 -3.20
(-4X)
600 +.20
240 -1.90
(-0.8X)
130 -1.00
(-0.8X)
660 -1.30
(-0.2X)
1,500 +.60
200 -.40
(-0.2X)
80 -3.20
(-4X)
Max.
Impact
Case
-3.30
(-1X)
-6.60
(-3X)
-5.60
(-7X)
-4.20
(-0.7X)
-9.60
(-4X)
-10.00
(-8X)
-5.30
(-0.8X)
-4.50
(-0.3X)
-.40
(-0.2X)
-4.00
(-5X)
Scenario 3
Avg.
Impact
Case
+0.70
(0.2X)
-132.00
(-60X)
-67.00
(-84X)
+18.00
<3X>
-12.00
(-5X)
-13.00
(-10X)
-6.60
(-1X)
-210.00
(-14X)
+ .10
-3.20
(-4X)
Max.
Impact
Case
-9.90
(-3X)
-163.00
(-74X)
-145.00
(-182X)
-54.00
(-9X)
-26.00
(-11X)
-27.00
(-21X)
-132.00
(-20X)
-240.00
(-16X)
-1.20
(-0.6X)
-4.80
(-6X)
I/ Net returns per acre are based on regional budget information, and arc
assumed constant over the period 1987-1996 in the base case, and are
in 1986 dollars.
42
-------
some of the proposed restrictions involve pesticides that are used
in some regions and not in others. Even though the results of this
study must be considered preliminary/ these figures show that EPA
actions could create economic problems for some specialty crop farms
and suggest that the. Agency exercise considerable caution in this
area.
Impacts on potato producers under Scenario 3 are significant,
although the absolute decreases are relatively small (approximately
$26 in each region) these decreases result in an 11 percent and a 21
percent reduction in net returns per acre in Minnesota/North Dakota
and Maine, respectively.
Impacts on pea producers are relatively modest. Even under the
maximum impact cases for the most expansive EPA scenario/ net
returns per acre are decreased by less than $5.00 in both of the
regions that were examined.
This study illustrates the advantages,, of examining the impacts of
environmental regulations at the farm'level as well as at the
aggregate national level. While national analyses provide useful
information concerning the total losses incurred by different
aggregate types of farmers (e.g./ corn farmers as a whole)/ the
impact of environmental regulations on farms' financial conditions
depends on the distribution of those losses among farmers and on the
initial financial conditions of the affected farms. In order to
determine the effect of EPA regulations on the ability of farms to
survive/ both aggregate and farm level analyses are necessary.
This study highlights the data and analytical requirements necessary
to determine the impacts of EPA actions on agriculture. Such
requirements include:
1. Accurate pesticide usage data/
2. Accurate pesticide efficacy data/
3. Improved information on how initial pesticide
cancellation effects change, over time/
4. Accurate incidence data for non-pesticide related impacts
(e.g./ underground storage tanks)/
5. Improved national price-quantity models to
predict commodity price changes due to EPA
actions/ and
6. Better information on the initial financial and
production conditions of agricultural producers
and farm level models for estimating changes in
these over time.
The need for better data and modeling capability is greatest for
specialty crops, where reliable pesticide usage and efficacy data/
often do not exist/ limited information is available on producers'
initial financial condition, and few models are available. EPA is
43
-------
currently compiling a directory of all specialty crop models.
Improvements in pesticide usage data might be obtained by increased
cooperation and cost sharing with USDA and states to fund additional
pesticide usage surveys or to add pesticide usage questions to
surveys designed for. other purposes. In addition, registrants of
pesticides might be required to provide usage information. Appendix
H provides a discussion of additional options that might be
considered for improving the data available to complete studies of
this type. Reliable pesticide usage data, efficacy data, national
price-quantity models, and farm level models are likely to become
increasingly important in the future, as EPA tries to reduce
environmental risks associated with agricultural production in a
cost-effective manner.
44
-------
AGRICULTURAL SECTOR STUDY
APPENDICES
Appendix A:
Appendix B:
Appendix C:
Appendix D:
Appendix E:
Appendix F:
Appendix G:
Appendix H:
EPA Actions Considered in This Study-
AGSIM Model and Results
National Price-Quality Model and Results
REPFARM Model and Results
Income Budget Analysis and Results
Data Problems and Assumptions
Cumulative Probability Cost Curve Distribution
Recommendations for Acquiring Better Pesticide
Usage Data
-------
APPENDIXiA
EPA Actions Considered In This Study
By
Terry Dinan ^/
and
Susan Slotnick 2/
\J Office of Policy Analysis, U.S. Environmental Protection
Agency
2/ Office of Standards and Regulations, U.S. Environmental
~ Protection Agency
-------
Appendix A
EPA Actions Considered in this Study
As part of this study, each of the program offices at EPA submitted
a description of the regulations that were passed during the past
five years and those that were being considered for the next five
years. These regulations were reviewed to determine which ones
were likely to have a direct economic impact on the agricultural
sector; regulations having an indirect economic impact were not
included in this analysis because of the difficulty in determining
what portion of their cost would be passed on to agricultural
producers. The set of potential direct impacts included:
Air Lead Phasedown: If lead is banned from gasoline/ farmers
that use gasoline powered tractors, combines and trucks
would have to use a fuel additive or rebuild their
valves. These costs were incorporated into Scenario 3.
Air Agricultural Burning Restrictions: Agricultural open
burning of crop residues may be restricted. Possible
control techniques include proper fire and fuel manage-
ment, appropriate burning operations under optimum
meteorological conditions, and alternative residue
disposal procedures. The impact of this regulation was
not quantified in -this study because of insufficient
information on its cost and incidence.
OPTS SARA Title III (jointly with OSWER): Title III of SARA
requires farmers to provide information on the chemicals
that they use and store. The cost of Sections 302-303
are estimated to be approximately $50 per farm, and
apply to 33% of all farms. Farms are.exempt from 311-
312 requirements provided that they dp not.employ more
than 10 full-time employees. This means that virtually
all farms are exempt from Section 311-312 requirements.
SARA Title III costs were incorporated into Scenarios
1-3.
OSWER Financial Responsibility Requirements for Petroleum
Underground Storage Tanks (USTs): Would require farms
with petroleum USTs of greater than a 1,100 gallon
capacity to carry insurance. This would cost farms
$2,500 per year. Information is available on the
number of covered USTs in each USDA production region;
however, no information is available concerning the
types of farms most likely to have them. Insurance
costs were incorporated into Scenarios 1-3.
OSWER Technical Standards for Design and Operation of USTs
Containing Petroleum or Hazardous Substances: By 1991,
farms having USTs will have to begin monitoring. This
A-l
-------
is estimated to cost $500 and will have to be repeated
at least every 3 years. If a leak is found, they will
have to be repaired and upgraded. No information is
available on the likelihood of finding leaks in farm
USTs or the cost of repairing or replacing the tanks.
By year 10, all USTs will have to be brought up to
standards, again. Monitoring costs were incorporated
into Scenarios 1-3. Although there is no information
specific to farm USTs, national data estimate that 15
percent of all USTs may be leaking. The estimated cost
of replacing a 4,000 gallon coated and cathodically
protected tank system is $21,000 and the cost of upgrading
an existing tank is $3,050.
OSWER Waste Oil Management: There is insufficient information
to determine whether this is relevant.
Water Nonpoint Source Guidance and Management Plans: Under
legislation passed in February 1987, states were given
grants to assess the magnitude of NFS problem and to
develop management plans. These plans will have to be
submitted by August 1988. , EPA has until February 1988
to approve the plans. Information from Office, of Water
indicates that this should not be considered a direct
affect on agriculture because EPA cannot force states
to implement their management plans, and because actions
on the part of farmers will be voluntary.
Water Wellhead Protection Program: Section ,1428 of SDWA as
amended in June 1986 mandated states to submit wellhead
protection programs to EPA. Although states are required
to submit plans, there are no federal sanctions for not
submitting except for the withholding of grant funds.
Twenty states have begun development of plans. The
cost question is difficult to address because there |are
no minimum federal standards or management strategies
which states must .include,as part of. an approvable WHP;
therefore, impacts are likely to vary considerably from
state to state. These costs were not quantified in
this study.
Water National Estuary Program: There are no national program
guidances and/or regulations yet associated with the
NEP. The first is expected in 1989. For agriculture,
use of pesticides in certain watersheds may be eliminated
or restricted. Target reductions of nutrient loadings
may be established and BMPs may be put into place by
SCS and state cost sharing programs. No information is
currently available to determine the impact of this
program on agriculture.
A-2
-------
Water Sewage Sludge Regulations: A proposed rule is planned
for October 1988. This rule may limit the amount of
municipal sludge farmers are allowed to use on their
fields. No information currently exists on the limits
that would be imposed or the costs that farmers would
bear as a result of this rule.
OPTS FIFRA/OPP Part 170 (Farm workers): The proposed rule
establishes requirements to improve the occupational
health and safety of workers performing hand labor in
the fields. Specific estimates on per acre production
cost increases for various crops were utilized in this
analysis and were incorporated into Scenarios 1-3.
OPTS Pesticides in Groundwater Strategy: Groundwater
protection may result in prohibitions of certain water
soluble pesticides in areas with vulnerable groundwater.
Three alternative sets of impacts associated with the
Pesticide in Groundwater Strategy were developed by
OPTS and used in Scenarios 1-3.
OPTS Endangered Species Act: Actions that bring EPA into
compliance with the Endangered Species Act will impose
some direct costs on agriculture. No information
currently exits to determine the extent of costs imposed
by the ESA; therefore, these costs were not included in
this analysis.
OPTS FIFRA/OPP Individual Actions: The following individual
actions were included in this study: cancellation of
EDB, toxaphene, dinoseb; restricted use of alachlor;
cancellation of yield enhancement of cholordimeform;
and an expansive, intermediate, and conservative scenario
for actions on the following groups of pesticides:
fungicides, corn rootworm insecticides, broad .spectrum
organophosphates, and grain fumigants.
Direct Impacts Included in the Empirical Analysis;
The objective of this study is to examine the cumulative impact
that EPA policies promulgated over the period 1983-1992 have on
the agricultural sector. It is obviously difficult to predict
what .future EPA policies might look like; therefore, we have
defined three alternative scenarios corresponding to a range of
future EPA policies. The scenarios can best be summarized as
follows:
SCENARIO 1: Past and current EPA actions plus a conservative
(low cost) set of assumptions about future
actions.
A-3
-------
SCENARIO 2: Past and current EPA actions plus an inter-
mediate (mid cost) set of assumptions about
future actions.
SCENARIO 3: Past and current EPA actions plus an expansive
(high cost) set of assumptions about future
actions.
Past and Near Term Actions Included in Scenarios 1-3;
Actions that the Agency has undertaken in the past five years or
plans to undertake .in the very near future were included in all
three scenarios. These actions are:
EDB - cancellation
Toxaphene - cancellation
Dinoseb - cancellation
SARA Title III
Leaking Underground Storage Tanks
Farm Worker Protection Standards
Chlorodimeform - cancellation of yield enhancement
Alachlor - restricted use.
For actions that there is a great deal of uncertainty over, three
alternative plans were considered, with the most conservative
plan being incorporated into Scenario 1, the intermediate plan
into Scenario 2, and the most expansive plan into Scenario 3.
These actions and the alternative plans are listed below:
Fungicides
Scenario 3: EPA would cancel the use of all EBDCs and
chlorothalonil. Captan would not be
cancelled.
Scenario 2: EPA would cancel.the use of all EBDCs.
Chlorothalonil and captan would not be
cancelled.
Scenario 1: EPA would put additional restrictions on the
use of all EBDCs, chlorothalonil and captan
(e.g., restricted use, pre-harvest
restrictions, limited number of
applications).
Corn Rootworm Insecticides
Scenario 3: EPA would cancel all of the corn rootworm
insecticides.
A-4
-------
Scenario 2: EPA would cancel all of the corn rootworm
insecticides with the exception of one of the
organophosphates and one of the carbamates.
Scenario 1: EPA would cancel soil use/ but not foliar
use, of all of the corn rootworm
insecticides.
Broad Spectrum Organophosphates
Scenario 3: EPA would cancel three-quarters of all of the
broad spectrum OPs. The most toxic ones
would be cancelled.
Scenario 2: EPA would cancel one-half of all of the broad
spectrum OPs. The most toxic ones would be
cancelled.
Scenario 1: EPA would place restrictions on the use of
OPs (e.g./ closed cabs).
Grain .Fumigants
Scenario 3: EPA would cancel methyl bromide. Aluminum
phosphine and magnesium phosphine would hot
be cancelled.
Scenario 2:. EPA would put additional restrictions on the
use of methyl bromide, aluminum phosphine,
and magnesium phosphine..
Scenario 1: No action.
Pesticides in Groundwater Strategy
Scenario 3: EPA would cancel the use of aldicarb, alachlor,
and three triazines over the next five years
In^'all counties haying high drastic scores
and 20% of the counties having medium drastic.
scores.
Scenario 2: EPA would cancel the use of aldicarb, alachlor,
and three triazines over the next five years
in 25% of the counties having high drastic
scores.
Scenario 1: EPA would cancel the use of aldicarb in 25%
of the counties having high drastic scores.
Restricted use would be instituted for alachlor
and the triazines. Monitoring would be required
for the triazines that have not yet had
monitoring required.
A-5
-------
Lead Phaseout
Scenario 1,2: A total ban of lead in gasoline (for agricul-
tural use) was not assumed in these two
scenarios.
Scenario 3: EPA would eliminate lead in gasoline for
agricultural use.
Risk Reductions Corresponding to the Actions Considered;
The objective of the preceding report is to estimate cumulative
costs associated with EPA actions. To provide some background as
to why EPA has undertaken, or might consider, the actions listed
above, the following section describes the health .and environmental
risks and exposure pathways associated with the substances those
actions are meant to control.
EDB:
Health effects were the primary concern that motivated the cancel-
lation of EDB. EDB is classified as a likely human carcinogen
and may cause adverse reproductive effects to exposed workers.
The exposure routes were: food consumption, drinking water, and
worker exposure. Cancer risk estimates due to occupational
inhalation of EDB range from 1 x 10~ to 3.6 x 10~ . Millworkers
and farmers had the largest populations of workers at risk, with
16,000 millworkers and 14,000 farmers estimated as being exposed
to EDB through inhalation. Dietary risks occurred through the
consumption of wheat products, citrus, and tropical fruits.
Cancer risks from EDB to the average U.S. consumer were estimated
to be 3.55 x 10~_-due to wheat product consumption and from 2.8 x
10~ to 1.7 x 10~ due to citrus fruit consumption, depending on
state requirements about fumigation.
Toxaphene:
Ecological qamages were the primary concern motivating the cancel-
lation of toxaphene. Toxaphene was found to cause adverse reproduc-
tive effects in fish populations at very low concentrations. It
may be carried for long distance in the upper atmosphere and find
its way into water bodies far from the locations where it was
used. In addition to the concern about fish populations, laboratory
experiments indicated that toxaphene has both acute and chronic
effects on several bird species. Finally, human exposure may
occur both through worker exposure (inhalation and dermal) and
dietary exposure. Estimates of lifetime probability of cancer to
toxaphene applicators (toxaphene was applied to several crops)
ranged from 2 x 10~ to 3 x 1C~ . Dietary risk was estimated to
be the greatest for local populations of fish consumers in areas
where significant fish contamination had been demonstrated.
A-6
-------
Dinoseb:
Exposure to dinoseb may cause a variety of hazards such as develop-
mental toxicity, reproductive toxicity, acute toxicity, induction
of cataracts, and immunotoxicity. An oncogenicity hazard (resulting
in benign tumors) may also exist. A particular concern that.led
to the emergency suspension of dinoseb was its potential to cause
birth defects. Exposure to dinoseb occurred through direct contact
by farm workers. Approximately 45,000 workers/ including up to
2,200 females, were involved in the application of dinoseb. A
large number of farm workers and bystanders had the potential to
be exposed to dinoseb during or shortly after application, and
other people had a chance of being exposed by a secondary route
(e.g., laundering of contaminated clothing). In addition, dinoseb
has been found in groundwater in several states, indicating that
exposure through drinking water is also possible.
Chlorodimeform:
The registrants of Chlorodimeform have voluntarily cancelled it
since the beginning of this project. Chlorodimeform was used
only on cotton. The health risk of concern was the possibility
of cancer in exposed workers.
Alachlor:
Risk of cancer is the primary concern associated with alachlpr.
There are multiple routes of exposure: worker exposure, consumption
of ground water and surface water, and residue on food products.
Farm Worker Safety:
The objective of farm worker safety requirements are to minimize
the acute and .chronic health effects for,pesticide handlers and
field workers. There are approximately 500,000 handlers and 1.8
million field workers. The regulations are directed .primarily
towards minimizing the risk of acute poisoning. There are 20,000
to 300,000 acute poisoning incidences estimated to occur annually
due to farm worker exposure.
Underground Storage Tank Regulations:
The proposed underground storage tank regulations would set
insurance and monitoring requirements for underground petroleum
tanks (with greater than 1,100 gallon capacity) on farms. The
primary health risks associated with leakage from these tanks are
cancer (caused by benzene, a component of petroleum) and fire and
explosion. Ecological damages may occur if leakages found their
way into streams. Risks are greatest in small streams where the
opportunity for dilution is less than in larger streams.
A-7
-------
SARA Title III:
Benefits associated with Title III take the form of "negative
reductions in damages". .Title III is expected to contribute to
human health and welfare in at least two ways: by helping to
prevent potentially harmful releases of hazardous substances, and
by making it possible to reduce the harm from those releases that
still occur.
Fungicides:
The fungicides OPP may consider for cancellation are classified
as probable human carcinogens. Exposure routes for fungicides
are: worker exposure, dietary, and groundwater. Worker exposure
is the primary concern associated with chlorothalonil at this
point, with dietary exposure the primary concern for both 'captan
and EBDCs; however, evidence of thyroid and teratogenic effects
(birth defects) have been found for EBDCs. Chlorothalonil and
EBDCs (or their breakdown products) have been found in groundwater.
Broad Spectrum Organophosphates:
There are both human health and ecological concerns associated
with broad spectrum organophosphates (OPs). The OPs are acutely
toxic. They depress an enzyme that causes an interference with
nerve transmission, and may result in nausea, diarrhea, dizziness,
or death. In addition, some OPs may result in adverse eye effects
(myopia) and neurological disorders. Worker exposure, dietary
exposure, and groundwater contamination are all of concern. '
Ecological impacts are also a concern, since broad spectrum OPs
are acutely toxic to birds and fish, as well as humans.
Corn Rootworm Insecticides:
The health and ecological concerns associated with corn rootworm
insecticides are similar to those for broad spectrum organophos-
phates. However, worker exposure is not thought to be a problem
with corn rootworm insecticides because they are applied in granular.
form, as opposed to a spray. Hazard to bird populations is a major
concern with corn rootworm insecticides.
Grain Fumigants:
Worker exposure is the primary concern with, grain fumigants.
Methyl bromide may result in acute toxicity (possibly causing
nausea, diarrhea, dizziness, or death) while aluminum phosphine
and magnesium phosphine are neurotoxins.
Pesticides in Groundwater:
Alachlor effects and exposure routes are discussed above.
A-8
-------
Aldicarb is an acutely toxic substance that may result in nausea,
diarrhea, dizziness, or death. The exposure paths of concern for
aldicarb are residues on food (mainly potatoes and citrus crops)
and groundwater contamination.
Triazine herbicide (cyanazine, atrazine, and simazine) exposure
may occur through groundwater and surface water. Health effects
are the primary concern for these substances. All of the triazines
are considered possible human carcinogens, and there is some
concern that the triazines can react with nitrites (also found in
groundwater) to form nitrosamines, which are potent animal carcin-
ogens. In addition, exposure to cynanzine may cause birth defects.
Lead in Gasoline:
Lead in gasoline has been shown to increase blood lead levels,
which in turn have been linked to a variety of serious health
effects, particularly in small children. Recent studies linking
lead to high blood pressure in adult males also are a source of
concern. People are exposed to lead from gasoline through a
variety of routes, including direct inhalation of lead particles
when they are emitted from vehicles, inhalation of lead contaminated
dust, and ingestion of lead contaminated food.
A-9
-------
APPENDIX B
AGSIM Model and Results
By
Fred Kuchler
and
Craig Osteen I/
I/ Resources and Technology Division, Economic Research Service,
U.S. Department of Agriculture
-------
Appendix B
AGSIM Model and Results
1.0 Introduction
In examining the impact of EPA actions on the financial condition
of agricultural producers, it is crucial to account for the crop
and livestock price increases that result from these actions.
Failure to account for these price changes would result in an
overestimation of the impact of EPA actions on farmers. The crop
and livestock price changes resulting from EPA policies were
predicted using AGSIM, a regional econometric-simulation model of
U.S. crop and livestock markets (Bales, Frank, Taylor 1987a,
1987b, 1987c). The new crop and livestock prices obtained from
AGSIM under each scenario, were then used as inputs to represen-
tative farm models (along wich additional information on production
costs and yield impacts) to determine the change in financial
condition caused by EPA actions. The set of crop and livestock
prices in the base run of AGSIM (no EPA actions) is presented in
Table B-2 (tables appear at the end of this appendix). The
change in these prices under Scenarios 1, 2, and 3, are presented
in Tables B-6, B-ll, and B-16, respectively.
In addition to providing information on price changes, AGSIM is
useful in predicting the impact of EPA actions on: crop acreage,
livestock production, and changes in aggregate producer and
consumer welfare. All of these impacts are examined in this
appendix; however, only the price changes are essential to the
preceding report. While the examination of these additional
impacts does not shed any further light on how representative
producers are impacted by EPA actions, it provides a more complete
picture of the cost these actions are likely to have on society
as a whole.
