AGRICULTURAL OUTLOOK                                        August 23, 2000
September 2000, ERS-AO-274
               Approved by the World Agricultural Outlook Board
---------------------------------------------------------------------------
AGRICULTURAL OUTLOOK is published ten times a year by the Economic Research
Service, U.S. Department of Agriculture, Washington, DC 20036-5831.  Please
note that this release contains only the text of AGRICULTURAL OUTLOOK --
tables and graphics are not included. 

Subscriptions to the printed version of this report are available from 
the ERS-NASS order desk.  Call toll-free, 1-800-999-6779 and ask for stock 
# SUB-AGO 4001, $65/year.  ERS-NASS accepts MasterCard and Visa.
---------------------------------------------------------------------------
CONTENTS

IN THIS ISSUE 

BRIEFS

Field Crops: Harvested Durum Area to Be Largest Since 1982

Livestock: Hog Producers Plan Modest Expansion

Technology: On the Upswing--Online Buying & Selling of Crop Inputs,
Livestock

COMMODITY SPOTLIGHT
Prices for Bumper U.S. Soybean Crop Hinge on China's Imports

POLICY
Weak Prices Test U.S. Sugar Policy

RESOURCES & ENVIRONMENT

Confined Animal Production Poses Manure Management Problems

Environmental Regulation & Location of Hog Production

SPECIAL ARTICLE
Transportation Bottlenecks Shape U.S.-Mexico Food & Agricultural Trade


IN THIS ISSUE

On the Upswing: Online Buying & Selling by Farmers

Increasing numbers of farmers and ranchers are doing business over the
Internet.  More than 600,000 U.S. farms and ranches accessed the Internet
in 1999, with 15 percent conducting e-commerce transactions, based on new
data from USDA. Of these farms, over 40 percent reported purchasing crop
inputs online in 1999, about one-third reported purchasing livestock
inputs, and a quarter reported selling livestock. Mitch Morehart (202)
694-5581; morehart@ers.usda.gov

Hog Producers Plan Modest Expansion

Hog producers indicate intentions to begin rebuilding breeding herds,
according to USDA's June Hogs and Pigs report. But before expansion takes
hold, pork production will fall 2 percent in second-half 2000 from a year
earlier. With lower pork production in the near term and prospects for only
modest expansion next year, hog prices are expected up from last year's $34
per cwt to average in the mid-$40's in 2000 and 2001. Leland Southard (202)
694-5187; southard@ers.usda.gov

U.S. Soybean Stocks to Build 

USDA forecasts a record U.S. soybean crop in 2000--nearly 3 billion
bushels--based on record-high acreage and relatively high yields.  Despite
the liberal supply expansion, U.S. soybean exports in 2000/01 are projected
to rise only slightly, primarily because of larger soybean harvests in
China (a major importer) and in Brazil and Argentina (major export
competitors), as well as shrinking imports by the European Union.  With
U.S. soybean demand expected to lag supply growth, ending stocks in 2000/01
are projected to swell, and the U.S. farm price of soybeans is expected to
average $3.90-$4.80 per bushel in 2000/01, a drop from $4.65 in 1999/2000.
Thus marketing loan benefits will continue to be important for soybean
producers. Mark Ash (202) 694-5289; mash@ers.usda.gov

U.S. Sugar Policy: Sticky Issues

Rising domestic sugar production as well as prospects for higher imports
are testing the government's ability to prevent sugar prices from dipping
below support levels.  In June, USDA entered the sugar market for the first
time since 1986, purchasing 132,000 tons of refined sugar at a cost of $54
million.  With this move, USDA projected savings of as much as $6 million
in administrative costs that the government might otherwise incur from
expected sugar program loan forfeitures later in the fiscal year. With
domestic sugar production plus imports exceeding domestic consumption in
the foreseeable future, it will be difficult to keep prices above support
levels without reducing output through a domestic supply control program or
incurring large Treasury costs. On August 17, USDA announced a 2-week
signup period for the Sugar Payment-In-Kind (PIK) Program, which offers
sugar beet producers the option of diverting a portion of this year's crop
from production in exchange for government-held sugar.  Stephen Haley (202)
694-5247;  shaley@ers.usda.gov