2.0 Description of AGSIM
AGSIM simulates regional production of major field crops and
livestock as well as the demand for those commodities. Together
the demand and supply systems provide estimates of commodity
production, distribution, prices, and the economic welfare of
producers and consumers. Initial impacts of EPA actions under
each scenario are expressed as inputs to AGSIM in the form of
increased costs of crop production and reduced crop yields.
The crop supply component of AGSIM is comprised of a set of
supply equations for each of 11 regions. Results from only 10
regions are presented here to correspond to the principal produc-
tion regions. Crops included in the model are corn, grain sorghum,
barley, oats, wheat, soybeans, cotton, and hay. Cultivated
B-l
-------
summer fallow is treated as another land use in semi-arid regions.
Region definitions are presented below.
Corn Belt: Iowa, Illinois, Indiana, Missouri, Ohio
Lake.States: Michigan, Minnesota, Wisconsin
Northern Plains: Kansas, Nebraska, North Dakota, South
Dakota
Southern Plains: Oklahoma, Texas
Mountain States: Arizona/ Colorado, Idaho, Montana,
Nevada, New Mexico, Utah, Wyoming
Pacific States: California, Oregon, Washington
Delta States: Arkansas, Louisiana, Mississippi
Southeast: Alabama, Florida, Georgia, South Carolina
Appalachia: Kentucky, North Carolina, Tennessee,
Virginia, West Virginia
Northeast: Mid-Atlantic States and New England
For each region, the model first determines total acreage planted
or placed in summer fallow and total acreage diverted or set-aside
under farm programs. Then, a set of equations determine the
proportion of acreage planted to each crop. Acreage is modeled
as a function of expected returns, which account for target
prices. Yield per acre, modeled as a time trend.for each crop in
each region is held constant after.1987 (except as altered by EPA
actions). Yield per acre is multiplied by acreage to calculate
production. Summing crop production across regions and adding
inventories determines.crop supply.
Crop demands are estimated for cotton lint, hay, grain exports,
grain stocks, food, soybeans, feed, and cottonseed. The soybean
demand component consists of a crushing, export, and stock demand
function as well as demands for the derivative meal and oil
products. These functions are primarily determined by relative
prices.
Equating crop supply and demand functions and solving the system
of price-dependent equilibrium excess supply equations provides
annual equilibrium prices; Prices from one simulated year are
used to calculate net returns for that year. The system is
recursive. A price from one year may affect acreage response
the following year. Expected net returns drive the acreage
response functions. The maximum of price from the previous
simulated year and the effective support price is used to calculate
expected net returns. That is, price from the previous year
serves as a price expectation for the following year.
The livestock sector of AGSIM is linked to the crop sector through
feed and hay prices which determine the supply and inventory of
livestock products: beef, veal, pork, chicken, and milk.. Also,
quantities of feed demanded are influenced by livestock prices.
B-2
-------
The model runs twice to simulate a technological change. The
initial, or base run simulates commodity market conditions without
any technological change. A second run simulates market condi-
tions under the new,technology, showing differences attributable
to the technology. The three scenarios were simulated by changing
the yields, and both fixed and variable production costs of
selected crops in particular regions.
The principal limitation on the interpretation of AGSIM results
is that the model is not specific and detailed enough to recognize
any particular technological change. That is, any two changes
having identical impacts on net returns would be treated identi-
cally by AGSIM and, thus, calculated economic impacts would be
identical. The factors that might limit use of any particular
technology may not be incorporated in AGSIM. Overestimating
impacts is a real possibility.
Income impacts may be overestimated because AGSIM does not account
for the effects of price changes on commodity program payments.
Commodity programs may stabilize farm income. When prices rise,
revenues derived from commodity sales rise, but deficiency payments
fall, thereby partially offsetting the revenue increase. AGSIM
calculates farm income based on a market price ignoring deficiency
payments and hence the reduction in payments likely to accompany
an increase in market price.
AGSIM simulates production and the operation of commodity markets
over a ten year .horizon. The year-by-year changes cannot be
considered market forecasts. .Instead, the multi-year infor-
mation is designed to provide a longrun description of the policy
impacts. AGSIM is designed to equilibrate supply and demand
forces in each simulated year. Actual commodity markets may
operate, at times, with much greater or lesser speed than AGSIM
suggests. For example, price expectations modeled in AGSIM do
not rapidly adjust to changed conditions. That expectations
mechanism is empirically adequate for historical data. Whether
that expectation formation mechanism will hold in the future is a
matter of .speculation. The particular type of equilibrium assumed
for commodity markets in AGSIM leads to stocks being rapidly
depleted. In recent years, stocks have demonstrated much more
inertia, suggesting that prices may not increase as rapidly as
the AGSIM simulations suggest. Again, these examples indicate
that the presented time paths variables follow are primarily
descriptive, rather than exact.
Information from the AGSIM base run and the three alternative
policy runs is presented in this appendix. Information from a
base run, which is common to all three policy scenarios, is
presented in Tables B-l through B-4. This information includes
crop acreage by commodity, commodity prices (farm level prices
for crops and retail prices for dairy and livestock commodities),
crop and livestock income, and livestock production. Crop income
B-3
-------
is calculated by subtracting most fixed and variable production
costs from gross revenue. Land costs and commodity program
payments are not considered in that calculation. Changes in
acreage,, prices, and income are presented for each policy scenario.
Also, several variables measuring income changes throughout the
agricultural sector and impacts on consumers are shown.
Gains and losses resulting from regulations affecting crop
productivity may go far beyond the farms for which yields and
costs of production are immediately affected. The crop sector
supplies the dairy and livestock sectors. Increased crop produc-
tion costs and reduced production may lead to higher feed costs
and hence, higher meat and dairy products. Other industries
depend on the success of crop enterprises. Industries that
process and market field crops as well as dairy and livestock
products depend on the price and volume of those products. AGSIM
provides, some estimates of the aggregate gains and losses to
industries up and'down the food and fiber marketing chain.
The heading, "Crop Consumer Effect" in the boxheads of Tables B-9,
B-14, and B-19 refers to the sum of gains or losses to consumers
(that is, the effects .of higher prices for all food and fiber
products) and to all industries beyond the farm gate that depend
on crop production. These industries include, but are not limited
to, processors, packers, retail grocers, and transportation
firms. One should expect that as crop production is carried out
less efficiently, farm prices will rise and output will fall.
The intermediate industries will have reduced business and the
price increase, representing higher input prices to processors,
will imply reduced profits for the various processing industries.
With higher input and output prices throughout the marketing
chain, consumers should face higher retail prices.
Similarly, the heading "Livestock Consumer Effect" refers to the
sum of gains and losses beginning with livestock purchasers and
ending with consumers of meat and dairy products. These gains
and losses are a subset, of those included in the "Crop Consumer
Effect". The "Livestock Consumer Effect" is smaller, in absolute
value, than the "Crop Consumer Effect" because the latter effect
includes crop uses that do not support livestock production.
Only a small portion of wheat supply, for example, is used for
livestock feed. Cotton is not used for livestock feed, although
cottonseed meal is used for feed.
Just and Hueth showed that in vertically related industries where
the output of each industry is an input for the industry one
step up the marketing chain, the welfare effects of an imposed
price distortion in an initial or intermediate market on all
forward industries can be captured by measuring the change in
consumers' surplus (the difference between what consumers are
willing to pay and what they are required to pay to acquire goods
and services). That is, if a calculation to compute changes in
B-4
-------
consumers' surplus were carried out on an initial or intermediate-
level general equilibrium demand function, the change should be
interpreted as the change in final consumers' surplus plus the
changes in all forward industry rents. Chavas and Collins
generalized this analysis to include technological change or
distortion. These ideas are incorporated in the AGSIM calcula-
tions presented here.
3.0 Results
As discussed above, the impact of EPA actions on crop producers1
are entered into AGSIM in the form of yield decreases and/or,
production cost increases. These impacts result in a decline in
crop production and an increase in crop prices - a cost for crop
purchasers. Yield and cost changes in Scenario 1 are the least
of the three scenarios. The changes induce losses for both crop
consumers and producers. As a result of higher costs of, feeding
livestock, livestock income decreases, but livestock purchasers.
are affected less since livestock prices change less then crop
prices.
Scenarios 2 and 3 generate greater effects than Scenario 1,
primarily because of larger corn yield declines beginning in
1992. Prior to 1992, these two scenarios have somewhat greater
cost changes than Scenario 1, while Scenario 3 has greater changes
'than Scenario 2. Thus, Scenarios 2 and 3 cause somewhat larger
price changes than Scenario 1, during that time period. As a
result, crop consumers, livestock producers, and livestock
consumers generally lose more and crop producers lose less than
in Scenario 1. Beginning in 1992, prices, in Scenarios 2 and 3
increase so much that crop income increases. In effect, the
relatively large yield and cost changes of Scenarios 2 and 3
cause an income transfer from crop consumers to crop producers.;
Crop consumers, livestock producers, and livestock consumers, lose
more while producers gain more for Scenario 3 than Scenario 2,
during 1992-96.
While crop producers gain in aggregate under Scenarios 2 and 3
during 1992-96, income does not increase for all crops and all
regions. The cost and yield changes cause a complex change of
acreages and prices for different crops. Income decreases for
some crops because price increases do not outweigh cost increases
and/or yield declines. Crop income declines in some regions.
For example, the Northeast and Appalachian States lose in Scenario
3 because they have the highest corn yield losses, despite higher
corn prices.
Scenario 1
Scenario 1 assumes the smallest initial direct changes in yields
and costs among the three scenarios. Only cotton, soybeans, and
B-5
-------
wheat yields decrease, all by less than 0.5 percent. Fixed
costs generally increase by less than $1 per acre, but never by
more than $1.50 per acre. Similarly, variable production costs
generally increase by less than $1 per acre. Thus, changes in
acreage, output, and prices are smaller than changes estimated
for Scenarios 2 and 3.
Acreage and Prices. Total crop acreage steadily decreases, but
never by more than 200,000 acres which is less than 0.1 percent
of baseline total crop acreage (Table B-5). The acreage of all
crops decreases in most years. Price changes for field crops
never exceed $0.022 per bushel and are generally less.than $0.01
per bushel (Table B-6). Retail prices for livestock products
either fail to change or change by less than $0.01 per pound.
Price decreases occur for soybean meal in 1991-93 because soybean
production increases. AGSIM predicts that higher hay prices
encourage the slaughter of cattle and calves. The result is that
beef and veal prices fall by less than $0.01 after 1991.
Income. Since the cost increases outweigh the price increases,
total crop income (net of fixed and variable costs) decreases in
all years (Table B-7). The greatest income loss, $339 million,
which is about 4 percent of baseline total crop.income, occurs' in
1988. The losses become smaller in succeeding years as cost and
yield changes decline. On average, the crop income losses are
less than $1 per baseline crop acre. However, income (net of
variable costs) increases for barley in 1992-93 and for hay from
1991-96. In 1987, there are crop income gains (net of fixed and
variable costs) in some regions. These gains are exceeded by
losses in the Delta States and Southeast (Table B-8). These two
regions have relatively high soybean and cotton yield losses from
1987-89. From 1988 on, income declines in all regions.
Consumer Effects. Crop consumers lose from higher.prices and
lower production (Table B-9). The losses become steadily larger,
varying from $23 million in 1987 and to $95 million in 1996.
Livestock producers generally suffer income losses due to higher
feed and hay costs and unchanged or lower livestock prices
beginning in 1988. The greatest loss, $42 million (less than 0.1
percent of baseline income for the 5 livestock products), occurs
in 1993. Additionally, livestock consumers would gain in some
years and lose in others.
Scenario 2
Scenario 2 has greater cost and yield changes than Scenario 1.
The largest differences from Scenario 1 are the corn yield declines
beginning in 1992 due to restrictions on soil insecticides. Corn
yield losses exceed 8 percent in the Corn Belt and Northern
Plains and vary from 2 to 6 percent in the remaining regions.
The yield losses moderate in later years. Some variable costs
increase noticeably in 1992. Cotton costs increase by $5.40 in
B-6
-------
the Delta States. However, corn costs decrease by less than $2
per acre in the Corn Belt, Northern Plains, Southern Plains, and
Pacific States.
Acreage and Prices. Prior to 1992, price and acreage changes are
greater than Scenario 1. As a result, total crop acreage declines
range from 19,000 in 1987 to 108,000 in 1991 (Table B-10). From
1988 to 1991, total crop acreage declines less for Scenario 2
than for Scenario 1. Higher crop prices in Scenario 2 seem to
explain this result. Soybean price increases by $0.18 per bushel
in 1988 and by lesser amounts in 1987 and 1989 (Table B-ll).
During these three years, the Appalachian, Delta, and Southeastern
States suffer greater soybean yield losses than in Scenario 1.
These initial soybean yield losses reduce soybean and increase
corn and cotton acreage, primarily due to similar changes in the
Southeast. The prices of meal and oil products of cotton and
soybeans increase during 1987-89 as a result of lower soybean
production and higher prices.
The larger corn yield losses (as compared to Scenario 1) beginning
in 1992 cause a noticeable change in results from Scenario 1.
Total crop acreage decreases by 300,000 in 1992, but increases by
46,000 in 1993 to 226,000 in 1996 (Table B-10). Corn price
increases by $0.51 per bushel in 1992 (Table B-ll). AGSIM predicts
an interesting pattern for corn and soybeans. Soybean acreage
increases and price decreases in 1992, because corn cost and
-yield changes reducfe expected corn returns and, hence, planted
acreage. The higher corn price in 1992 encourages farmers to
shift acreage from soybeans and other crops to corn. Corn price
rises less in following years, and the prices of barley, oats,
wheat, soybeans, and cotton also increase as the acreage and
production of those crops decrease. As a result, prices of meal
and oil products of cotton and soybeans also rise. Since sorghum
is a good feed substitute for corn, sorghum demand rises causing
its price and acreage to rise. Hay acreage increases in 1988 and
later years causing price decreases in most years. In 1992,
lower feed and hay prices reduce retail livestock prices. After
tha.t, higher feed costs increase all livestock prices in some
years. However, all price changes are less than $0.10 per pound.
Beef and veal prices increase in some years and decrease in others:
Income. Crop income (net of fixed and variable costs) rises $159
million in 1987, but falls $111 million in 1988 and $315 million
in 1991 after fixed costs increase in 1988 (by the same amount as
in Scenario 1) and groundwater regulations begin in 1990 (Table
B-12). All of those income changes are less than 2 percent of
baseline total crop income. Income (net of variable costs only)
decreases for all crops except soybeans, barley, oats, and hay in
1987 and soybeans in 1988, because cost increases outweigh price
increases. Beginning in 1992, crop income increases because .
price increases, particularly for corn and soybeans, outweigh
cost increases. Crop income increases the most in 1992, $1.9
B-7
-------
billion (12 percent of baseline crop income) or an average of
about $5 per crop acre, but the increases become smaller as cost
and yield changes decline over time. After 1992, income rises
for corn (the crop suffering the greatest regulation induced per-
acre production loss), and soybeans, but decreases for barley,
oats, wheat, and hay in some years.
Prior to 1992, crop income (net of fixed and variable costs)
falls in most regions (Table B-13). In 1987, before fixed costs
increase, income increases in the Corn Belt, Lake States, North-
east, and Appalachian States. From 1988 to 1991, crop income
decreases in all regions but the Corn Belt in 1988-89. In 1992
and later years, income increases in all regions except the
Delta States in 1992 because soybean prices fall and in the
Mountain and Northeastern States in 1996 because higher crop
prices no longer rise enough to outweigh cost increases and yield
losses.
Consumer Effects. Consumers lose much more in Scenario 2 than in
Scenario 1 (Table B-14}. Prior to 1992, crop consumer loss peaks
at $272 million and declines to $45 million in 1991. Because of
the large price.increases after 1992, the consumer loss peaks at
$2.8 billion in 1992 but falls to $1.5 billion in 1996. Due to
higher feed costs and modest livestock price increases, livestock
income declines after 1987. Before the corn yield losses have
their full effect on feed prices in 1993, livestock producer
losses do not exceed $100 million (less than 0.1 percent of
baseline income .for the 5 livestock products). In 1993 and some
later years, their losses exceed $1 billion (2 percent of livestock
income). Before the corn yield losses have their full effect,
livestock .consumers have losses of less than $100 million while
gaining in 1991-92 when beef prices fall. In 1993-94, livestock
consumers lose more than $2 billion. However, lower beef and
veal prices cause consumer gains in 1996.
Scenario 3
Scenario 3 has greater fixed cost changes throughout the simulation
than Scenario 2. Yield losses and variable cost changes are
greater during 1990-96. In particular, greater corn yield losses
occur after 1991, than in Scenario 2. Corn yield losses are
approximately 23 percent for the Northeast, 13 percent for the
Appalachian States, and 10 percent for the Corn Belt and Northern
Plains in 1992. Production costs are also greater than for
Scenario 2; cost increases approach $12 per acre in the Northeast
and Southeast and $14 per acre in Appalachian States in 1992.
The yield losses and cost changes moderate in later years.
Fixed costs also increase more than Scenario 2 but never by more
than $2.25 per acre. The result is greater price changes, income
changes, and consumer losses than for Scenario 2. The two
scenarios produce identical results for 1987.
B-8
-------
Acreage and Prices. Prior to 1992, total crop acreage in Scenario
3 decreases, ranging from 57,000 in 1988 to 141,000 in 1991.
These changes are greater than those in Scenario 2 for 1988-91,
but less than those in Scenario 1 for 1987-89 (Table B-15). The
pattern of individual crop acreage and price changes is very
similar to Scenario. 2 for 1988-91. However, acreage changes
tend to be greater for Scenario 3 than Scenario 2. Also, soybean
acreage increases in 1991 rather than decreases. Price changes
for Scenario 3 are also greater than Scenario 2, but soybean,
soybean meal, and cottonseed meal prices increase in 1990-91
rather than decrease (Table B-16). Some livestock prices do not
change during 1988-91, but increases of $0.003 per pound or less
occur for beef and pork.
For 1992 and later years when the larger corn yield losses occur,
total crop acreage is less for Scenario 3 than Scenario 2.
Total crop area decreases by 505,000 acres in 1992, decreases by
lesser amounts in 1993-94, and increases by less than 200,000
acres in 1995-96 (Table B-15). Scenario 3 shows the same pattern
of corn and soybean acreage changes as Scenario 2, but has greater
price changes. Corn price increases by $0.78 per bushel and
soybean price decreases by $0.26 per bushel in 1992, reflecting
greater corn acreage decreases and soybean increases for Scenario
3 (Table B-16). The higher corn prices encourage farmers to shift
acreage from other crops to corn causing the prices of the crops
to increase. As a result, the price of soybeans increases in
1993-96, with its greatest increase, $0.43, in 1994. : Barley
acreage increases from 1994-96, but did not in Scenario 2;
However, hay acreage does not begin to increase' until 1995, while
in Scenario 2 it began to increase in 1993. Higher feed prices
cause higher pork and chicken prices. Most livestock prices do
not change by more than $0.10 per pound, but pork price increases
by $0.13 per pound in 1994. Beef prices decrease $0.016 per
pound or less in 1992 and .1996. Veal prices decrease $0.067 per
pound or less through the entire time period.
Income. Total crop income (net of fixed, and variable, costs)
declines during 1988-1991, ranging from'$200 million in 1988 to
$303 million in 1991, approximately 2 percent of baseline crop
income (Table B-17). Income declines more than for Scenario 2 in
1988-90. In 1991, income decreases less for Scenario 3 than
Scenario 2 because of higher soybean prices in Scenario 3.
Income (net of variable costs only) decreases for all crops
except corn and sorghum in 1990-91 and barley and hay in 1991.
Beginning in 1992, crop income (net of fixed and variable costs)
increases because of price increases that outweigh yield and cost
changes. Crop income increases by $2.6 billion in 1992 (16
percent of baseline income), approaching an average of $7 per
crop acre, but increases are smaller in later years as cost and
yield changes decrease. These crop income increases are greater
than those in Scenario 2. After 1992, income (net of variable
costs only) increases for all crops in most years.
B-9
-------
From 1988-91, regional crop income (net of fixed and variable
costs) decreases for all regions except the Corn Belt in 1988-8-9,
when it benefits from higher soybean prices (Table B-18). From
1992-96, most regions gain, but some lose. The Delta States
lose in 1992 due to lower soybean prices. The Northeast and
Appalachian States lose in all those years, because they incur
relatively high corn yield losses. In most regions, corn replaces
soybeans as corn price rises. However, corn acreage in the
Northeast and Appalachian States is replaced by soybeans, resulting
in income declines. In Scenario 2, these two regions generally
did not lose although the Northeast lost in 1996.
Consumer Effects.; Consumers lose in all years due to higher
prices and lower production. Prior to 1992, the greatest consumer
loss is §280 million in. 1988. Consumer loss falls to $170 million
in 1989 but then rises to $206 million in 1991. After the
comparatively large price increases beginning in 1992/ consumer
loss peaks at $4.4 billion in 1992 declining to $2.2 billion in
1996. These consumer losses are larger than those in Scenario 2.
Livestock effects are identical for Scenarios 2 and 3 in 1988.
Livestock income falls more under Scenario 3 than Scenario 2 from
1989-96 due to. higher feed costs which outweigh livestock price
increases. Livestock income declines range from $3.5 million in
1989 to $122 million in 1992 (less than 0.2 percent of baseline
income for the 5 livestock products). After the corn yield
losses have their full effect on feed prices, 'livestock income
decreases by $2.5 billion in 1993 (about 3 percent of baseline
livestock income), ranging from about $1 billion to $2 billion in
later years. From 1989 to 1993, livestock consumers incur losses
of less than $86 million while gaining in 1992 when beef prices
fall slightly. After 1992, livestock consumers suffer greater
losses than under. Scenario 2, exceeding $3 billion in 1993-94.
B-10
-------
REFERENCES
Chavas/ Jean-Paul and Glenn S. Collins. "Welfare Measures from
Technological Distortions in General Equilibrium," Southern
Economic Journal, Vol. 47, January 1982, pp. 745-53.
Eales, James. "AGSIM: The Grain and Oilseed Demand Model,"
University of Illinois, Department of Agricultural Economics Staff
Paper No. 87 E-392, July 1987.
Frank, M.D. "AGSIM: The Livestock.Econometric Model," University
of Illinois, Department of Agricultural Economics Staff Paper No.
87 E-395, August 1987.
Just, Richard E. and Darrell L. Hueth. "Welfare Measures in a
Multimarket Framework," American Economic Review, Vol. 79, December
1979, pp. 947-54.
Kuchler, Fred and Michael Duffy. "Control of Exotic Pests—
Forecasting Economic Impacts." AER-518. U.S. Dept. Agriculture,
Economic Research Service, August 1984.
Osteen, Craig and Fred Kuchler. "Potential Bans of Corn and
Soybean Pesticides—Economic Implications for Farmers and Con-
sumers.!1 AER-546. U.S. Dept. Agriculture, Economic Research
Service, April 1986.
Taylor, C. Robert. "AGSIM: The Crop Supply Component," University
of Illinois, Department of Agricultural Economics Staff Report 87
E-386, July 1987a.
Taylor, C. Robert. "AGSIM User's Manual Version 87.3," University
of Illinois, Department of Agricultural Economics Staff Report 87
E-394, August 1987b.
Taylor, C. Robert. "AGSIM: Deterministic and Stochastic Simula-
tion Models and Benchmark Results for Version 87.3, " University
of Illinois, Department of Agricultural Economics Staff Report 87
E-396, September 1987c.
U.S. Department of Agriculture, Economic Research Service.
"Beltwide Boll Weevil/Cotton Insect Management Programs—Economic
Evaluation." U.S. Dept. Agriculture, Economic Research Service
Staff Report No. AGESS810518, 1981.
B-ll
-------
Table B-l. AGS1M baseline U.S. crop acreage.
OO
I
ro
Crop
Corn
Grain aorghun
Barley
Oats
Wheat
Soybeans
Cotton
All hay
Fallow
Diverted
Conservation reserve
U.S. total
19B7
74784.
12253.
11859.
12982.
76596.
61090.
11472.
57219.
27462.
11259.
16600.
373577.
1988
76765.
12292.
11395.
11921.
79614.
62067.
12143.
58369.
28091.
8893.
16600.
378150.
1989
76152.
12177.
10720.
12595.
79888.
64536.
12323.
57535.
27642.
9418.
16600.
379586.
Table B-2.
Commodity
Farm prices i
Corn (6/bu. >
Grain sorghum (6/bu.
Barley (O/bu.)
Oata (8/bu. )
Wheat
Cottonseed oil (8/T>
Soybean meal (e/T>
Soybean oil (9/T)
Retail prlceai
Beef (3/lb.)
Pork (8/lb.)
Chicken (8/lb.)
Milk (8/lb.)
Veal (S/lb.)
1987
1.578
) 1.738
1 .502
i.aia
2.495
5.117
.582
95.499
55.207
102.150
.262
110.021
.302
2.599
1 .665
.524
.163
4.157
1988
1.980
2. .001
1.579
2.921
2.737
5.856
.563
88.346
63.039
93.869
.252
105.160
.343
2.408
1.382
.515
. 168
4.330
1989
2.371
2.284
1 .757
•3 . 307
3.028
5.701
.572
92.260
63.075
94.771
.246
104.608
.341
2.657
1 . 577
.623
.170
4.239
1990
75558.
12175.
10512.
13203.
81123.
66871.
12465.
57891.
2820O.
• 10068.
1660O.
384666.
1991
Thousand
76028.
12169.
10184.
13419.
B117O.
67480.
12478.
5777B.
27955.
10364.
16600.
1992
acres
77036.
12414.
10140.
13519.
82045.
67452.
12391.
58369.
28343.
10464.
16600.
385625. 388774.
1993
77683.
12515.
10035.
13685.
82192.
67324.
12321.
58329.
28216.
10900.
16600.
389802.
1994
78241.
12608.
10199.
13821.
82766.
67340.
12310.
58657.
28495.
11171.
16600.
392208.
1995
78564.
12637.
10303.
13932.
82827.
67384 .
12291.
58640.
28443.
11495.
16600.
393115.
1996
78967.
12717.
1O482.
14008.
B32O1.
67437.
12262.
58895.
28655.
11592.
16600.
394816.
AOSIH baseline commodity prices
1990
2.669
2.555
1.931
3.124
3.212
5.206
.581
92.408
57.457
94.092
.232
101 .289
.304
2.831
1.BS4
.699-
.169 ~
4.J65
1991
2.801
2.742
2.163
2.966
3.324
4.963
.598
93.561
54.878
92.516
.227
97.728
.290
2.884
1 . 948
.723
— .170
4.426
1992
2.791
2.770
2.354
2.873
3.329
4.856
.618
89.914
53.695
89.341
.229
93.391
.293
.
2.782
1.613
. .699
.174
4.442
1993
2.786
2.765
2.553
2.763
3.325
4.841
.640
88.761
53.663
87.137
.235
9O.167
.306
2.671
1 .728
.684
.179
4.480 -
1994
2.799
2.771
2.636
2.674
3.310
4.867
.660
86. 480
53.673
86.062
.242
88. 195
.322
2.584
1 .757
.".683
.181
4.494
1995
2.846
2.810
2.683
2.606
3.325
4.933
.681
86.056
54.439
85.892
.250
86.978
.342
2.546
1 .832
.695
. 180
4.478
1996
2.882
2.845
2.667
2.562
3.337
5.003
.703
84.635
55. 191
85.447
.260
85.541
.365
2.523
1 .858
.703
. 1UO
4.423
-------
Table B-3. AQSIH baseline crop Income net of fixed and variable costs. I/
Region
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
Myj,Aon_
-------
Table'B-S. Change in U.S. crop acreages, scenario 1,
O>
Crop
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
Thousand acres
Corn
Grain sorghum
Barley
Oata
Wheat
Soybeans
Cotton
All hay
Fallow
Total
Commodity
Corn
Grain sorghum
Barley
Oats
Wheat
Soybeans
Cotton
All hay
Cottonseed
Cottonseed meal
Cottonseed oil
Soybean meal
Soybean oil
Beef
Pork
Chicken
Milk
Veal
21.92
2.54
.31
-.07
-.70
-26.83
-1.21
-.67
-.04
-4.82
1987
-.000
-.000
.OOO
.000
.000
.012
.000
.004
.225
.112
.000
.080
.001
.000
.000
.OOO
.000
.000
-5.86
.12
-10.04
-16.87
-3.97
-23.25
-2.71
9.52
-3.19
-59.42
Table B-6
1988
.000
.000
.002
.009
.OOO
.015
.000
-.049
.263
. 117
.000
.081
.002
.OOO
.000
.000
. 000
.000
-23.73
-2.28
-16.27
-17.16
-4.0V
-8.24
-5.35
8.48
-3.84
-76.28 -
Change
1989
.002
.000
.004
.015
.000
.009
.001
-.062
.209
. 104
.OOO
.033
.000
.000
.000
.000
.000
.000
-43.82
-3.48
-22.46
-16.28
-6.07
4.23
-9.20
-1.13
-6.29
110.79
-45.72
-2.75
-2S.41
-21.19
-7.49
3.87
-12. 13
-2.73
-6.38
-126.32
-45.12
-2.29
-29.02
-25.53
-11.35
-.11
-13.20
-4.48
-8.78
-148.68
-47.36
-2.54
-20.11
-28.51
-23.26
-.94
-12.22
-5.10
-9.09
-158.25
-51.27
-3.48
-12*. 85
-33.11
-33.87
-1 .54
-9.89
-7.19
-10.18
-173.58
-51.68
-3.70
-10.49
-36. 18
-39.75
-3.05
-7.42
-8. 10
-1 1 .08
-182.51
-42.83
-4.09
-6.89
-38.58
-47.00
-12.90
-5.61
-9.31
-11.99
-191.20
In commodity prices, scenario 1.
1990
.004
.002
.008
.013
.000
.004
.001
.004
.163
.098
.000
- .000
.000
.000
.000
.000
.000
.000
1991
.005
.003
.013
.014
.000
.003
.002
.043
.138
.087
.000
-.017
.OOO
.000
.000
.000
.000
.000
1992
.004
.003
.018
.016
.000
.003
.002
.068
.128
.080
.000
-.015
.000
-.000
.000
.OOO
.000
-.001
1993
.004
.003
.019
.017
.002
.003
.001
. OB4
.111
.070
.OOO
-.009
.000
-.001
.000
.OOO
.000*
-.002
1994
.005
.003
.016
.019
. OO3
.003
.001
. 103
.087
.060
.000
.OOO
.OOO
-.000
.000
.OOO
.000
-.003
1995
.005
.003
.013
.020
.003
.OO3
.000
. 1 18
.079
.056
.OOO
.008
.OOO
- .000
.001
.000
.000
- .003
1996
.004
.003
.010
.022
.004
.006
.OOO
. 13O
.098
.064
.000
.030
.OOO
.000
.001
.000
.000
-.002
-------
Table B-7. Change In crop income over variable coata. scenario 1. I/
OD
I
CJl
Crop
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
Hillion dollara
Crop
Grain sorghum
Barley
Data
Wheat
Soybeans
Cotton
All hay
Total, net of fixed
and variable costs
-3.14
-.12
-.07
.00
-.16
-3.81
-2.62
.38
-9.32
-76.39
-4.19
-7.48
-5.12
-26.17
-33.09
-10. OS
-5.01
-338.62
-53.10
-3.69
-5.86
-3.43
-25.75
-33.92
-7.81
-6.35
-309.65
-50.88
-2.84
-4.08
-4.64
-25.56
-36.36
-S.79
-.20
-297.93
-47.
-2.
-2.
-4.
-25.
-36.
-3.
3.
-291.
83
43
00
13
26
50
94
62
17-
-48.50
-2.54
.23
-3.42
-25.13
-33.43
-2.36
6.15
-271.49
-46.62
-2.62
.35
-3.17
-23.98
-30.34
-1.57
7.72
-260.55
-43.51
-2.52
-.99
-2.49
-22.99
-27.49
-1.02
9.62
-257.84
-40.83
-2.34
-2.18
-2.01
-21.81
-27.68
-1.03
11 .15
-244.76
-42.89
-2.40
-3.62
-1.57
-21.13
-24.02
-1.97
12.42
-243.35
I/ Excluding changes In commodity program payments.
Table B-8. Change in crop income by region, scenario 1. !_/
Region
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
Hillion dollara
Corn Belt
Lake States
Northern Plains
Southern Plains
Delta States
Mountain
Pacific States
Northeast
Appalachian
Southeast
1O
1
1
-19
-
-6
.63
.88
.96
.57
.09
.27
.88
.22
.07
. 17
-75.90
-43.70
-si .09
-21.73
-34.58
-36.08
-17.49
-18.22
-20.64
-19.18
-65.95
-39.97
-45.77
-21.04
-31.91
-34.90
-17.03
-17.90
-18.58
-16.61
-67.95
-39.56
-44.37.
-19.78
-28.41
-33.13
-15.55
-17.76
-17.95
-13.47
-69 .45.
-41.96
-43.12
-20.07
-25.58
-19.55
-16.69
-20.27
-19.66
-14.62 •
-64.84
-37.50
-38.61
-18.08
-2O.34
-30.50
-13.37
-17.85
-17.54
-12.86
-64.42
-36.68
-36.30
-17.92
-16.22
-30.42
-12. .82
-17.93
-17.24
-12.60
. -62.92
-39.27
'-36.54
-19.00
-13.17
-18.84
-15.18
-2O.34
-18.67
-13.90
-57.07
-34.98
-32.83
-17.34
-11 .90
-30.82
-12.96
-17.93
-16.48
-12.46
-55.38
-35.11
-32.17
-17.52
- 1 1 . 67
-31.38
-13.32
-18. 11
-16.14
-12.55
I./ Excluding changes in commodity program payments.
-------
Table B-9. Important welfare effects, scenario 1,
1987
198B
1989
1990
1991
1992
1993
1994
1995
1996
Crop consumer effect
Livestock income
change
Livestock consumer
effect
-22.72 -32.44 -36.65 -56.26
.00 -.45 1.21 -5.21
.00 .00 -24.01 -35.01
Million dollars
-69.11 -78.34 -82.43 -88.57
-25.90 -39.62 -44.45 '-41.91
-14.27 5.43 10.81 6.03
-91.85 -95.44
-37.18 -29.86'
-8.57 -25.99
Table B-10. Change in U.S. crop acreages, scenario 2.
Crop
1987
1988
1989 1990
1991
1992
1993
1994
1995
TO
I
1996
Thousand acres
Corn
Grain sorghum
Barley
Oats
Wheat
Soybeans
Cotton
All hay
Fallow
Total
1S7
4
-
-
-204
27
-3
-
-19
.22
.19
.08
.43
.47
.99
.85
.69
.02
.52
42.39
-3.96
-9.51
-15.66
-17.58
-98.40
41.07
23.78
-2.01
-41.86
-84. 3O
-12.05
-15.34
-11.46
-27.06
50.78
36.13
22.89
-3.16
-46.71
-120.60
-12.36
-20.57
-9.00
-17.00
90.41
16.22
7.02
-5.67
-77.19
-39.80
-2.82
-23. 7O
-IS. 48
-3.27
-7.75
.37
-2.01
-6.78
-108.02
-800.95
31.99
-25.74
-20.70
-11-. 04
607.04
-41.71
-25.83
-8.97
-304.88
1115.13
138.12
-18.38
-51.36
18.35
-999.43
-152.68
11.46
-7.47
46.27
1023.33
197.58
-14.20
-54.04
-17.31
-933.60
-217.99
58.31
-7.40
27.28
841.48
183.52
-7.93
-33.75
-22.34
-601 .20
-232.57
99.06
-.08
226.13
706.48
146.83
-3.57
-15.32
-23.69
-445.06
-215.87
80.95
-2.17
226.42
-------
Table B-ll. Change In commodity prices, scenario 2.
CO
I
Comnodity
Corn
Grain sorghum
Barley
Oata
Wheat
Soybeans
Cotton
All hay
Cottonseed
Cottonseed meal
Cottonseed oil
Soybean meal
Soybean oil
Beef
Pork
Chicken
Milk
Veal
1987
-.OO2
-.000
.OOO
.000
.000
.156
-.000
.022
1 . 4O3
.822
.004
1.441
.012
.000
.000
.000
.000
.000
1988
.000.
.000
.002
.009
.001
.180
-.002
-.124
1.727
.759
.005
1 .393
.017
.OOO
.000
.000
.000
-.001
1989
.008
.003
.004
.012
.003
.076
-.001
-.172
.827
.344
.002
.617
.008
.002
.002
.000
.OOO
.000
1990
.012
.007
.008
.009
.003
-.012
-.OOO
-.079
- .000
-.028
.000
-.101
.000
.002
.002
.000
.000
.002
1991
.007
.006
.012
.011
.002
-.012
.000
.010
-.039
-.048
.000
-.159
-.000
-.002
.002
.000
.000
.000
1992
. SOB
.149
.017
.013
.001
-.153
.003
.198
-1.198 ,
-.660
-.003
-1.330
-.012
-.005
.000
-.000
.000
-.002
1993
.339
.163
.017
.027
.OO5
.216
.ooa
.146
2.713
2.048
.OO7
2.007
.016
.045
.048
.027
.000
-.047
1994
.286
.137
.016
.030
' .010
.367
.011
-.086
4.756
3.101
.015
3.141
.036
.033
.088
.025
.000
. OO3
1995
.227
.106
.012
.020
.010
.322
.013
-.479
4.491
2.552
.017
2.339
- .037
.009
.056
.016
.000
.002
1996
.161
.073
.007
.008
.008
.243
.013
-.581
3.547
1.699
.016
1.339
.032
-.013
.013
.007
.000
-.007
Table B-12. Change in crop income over variable coats, scenario 2. !_/
Crop
1987
1988
1989
1990
.1991
1992
1993
1994
199S
1996
Million dollars
Corn
Grain sorghum
Barley
Oata
Wheat
Soybeans
Cotton
All hay
-21
-
-
189
-11
2
.51
.41
.00
.05
.01
.94
.57
.14
-83.82
-4.28
-7.52
-5.25
-25.96
209.05
-23.22
-12.21
-21.77
-2.34
-5.94
-4.69
-23.21
68.66
-20.37
-17.26
1 .28
-.15
-3.88
-6.27
-22.19
-58.29
-14.16
-8.74
-35.16
-.71
-1.98
-5.44
-24.24
-56.26
-9.23
- .00
2244.83
1O0.50
.03
-4.52
-25.75
-292.86
-17.35
17.96
1010.27
113. 4O
-.03
1 .00
-15.26
333.95 .
9.46
17.80
973.92
100.51
-1.01
1.69
-6.43
591 .33
30. SO
-4.46
747.64
79.15
-2.68
-2.45
-8.06
S14.60
42.01
-44.74
472.29
54.98
-4.27
-6.74
-15.95
380.52
44.28
-57.97
Total, net of fixed
and variable coats
1S9.46 -111.34 -186.60 -272.97 -314.49 1864.8O 129H.OS 1469.1O 11UO. 1)7
611.63
\_l Excluding changes in conocidity program, payments.
-------
Table B-13. • Change in crop income by region, scenario 2. I/
Region
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
Million dollars
Corn Belt
Lake States
Northern Plains
Southern Plains
Delta States
(fountain States
Pacific States
Northeast
Appalachian
Southeast
ISO
27
23
-
-4
-
-
3
2
-42
.16
.98
.03
.00
.15
.86
.83
.26
.87
.00
88.40
-12.54
-26.48
-24.37
-15.72
-24.33
-21.13
-14.69
-12.64
-47.83
21.17
-20.44
-30.59
-24.10
-28.91
-22.21
-20.56
-15.45
-14.89
-30.62
-57.75
-34.18
-36.32
-22.23'
-37.01
-19.13
-17.69
-16.81
-19.41
-12.46
-81.11
-44.42
-42.35
-21 .85
-31.56
-19.77
-18.21
-2O.57
-21.94
-12.71
832.22
486.61
246.97
84.36
-47.62
6O.30
21.85
54.20
88.96
36.95
549.58
288.87
130.53
86.67
52.42
32.44
17. 03
15.49
78.48
46.55
666.42
277 . 27
141.59
88.16
89.80
24.79
15.25
15.51
93.41
56.89
515.54
199.98
89.38
72.59
"74.91
11.42
1O.O5
7.71
72.91
45.58
306 . 47
107.03
28.80
48.74
51.85
-1.74
2.46
-4.O4
43.74
28.31
!_/ Excluding changes in commodity program payments.
03
I
»-»
00
Table B-14. Important welfare effects, scenario 2.
Crop
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
Crop consumer effect -235.82 -272.95 -158.41 -6O.75
Livestock income
change .00 -6.53 -.63 -29.99
Livestock con«un«i
effect .00 -38.95 -87.96 -81.74
Million dollars
-44.76 -2881.31 -26O7.44 -2564.88 -2062.60 -1477.10
-78.33 -99.18 -1601.30 -621.24 -1067.O3 -1421.79
11.31 109.22 -2251.68 -2626.22 -13S0.11 16.92
-------
Table-B-IS. Change in U.S. crop acreages, scenario 3.
Crop
1987
1988
1989
199'0
1991
1992
1993
1994
199S
1996
Thousand acrea
Corn
Grain sorghum
Barley
Data
Wheat
Soybeans
Cotton
All hay
Fallow
Total
157
4
-
-
-204
27
-3
-
-19
.22
.19
.08
.43
.47
.99
.as
.69
.02
.52
38.23
-4.11
- 9 . 98
-16.27
-19.73
-101.00
40.87
20.26
-2.63
-56.97
-90.93
- 12 . 30
-16.01
-12.91
-30.45
48.62
35.84
19.17'
-3.86
-66.72
-139.25
-11.93
-27.02
-11.15 '
-14.20
82.16
19.73
.62
-6.84
-114.72
-1O6..42
-3.26
-33.04
"-17.83
-7.50
4S.79
3.92
-7.26
-7.71
-140.99.
-1639.77
43.02
-30.21
-23.73
-26.08
1358.06
-91.45
-72.90
-10.78
-504.65
1098.16
212.86
-4.92
-82.43
-9.36
-867.56
-287.44
-61.61
-7.61
-17.52
. 1003.95
308.10
8.08
-104.37
-85.58
-776.59
-400.34
-21.95
-4.59
-77.89
797. 6O
285.68
16.47
-85.23
-112.37
-327.85
-424.51
' 31.78
8.96
199.51
664.47
222.23
' 16.97
r62.61
-127.91
-161.86
-385.31
8.83
8.41
191.65
CO
I
Table B-16. Change in commodity prices, scenario 3.
Commodity
Corn
Grain sorghum
Barley
Oats
Wheat
Soybeans
Cotton
All hay
Cottonseed
Cottonseed meal
Cottonseed oil
Soybean meal
Soybean oil
Beef
Pork
rhickan
Milk
Veal
1987
-.002
-.000.