U.S.-Mexico Faces Border Bottlenecks 

The high volume of traffic at U.S.-Mexico border crossings reflects the
dynamic and fast-growing trade relationship between the U.S. and Mexico.
But rising agricultural and other trade between the U.S. and Mexico has led
to congestion and, in some instances, to costly delays at the border. A
major source of delay is a multi-step process for transferring cargo across
the border, because long-haul trucks destined for the interior of the U.S.
or Mexico are not allowed to travel beyond a border zone. A broad spectrum
of incremental measures--e.g., enhancement of physical
facilities/infrastructure at crossing points and use of new technologies
for checking cargo--is advancing the capacity and efficiency of the
increasingly integrated U.S.-Mexico transportation system. Freer truck
access and the upgrading of Mexico's rail system are key factors in future
growth in U.S.-Mexico food and agricultural trade.    William T. Coyle
(202) 694-5216; wcoyle@ers.usda.gov

Confined Animal Production Poses Manure Management Problems

Livestock and poultry manure applied to farmland provide a valuable source
of organic nutrients, but nitrogen and phosphorus from manure in excess of
the farm's crop requirements can compromise water quality.  Many confined
animal operations are unable to utilize all manure nutrients produced on
the farm--i.e., apply the animal waste to crops on land under their
control.  For example, some locations in the Southeast, where poultry
production is a dominant ag industry, have high levels of excess nitrogen
because poultry manure has a high nitrogen content, and because land
available for spreading is limited.

For areas with excess manure, initiatives to encourage land application on
other farms or to provide incentives for alternative manure treatment
strategies may be necessary.  USDA's Environmental Quality Incentives
Program (EQIP) provides technical, educational, and financial assistance to
farmers and ranchers for adopting practices that protect or enhance
environmental quality.  Noel Gollehon (202) 694-5539; gollehon@ers.usda.gov

Environmental Regulation & Location of Hog Production 

Increasing concentration of hog production and manure waste in certain
geographic areas of the U.S. has heightened interest in the potential links
between stringency of environmental regulation and location of animal
production.  Policies regulating environmental pollution from confined
animal farming may vary geographically, partly because Federal water policy
laws allow states to have authority and flexibility to design and implement
their own environmental laws. 

Costs associated with environmental regulation compliance may be a
consideration in choosing a business location.  Producers may respond to
existing or impending costs of regulation by exiting the industry or
changing the scale and/or location of production.  Hog production has
expanded in recent years in areas in the South and in nontraditional areas
of the West, prompting speculation that large operations moved to those
areas because of possibly less stringent environmental regulations. John
Sullivan (202) 694-5493; johnp@ers.usda.gov 


BRIEFS

Field Crops
Harvested Durum Area To Be Largest Since 1982

The harvested area of durum wheat-used mainly for pasta production-is
forecast at 4 million acres in 2000, up 12 percent from 1999 and the
largest since 1982. Plantings are unchanged compared with a year earlier,
but abandonment rates should return to normal after increasing sharply last
year due to late maturation of the crop and unusually cold and wet
conditions at harvest.

In North Dakota, the leading durum state with almost 82 percent of
harvested area, planted area is 12 percent above farmer intentions
published in USDA's March 31 Prospective Plantings report, primarily
reflecting favorable weather at planting time and relatively strong futures
prices for durum. Acres planted to all major field crops in North Dakota
were 2.4 million above planting intentions (including a 900,000 acre gain
for "other" spring wheat). Durum acreage intended for harvest is forecast
at 3.25 million acres, 8 percent above last year. 

In Montana, the second-ranked durum producing state, producers followed
through on their large 2000 planting intentions by seeding 53 percent more
acres to durum than in 1999. The forecast harvested area in Montana is the
largest since 1957 and the third largest on record. Drought conditions in
Montana led producers to plant 1.2 million fewer acres than intended to
major crops, primarily other spring wheat and barley. Most of this land
will be in summer fallow this year. The more favorable price outlook prior
to planting for durum relative to other spring wheat likely convinced
Montana producers to stick with earlier intentions. Soil moisture
conditions have been generally favorable in northeast Montana where durum
is grown.