.000
.000
.000
.156
- .OOO
.022
1 .403
.822
.004
1.441
.012
.000
.OOO
.OOO
.000
.000
1988
.000
.000
.002
.009
.001
.101
-.002
-.101
1 .739
.764
.005
1.399
.017
.000
.000
.OOO
.000
-.001
1989
.008
.004
.004
.013
.003
.077
-.001
-.137
.841
.349
.002
.623
.008
.002
.002
.OOO
.000
.OOO
1990
.018
.009
.018
.011
. OO7
.031
-.000
-.022
.410
.173
.001
.242
.004
.002
.002
.OOO
.000
.OO1
1991
.016
.010
.030
.013
.008
.025
.OOO
.074
.352
. 162
.001
'. 167
.003
- .000
.003
.OOO
• .000
- .000
1992
.777
.237
.038
.017
.008
-.257
.010
.547
-1.450
-.855
- .003
-2.283
- .020
-.003 .
.003 \
.000
.000
-.003
1993
.513
.254
.033
.043
.013
.222
.018
.755
3.6O6
2.873
.011
2.06B
.016
.067.
.072
.039
.000
-.067
1994
.430
. 2O8
.022
.058
.021
.425
.025
.632
6.376
4.353
.022
3.596
.041
.048
.130
.037
.000
-.000
1995
.339
.159
.010
.051
.021
.354
.029
.140
5.852
3.537
.024
2.461
.042
.013
.082
.024
.000
- . OO6
1996
.240
.110
.002
.038
.017
.243
.027
-.020
4.283
2.263
.021
1 .145
.033
-.016
..02O
.012
.000
-.019
-------
Table B-17. Change in crop income above variable coats, ecenario 3. !_/
Crop
Corn
Grain aorghum
Barley
Oats
Wheat
Soybeans
Cotton
All hay
Total, net of fixed
and variable costs
1987
-21.51
-.41
.00
.OS
-.01
189.94
- 1 1 . 57
2.14
159.46
1988
-82.72
-4.27
-7.48
-5.14
-25.83
210.46
-23.23
-10.00
-199.53
1989
-19.55
-2.29
-5.86
-4.26
-22.85
70.35
-20.39
-13.85
-271.62
1990
Million
16. 07
.52
-1.35
-5.63
-20.84
-2.88
-14.72
-3.15
-285.89
1991-
dollars
3.74
1.63
3.63
-4.73
-18.26
-9.25
-10.11
6.22
-302.70
1992
3136.56
149.84
7.40
-3.41
-17.43
-480. 3O
-27.97
50.56
2561 .71
1993
1308.52
169.17
4.99
7.35
-.58
330.17
23.03
77.93
1645.41
1994
1283.34
147.73
.37
12.12
13.81
672.07
62.50
66.93
1930.67
1995
982
115
-4
8
12
553
83
16
1421
.40
. 10
.35
.64
.65
.46
.96
.76
.98
1996
620.88
80.59
-7.60
3.75
2.23
366.99
85.64
-3.31
763.67
I/ Excluding changes in commodity program payments.
DO
ro
o
Table B-1B. Change in crop income by region, scenario 3. !_/
Region
1987
1988 1989
199O
1991
1992
1993
1994
1995
1996
Million -dollar*
Corn Belt
Lake States
Northern Plains
Southern Plains
Delta States
Mountain States
Pacific States
Northeant
Appalachian
.Soul hi*ant
ISO
27
23
-
-4
-
-
3
2
-A?
.16
.98
.03
.00
.15
.86
.83
.26
.87
.00
66.23
-26.57
-39.76
-29.10
-17.16
-29.80
-24.67
-22.95
-25.90
-49.85
. 11
-33.96
-43.14
-28.59
-30.27
-27.41
-23.80
-23.80
-28.07
-32. S9
-15.60
•43. 95
-41.44
-23.76
-27.80
-20.05
-10.36
-23.02
-59.75
-12.09
-33.74
-47.97
-40.95
-22.23
-24.70
-1.8.46
-17.31
-26.01
-SO.B3
-12.49
1391 .40
641.81
483.86
151.51
-93.58
112.36
46.50
-137.04
-48.75
13.55
876.73
354.27
281.46
159.98
43.84
73.83
45.93
- 164.05
-51.86
25.28
1009. 11
353.69
284.99
164.13
99.35
66.26
49.56
-133.20
-7.09
43.86
76O.24
255.06
193.00
139.59
80.77
45.70
42.41
-115.05
-11.48
.11 .74
438.96
134.95
91 .53
100.3.2
50.84
26.98
29.49
-99.70
-23.57
13.88
.!/ Excluding chan<|pB in commodity protjrau payments.
-------
00
fvi
Table B-19. Important welfare effects, scenario 3.
Crop 1987 1986 1989 199O 1991 1992 1993 1994 1995 1996
Crop consumer effect -235.82 -280.22 -169.7O -195.30 -2O5.B3 -44O3.68 -3U93.57 -3805.28 -30S5.04 -2188.32
Livestock Income
change .00 -6.53 -3.46 -33.35 - -97.74 -121.54 -2501.32 -995.59 -1597.77 -2053.40
Livestock consumer
effect .00 -38.95 -85.67 -79.43 -35.48 19.78 -3331.59 -3822.66 -1974.73 -60.06
-------
APPENDIX C
National Price-Quantity Model and Results
By
Craig Simons
and
Roger Lloyd I/
I/ OPRA Incorporated
-------
Appendix C
National Price-Quantity Model and Results
1.0 Model Description
The model used to estimate national commodity price-quantity
impacts closely follows the model developed by Lichtenberg et
al., I/ with some modifications required to overcome data defic-
iencies. With estimates of national impacts on production for
each commodity—through both increased costs and decreased yields-
-changes in marginal costs were estimated. The resulting changes
in commodity production and price at the national level were then
assessed with consideration of supply and demand elasticities.
Specific algebraic equations used to define the model are as
follows:
(1) PQ = MCQ
P (dY/Y.J + (dC/Y,J
(2) dMC = -=-
1 - (dY/YQ)
(3) dP/PQ = [es/(egeD)](dMC/MC)
(4) dQ/Q = [eDes/(es - eD)](dMC/MC)
where:
P = commodity baseline price, farm level
MC = baseline commodity marginal cost of production
dY = change in yield per acre of crop production from
the regulatory scenario
dC = change in variable cost per acre from the regula-
tory scenario
ec = elasticity of supply
3
e_ = elasticity of demand
Q = total baseline quantity of commodity production
Changes in producer and consumer surplus were then approximated.
To estimate changes in producer surplus, it was assumed that all
planned reductions in output would be achieved by shifts in
marginal production inputs (where zero economic profits were
I/ Lichtenberg, Erik; Douglas Parker and David Zilberman.
Economic Impacts of Cancelling Parathion Registration -for
Almonds, Western Consortium for the Health Professions,
Inc., January 1987.
C-l
-------
being earned in the baseline) to an alternative equally profitable
crop. Economic profits on this marginal production would be the
same before and after the regulatory scenarios. The change from
the baseline in total revenue earned by producers would be:
(5) dR = P1Ql - PQQo
and since price equals marginal cost, the cost savings would be:
(6) CTS = PQ(Q0 - C^).
The change in costs for the acreage remaining in production is
(7) dTC = AidC.
Accordingly, the change in producer surplus from the baseline is
defined as
(8) dPS = dR + CTS - dTC.
The change in consumer surplus from the baseline was approximated
using the following relationship:
(9) dCS = -(P. - P0)(Qi * Q0)/2
where:
dR = change in total revenue
CTS = cost savings
Q. = production in year i
dTC = change in total production cost
A. = commodity acreage in year i
dC = change in cost per acre from the regulatory
scenario
dPS = change in producer surplus
dCS = change in consumer surplus.
This model presumes that all other variables not considered will
remain constant and thus have no affect on the model results.
2.0 Data Inputs
National information was compiled on baseline price, harvested
acreage, production, farm size, and yield for each of the six
specialty crops. The baseline commodity prices, harvested acre-
ages, and production quantities used in this study are an average
from 1981-1985 as obtained from various issues of Agricultural
Statistics (Table C-l). Commodity prices were adjusted by the
GNP Implicit Price Deflator to reflect constant 1986 dollars.
C-2
-------
Table C-l. Average prices, production and acreages
Irish Potatoes
U.S.
ID-UA
ND-MN
HE
Green Peas
U.S.
Ul
UA
Apples
WA '
NY
Ml
Peanuts
Additional!
U.S.
GA-AL
NC-VA
TX-OK
Quotas
U.S.
GA-AL
NC-VA
TX-OK
Caneberries
(Raspberries)
U.S.
WA
OR
Tomatoes
Processing
CA
U.S.
Fresh
FL
U.S.
Average
price
1981-1985
(1986 dollars)
5.02/cwt.
4.71/cwt.
4.77/cwt.
4.12 cwt.
253. OO/ ton
239.64/ton
250. OO/ ton
264. OO/ ton
287. CO/ ton
246. OO/ ton
193. OO/ ton
599.14/ton
549.S6/ ton
579.07/ton
56S.47/ ton
587. BO/ ton
587. 80/ ton
587. 80/ ton
587. BO/ ton
.641/lb.
.643/1b.
.638/lb.
72.40/ton
75.68/ton
559. 70/ ton
522. OO/ ton
Average
acreage
harvested
1981-1985
(1000)
1.280
437
194
98
3,180
857
638
N.A.
100.8
64.1
46.6
1.112.6
603.3
199.2
250.4
314.9
162.7
53.7
67.5
107.5*
29.0
25.0
225.3
280.4
45.4
123.7
Average
production
1981-1985
359.282.000 cut.
146,083.000 cwt.
33,031.000 cwt.
24.926.000 cwt.
490,040 tons
134.400 tons
100.430 tons
4.064.500 tons
1.343,000 tons
517,000 tons
426,000 tons
1,500.053 tons
916,799 tons
179,649 tons
218,555 tons
424.564 tons
247,237 tons
75.416 tons
58,940 tons
38.979.000 Ibs.
15.934,000 Ibs.
13,360.000 Ibs.
5.944.000 tons
6.981.000 tons
660,000 tons
1.385.000 tons
Typical
farm
size
(acres)
—
725
1.000
600
—
540
1.500
200
150
200
...
500
400
1.100
—
500
400
1.100
—
30
11
1.200
—
500
—
Average
yield/acre
1981-1985
280.7 cwt.
386.0 cwt.
170.3 cwt.
254.4 cwt.
1.54 tons
1.57 tons
1.57 tons
8.70 tons
13.32 tons
8.06 tons
9.14 tons
1.35 tons
1.52 tons
1.40 tons
.88 tons
1.35 tons
1.52 tons
1.40 tons
.88 tons
3.625 Ibs.*
5.494 Ibs.
5.344 Ibs.
26.6 tons
24.9 tons
14.6 tons
11.2 tons
1982.
-------
In order to assess -the impacts of regulatory costs on per acre
net returns, a definition of a typical commercial farm, in terms
of acreage, was necessary. Such estimates were obtained from a
poll of extension crop production specialists (a DELPHI approach)
and from estimates obtained in crop enterprise production budgets.
Because farm size is highly variable within each region, the
estimates presented in Table C-l and used in the impact analysis
must be interpreted with caution.
Estimates of supply and demand elasticities were obtained from
several sources, both published and unpublished. Elasticity
estimates are presented in Table C-2.
National estimates of variable cost and yield changes associated
with environmental regulations for each specialty crop under
three scenarios were provided by EPA. The yearly estimates are
provided as the change from a base year prior to the initiation
of regulatory impacts (Table C-3).
3.0 Model Results
Results of the National Price-Quantity Model are presented in
Tables C-4 through C-18 as the percent change in production,
price, consumer surplus and producer surplus from a base year of
no regulatory impacts. Effects of each policy scenario are
examined under each of the four specialty crops. Data limitations
prevented analyses of peanuts and caneberries.
C-4
-------
Table C-2. Supply and demand elasticities
Demand Elasticities I/
Potatoes -.3688
Apples -.2015
Tomatoes (fresh) -.5584
Tomatoes (processing) -.3811
Other fresh vegetables (peas) -.2102
Supply Elasticities
Short-run
Peas .31 2/
Tomatoes 1.35 3_/
Potatoes .87 4/
Apples .11 4/
Sources: !_/ USDA, ERS, By Kuo S. Huang, U.S. Demand for Food;
A Complete System of Price and Income Effects.
Technical Bulletin Number 1714, December 1985.
2/ Askari, Hedsein, and Jonn T. Cummings, Estimating
Agricultural Supply Response with the Nerlove
Model; A Survey, International Economic Review,
Vol. 18, No. 2, June 1977.
3_/ Chern, W.S. "Acreage Response and Demand for
Processing Tomatoes in California". American
Journal of Agricultural Economics. May 1976.
4_/ Unpublished estimates provided by USDA.
C-5
-------
Table C-3. Regulatory cost and yield impact estimates for specialty crops
Change in variable cost from base year ())
Vear
Scenario 1
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
Scenario 2
1983
1984
<"> 1985
en 1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
Scenario 3
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
Apples
0
0
0
0
0
4.86
4.86
4.86
4.86
6.78
6.51
6.23
5.96
5.68
0
0
0
0
0
4.86
4.86
-1.71
.12
13.30
12.34
11.39
10.43
9.48
0
0
0
0
0
4.86
4.86
-1.71
3.45
18.32
16.58
14.84
13.10
11.36
Potatoes
0
.26
.23
.19
4.41
5.05
.40
.88
.08
.05
.17
2.30
2.03
1.76
0
.69
.66
.62
4.84
5.48
4.83
13.77
11.75
13.78
11.23
8.68
6.74
4.80
0
.86
.83
.79
5.01
5.65
5.00
10.17
8.70
12.44
10.26
8.08
6.51
4.94
Tomatoes
Fresh
0
.66
.56
.47
.38
7.03
6.94
6.84
6.75
6.75
6.75
6.75
6.75
6.75
0
.66
.56
.47
.38
7.03
6.94
22.95
20.56
18.26
15.95
13.65
11.35
9.05
0
.66
.56
.47
.38
7.03
6.94
-13.08
-10.33
-7.48
-4.64
-1.79
1.05
3.90
Proc.
0
.66
.56
.47
.38
7.03
6.94
6.84
6.75
6.75
6.75
6.75
6.75
6.75
0
.66
.56
.47
.38
7.03
6.94
6.97
6.86
6.84
6.82
6.80
6.78
6.76
0
.66
.56
.47
.38
7.03
6.94
5.99
6.02
6.14
6.26
6.38
6.50
6.62
Peas
0
0
0
0
3.46
3.74
3.25
2.75
2.26
2.06
1.53
.99
.95
.90
0
0
0
0
3.46
3.74
3.25
2.75
2.26
2.64
2.02
1.40
1.27
1.15
0
0
0
0
3.46
3.74
3.25
2.75
2.26
2.06
1.53
.99
.95
.90
Peanuts
.01
1.66
1.42
1.19
1.60
2.21
1.58
5.13
4.25
3.60
2.95
2.30
1.75
1.19
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
.01
1.66
1.42
1.19
1.60
2.21
1.58
26.09
22.21
22.12
17.97
13.82
9.77
5.71
Apples
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
.050
.043
.036
.028
.021
.014
.007
0
0
0
0
0
0
0
.050
.043
.036
.028
.021
.014
.01
Change in yield from base year
Potatoes
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
.048
.041
.034
.027
.021
.014
.007
Tomatoes
Fresh
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
.196
.168
.140
.112
.084
.056
.028
Proc.
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
.050
.043
.036
.029
.021
.014
.007
Peas
0
0
0>
o|
o1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Peanuts
0
.013
.011
.010
.080
.067
.055
.045
.033
.022
.011
.001
.001
.001
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
0
.013
.011
.010
.080
.067
.055
.253
.211
.279
.223
.168
.122
.077
-------
Table C-4. Production and welfare impacts from Scenario I
environmental regulations affecting apples
Percent change
from Base Year 1987
Change in welfare
from Base Year 1987
Year Production
Price
Consumer
Surplus
Producer
Surplus
Net
1988
1989
1990
1991
1992
1993
1994
1995
1996
-0.015
-0.015
-0.015
-0.015
-0.021
-0.020
-0.019
-0.018
-0.018
0.0747
0.0747
0.0747
0.0747
0.1042
0.1000
0.0958
0.0916
0.0874
-799,261
-799,261
-799,261
-799,261
-1,114,985
•1,069,880
•1,024,780
-979,676
-934,574
•1,463,990
•1,463,990
•1,463,990
•1,463,990
•2,042,235
•1,959,628
•1,877,029
•1,794,423
•1,711,818
-2,263,251
-2,263,251
-2,263,251
-2,263,251
-3,157,220
-3,029,508
-2,901,809
-2,774,099
-2,646,392
C-7
-------
Table C-5. Production and welfare impacts from Scenario II
environmental regulations affecting apples
Percent change
from Base Year 1987
Change in welfare
from Base Year 1987
Year Production
Price
Consumer
Surplus
Producer
Surplus
Met
1988
1989
1990
1991
1992
1993
1994
1995
1996
-0.015
-0.015
-0.367
-0.318
-0.305
-0.248
•0.191
-0.135
-0.081
0.0747
0.0747
1.8230
1.5764
1.5144
1.2296
0.9489
0.6724
0.3997
-799,261
-799,261
•19,465,134
-16,836,028
•16,174,993
•13,136,997
•10,141,456
-7,187,490
-4,274,242
-1,463,990
-1,463,990
•35,590,977
•30,791,483
•29,584,370
•20,034,741
•18,559,526
•13,157,250
-7,826,479
-2,263,251
-2,263,251
-55,056,111
-47,627,511
-45,759,363
-37,171,738
-28,700,982
-20,344,740
-12,100,721
C-8
-------
Table C-6. Production and welfare impacts from Scenario III
environmental regulations affecting apples
Percent change Change in welfare
from Base Year 1987 from Base Year 1987
Consumer Producer
Year Production Price Surplus Surplus Net
1988 -0.015 0.0747 -799,261 -1,463,990 -2,263,251
1989 -0.015 0.0747 -799,261 -1,463,990 -2,263,251
1990 -0.367 1.8230 -19,465,134 -35,590,977 -55,056,111
1991 -0.328 1.6299 -17,406,795 -31,833,637 -49,240,432
1992 -0.321 1.5944 -17,028,442 -31,142,821 -48,171,263
1993 -0.261 1.2966 -13,852,108 -25,341,353 -39,193,461
1994 -0.202 1.0031 -10,720,085 -19,617,380 -30,337,465
1995 -0.144 0.7139 -7,631,449 -13,969,365 -21,600,814
1996 -0.086 0.4288 -4,585,308 -8,395,821 -12,981,129
C-9
-------
Table C-7. Production and welfare impacts from Scenario I
environmental regulations affecting potatoes
Percent change
from Base Year 1983
Year
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
Production
-0.005
-0.004
-0.004
-0.088
-0.100
-0.088
-0.097
-0.081
-0.080
-0.063
-0.046
-0.040
-0.035
Price
0.0142
0.0122
0.0101
0.2375
0.2721
0.2373
0.2632
0.2198
0.2182
0.1711
0.1239
0.1095
0.0952
Change-in welfare
from Base Year 1983
Consumer
Surplus
-236,332
-202,571
-168,810
-3,958,133
-4,534,309
-3,954,689
-4,386,196
-3,662,095
-3,636,317
-2.850.87L
-2,065,289
-1,825,783
-1,586,266
Producer
Surplus
-100,181
-85,870
-71,559
-1,677,149
-1,921,165
-1,675,690
-1,858,441
-1,551,762
-1,540,844
-1,208,126
-875,292
-773,808
-672,31-3
Net
-336,513
-288,441
-240,369
-5,635,282
-6,455,474
-5,630,379
-6,244,637
-5,213,857
-5,177,161
-4,058,997
-2,940,581
-2,599,591
-2,258,579
c-io
-------
Table C-8. Production and welfare impacts from Scenario II
environmental regulations affecting potatoes
Percent change
from Base Year 1983
Year
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
Production
-0.014
-0.013
-0.012
-0.096
-0.109
-0.096
-0.274
-0.234
-0.274
-0.223
-0.173
-0.134
-0.095
Pri ce
0.0374
0.0353
0.0333
0.2607
0.2953
0.2605
0.7424
0.6338
0.7428
0.6054
0.4680
0.3634
0.2588
Change in welfare
from Base Year 1983
Consumer
Surplus
-622,822
-589,064
-555,306
-4,344,305
-4,920,431
-4,340,861
-12,359,750
-10,553,621
-12,366,205
-10,081,485
-7,795,613
-6,054,275
-4,312,263 .
Producer
Surplus
-264,001
-249,693
-235,384
-1,840,700
-2,084,674
-1,839,241
-5,232,215
-4,468,530
-5,234,943
-4,268,846
-3,301,768
-2,564,735
-1,827,130
Net
-886 ,823
-838,757
-790,690
-6,185,005
-7,005,105
-6,180,102
-17,591,965
-15,022,151
-17,601,148
-14,350,331
-11,097,381
-8,619,010
-6,139,393
C-ll
-------
Table C-9. Production and welfare impacts from Scenario III
environmental regulations affecting potatoes
Rercent change
from Base Year 1983
Year
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
Production
-0.017
-0.016
-0.016
-0.100
-0.112
-0.099
-1.518
-1.292
-1.176
-0.940
-0.708
-0.491
-0.278
Price
0.0465
0.0445
0.0425
0.2699
0.3045
0.2697
4.1172
3.5024
3.1878
2.5494
1.9199
1.3324
0.7530
Change in welfare
from Base Year 1983
Consumer
Surplus
-775,612
-741,855
-708,097
-4,496,968
-5,073,075
-4,493,524
-68,115,431
-58,010,709
-52,830,082
-42,299,748
-31,892,323
-22,157,065
-12,535,455
Producer
Surplus
-328,760
-314,452
-300,145
-1,905,352
-2,149,309
-1,903,893
-28,653,782
-24,431,352
-22,262,674
-17,846,511
-13,471,380
-9,369,425
-5,306,492
Net
-1,104,372
-1,056,307
-1,008,243
-6,402,320
-7,222,384
-6,397,417
-96,769,213
-82,442,061
-75,092,756
-60,146,259
-45,363,703
-31,526,490
-17,841,947
C-12
-------
Table C-10. Production and welfare impacts from Scenario I
environmental regulations affecting fresh tomatoes
Percent change
from Base Year 1983
Year
1984
1985
1986
1987
.1988
1989
1990
1991
1992
1993
1994
1995
1996
Production
-0.004
-0.004
-0.003
-0.003
-0.048
-0.047
-0.046
-0.046
-0.046
-0.046
-0.046
-0.046
-0.046
Price
0.0079
0.0068
0.0057
0.0045
0.0851
0.0839
0.0828
0.0817
0.0817
0.0817
0.0817
0.0817
0.0817
Change in welfare
from Base Year 1983
Consumer
Surplus
-57,473
-49,263
-41,052
-32,842
-615,108
-606,901
-598,694
-590,487
-590,487
-590,487
-590,487
-590,487
-590,487
Producer
Surplus
-23,772
-20,376
-16,980
-13,584
-254,366
-250,973
-247,580
-244,187
-244,187
-244,187
-244,187
-244,187
-244,187
Net
-81,245
-69,639
-58,032
-46,426
-869,474
-857,874
-846,274
-834,674
-834,674
-834,674
-834,674
-834,674
-834,674
U'S-EPi ^quarters Library
no a"COde3201
° Pennsylvania Avenue NW
Washington DC 20460
C-13
-------
Table C-101. Production and welfare impacts from Scenario I
environmental regulations affecting processing tomatoes
Percent change
from Base Year 1983
Year
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
Production
-0.010
-0.009
-0.007
-0.006
-0.111
-0.109
-0.108
-0.106
-0.106
-0.106
-0.106
-0.106
-0.106
Price
0.0272
0.0233
0.0194
0.0155
0.2910
0.2871
0.2832
0.2793
0.2793
0.2793
0.2793
0.2793
0.2793
Change in welfare
from Base Year 1983
Consumer
Surplus
-143,609
-123,094
-102,579
-82,064
-1,536,544
-1,516,050
-1,495,556
-1,475,061
-1,475,061
-1,475,061
-1,475,061
-1,475,061
-1,475,061
Producer
Surplus
-150,002
-128,574
-107,145
-85,717
-1,605,324
-1,583,907
-1,562,490
-1,541,073
-1,541,073
-1,541,073
-1,541,073
-1,541,073
•-1,541,073
Net
-293,611
-251,668
-209,724
-167,781
-3,141,868
-3,099,957
-3,058,046
-3,016,134
-3,016,134
-3,016,134
-3,016,134
-3,016,134
-3,016,134
C-14
-------
Table C-12. Production and welfare impacts from Scenario II
environmental regulations affecting fresh tomatoes
Percent change
from Base Year 1983
Year
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
Production
-0.004
-0.004
-0.003
-0.003
-0.048
-0.047
-0.155
-0.139
-0.123
-0.108
-0.092
-0.077
-0.061
Price
0.0079
0.0068
0.0057
0.0045
0.0851
0.0839
0.2777
0.2488
0.2209
0.1930
0.1652
0.1373
0.1095
Change in welfare
from Base Year 1983
Consumer
Surplus
-57,473
-49,263
-41,052
-32,842
-615,108
-606,901
-2,006,558
-1,797,741
-1,596,756
-1,394,866
-1,193,818
-992,739
-791,629
Producer
Surplus
-23,772
-20,376
-16,980
-13,584
-254,366
-250,973
-829,328
-743,082
-660,058
-576,647
-493,571
-410,469
-327,341
Net
-81,245
-69,639
-58,032
-46,426
-869,474
-857,874
-2,835,886
-540,823
-256,814
-1,971,513
-1,687,389
-1,403,208
-1,118,970
C-15
-------
Table C-13. Production and welfare impacts from Scenario II
environmental regulations affecting processing tomatoes
Percent change
from Base Year 1983
Year
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
Production
-0.010
-0.009
-0.007
-0.006
-0.111
-0.109
-0.110
-0.108
-0.108
-0.108
-0.107
-0.107
-0.107
Price
0.0039
0.0034
0.0028
0.0023
0.0422
0.0416
0.0418
0.0412
0.0410
0.0409
0.0408
0.0407
0.0406
- Change- in welfare
from Base Year 1983
Consumer
Surplus
-143,609
-123,094
-102,579
-82,064
-1,536,544
-1,516,050
-1,523,111
-1,499,086
-1,494,718
-1,490,350
-1,485,982
-1,481,614
-1,477,245
Producer
Surplus
-150,002
-128,574
-107,145
-85,717
-1,605,324
-1,583,907
-1,591,286
-1,566,179
-1,561,615
-1,557,050
-1,552,485
-1,547,920
-1,543,355
Net
-293,611
-251,668
-209,724
-167,781
-3,141,868
-3,099,957
-3,114,397
-3,065,265
-3,056,333
-3,047,400
-3,038,467
-3,029,534
-3,020,600
C-16
-------
Table C-14. Production and welfare impacts from Scenario III
environmental regulations affecting fresh tomatoes
Percent change
from Base Year 1983
Year
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
Production
-0.004
-0.004
-0.003
-0.003
-0.048
-0.047
-9.520
-7.892
-6.372
-4.947
-3.609
-2.351
-1.165
Price
0.0079
0.0068
0.0057
0.0045
0.0851
0.0839
17.0482
14.1338
11.4106
8.8589
6.4634
4.2099
2.0863
Change in welfare
from Base Year 1983
Consumer
Surplus
-57,473
-49,263
-41,052
-32,842
-615,108
-606,901
-120,000,000
-98,000,000
-80,000,000
-62,000,000
-46,000,000
-30,000,000
-26,000,000
Producer
Surplus
-23,772
-20,376
-16,980
-13,584
-254,366
-250,973
-46,000,000
-39,000,000
-32,000,000
-25,000,000
-19,000,000
-12,000,000
-6,167,852
Net
-81,245
-69,639.
-58,032
-46,426
-869,474
-857,874
-166,000,000
-137,000,000
-112,000,000
-87,000,000
-65,000,000
-42,000,000
-32,167,852
C-17
-------
Table C-15. Production and welfare impacts from Scenario III
environmental regulations affecting processing tomatoes
Percent change
from Base Year 1983
Year
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
Production
-0.010
-0.009
-0.007
-0.006
-0.111
-0.109
-1.664
-1.430
-1.201
-0.976
-0.753
-0.535
-0.319
Price
0.0272
0.0232
0.0194
0.0155
0.2910
0.2871
4.3654
3.7515
3.1515
2.5602
1.9768
1.4027
0.8367
Change in welfare
from Base Year 1983
Consumer
Surplus
-143,609
-123,094
-102,579
-.82,064
-1,536,544
-1,516,050
-23,000,000
-20,000,000
-17,000,000
-13,000,000
-10,000,000
-7 ,400 ,000
-4,413,759
Producer
Surplus
-150,002
-128,574
-107,145
-85,717
-1,605,324
-1,583,907
-24,000,000
-21,000,000
-17,000,000
-14,000,000
-11,000,000
-7,700,000
-4,613,538
Net
-293,611
-251,668
-209,724
-167,781
-3,141,868
-3,099,957
-47,000,000
-41,000,000
-34,000,000
-27,000,000
-21,000,000
-15,100,000
-9,027,297
C-18
-------
Table C-16. Production and welfare impacts from Scenario I
environmental regulations affecting peas
Percent change
from Base Year 1986
Year
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
Production
-0.111
-0.120
-0.104
-0.089
-0.073
-0.066
-0.049
-0.032
-0.030
-0.029
Price
0.5297
0.5724
0.4967
0.4211
0.3454
0.3156
0.2334
0.1512
0.1446
0.1380
Change in welfare
from Base Year 1986
Consumer
Surplus
-656,421
-709,320
-615,598
-521,862
-428,111
-391,204
-289,309
-187,396
-179,271
-171,145
Producer
Surplus
-444,848
-480,675
-417,197
-353,700
-290,182
-265,174
-196,122
-127,046
-121,539
-116,031
Net
-1,101,269
-1,189,995
-1,032,795
-875,562
-718,293
-656,378
-485,431
-314,442
-300,810
-287,176
C-19
-------
Table C-17. Production and welfare impacts from Scenario II
environmental regulations affecting peas
Percent change
from Base Year 1986
Year
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
Production
-0.111
-0.120
-0.104
-0.089
-0.073
-0.085
-0.065
-0.045
-0.041
-0.037
Price
0.5297
0.5724
0.4967
0.4211
0.3454
0.4037
0.3089
0.2141
0.1949
0.1758
Change in welfare
from Base Year 1986
Consumer
Surplus
-656,421
-709,320
-615,598
-521,862
-428,111
-500,359
-382,887
-265,392
-241,669
-217,946
Producer
Surplus
-444,848
-480,675
-417,197
-353,700
-290,182
-339,131
-259,538
-179,912
-163,834
-147,754
Net
-1,101,269
-1,189,995
-1,032,795
-875,562
-718,293
-839,490
-642,425
-445,304
-405,503
-365,700
C-20
-------
Table C-18. Production and welfare impacts from Scenario III
environmental regulations affecting peas
Percent change
from Base Year 1986
Year
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
Production
-0.111
-0.120
-0.104
-0.089
-0.073
-0.066
-0.049
-0.032
-0.030
-0.029
Price
0.5297
. 0.5724
0.4967
0.4211
0.3454
0.3156
0.2334
0.1512
0.1446
0.1380
Change in welfare
from Base Year 1986
Consumer
Surplus
-656,481
-709,320
-615,598
-521,862
-428,111
-391,204
-289,309
-187,396
-179,271
-171,145
Producer
Surplus
-444,848
-480,675
-417,197
-353,700
-290,182
-265,174
-196,122
-127,046
-121,539
-116,031
Net
-1,101,269
-1,189,995
-1,032,795
-875,562
-718,293
-656,378
-485,431
-314,442
-300,810
-287,176
C-21
-------
APPENDIX D
REPFARM Model and Results
By
Mike Salassi !_/
Terry Dinan 2/
I/ Agriculture and Rural Economics Division, Economic Research
Service, U.S. Department of Agriculture.
2/ Office of Policy Analysis, U.S. Environmental Protection
Agency.
-------
Appendix D
REPFARM Model and Results
1.0 Description of REPFARM Model
REPFARM is a whole-farm, recursive programming-simulation model
which is capable of using a wide variety of farm policy, produc-
tion, and market environments in order to provide financial
impact information for a variety of representative farms across
the United States. REPFARM essentially links a set of accounting
decision subroutines with a set of optimizing subroutines. The
optimizing subroutines annually adjust the mix of crop enterprises
produced on the farm based upon estimated returns for each
enterprise. The accounting subroutines calculate farm income and
expenses, value of assets and liabilities, as well as other
financial information associated with the production decisions
made each year.
REPFARM is capable of simulating the annual production and
financial operations of a representative farm for a period of
1-10 years. The model utilizes user-specified data sets which
contain information relative to the partiqular representative
farm being simulated. Information about a particular farm
contained in a data set includes farm size, acres owned and
leased, initial values of farm assets and liabilities, off-farm
income, family living expenses, itemized expenses for the farm
such as taxes and insurance, as well as acreages, yields, produc-
tion costs, and labor requirements of each crop enterprise produced
on the farm and herd size, input costs, and labor requirements of
each livestock enterprise produced on the farm. Additional
information which must also be supplied by the user on an annual
basis includes itemized inflation indexes for various production
expense items, interest rates for short-term, intermediate-term,
and long-term loans, machinery depreciation rates, income tax
rates, market prices for all crop and livestock enterprises
included on the farm as well as farm policy -data such as loan
rates, target prices, crop set-asides, diversion payment rates,
and payment limitations.
REPFARM can simulate a representative farm in a deterministic or
stochastic mode. In the deterministic mode, the farm is simulated
with specified crop and livestock market prices and crop yields
for each year of simulation. Model output consists of annual
financial statements for the farm. These financial statements
include itemized income statements, cashflow statements, and
balance sheets. Additional production information is also provided
relating to the acreage and production of each crop enterprise.
In the stochastic mode, several iterations are performed for each
year of simulation using variable crop yields and crop and
livestock market prices. Model output in this mode consists
D-l
-------
primarily of annual mean and variance estimates of selected
financial measures and production items. REPFARM was simulated
in the deterministic mode in this study.
Three key assumptions that were made in the baseline projections
of each of the REPFARM models are:
1) production costs were assumed to increase at two percent
per year,
2) crop yield.was assumed to increase at two percent per
year, and
3) the current farm bill was assumed to be in effect
through 1990 and policy variables were held constant at
the 1990 level for the remaining forecast period.
If these assumptions overestimate the financial well-being of the
representative producers in the baseline, then the ability of the
producers to bear the costs of environmental regulations will be
overestimated. Likewise, if these assumptions result in an
underestimation of producers well-being, then the ability of
producers to bear the costs of environmental regulations will be
underestimated.
2.0 Description of Representative Farms
Representative farms evaluated in this study were developed from
data obtained from the USDA's 1986 Farm Costs and Returns Survey.
Three general types of farms considered included a Mississippi
cotton soybean farm, and Illinois corn soybean farm, and a Kansas
wheat cattle farm. For each one of these general farm types, two
representative farm data sets were constructed: one representing
a farm in an average financial position and another representing
a farm in a vulnerable financial position. Representative farm
data sets for farms in an average financial position were developed
from data on all farms meeting the specified state/enterprise
definition. Representative farm data sets for farms in a vul-
nerable financial position were developed from data on all farms
meeting the state/enterprise definition plus the additional
requirements of a negative net cash income and a debt to asset
ratio greater than 0.40.
2.1 Illinois Corn Soybean Farms
The two representative Illinois corn soybean farms were developed
from survey information on farms in Illinois which were classified
as cash grain farms (cash grain sales represented the largest
portion of gross income for the farm) and produced corn and
soybeans. Survey observations fitting this description represent
D-2
-------
an expanded number of 30,837 farms in Illinois (Table D-l) and
were used to estimate the characteristics of the corn soybean
farm in an average financial position (Table D-2). Of these
30,837 farms, approximately 9.9% were determined to be in a
vulnerable position (as defined above) and survey observations
relating to this group of farms were used to develop the charac-
teristics of the corn soybean farm in a vulnerable financial
position (Table 0-2).
2.2 Mississippi Cotton Soybean Farms
The two representative Mississippi cotton soybean farms were
developed from survey information on farms in Mississippi which
were classified as field crop farms (field crop sales represented
the largest portion of gross income for the farm) and produced
cotton and soybeans. Survey observations fitting this description
represent an expanded number of 1,798 farms in Mississippi (Table
0-1) and were to estimate the characteristics of the cotton
soybean farm in an average financial position (Table 0-3). Of
these 1,798 farms, approximately 14.2% were determined to be in a
vulnerable financial position (as defined above) and survey
observations relating to this group of farms were used to develop
the characteristics of the cotton soybean farm in a vulnerable
financial position (Table 0-3).
2.3 Kansas Wheat Cattle Farms
The two representative Kansas wheat cattle farms were developed
from survey information on farms in Kansas which produced wheat
and had sales of cattle. Survey observations fitting this
description represent an expanded number of 19,966 farms in
Kansas (Table D-l) and were used to estimate the characteristics
of the wheat cattle farms in an average financial position (Table
D-4). Of these 19,966 farms, approximately 7.1% were determined
to be in a vulnerable financial position (as defined above) and
survey observations relating to this group of farms were used to
develop the characteristics of the wheat cattle farm in a vulner-
able position (Table 0-4).
3.0 EPA Supplied REPFARM Inputs
EPA actions are entered into the REPFARM model as:
* changes in variable production costs,
* changes in fixed production costs,
* changes in crop yields, and
* changes in crop and livestock prices.
The changes in crop and livestock prices were obtained from AGSIM
and are described in Appendix B. The first year cost and yield
impacts assumed for each of the REPFARM models are described in
0-3
-------
Table D-l
1986 Farm Numbers
Illinois Corn Soybean;
Corn Belt —
Illinois
345,871 total farms
220,763 farms produce corn for grain
112,489 classified as cash grain farms
producing corn and soybeans I/
65,672 total farms
49,083 farms produce corn for grain
30,837 classified as cash grain farms producing
corn and soybeans I/
Mississippi Cotton Soybean;
Delta States - 73,747 total farms
- 7,438 farms produce cotton
- 3,576 classified as field crop farms producing
cotton and soybeans 2/
Mississippi
- 27,542 total farms
- 3,435 farms produce cotton
- 1,798 classified as field crop farms producing
cotton and soybeans 2_/
Kansas Wheat Cattle;
Northern
Plains
Kansas
153,884 total farms
84,097 farms produce wheat
50,143 produce wheat and raise cattle
54,024 total farms
31,000 farms produce wheat
19,966 produce wheat and raise cattle
I/ Cash grain farms are farms on which the largest portion of
gross income is accounted for by sales of cash grains such as
corn, soybeans or wheat.
2_/ Field crop farms are farms on which the largest portion of
gross income is accounted for by sales of field crops such as
cotton or tobacco.
Source: 1986 Farm Costs and Returns Survey
D-4
-------
Table D-2
Initial Characteristics of Representative Farms
Simulated for EPA's Agricultural Sector Study
Illinois Corn Soybean Farms
Farm acreage:
Cropland owned
Cropland rented
Pastureland owned
Pastureland rented
Total land operated
Cropland, percent tillable
Average
Financial
Position
160
363
0
0
523
98%
Number of full-time hired workers
Value of assets ($) I/:
Cropland & buildings 194,293
Pastureland 0
Farm machinery 86,920
Livestock 0
Non-farm investments 12,777
Beginning cash reserve 2,000
Debt to Asset Ratio .28
Off-farm income ($) 17,766
Family living expenses ($) 15,500
Crop acreage 2/:
Corn " 325
Soybeans 190
Crop yields (bu.) 3_/:
Corn 122.4
Soybeans 36.8
Vulnerable
Financial
Position
92
445
0
0
537
84%
130,656
0
85,980
0
6,736
2,000
.67
36,072
15,500
280
173
109.5
32.8
As of January 1, 1987.
Planted acreage plus set-aside acreage.
State average yields (1981-1987) were used for representative
producers in average financial condition. (Source: Crop
Production, 1983, 1986, and 1987 Annual Summaries). These
yields were adjusted (based on survey information) for
vulnerable producers.
Source: Data developed from 1986 Farm Costs and Returns Survey
D-5
-------
Table D-3
Initial Characteristics of Representative Farms
Simulated for EPA's Agricultural Sector Study
Mississippi Cotton Soybean Farms;
Farm acreage:
Cropland owned
Cropland rented
Pastureland owned
Pastureland rented
Total land operated
Cropland, percent tillable
Average
Financial
Position
413
1,016
0
0
1,429
81%
Number of full-time hired workers 2
Value of assets ($) I/:
Cropland & buildings
Pastureland
Farm machinery
Livestock
Non-farm investments
Beginning cash reserve
Debt to Asset Ratio
Off-farm income ($)
Family living expenses ($)
Crop acreage 2_/:
Cotton
Soybeans
Crop yields 3/:
Cotton (Ib.)
Soybeans ( bu . )
429,943
0
140,557
0
11,506
2,000
.33
16,856
15,500
545
611
722.5
22.0
Vulnerable
Financial
Position
409
1,442
0
0
1,851
84%
2
340,204
0
153,280
0
15,069
2,000
.83
5,193
15,500
657
889
722.5
18.7
I/
As of January 1, 1987.
Planted acreage plus set-aside acreage.
State average yields (1981-1987) were used. (Source
Production, 1983, 1986, and 1987 Annual Summaries).
Crop
Source: Data developed from 1986 Farm Costs and Returns Survey
D-6
-------
Table 0-4
Initial Characteristics of Representative Farms
Simulated for EPA's Agricultural Sector Study
Kansas Wheat Cattle Farms:
Farm acreage:
Cropland owned
Cropland rented
Pastureland owned
Pastureland rented
Total land operated
Cropland, percent tillable
Average
Financial
Position
326
431
224
296
1,277
77%
Vulnerable
Financial
Position
318
743
176
409
1,646
78%
Number of full-time hired workers
Value of assets (§) I/:
Cropland & buildings 145,356
Pastureland 50,176
Farm machinery 69,740
Livestock 9,390
Non-farm investments 15,187
Beginning cash reserve 2,000
Debt to Asset Ratio .31
Off-farm income (?) 20,123
Family living expenses ($) 15,500
114,326
39,424
80,143
24,540
8,571
2,000
.85
15,366
15,500
Crop acreage 2/:
Wheat
Soybeans
Sorghum
Corn
342
39
165
37
430
123
223
52
Crop yields (bu.) 3/:
Wheat
Soybeans
Sorghum
Corn
35.4
26.5
62.8
120.8
32.2
15.4
60.9
97.0
Continued..