The hike in harvested U.S. durum area, combined with a larger harvested
area of other spring wheat (up 2 percent from last year), will reverse the
downward trend in harvested wheat area that began in 1997. Durum and other
spring wheat planting progressed much faster than normal in the northern
Plains in 2000 because of favorable weather during planting.

USDA's August 1 forecast indicates that U.S. farmers will harvest 115
million bushels of durum in 2000, up 16 million from the weather-plagued
1999 crop. Higher forecast yield (28.9 bushels per harvested acre vs. 27.7
bushels last year) and greater harvested area will push up production this
year (however, the yield forecast for North Dakota is down 4 bushels per
acre since early July). With beginning stocks on June 1 estimated at 50
million bushels and imports (grain and products) projected at 30 million,
durum supply is forecast at 195 million bushels in 2000/01, up 13 million
from last year.

While supplies expand, total use is projected to decline 2 million bushels
to 131 million in 2000/01. This includes a projected 4-million-bushel
decline in exports. Ending stocks are forecast up 14 million bushels to 64
million, the highest since 1987/88. 

Larger supplies and weak export demand will keep downward pressure on farm
prices for durum in 2000/01. With few alternative uses, large supplies and
a huge crop expected in Canada will limit the price premium producers
usually receive for durum relative to other spring wheat. 

In 1999/2000, large supplies and significant quality problems drove the
average farm price for durum to a 9-year low and 14 cents per bushel below
the average for other spring wheat. The last time the farm price for durum
was at a discount to other spring wheat was in the 1992/93 marketing year.
In contrast, high-quality U.S. No. 1 hard amber durum at the Minneapolis
cash market in 1999/2000 commanded an average premium of 57 cents per
bushel over U.S. No. 1 dark northern spring wheat (14 percent protein). The
cash market premium for durum is expected to narrow in 2000/01 because
large durum crops are forecast for the U.S., Canada, and the European Union
(a major importer and exporter in most years). Canada's durum production is
forecast up 41 percent from last year. The EU crop is forecast up 21
percent.

Persistent drought is cutting the 2000 durum crop in the key North African
market--Algeria, Morocco, Tunisia, and Libya--which usually accounts for
over 40 percent of world imports. In the North African region, durum is
consumed primarily as couscous, a traditional durum-semolina-based dish.
U.S. exports are not expected to benefit significantly from the region's
production shortfall since Canada and the EU are primary suppliers to the
region. Canada is expected to be a major competitor in other key import
markets, including the EU, Japan and Venezuela.

Mack N. Leath (202) 694-5302
mleath@ers.usda.gov


Livestock
Hog Producers Plan Modest Expansion

Viewing prospects for favorable returns, hog producers indicate they intend
to begin rebuilding breeding herds, according to the USDA's June Hogs and
Pigs report. After signaling a 2 percent decline for June-August 2000
compared with a year ago, producers have reversed direction and are
planning to have 1 percent more sows farrow (produce litters) in
September-November than a year ago. Hog producers' returns moved well above
the economic breakeven point (receipts less costs) in first-half 2000, as
prices rallied into the low $50's per cwt and feed costs remained the
lowest in several years.

Producers have been reducing herds due to poor returns in late 1998 and
1999. The June 1 inventory of all hogs and pigs totaled 59.4 million head,
and the 6.23 million breeding hogs in that total was a 4-percent drop from
last year.

The additional farrowings, along with a rising number of pigs per litter,
would increase the pig crop 1-2 percent for September 2000-February 2001
over the same period a year earlier. Pork production, in turn, is forecast
up about 1 percent in calendar 2001, with the first quarterly
year-over-year rise in March-May 2001 since third-quarter 1999. The hog
production cycle is 10 months-4 from conception to birth, 1 from birth to
weaning, and 5 from weaning until slaughter.

But before expansion takes hold, pork production will fall 2 percent in
second-half 2000 from a year earlier, based on a 4-percent decline in hog
slaughter and a continued rise in average dressed weight. The forecast drop
in slaughter results from a 2-percent-smaller pig crop for December
1999-May 2000 compared with a year earlier and slightly higher gilt
(female) retention for the breeding herd. Farrowings during the period were
down from a year earlier, but pigs per litter were up. (The move in the
industry to larger specialized operations has been partly responsible for
the larger litters. For example, farms with 5,000 or more hogs averaged 9
pigs per litter in March-May 2000, compared with 7.8 for farms with 1-99
head.)