D-7
-------
Table D-4. (Continued)
Kansas Wheat Cattle Farms:
Livestock inventory:
Cows
Replacement heifers
Feeder steers 4/
Average
Financial
Position
15
3
75
Vulnerable
Financial
Position
40
6
50
I/ As of January 1, 1987.
2/ Planted acreage plus set-aside acreage.
3/ State average yields (1981-1987) were used for representative
producers in average financial condition. (Source: Crop
Production, 1983, 1986, and 1987 Annual Summaries). These
yields were adjusted (based on survey information) for
vulnerable producers.
4/ Feeder steers are purchased and sold within the calendar year,
Source: Data developed from 1986 Farm Costs and Returns Survey
D-8
-------
Tables D-5 through D-7. These cost and yield effects were provided
by EPA Program Offices. Impacts of pesticide cancellations were
assumed to dissipate evenly over a seven year period.
4.0 REPFARM Output
The impact of EPA actions on the financial condition of each of
the representative farms was determined by examining:
* the change in net cash farm income due to EPA actions, and
* the change in debt asset ratios due to EPA actions.
Three major field crop and livestock farms in two financial
conditions were created, resulting in a total of six different
representative farms:
* an Illinois Corn Soybean Farm
- in average financial condition
- in vulnerable financial condition
* a Mississippi Cotton Soybean Farm
- in average financial condition
- in vulnerable financial condition
* a Kansas Wheat Cattle Farm
- in average financial condition
- in vulnerable financial condition
For each REPFARM in each scenario, two alternative sets of impacts
were considered:
* A Maximum Impact Case; In this case it is assumed that
the producer is impacted by every regulation that may
possibly affect a producer of that type.
* An Average Impact Case; In this case it is assumed that
the producer experiences the average impact of producers
of that type - e.g., if 50% of all producer of a given
type experience a $2.00/acre cost, we would assume a
$1.00/acre cost for the average impacted producer.
The net cash farm income and debt to asset ratios of each of these
farms is examined for each of the three alternative EPA scenarios
defined in this study. This output is presented in Figures Dl -
018.
D-9
-------
Table D-5
Potential Impacts on Illinois Corn Soybean Farm I/
Variable Cost; First Year Impacts
Scenario Action Crop Cost 2/ Yield(%) Acres(%)3/
1-3 Alachlor-restricted
use
corn .50 0 38.6
soybeans .50 0 25.4
1-3 Farm Worker Safety
corn .98 0 90
soybeans .62 80
1 Corn Rootworm
Insecticides Plan
I
corn .70 0 20
2 Groundwater Plan II:
alachlor
corn 1.80 0 1.5
soybeans 1.60 0 1
2 Groundwater Plan II:
cyanazine
corn 17.87 -11.07 0.2
2 Groundwater Plan II:
atrazine
corn 17.87 -11.07 1.6
2 Corn Rootworm
Insecticides Plan
II
corn -8.50 -24.0 34
3 Groundwater Plan III:
alachlor
corn 1.80 0 6.1
soybeans 1.60 0 8.3
3 Groundwater Plan III:
cyanazine
corn 17.87 -11.07 4.3
3 Groundwater Plan III:
atrazine
corn 17.87 -11.07 14.6
Continued...
D-10
-------
Scenario
Action
Table D-5 (continued)
Crop Cost y Yield(%) Acres(%)3/
Corn Rootworm
Insecticides Plan
III
corn
-8.50 -24.0
34
Fixed Costs:
Scenario
1-3
Action
Underground Storage Tank
1-3
Enclosed Cabs
Lead Ban
1-3
SARA Title III,
Section 302-304
Impact
Insurance: $2,500/yr.
2 tank tightness test @
$500, there are 5,428
USTs in the cornbelt
distributed over 310,000
farms.
Cost of enclosing cab =
$2,500. Assumed the 1/3
of all cabs must be
enclosed.
Assumed impacted farm
incurred 1,000 cost to
rebuild a tractor, truck
or combine engine.
Predicted 7,280 trucks,
4,865 combines and 23,112
tractors in cornbelt
would need to be rebuilt.
Cost = $50/covered farm.
Assumed 1/3 of all farms
covered.
I/ Supplied by EPA Program Offices.
2/ Cost per acre (1986$).
I/ Percent of indicated crop acres in the cornbelt likely to be
~~ affected.
D-ll
-------
Table D-6
Potential Impacts for Mississippi Cotton Soybean Farm I/
Variable Costs; First Year Impacts
Scenario Action Crop Cost 2/ Yield(%) Acres(%)3/
1-3 Dinoseb Cancellation
cotton 5.00 -1.5 24.1
soybeans 16.00 0 10.5
1-3 Toxaphene
cancellation
soybeans 6.8 0 1.2
1-3 Chlorodimeform -
cancellation of
yield enhancement
cotton 3.88 0 24
1-3 Alachlor-restricted
use
soybeans .50 0 10
1-3 Farm Worker safety
cotton .44 0 95
soybeans .65 0 85
1-2 Groundwater Plan I
& II: aldicarb
cotton 6.42 0 0.4
1 Groundwater Plan II:
alachlor
soybeans 1.60 0 1
2 Organophosphates
Plan II
cotton 4.15 0 1
2 Groundwater Plan II:
cynazine
cotton 5.00 6 1.3
3 Groundwater Plan III:
alachlor
soybeans 1.60 0 5
3 Organophosphates
Plan III
cotton 8.92 0 93.5
Continued...
D-12
-------
Table D-6 (continued)
Scenario Action Crop Cost 2/ Yield(%) Acres(%)3/
Groundwater Plan III:
aldicarb
cotton 6.42 0
Groundwater Plan III:
cyanazine
cotton 5.00 6
2.4
23.1
Fixed Costs:
Scenario
1-3
Action
1-3
1-3
Underground Storage Tank
Enclosed Cabs
SARA Title III,
Sections 302-304
Lead Ban
Impact
Insurance = $2,500/yr
Tank tightness test (2) =
$500. There are 2,099 UST
in the Delta distributed
over 132,000 farms.
Cost of Enclosing Cab =
$2,500. Assumed that 1/3
of all cabs must be
enclosed.
Cost = $50/covered farm.
Assumed 1/3 of all farms
covered.
Assumed impacted farm
incurred $1,000 cost to
rebuild a tractor, truck
or combine engine.
Assumed 1,150 tractors,
1,124 trucks and 303
combines in Delta need
to be rebuilt.
I/ Supplied by EPA Program Offices.
|/ Cost per acre (1986$).
3/ Percent of indicated crop acres in the cornbelt likely to be
affected.
D-13
-------
Table D-7
Potential Impacts for Kansas Wheat Cattle Farm I/
Variable Costs: First Year Impacts
Scenario Action Crop Cost 2/ Yield(%) Acres(%)3/
1-3 Alachlor-Restricted
Use
corn .50 0 37.1
soybeans .50 0 19
1-3 Farm Worker
corn .98 0 90
soybeans .65 0 75
wheat .45 0 80
1 Corn Rootworm
Insecticides Plan I
corn .70 0 35
2 Groundwater Plan II:
alachlor
corn 1.82 0 0.3
soybeans 1.60 0 0.1
sorghum 1.82 0 0.2
2 Groundwater Plan II:
atrazine
corn 18.41 -1 0.5
sorghum 18.41 -1 0.5
2 Groundwater Plan II:
cyanazine
corn 18.41 -1 0.2
2-3 Corn Rootworm
Insecticides Plan
II, III
corn -8.50 -16 58
2-3 Fungicides Plan II,
III
wheat -3.71 -44 0.7
3 Groundwater Plan III:
alachlor
corn 1.82 0 1.3
soybeans 1.60 0 0.5
sorghum 1.82 0 3.4
Continued...
D-14
-------
Scenario
Table D-7 (continued)
Action Crop Cost 2/ Yield(%) Acres(%)3/
Groundwater Plan III:
atrazine
corn 18.41
sorghum 18.41
Groundwater Plan III:
cyanazine
corn 18.41
sorghum 18.41
Fixed Costs;
Scenario
1-3
Action
Underground Storage Tanks
1-3
Enclosed Cabs
1-3
SARA Title III:
Sections 302-304
Lead Ban
-1 9.6
-1 11.4
-1 2.7
-1 0.10
Impact
Insurance = $2,500/yr
Tank Tightness Test =
$500/each (need 2)
There are 4,045 UST
in the Northern Plains
distributed over
196,000 farms.
Cost of Enclosing cab
= $2,500. Assumed 1/3
of all cabs must be
enclosed.
Cost = $50/covered
farms. Assumed 1/3
of all farms are
covered.
Assumed impacted farm
incurred $1,000 cost
to rebuild a tractor,
truck or combine
engine. Assumed
8,580 trucks, 8,380
tractors and 3,015
combines in the Northern
Plains would need to
be rebuilt.
I/ Supplied by EPA Program Offices.
2 Cost per acre (1986$).
3_/ Percent of indicated crop areas in the cornbelt likely to
be affected.
D-15
-------
Illinois Corn Soybean Farm: Scenario 1
Average Financial Condition
CO
oo
CD
E-3T
0~O
O C
C CO
— c/)
(A
CO
O
00
45 r-
40
35
30-
25
Average
Maximum
Base
1987 1989 1991 1993 1995
1988 1990 1992 1994 1996
Year
CO
DC
0)
(/)
0)
CD
Q
0.45r-
0.4
0.35
0.3
0.25
0.2
1987 1989 1991 1993 1995
1988 1990 1992 1994 1996
Year
Figure D-l. EPA impacts on net cash farm income and debt
asset ratio for a representative Illinois corn soybean farm
in average financial condition: Scenario 1
n-i fi
-------
Illinois Corn Soybean Farm: Scenario 1
Vulnerable Financial Condition
10,-
CO
00
CD
E-5T
OT3
O C
C CTJ
— tf)
en
CD
O
-5
-10-
-15
Average
Maximum
Base
1987 1989 1991 1993 1995
1988 1990 1992 1994 1996
Year
g
"to
QC
0)
in
3
CD
Q
0.75
0.7
0.65
o.e|
o.ssl
1987
1989 1991 1993 1995
1988 1990 1992 1994
Year
1996
Figure D-2. EPA impacts on net cash farm income and debt
asset ratio for a representative Illinois corn soybean farm
in vulnerable financial condition: Scenario 1
D-17
-------
Illinois Corn Soybean Farm: Scenario 2
Average Financial Condition
Q
0.44 r-
0.4
0.36
0.32
0.28
0.24
0.2
1987 1989199119931995
1988 1990 1992 1994 1996
Year
Figure D-3. EPA impacts on net cash farm income and debt
asset ratio for a representative Illinois corn soybean farm
in average financial condition: Scenario 2
D-18
-------
Illinois Corn Soybean Farm: Scenario 2
Vulnerable Financial Condition
-------
Illinois Corn Soybean Farm: Scenario 3
Average Financial Condition
CO
at
«
Q
0.44
0.4
0.36
0.32
0.28
0.24-
0.2
1987 1989 1991 1993 1995
1988 1990 1992 1994 1996
Year
Figure D-5. EPA impacts on net cash farm income and debt
asset ratio for a representative Illinois corn soybean farm
in average financial condition: Scenario 3
D-20
-------
Illinois Corn Soybean Farm: Scenario 3
Vulnerable Financial Condition
to
00
o>
-------
MS Cotton Soybean Farm: Scenario 1
Average Financial Condition
(O
CO
O)
Q
0.4,-
0.36
0.32
0.28
0.24
0.2
1987 1989 1991 1993 1995
1988 1990 1992 1994 1996
Year
Figure D-7. EPA impacts on net cash farm income and debt
asset ratio for a representative Mississippi cotton soybean
farm in average financial condition: Scenario 1
D-22
-------
MS Cotton Soybean Farm: Scenario 1
Vulnerable Financial Condition
o>
Q
0.9r-
0.85
0.8
0.75
0.7
0.65
0.6
0.55
1987 1989 1991 1993 1995
1988 1990 1992 1994 1996
Year
Figure D-8. EPA impacts on net cash farm income and debt
asset ratio for a representative Mississippi cotton soybean
farm in vulnerable financial condition: Scenario 1
D-23
-------
MS Cotton Soybean Farm: Scenario 2
Average Financial Condition
o>
0>
OT3
O C
C (0
— (0
0)
(0
O
09
100
90
80
70
60
50
40
30
20
Average
Maximum
Base
1987 1989 1991 1993 1995
1988 1990 1992 1994 1996
Year
CO
CC
GO
(A
cn
-O
o>
Q
0.4r-
0.36
0.32
0.28
0.24
0.2
1987 1989 1991 1993 1995
1988 1990 1992 1994 1996
Year
Figure D-9. EPA impacts on net cash farm income and debt
asset ratio for a representative Mississippi cotton soybean
farm in average financial condition: Scenario 2
D-24
-------
E-ST
81
C (0
— c/>
§1
"t
£
(O
(0
O
0)
z
MS Cotton Soybean Farm: Scenario 2
Vulnerable Financial Condition
80
70
60
50
40
30
20
10
Average
Maximum
Base
1987 1989 1991 1993 1995
1988 1990 1992 1994 1996
Year
CO
CC
.O
-------
MS Cotton Soybean Farm: Scenario 3
Average Financial Condition
-------
MS Cotton Soybean Farm: Scenario 3
Vulnerable Financial Condition
8
o>
80
70
60
50
40
30
20
10
0
Average
Maximum
Base
1987 1989 1991 1993 1995
1988 1990 1992 1994 1996
Year
OJ
QC
0)
(/)
(A
0)
Q
0.95
0.9
0.85
0.8
0.75
0.7
0.65
0.6
0.55
1987 1989 1991 1993 1995
1988 1990 1992 1994 1996
Year
Figure D-12. EPA impacts on net cash farm income and debt
asset ratio for a representative Mississippi cotton soybean
farm in vulnerable financial condition: Scenario 3
D-27
-------
Kansas Wheat Cattle Farm: Scenairo 1
Average Financial Condition
co
CO
o>
-------
Kansas Wheat Cattle Farm: Scenario 1
Vulnerable Financial Condition
2r-
tO
00
O)
OT3
O C
C flj
— o
il
(A
(0
O
CB
z
-2
-4
-6
-8
•10
•12
1987 1989 1991 1993 1995
1988 1990 1992 1994 1996
Year
.0
QC
i
-------
Kansas Wheat Cattle Farm: Scenairo 2
Average Financial Condition
CO
GO
O)
0)
E«T
OT3
O C
C (Q
CO
nj
O
GO
20
15
10
-5
-10
-15
Average
Maximum
Base
1987 1989 1991 1993 1995
1988 1990 1992 1994 1996
Year
OJ
-------
Kansas Wheat Cattle Farm: Scenario 2
Vulnerable Financial Condition
CD
GO
O)
<0
E-W
8?
C (0
— w
§1
OJ
O
CD
5
0
-5
-10
-15
-20
-25
-30
-35
-40
Average
Maximum
Base
1987 1989 1991 1993 1995
1988 1990 1992 1994 1996
Year
1.2,-
(0
DC
09
en
in
-------
Kansas Wheat Cattle Farm: Scenairo 3
Average Financial Condition
CO
GO
cn
0-
-5-
-10-
Average
Maximum
Base
-15
1987 1989 1991 1993 1995
1988 1990 1992 1994 1996
Year
0.5r-
.g
ra
CC
(A
(/I
2
0)
Q
0.45
0.4
0.35
0.3
0.25
0.2
1987 1989 1991 1993 1995
1988 1990 1992 1994 1996
Year
Figure D-17. EPA impacts on net cash farm income and debt
asset ratio for a representative Kansas wheat cattle farm
in average financial condition: Scenario 3
D-32
-------
Kansas Wheat Cattle Farm: Scenario 3
Vulnerable Financial Condition
CD
00
o>
0)
8 -a
c
C CO
— v)
il
(A
0)
O
5
0
-5
-10
-15
-20
-25
-30
-35
Average
Maximum
Base
1987 1989 1991 1993 1995
1988 1990 1992 1994 1996
Year
1.2r-
o
"a
CC
«
(/)
w
1
0.8
0.6
.a
0)
a
0.4
0.2
1987 1989 1991 1993 1995
1988 1990 1992 1994 1996
Year
Figure D-18. EPA impacts on net cash farm income and debt
asset ratio for a representative Kansas wheat cattle farm
in vulnerable financial condition: Scenario 3
D-33
-------
APPENDIX E
Income Budget Analysis and Results
By
Craig Simons
and
Roger Lloyd I/
I/ OPRA Incorporated
-------
Appendix E
Income Budget Analysis and Results
1.0 Budgeting Analysis
To more clearly assess regulatory impacts on an individual unit
of production for a given commodity and region, a budgeting analysis
was used. Baseline conditions were defined as net returns to
management and land for one acre of production prior to any
regulatory action. These conditions were calculated from regional
production cost and yield estimates and national price estimates.
Total production cost estimates were obtained from crop enterprise
budgets compiled by the USDA Cooperative Extension Service in
each appropriate state. Crop enterprise budgets typically catego-
rize total costs as variable and fixed. Variable costs are those
which vary according to the level of production. Fixed costs are
those which (in the short run) are unrelated to production levels.
Enterprise budgets vary in their treatment of expensing the cost
of owner provided inputs. For this study, the cost of owner
provided land and management were excluded. Any net returns
would then be attributable to these factors of production. To
the extent possible, all budgets were adjusted to be comparable.
In instances where a production region consisted of two or more
states (e.g., Idaho and Washington potatoes) a production weighted
total cost of production was calculated. All costs were adjusted
by the Index of Prices Paid by Farmers to reflect 1986 dollars.
The baseline-conditions were then adjusted by the cost and yield
impact estimates and the national price change estimates (developed
from the national price-quantity model and adjusted for regional
differences) to estimate the post-impact net returns per acre
for each regulatory scenario by region and crop. It is expected
net returns per acre will typically decrease from the influence
of regulatory impacts because of:
1. increased variable costs per acre of production, and
2. decreases in yield which lowers production and thus lowers
revenue per acre.
Ameliorating these negative effects on net revenue would be an
increase in price caused by a national decline in supply due to
decreased production nationwide.
Algebraically, the farm income budgeting model can be expressed as:
NR. = NR + dTR - dC.
1 O
E-l
-------
Since TR is dependent on price and production,
dTR = P.Q. - PoQQ.
Thus,
NR. = NRQ
P.Q. - PQQo - dC.
Where:
NR. = Net returns per acre of commodity production
after the regulatory scenario,
NR
Net returns per acre or commodity production
before the regulatory scenario,
dTR = change in total revenue,
dC = change in total costs;
P. = commodity price after the regulatory scenario,
P = commodity baseline price
Q. = commodity, production per acre after the regulatory
scenario, and
Q = commodity production per acre under baseline
conditions.
2.0 Data Inputs
Production cost estimates and baseline net returns for each
specialty crop production region (Table E-l) along with an estimate
of an average price and production (Appendix C, Table C-l) were
required to complete this analysis. Regional estimates of average
and maximum variable cost and yield changes associated with
environmental regulations for each specialty crop under each
scenario were provided by EPA. First year production cost and
yield changes are presented in Tables E-2 through E-5.
3.0 Model Results
Regulatory impacts on net returns which consider effects on product
price, quantity of production and production costs are presented
graphically in Figures E-l through E-9. Average and maximum
impacts are measured from a baseline net return (no regulatory
impact) for each of the specialty crops under the three policy
scenarios.
E-2
-------
Table E-l.
Baseline production costs and net returns
Crop/Region
Per acre production costs
Variable Fixed Total
costs costs costs
Baseline
net returns
(1986$)
Irish Potatoes
ID -
ND -
ME
WA
MN
983
332
762
.14
.90
.67
229
235
149
.22
.19
.88
1,212
568
912
.36
.09
.55
606
243
134
.00
.00
.00
Green Peas
WI
WA
Apples
WA
NY
MI
Peanuts
GA -
NC -
TX -
I/
AL
VA
OK
132
245
2,593
1,785
1,112
322
338
222
.35
.81
.41
.00
.70
.16
.65
.27
47
59
897
162
544
126
185
88
.20
.68
.66
.07
.44
.84
.98
.99
179
314
3,491
1,947
1,657
449
524
311
.55
.49
.07
.07
.14
.00
.63
.26
197
78
327
217
76
286
386
186
.00
.00
.00
.00
.00
.00
.00
.00
Caneberries
(Red Raspberries)
WA
OR
Tomatoes
FL (Fresh)
CA (Processing)
3,274.21 1,588.81 4,863.02
3,962.45 1,922.78 5,885.23
6,310.31
1,092.05
351.59 6,661.90
174.50 1,266.55
NA
NA
1,510.00
659.00
I/ Net returns are for additional peanuts. Net returns for
~ quota peanuts are $298, $444 and $206 for GA-AL, NC-VA and
TX-OK, respectively.
Source: Crop enterprise budgets from the individual states.
E-3
-------
Table E-2
Potential Impacts for Selected Apple Producers
Variable Cost; First Year Impact
Scenario Action
1-3 Farm Worker Safety
1 Organophosphates Plan
2 Organophosphates Plan
3 Organophosphates Plan
1 Groundwater Plan I
2 Groundwater Plan II
3 Groundwater Plan II
1 Fungicides Plan I
2 Fungicides Plan II
3 Fungicides Plan II
Region
WA
NY
MI
I WA
NY
MI
II WA
NY
MI
III WA
NY
MI
WA
NY
MI
WA
NY
MI
WA
NY
MI
WA
NY
MI
WA
NY
MI
WA
NY
MI
Cost I/
5.40
5.40
5.40
2.00
2.00
2.00
25.08
14.38
14.38
33.08
9.39
9.39
0.0
0.0
0.0
11.83
10.90
10.90
11.83
10.90
10.90
0.0
0.0
0.0
0.0
-13.06
-13.06
0.0
-13.06
-13.06
Yield(%)
0
0
0
0
0
0
0
0
0
-2
-2
-2
0
0
0
0
0
0
0
0
0
0
0
0
0
-20
-20
0
-20
-20
Acres
90
90
90
86
100
100
62
75
75
86
100
100
0
0
0
5
10
10
25
45
45
0
0
0
0
83
58
0
83
58
Continued,
E-4
-------
Table E-2 (continued)
Fixed costs:
Scenario
1-3
1-3
1-3
Action
SARA Title III
Section 302-304
Enclosed Cabs
Underground Storage Tanks
Lead Phasedown
Impact
Cost = $50/covered farm.
Assumed 1/3 of all farms
covered.
Cost = $2,500. Assumed
1/3 of all cabs must be
enclosed.
Some farms may incur costs
due to Underground Storage
Tank regulations, however,
due to the significant
amount of uncertainty as to
whether specialty crop farms
would have covered UST's.
These costs were not included.
Under a total ban of lead
in gasoline for agricultural
use, farmers having gasoline
powered tractors, combines,
and trucks may incur a cost
to rebuild the valves.
This cost would be approxi-
mately $1,000 for a combine
and a truck, and $750 for a
tractor. These costs were
not included in the budget
analyses for apple producers.
I/ Cost per acre (1986$)
E-5
-------
Table E-3
Potential Impacts for Selected Potato Producers
Variable Cost; First Year Impacts
Scenario Action
1-3 EDB Cancellation
1-3 Dinoseb Cancellation
1-3 Farm Worker Safety
1 Groundwater Plan I
2 Groundwater Plan II
3 Groundwater Plan III
1 Organophosphates Plan
2 Organophosphates Plan
3 Organophosphates Plan
1 Fungicides I
2 Fungicides II
Region
WA/ID
MN/ND
ME
WA/ID
MN/ND
ME
WA/ID
MN/ND
ME
WA/ID
MN/ND
ME
WA/ID
MN/ND
ME
WA/ID
MN/ND
ME
I WA/ID
MN/ND
ME
II WA/ID
MN/ND
ME
III WA/ID
MN/ND
ME
WA/ID
MN/ND
ME
WA/ID
MN/ND
ME
Cost I/
16.80
18.48
18.48
8.51
8.51
8.51
1.43
1.43
1.43
0.00
10.00
11.00
0.00
10.00
11.00
39.13
10.00
11.00
1.00
1.00
1.00
5.88
5.88
5.88
7.00
7.00
7.00
0.00
0.00
0.00
8.81
6.61
11.05
Yield(%)
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
-8
-8
-8
0
0
0
0
0
0
Acres(%)
2.2
1.1
1.1
50.0
50.0
50.0
90.0
90.0
90.0
0.0
3.5
1.9
0.0
3.5
1.9
12.4
14.6
7.5
74.0
74.0
74.0
68.0
68.0
68.0
74.0
74.0
74.0
0.0
0.0
0.0
7.0
54.0
80.0
Continued..
E-6
-------
Table E-3 (continued)
Fungicides III
WA/ID
MN/ND
ME
-0.60
-0.45
-0.75
-8
-8
-8
12.0
80.0
80.0
Fixed costs:
Scenario
1-3
1-3
1-3
Action
SARA Title III
Section 302-304
Enclosed Cabs
Underground Storage Tanks
Lead Phasedown
Impact
Cost = $50/covered farm.
Assumed 1/3 of all farms
covered.
Cost = $2,500. Assumed
1/3 of all cabs must be
enclosed.
Some farms may incur costs
due to Underground Storage
Tank, regulations, however,
due to the significant
amount of uncertainty as to
whether specialty crop farms
would have covered UST's.
These costs were not included.
Under a total ban of lead
in gasoline for agricultural
use, farmers having gasoline
powered tractors, combines,
and trucks may incur a cost
to rebuild the valves.
This cost would be approxi-
mately $1,000 for a combine
and a truck, and $750 for a
tractor. These costs were
not included in the budget
analyses for potato producers.
I/ Cost per acre (1986$)
E-7
-------
Table E-4
Potential Impacts for Selected Pea Producers
Variable Costs; First Year Impacts
Scenario Action
1-3 Dinoseb Cancellation
1-3 Farm Worker Safety
1 Organophosphates Plan I
2 Organophosphates Plan II
3 Organophosphates Plan III
Region Cost I/ Yield(%) Acres(%)
WA
WI
WA
WI
WA
WI
WA
WI
WA
WI
10.40
0.00
0.86
0.86
1.00
1.00
2.92
2.92
3.08
3.08
0
0
0
0
0
0
0
0
0
0
75
0
90
90
30
30
30
30
35
35
Fixed costs:
Scenario
1-3
1-3
1-3
Action
SARA Title III
Section 302-304
Enclosed Cabs
Underground Storage Tanks
Impact
Cost = $50/covered farm.
Assumed 1/3 of all farms
covered.
Cost = $2,500. Assumed
1/3 of all cabs must be
enclosed.
Some farms may incur costs
due to Underground Storage
Tank regulations, however,
due to the significant
amount of uncertainty as to
whether specialty crop farms
would have covered UST's.
These costs were not included.
Continued..
E-8
-------
Table E-4 (continued)
Lead Phasedown
Under a total ban of lead
in gasoline for agricultural
use, farmers having gasoline
powered tractors/ combines,
and trucks may incur a cost
to rebuild the valves.
This cost would be approxi-
mately $1,000 for a combine
and a truck, and $750 for a
tractor. These costs were
not included in the budget
analyses for pea producers.
I/ Cost per acre (1986$)
E-9
-------
Table E-5
Potential Impacts for Selected Tomato Producers
Variable Costs; First Year Impacts
Scenario
1-3
1-3
1
2
3
Action
EDB Cancellation
Farm Worker Safety
Fungicides Plan I
Fungicides Plan II
Fungicides Plan III
Region
CA
FL
CA
FL
CA
FL
CA
FL
CA
FL
Cost
22.65
22.65
7.50
7.50
0.00
0.00
1.50
20.93
-3.39
-20.34
0
0
0
0
0
0
0
0
-20
-20
2.9
2.9
90.0
90.0
0.0
0.0
9.0
77.0
25.0
98.0
Fixed costs:
Scenario
1-3
1-3
1-3
Action
SARA Title III
Section 302-304
Enclosed Cabs
Underground Storage Tanks
Impact
Cost = $50/covered farm.
Assumed 1/3 of all farms
covered.
Cost = $2,500. Assumed
1/3 of all cabs must be
enclosed.
Some farms may incur costs
due to Underground Storage
Tank regulations/ however,
due to the significant
amount of uncertainty as to
whether specialty crop farms
would have covered UST's.
These costs were not included.
Continued..
E-10
-------
Table E-5 (continued)
Lead Phasedown
Under a total ban of lead
in gasoline for agricultural
use, farmers having gasoline
powered tractors, combines,
and trucks may incur a cost
to rebuild the valves.
This cost would be approxi-
mately $1,000 for a combine
and a truck, and $750 for a
tractor. These costs were
not included in the budget
analyses for tomato producers.
I/ Cost per acre (1986$)
E-ll
-------
Impacts on WA Apple Net Returns
330|-
325-
320-
315
Impacts on NY Apple Net Returns
220,-
216-
210
1887 1888 1881 1893 1895
1888 IBM i&82 is9< IBM
Yeai
20!
1887 1868 1881 1893 1895
1888 1890 1882 1894 1896
Yew
I
i—«
ro
Impacts on Ml Apple Net Returns
70-
Average
Maximum
Base
1967 1989 1881 1993 1895
1868 1890 1892 1894 1896
Year
Figure E-l. Scenario 1. reBulatory impacts on apple net returns
-------
Impacts on WA/ID Potato Net Returns
Impacts on MN/ND Potato Net Returns
2
g
610,-
605 -
600 -
695
i
245r-
240-
235 -
230
225
-TB87 1989 1991 1993 1995
1968 1990 1992 1994 1996
Yeai
220
1087 1989 1991 1993 199S
IB88 1990 1992 1994 1996
Yflitf
Impacts on ME Potato Net Returns
MOr-
U
Aveiaga
Maximum
Basa
1987 1989 1991 1993 1995
1988 1990 1992 1994 1996
Yea/
Figure E-2. Scenario 1, regulatory impacts on potato net returns
-------
Impacts on CA Tomato Net Returns
670.-
660 -
e&o
640
1987 IMS 1991 1993 1995
1090 1992 1994 1990
Yea/
Impacts on FL Tomato Net Returns
1520.-
1510 -
1500-
1490
fc fc fc h h -*
1987 1989 1991 1993 1995
1990 1992 1994 1996
Vaar
Avuidge
Maximum
Baso
Impacts on Wl Pea Net Returns
205r-
200 -
195
1901
1987 1989 1991 1993 1995
1988 1990 1992 1994 1996
Year
i
I
dk
a
t
Impacts on WA Pea Net Returns
K -
70
1987 1989 1991 1993 1995
1980 1990 1992 1994 1996
Year
Figure E-3. Scenario 1, regulatory impacts on tomato and pea net returns
-------
Impacts on WA Apple Net Returns
Impacts on NY Apple Net Returns
1987
1089 1991 1993 1995'
1990 1992 1994 1998
Yew
250r-
200-
160-
100-
60-
0 -
•SO -
•100 -
•ISO -
•200
1987 1989 1991 1993 1995
1968 1990 1092 1094 I09B
Yaw
I
i
Impacts on Ml Apple Net Returns
ISOr-
100-
50 -
0 -
•50-
•100 -
•150-
200-
•250
Average
Bast
1987 1989 1991 1993 1995
1988 1990 1992 1994 1996
Yew
Fiyiire E-4. Scenario 2, regulatory impacts on apple net returns
-------
Impacts on WA/ID Potalo Net Returns
Impacts on MN/ND Potato Net Returns
615,-
610
605-
600
595 -
590
i
1
245,-
240
235
230 -
225
1987 1989 1991 1993 1995
1988 1990 1992 1994 1996
220
1687 1989199119931995
1988 1990 1992 1994 19-JB
Year
Yuar
Impacts on ME Potato Net Returns
Avewje
MaikiHiro
Base
1987 1989 1991 1993 1995
1988 1990 1992 1994 1996
Yeai
Figure E-5. Scenario 2 repulnrnru imm
-------
Impacts on CA Tomato Net Returns
670,-
660 -
650 -
Impacts on FL Tomato Net Returns
1520r-
1510-
1500 -
1400
Year
107 1989 1001 1693 1895
toaa loao 1092 1094
Yuaf
Average
Maximum
Base
Impacts on Wl Pea Net Returns
200,-
195-
190
1987 1989 1991 1993 1995
1988 1090 1992 1994 1096
Ydar
i
Impacts on WA Pea Net Returns
80r-
70 -
fiS
1987 1989 1991 1993 1995
1988 1090 1992 1994 1006
Yaar
Figure E-6. Scenario 2, regulatory impacts on tomato and pea net returns
-------
Impacts on WA Apple Net Returns
Impacts on NY Apple Net Returns
250r-
200-
150-
100-
60-
0 -
1987 1989 1991 1993 1995
1988 1990 1992 1994 199ft
YMT
•100-
-160 -
•200
1987 1989 1991 1993 1995
I9B8 1990 1992 1994 1996
Y«ar
I
!-•
00
Impacts on Ml Apple Net Returns
I00|-
so -
0-
-50 -
•100 -
-150-
200-
-250-
•300
Mainuim
Base
1987 1989 1991 1993 1995
1988 1990 1992 1994 1996
Yeai
Figure li-7. Scenario 3, regulatory impacts on apple net returns
-------
Impacts on WA/ID Potato Net Returns
Impacts on MN/ND Potato Net Returns
700,-
650-
600 -
550
500
450
UJ
a
u
S
I
255r-
240
225
210-
195-
1987 1989 1991 1993 1995
1988 1990 1992 1994 1996
1987 1989 1991 1993 1995
1988 1990 1992 1994 1998
l
•-•
10
Year
Year
Impacts on ME Potato Net Returns
!
Avaiaga
Maximum
Base
1987 1989 1991 1993 1995
1988 1990 1992 1994 1996
Year
Figure E-3. Scenario 3, regulatory Impacts on potato net returns
-------
Impacts on CA Tomato Net Returns
Impacts on FL Tomato Net Returns
1987 1989 1991 1993 1995
1988 1990 1992 1994 1998
Average
Maximum
Base
1907 19B9 1991 1993 1995
1988 1990 1992 1994 1996
Year
i
ro
o
Impacts on Wl Pea Net Returns
Impacts on WA Pea Net Returns
205,-
200-
195
1987 1989 1991 1993 1995
1988 I9a0 1992 1994 1996
u
5
80,-
75-
70-
65
1987 1989 1991 1993 1995
1988 1990 1992 1994 1996
Year
Yeai
Figure E-9. Scenario 3, regulatory impacts on tomato and pea net returns
-------
APPENDIX F
Data Problems and Assumptions
By
Robert Torla I/
I/ Office of Pesticide Programs, U.S. Environmental Protection
Agency
-------
Appendix F
Data Problems and Assumptions
The agricultural sector study relied on a wide range of information
sources of varying quality. This section summarizes the data
sources and briefly discusses the limitations of the data.
1.0 Basic Crop Production Information
Basic crop production data was obtained from annual publications
of the USDA National Agricultural Statistics Service (NASS) where
data were available. For apples and caneberries there was not a
consistent data source. Production and price information for
apples was obtained from USDA, while information on acres harvested
was obtained from the Bureau of Census. Different estimation
techniques were used in these two sources and they were collected
in different time periods. However, apples are a relatively slow
growing perennial crop, so differences in time frames of a few
years are probably not particularly important. There were limited
caneberry data available in statistical publications from some
important states. The production data sources used in this study
are listed below.
A. Crop Production, Annual Summary for relevant years,
National Agricultural Statistics Board, USDA.
B. Vegetables, Annual Summary for relevant years, National
Agricultural Statistics Board, USDA.
C. 1982 Census of Agriculture, Bureau of Census, USDC.
D. Non Citrus Fruits and Nuts, Annual Summary for relevant
years, National Agricultural Statistics Board, USDA.
E. Various state annual reports of agricultural statistics
for relevant years.
2.0 Time Frames for Actions
We attempted to project the year in which actions might take
place and, for past actions, relied on historical information as
to when actions actually occurred. Projections for future actions
were based on an examination of likely dates for actions to take
place.
For all pesticide specific actions we projected that impacts
would dissipate evenly over a seven year period as users adjusted
their practices and new pest control products became available.
F-l
-------
There is some question regarding the accuracy of this assumption.
Clearly, if new technologies exist to ameliorate the impacts of a
regulatory action, they would tend to be registered (if necessary
and they meet the criteria) and adopted within a seven year
period. In addition, the cancellation of a pesticide would
create some incentive to replace it. However, there is no
certainty that such new technologies exist or if they do not
currently exist, would be developed, registered, marketed, and
adopted within a seven year time frame. The incentive to develop
and market new technologies would tend to be greater for the
major field crops, where large potential markets exist. There
are also some data which suggest that new pesticides would be
more expensive than older ones which have been cancelled.
3.0 Pesticide Usage Data
Quality of pesticide usage data vary widely. There are adequate
regional (multi state) level usage data for most major field
crops (corn, cotton, sorghum, wheat, and soybeans). Pesticide
usage data for barley, oats, and hay are sporadic, with the most
recent data being from the 1970's. Therefore, usage estimates
developed by the registrants were used for these crops. In
general, the usage data bases for major field crops are designed
to be statistically reliable at the 10 percent level for the
sample region. USDA has on occasion, collected statistically
reliable state level data for selected major field crops in
selected states.
Specialty crop pesticide usage data are highly erratic. USDA
last collected pesticide usage data for tomatoes, green peas,
apples, and potatoes in the 1970's. Latest USDA peanut pesticide
usage data are for 1982 and there are no data for caneberries.
State collected pesticide usage data were utilized when available.
However, there are no regular periodic state usage surveys.
California collects and reports all pesticide usage for restricted
use materials and commercial applicators. This results in usage
data which should be very reliable for restricted use materials;
but are of questionable usefulness for unrestricted use materials.
The Pesticide Program has access to some proprietary pesticide
usage estimates for major field crops and selected specialty
crops. However, the reliability of these estimates is largely
unknown. For major pesticides on major crops, these estimates
agree with available data collected in statistically designed
surveys. However, for minor pesticides and specialty crops, usage
estimates obtained from proprietary sources are often inconsistent
with available statistically designed surveys.
Analysis of the proposed pesticides in groundwater actions required
projections of pesticide use at the county level. However, there
are no public data collected to be statistically reliable at the
F-2
-------
county level. Data provided by a contractor was used to predict
pesticide usage at the county level. However, this data base is
composed of information drawn from available reports and expert
opinion or local Cooperative Extension Service personnel and 'is not
based on a statistically valid sample. The Federal government
does not have data to check the reliability of any of these
estimates.
4.0 Comparative Efficacy and Costs of Alternative Controls
Inputs developed and cleared by the program offices were used
for past and near actions. The rigor of these analyses varied
considerably. In some instances, potential yield impacts were
not investigated and a zero yield loss was assumed. In other
situations, rigorous analyses of the magnitude of possible yield
losses were available.
In general, available pesticide crop trials are not designed to
generate statistically reliable estimates of the differences in
yields among substitute chemicals. The objective of the crop
trials is to demonstrate that the pesticide provides some control
of the pest and not to reveal how pesticides compare with each
other.
For actions expected to take place further in the future (generally
beyond about one year), various sources of information were
employed. The following reports generated by, or for, and cleared
by the program offices were used:
Preliminary Benefit Analysis of EOB
Preliminary Benefit Analysis of Toxaphene
Preliminary Benefit Analysis of EPN
Preliminary Benefit Analysis of 2,4,5-T
Preliminary Benefit Analysis of Silvex
Preliminary Benefit Analysis of Carbon Tetrachloride
Regulatory Impact Analysis: Worker Protection Standards for
Agricultural Pesticides
Regulatory Impact Analysis in Support of Rulemaking Under
Sections 302, 303 and 304 of Title III of the Superfund
Amendments and Reauthorization Act of 1986
Regulatory Impact Analysis of Proposed Technical Standards for
Underground Storage Tanks
F-3
-------
Regulatory Impact Analysis of Proposed Financial Responsi-
bility Requirements for Underground Storage Tanks Containing
Petroleum
Preliminary Benefit Analysis of Dinocap
Preliminary Benefit Analysis of Chlordimeform
Preliminary Benefit Analysis of Ethyl Parathion
Preliminary Benefit Analysis of Aldicarb
Abbreviated Benefit Analysis of Dinoseb.
4.1 Corn and Soybeans
Publications from the USDA Commodity Assessment of Pesticide Use
on Corn and Soybeans and Potential Bans of Corn and Soybean
Pesticides, by Craig Osteen and Fred Kuchler USDA, ERS, Agricul-
tural Economic Report Number 546 as well as some unpublished
supporting commodity assessment data information (made available
by the USDA) provided comparative efficacy for corn and soybeans.
This provided a consistent data base which appears reasonable for
the actions proposed for the future. The commodity assessment data
base was constructed by obtaining expert opinion of estimates of
product cost and yield effects due to losses of pesticides. The
USDA has not updated this report and the estimates are somewhat
dated. In some cases, the cost of alternatives provided in the
Commodity Assessment was not appropriate for this analysis. In
these cases the Commodity Assessment was supplemented with
information from the Economic Analysis Branch (EAB) price files.
Efficacy data for corn and soybeans is probably the most reliable
of all crops considered in this analysis.
Concerns about groundwater contamination were assumed to result
in the cancellation of both alachlor and the triazines in selected
areas. In reality alachlor and the triazines are partial sub-
stitutes; however, the Commodity Assessment never considered the
question of the loss of both alachlor and the triazines. In the
absence of any information on how production costs and yields would
change under the cancellation of both alachlor and the triazines,
we used the commodity assessment data, which indicate the efficacy
information associated with the cancellation of each one, assuming
the other remains on the market. Logic indicates that the simple
addition of impacts probably underestimated the impact of cancel-
ling both, but the degree of underestimation is unknown.
F-4
-------
4.2 Remaining Major Field Crops (Wheat/ Cotton, Sorghum, Barley,
Oats, Hay)
4.2.1 Wheat, Barley, Oats
There was only one significant future action that affected wheat.
Yield change estimates developed for EPA by the registrants were
used. There was no significant Agency review of these estimates
(Benefits Estimates for Maneb, Pennwalt Corporation, December
1987 & Response of the Rohm and Haas Company to the Special
Review for EBDC Fungicides, Rohm and Haas Company, October 1987).
4.2.2 Cotton
EPA policy actions assumed in this analysis have potentially
significant affects on cotton production. Estimates of impacts
were developed rather rapidly using judgments of EAB staff members
Possible actions are in areas where a number of alternative
controls exist. Therefore, it is likely that the estimates
developed are reasonable.
4.2.3 Sorghum
No efficacy data were available for sorghum. For herbicides it
was assumed that the cost and percent yield changes would be the
same as those for corn since the crops, pesticides, and pest
spectra are similar. This could be a significant limitation
since sorghum tends to be grown in drier and warmer areas than
corn. The actual performance of the herbicides could be different
under these conditions. The impacts of other actions were
developed internally based on judgement. Other pesticides are
of limited importance in the production of sorghum, therefore,
our estimates are probably within reason even though not well
documented.
4.2.4 Hay
Possible actions were very limited. Only a small portion of the
acres planted are impacted (less than one percent).
4.3 Specialty Crops
4.3.1 Peanuts
Most information for impact estimates for alachlor and aldicarb
(groundwater) were available from reports previously cleared by
the program office (see above). We estimated portions of acres
that would be affected based on knowledge of the soils where the
crop is grown. Industry estimates of fungicide cost and yield
impacts were used, although they had not been subject to internal
review. Insecticide cost and yield effects were developed intern-
ally based on information on alternatives and possible target
F-5
-------
pests. Although we feel reasonably comfortable with estimates
for the individual actions/ we feel very uncomfortable with the
simple addition as a means of aggregating yield impacts across
chemicals. This problem, in addition to lack of information on
supply elasticities for peanuts, prevented us from providing a
complete analysis of the impact of EPA actions on peanut growers.
4.3.2 Apples
Cost and yield impact information provided by industry was utilized
for fungicides. Cost information for other pesticides used on
apples was estimated internally based on knowledge of registered
materials and labeled target pests. Yield impacts were estimated
internally based on limited information on yield impacts from
selected pesticides.
4.3.3 Potatoes
Aldicarb (pesticide- in groundwater) information was available
from an existing Agency study. Fungicide information was available
from an industry report submitted to the Agency. Remaining impacts
were estimated internally as they were for apples.
4.3.4 'Green Peas and Tomatoes
Pesticide industry estimates were available for fungicides.
Only limited information (primarily materials registered and
target pests) was available to estimate cost and yield impacts
associated with other future actions. We had some limited
estimates from a contract publication (with no knowledge of how
these estimates were obtained) on most common target pests and
usage of various materials. Yield and cost impacts were estimated
internally with little or no foundation, other than past experience
on larger crops.
4.3.5 Caneberries
Virtually no information was available except for pesticide
registrations and target pests on labels. This was the situation
for most past actions as well as possible future actions. The
following informational reports were used:
Abbreviated Benefit Analysis of Dinoseb (Since the dinoseb
action was still in litigation at the time inputs were
developed for the study, estimates of impacts as developed
for the regulatory action were used for this analysis).
Preliminary Benefit Analysis of Aldicarb
Preliminary Benefit Analysis of Alachlor
Regulatory Impact Analysis: Registration fees under FIFRA
F-6
-------
Regulatory Impact Analysis: Data requirements for Registering
Pesticides
Benefit Estimates for Maneb, Pennwalt Corporation, December
1987
Response of the Rohm and Haas Company to the Special Review
for EBDC Fungicides, Rohm and Haas Company, October 1987.
5.0 Elasticities
Price elasticities used for the major field crops were those
contained within the simulation model (AGSIM). While the estimated
elasticities may be subject to criticism, they were generated in
a consistent manner within the same model. Price elasticities
for the specialty crops were short-run farm level elasticities
and were obtained from whatever reasonable sources were available.
These estimates of supply and demand elasticities may have been
estimated from different data bases using different techniques.
5.1 Apples
Obtained elasticities of supply from a USDA/ERS report "An
Econometric Model of the U.S. Apple Market," June 1985. Elasticity
of demand estimates from K. Huang, USDA/ERS, 1985.
5.2 Caneberries
Estimates of elasticities were not found.
5.3 Peanuts
Discussions with economists familiar with peanut production
(both with USDA and in major peanut production areas) indicated
that there are no reasonably reliable peanut elasticity of supply
estimates available. Elasticities of demand are from K. Huang,
USDA/ERS. However, these are questionable due to the nature of
perceived demand for domestic peanuts produced under quota and
additional peanuts (peanuts for export and oil).
5.4 Peas, Potatoes and Tomatoes
Elasticities of demand were obtained from K. Huang, USDA/ERS,
1985. Elasticities of supply for peas were obtained from Ascari and
Gummings, International Economic Review, 1977. Elasticity of
supply for potatoes was obtained from unpublished work by G.
Zepp, USDA/ERS, 1987. Elasticity of supply for tomatoes was
obtained from Churn and Just, Giannini Monograph, 1978.
F-7
-------
APPENDIX G
Cumulative Probability Cost Distribution
By
Terry Dinan I/
I/ Office of Policy Analysis, U.S. Environmental Protection Agency
-------
Appendix G
Cumulative Probability Cost Distribution
Since we are simultaneously examining the impact of several EPA
policies, a fundamental issue that had to be determined was:
how do we define an "impacted" farmer? For example, Illinois
corn soybean farmers may be affected by the cancellation of several
different pesticides, may incur insurance costs if they have an
underground storage tank that meets certain criteria, and may
incur an expense to rebuild their tractor engine if all lead is
banned from gasoline and they have a leaded gasoline tractor.
How many of these potential costs do we assume that the "impacted"
farmer incurs? For each producer we examine two alternative sets
of impacts:
* A Maximum Impact Case; In this case it is assumed that
the producer is affected by every regulation that may
possibly affect a producer of that type.
* An Average Impact Case; In this case it is assumed
that the producer experiences the average impact of
producers of that type - e.g., if 10% of all producers
of a given type experienced a cost of $1,000, we would
use a cost of $100 ($1,000 x 0.10) for the average
impact case.
Examining these two cases, however, only provides two snapshots
of possible impacts without providing the full picture of how
cost and yield impacts are likely to be distributed across
producers. To provide more insight into the likely distribution
of these initial cost and yield impacts, we constructed a cumula-
tive probability cost curve for each representative farm in
average financial position. The following example demonstrates
what these cumulative probability cost curves reveal.
Suppose a given farmer may be affected by three possible regula-
tions, each having the following associated cost and probability
of affecting a given producer:
Probability Probability
Regulation Cost of Impact of No Impact
A $100 .30 .70
B $200 .20 .80
C $300 .10 .90
Provided the probabilities of incurring the costs of the three
regulations are independent, the possible set of outcomes and
associated costs and probabilities may be defined as:
G-l
-------
Regulations
Affected by;
A
B
c
NONE
A,B
B,C
A,C
ALL
Cost
$100
$200
$300
$0
$300
$500
$400
$600
Probability I/
.216
.126
.056
.504
.054
.014
.024
.006
I/ Note the probability of being impacted by Regulation A =
P(A) x P(NB) x P(NC), where P(A) = the probability of being
affected by regulation A, and P(NB), P(NC) = the probability
of not being affected by B and C, respectively.
By ranking these possible outcomes in order of cost, and adding
up the associated probabilities, we can arrive at the following
cumulative probabilities:
Regulations
Affected by;
NONE
A
B
C
A,B
A,C
B,C
ALL
Cumulative
Probability
.504
.720
.846
.902
.956
.980
.994
1.00
Then, plotting the cost on the x-axis and the cumulative probabil-
ity on the y-axis, we can use this information to generate the
following cumulative probability cost curve:
Cumulative Probability Cost Curve
0.9
0.8
0.7
0.6
0.5
0.4
OJ
0.2
0.1
0
100 200 300 400 SOO GOO
Cost
G-2
-------
This cost curve indicates the probability of incurring a cost less
than or equal to a given level. For example, it indicates that
any given farmer has a probability of .846 of incurring a cost
that is less than or equal to $200.
To shed insight into the probability that the farms examined in
this report would actually incur any given level of cost, we
generated a cumulative probability cost curve for each of the
representative farms in average financial position. In the above
example, all of the costs were assumed to be independent. In
reality, however, this may not be the case. For example, farmers
who use a certain type of pesticide on their corn may very likely
be using the same pesticide on their soybeans, if the pesticide
is used on a certain pest that is found on both corn and soybeans.
In generating the cumulative probability cost curve for each
representative farm, we tried to account for the correlation
among different costs. The assumptions we used for each represen-
tative farm are outlined below:
Illinois corn soybean farm assumptions:
1. If a farmer is using any chemical, then he incurs Farm
Worker Safety Costs.
2. If a farmer is using alachlor on his soybeans, then he
is using alachlor on his corn.
3. If a farmer is using a corn rootworm insecticide on his
corn, then he is using a triazine on his corn.
4. If a farmer is using alachlor on his corn, then he is
using a triazine on his corn.
Mississippi cotton soybean farm assumptions:
1. If a farmer is using any chemical, then he incurs Farm
Worker Safety Costs.
2. If a farmer is using dinoseb on his soybeans, then he
is using dinoseb on his corn.
Kansas wheat cattle farm assumptions:
1. If a farmer is using any chemical, then he incurs Farm
Worker Safety Costs.
2. If a farmer is using alachlor on his soybeans, then he
is using alachlor on his corn.
3. If a farmer is using a triazine on his corn, then he is
using a triazine on his sorghum.
4. If a farmer is using alachlor on his corn, then he is
using a triazine on his corn.
Incorporating these assumptions into the method described in the
above example, we generated a cumulative probability cost curve
for each representative farm in each scenario (Figures G-l through
G-5). Any given point on the curve may be interpreted as the
G-3
-------
ILLINOIS CORN SOYBEAN FARM: SCENARIO 1
5
3
AVERAGE FINANCIAL POSITION
9.8 10 12
(Thousands)
DISCOUNTED PRESENT COST (1987-1998)
1 4
18
Maximum
Impact
Case
Figure G-la.
Scenario 1, cumulative probability cost curve for the representative
Illinois corn soybean farm in average financial condition
2
3
ILLINOIS CORN SOYBEAN FARM: SCENARIO 2
0.1 -
AVERAGE FINANCIAL POSITION
20 40
(Thousands)
DISCOUNTED PRESENT COST (1987-1998)
Av«ragt
Impact
Case
Maxim*
Impact
Case
Figure G-lb. Scenario 2, cumulative probability cost curve for the representative
Illinois corn soybean farm in average financial condition
G-4
-------
I
a
£
ILLINOIS CORN SOYBEAN FARM: SCENARIO 3
0.9 -
0.1 -
AVERAGE FINANCIAL POSITION
20 4O
(Thousomla)
DISCOUNTED PRESENT COST (1987-1998)
Avcraqt
[•Met
Cist
SO
[moact
Case
:igure G-2a. Scenario 3, cumulative probability cost curve for the representative
Illinois corn soybean farm in average financial condition
2
o
£
MS COTTON SOYBEAN FARM: SCENARIO 1
AVERAGE FINANCIAL POSITION
1
0.9
0.3
0.7
0.8
0.* -
0.3 -
02 -
0.1
20
60
4O
(Thousonds)
DISCOUNTED PRESENT COST (1987-1996)
Average
[•pact
Cast
80
Maximum
Imoact
Case
Figure G-2b. Scenario 1, cumulative probability cost curve for the representative
Mississippi cotton soybean farm in average financial condition
G-5
-------
MS COTTON SOYBEAN FARM: SCENARIO 2
AVERAGE FINANCIAL CONOfTlCN
*O
80
80
(Thousands)
DISCOUNTED PRESENT COST (1987-1998)
I«W«ct
100
120
.Maximum
Impact
Case
Figure G-3a. Scenario 2, cumulative probability cost curve for the representative
Mississippi cotton soybean farm in average financial condition
MS COTTON SOYBEAN FARM: SCENARIO 3
i -i
AVERAGE FINANCIAL POSTHON
2O
«O
80
Average
I"Pact
Cast
80
(Thousands)
DISCOUNTED PRESENT COST (1987-1998)
100
Maximu
Case
Figure G-3b.
Scenario 3, cumulative probability cost curve for the representative
Mississippi cotton soybean farm in average financial condition
G-6
-------
KANbAb WHt/J UAI I i_L fAKiV;
AVERAGE FINANCIAL POSITION
0
I
Q.
\
I
1 -
0.9 -
o.a -
0.7 -
0.9 -
0.3 -
0.* -
0.3 -
0.2 -
0.1 -
0 -
(
a — mxoBfr^*""^
J
j
f
go0
r*
i
i
i
i
i
i
1 IT 1 1 1 1 1 1 1 1 1 1 1 r i i i
J 2 * 8 fl 10 12 »* '8 1
(Thousands)
DISCOUNTED PRESENT COST (1987-1998)
Avvrtq* Max in
[npict ["KM!
Cue Case
a
turn
t
Figure G-Aa.
Scenario 1, cumulative probability cost curve for the representative
Kansas wheat cattle farm in average financial condition
KANSAS WHEAT CATTLE FARM: SCENARIO 2
AVERAGC FINANCIAL POSITION
0.9 -
o.a -
0.7 -
0.9 -
0.3 -
0.*
0.3
0.2
0.1
3
Average
Immct
Case
—1-
10
-r
—r— —i—
20 30
(Thousands)
DISCOUNTED PRESENT COST (1987-1998)
Maximum
Imoact
Case
Figure G-4b.
Scenario 2, cumulative probability cost curve for the representative
Kansas wheat cattle farm in average financial condition
G-7
-------
KANSAS WHEAT CATTLE FARM: SCENARIO 3
>
5
£
o
Q.
s
2
U
Average
Impact
Cast
AVERAGE FINANCIAL POSITION
10
20 30
(Thousands)
DISCOUNTED PRESENT COST (19B7-1996)
Maximum
Impact
Case
Figure G-5a.
Scenario 3, cumulative probability cost curve for the representative
Kansas wheat cattle farm in average financial condition
G-8
-------
probability that the representative farm will incur a cost equal
to or less than a given level. For example, the curve in Figure
G-la indicates that the representative Illinois corn soybean farm
in Scenario 1 has a .50 probability of incurring a discounted
present value of cost and yield impacts (1987-1996) of less than
or equal to $2", 000. The discounted present value of cost and
yield impacts corresponding to the average and maximum impact
cases are indicated on each curve.
If all Illinois corn soybean farms had the same number of acres
of each crop as the representative farm, Figure G-la could be
interpreted as the percent of farms likely to incur cost and
yield impacts less than or equal to a given level. Since farms
will vary in the number of crop acres that they plant, their
present discounted value of impacts under any particular combina-
tion of regulations will vary from the representative farm.
(Recall that the representative farm does not truly represent all
farms but is only a composite of farms of a given type.) These
curves, therefore, are only meant to provide some insight into
the distribution of cost and yield impacts for farms of a given
type, but do not represent accurate cost and yield impacts for
any particular farm (other than the average farm), or the true
distribution of impacts across farms.
G-9
-------
APPENDIX H
Recommendations for Acquiring Better Pesticide Usage Data
By
Susan Slotnick I/
I/ Office of Standards and Regulations, U.S. Environmental
Protection Agency
-------
Appendix H
Recommendations for Acquiring Better Pesticide Usage Data
In this agricultural sector study, the lack of current and reliable
pesticide usage data has limited the ability to accurately assess
the economic impact of EPA actions, particularly on the specialty
crops. The quality of the usage data used in the report is
described in Appendix F. To summarize, data for the major crops
were usually adequate only at the regional level. For small-area
crops, the data were old and/or of unknown statistical validity.
For no crop was information available nationwide at the county
level which is the minimum level of disaggregation needed for
measuring the impact of ground water regulatory actions. The
gaps identified in Appendix F could affect the study results
because the measurement of economic impacts of EPA actions depends
on the cost and yield effects of pesticide cancellation which in
turn depend on usage data.
The agricultural sector study is only one example of the many EPA
analyses that depend on basic pesticide data for accurate estima-
tion of economic and other effects of pesticide regulation.
Because this study is an excellent illustration of the difficulty
the data limitations present, it is an opportunity to discuss
those limitations, their consequences for economic and risk
analyses of pesticide use, and what can be done to improve the
situation.
As seen in the agricultural sector study, two types of basic
pesticide data are fundamental to assessing a pesticide's economic
importance: performance and usage. A current project in the
Office of Pesticide Programs directly addresses the incompleteness
of the performance data by strengthening data requirements placed
on pesticide manufacturers. For that reason, the discussion here
is limited to usage data, defined roughly as the amount a par-
ticular pesticide and its alternatives are used on a crop, how
many acres are treated with each pesticide, in which locations, at
what rate, and by what methods. For the sake of brevity, the
focus is on agricultural pesticide use, although data problems
exist with nonagricultural use as well.
1.0 Why Pesticide Usage Data are Important
The agricultural sector study is just one of several recent
special analyses relying on pesticide usage data. Some of the
special studies could be of far-reaching importance for future
pesticide use, for example, preparation for the Agency's En-
dangered Species Program and targeting of water wells for the
national groundwater monitoring program. For risk/benefit analyses
on individual pesticides and for other regular pesticide assess-
ments (e.g., exemptions for local use), usage data and performance
data form the foundation upon which scientists and economists
build their quantitative estimates of a pesticide's importance.
H-l
-------
Without complete information, often the case with small area
crops, analysts must rely on educated guesses, adding uncertainty
to their final conclusions. In the recent case of the herbicide
dinoseb, usage information on alternatives was not readily
available and analysts had inadequate time to gather it. This
lack of data contributed to a successful legal challenge by
growers of some small crops, causing EPA to exempt those crops
from the suspension decision already made. Furthermore, usage
data are an integral part of exposure assessments, which in turn
play a key role in deciding whether a pesticide is placed in
Special Review.
2.0 Current State of Usage Data
The agricultural pesticide usage data currently available are
very uneven in quality and coverage. For the major crops such as
corn, soybeans, cotton, and wheat, current survey data are
available from USDA and private sources and are likely to be
collected periodically in the foreseeable future. Information on
major crops falls short of OPP's needs because it often excludes
minor producing areas and are often not disaggregated to a small
enough geographic level. Considerably greater problems occur
with small-area crops, for example, there has been .no publicly-
available survey of pesticide use on citrus since 1977. For the
specialty crops studied in this report as well as the whole
spectrum of fruits, vegetables, and other crops, usage data are
rarely what they need to be: current, reliable, disaggregated at
least to the state level, and publicly available.
3.0 Recommendations for Acquiring Better Data
The Benefits and Use Division (BUD) of the Office of Pesticide
Programs has made a concerted effort to upgrade its usage data,
but is often met with budgetary constraints. BUD recently
estimated that it would cost $3 million to acquire adequate
survey usage data on crops and nonagricultural sites of importance
to OPP. That expenditure would be needed every three or five years.
However, the Office of Pesticide Programs is not the only organi-
zation needing pesticide usage data, and the list is growing
because of heightened concern about pesticide health and environ-
mental effects, for example groundwater contamination. Other
organizations which recently used pesticide usage data are:
Department of Agriculture,
EPA Office of Drinking Water, Non-Point Source Branch,
EPA Office of Ground Water Protection,
individual registrants,
Food and Drug Administration,
National Agricultural Chemicals Association,
H-2
-------
* state environmental, water quality, and public health
programs, and
* U.S. Geological Service, Water Resources Division.
For some of the options that follow, a cost-sharing arrangement
between EPA an~d~other interested organizations could make the
data acquisition far more affordable.
Below are possible options for generating better pesticide usage
data. Each has different costs and benefits.
1. Conduct a set of jointly-funded periodic surveys of
pesticide users
Each set would cover certain sites, such as major
crops, small area crops, crops in certain regions,
pesticide-intensive crops in areas of groundwater
vulnerability, or nonagricultural sites. A different
group of sponsoring organizations would fund each set.
Fees would be charged to non-sponsoring users.
2. Set up cost-sharing between EPA and states to conduct
surveys
This is a more limited version of option #1. In order
to receive EPA funds, states would have to design the
surveys to meet certain specifications so the data.
would fit EPA's needs. This might be the most efficient
approach for small crops.
3. "Socialize" private data collection services
These services currently poll farmers nationwide on
pesticide usage. EPA and other interested parties
could contract to completely fund the data collection,
in order to be able to control the survey methods and
site coverage, and to ensure the data is public.
4. Attach questions to existing USDA surveys currently
used for other purposes
This is already being done to a limited extent; the
new questions would be much more detailed.
5. Attach questions to the U.S. Census of Agriculture
The Census currently asks farmers questions on all
crops as well as usage of pesticide in broad categories.
To be useful for most EPA analyses, additional questions
would be added that are detailed at the active ingredient
level.
H-3
-------
6. Require data from registrants
Registrants are required to generate pesticide toxicity
and performance data to support pesticide registrations.
If usage data were also required, the cost to the
government would be lower than with other options,
though there could be problems with confidentiality.
7. A combination of the above
Existing USDA surveys cover only a subset of the crops
relevant to EPA. Pesticide usage questions could be
attached to those surveys while data on remaining
crops could be collected jointly by a consortium as in
#1 and #3.
An interagency committee composed of EPA, USDA, FDA, and DOI,
meets on occasion to share pesticide usage data. To date, there
has been no joint funding of data. Working through the committee,
the OPP Benefits and Use Division and the OPPE Office of Policy
Analysis have begun an initiative to acquire better data.
4.0 Summary
There is a clear need for more detailed, precise estimates of
pesticide usage, both agricultural and non-agricultural. Recent
renewed interest in pesticide-related environmental and health
problems has increased the number of organizations needing such
information. Because there are many hundreds of different
pesticidal active ingredients and hundreds of different crops and
nonagricultural sites across the country, acquiring high quality
information on a regular basis is expensive. Yet without it, the
accuracy of economic valuation of pesticides is uncertain. If
such accuracy is deemed important enough, some increased effort
will be needed to acquire the necessary data.
There are several ways to generate better usage data. Detailed
questions could be attached to existing surveys designed for other
purposes, EPA could require the data from registrants, or a consor-
tium of interested private, federal, and state organizations
could be formed to share the costs of new surveys. Since there
is a wide variety of use sites, a different arrangement might be
made for different types of sites.
Each approach would differ from a cost-benefit standpoint. To
the extent EPA can pool resources with other users of pesticide
data, costs can be lowered. The benefit of better data will be
greater efficiency in the assessments of pesticide use, a higher
quality of analysis, and subsequently, more informed decisions on
pesticide regulation.
H-4
-------