Spring's seasonal decline in slaughter rates, the public's taste for bacon,
and the rising price of beef ratcheted up hog prices into the low $50's per
cwt in late spring and early summer 2000. Demand for bacon continued
strong, especially for fast-food industry use in sandwiches. As beef prices
rose, food retailers featured the more price-attractive pork loins,
strengthening loin prices.

With lower pork production in the near term, prospects for only modest
expansion next year, and ongoing healthy retail demand, the hog market
promises to be relatively strong into 2001. While there is some uncertainty
about whether this year's strong demand will persist, hog prices are
expected to average in the mid-$40s per cwt in 2000. In late fall 2000,
when slaughter reaches a seasonal peak, hog prices could average around $40
per cwt. Given expected expansion in production in 2001, slaughter capacity
could be strained late in the year, putting downward pressure on hog
prices. Hog prices are expected to average $42-46 per cwt in 2001.

Retail price hikes usually lag behind increases in farm prices; with rising
hog prices in first-half 2000, the farm-to-retail price spread can be
expected to narrow, then widen. By second-quarter 2000, the price spread
had shrunk to $1.68 per pound after averaging $1.81 in 1998 and 1999.
Following the 2-year price decline, tighter pork supplies will probably
push retail pork prices up 5-6 percent this year. In 2001, retail prices
may increase about 1 percent, widening the spread again.

Producers will likely stay in a mood to expand, with hog prices much higher
than a few years ago and lower feed prices anticipated over the next 18
months (record corn and soybean crops are projected for 2000). As producers
rebuild their equity positions and as large facilities take extra time to
get up to speed, however, growth will probably be gradual.

For further information, contact: Leland Southard, coordinator; Ron
Gustafson, cattle; Leland Southard, hogs; Mildred Haley, world pork; Dale
Leuck, world beef; David Harvey, poultry. All are at (202) 694-5180.


Technology
On the Upswing: Online Buying & Selling Of Crop Inputs, Livestock

As online possibilities for agricultural commerce expand, more and more
farmers and ranchers are doing business over the Internet. The share of
farms with Internet access more than doubled to 29 percent between 1997 and
1999, according to USDA's National Agricultural Statistics Service. More
than 600,000 U.S. farms and ranches accessed the Internet in 1999, with 15
percent conducting e-commerce transactions, based on USDA's Agricultural
Resource Management Study. This means that roughly 1 of every 25 farms and
ranches in the country bought or sold agricultural products on the Net. 

Many agricultural e-commerce ventures were just getting started in 1999,
and farmers have adopted e-commerce at about the same rate at which they
opted to try biotech crops when they were first introduced.

Farms that bought or sold online in 1999 were more likely to be younger,
more educated operators than the national average. Almost three-quarters of
active e-commerce users were between 35 and 54 (over 70 percent), and just
over a third had completed college or graduate school. Only 46 percent in
the farm population as a whole are between 35 and 54 years old, and 21
percent have completed college or graduate school. Higher rates of adoption
among these groups are to be expected, because both age and education level
are often strong predictors of willingness to adopt a new technology.

More than half (55 percent) of 1999 agricultural e-commerce came from the
Heartland, Prairie Gateway, and Fruitful Rim regions, which together
account for 47 percent of total farms. In addition, many farms that bought
or sold online in 1999 were small (gross sales below $250,000), although
small farms accounted for only 60 percent of online agricultural business
compared with 94 percent of all U.S. farms.

Over 42 percent of online market activity in 1999 involved purchasing crop
inputs, and online buying was related to farm size. A higher share of
larger small farms (sales of $100,000 or more) bought crop inputs over the
Internet than smaller farms (17 vs. 4 percent). Farms with operators going
online to buy crop inputs accounted for almost one-tenth of U.S. corn and
soybean acreage, and the same share of total seed, fertilizer, and chemical
expenses. Crop-input purchases for these farms (all transaction methods)
totaled $2.2 billion in 1999. In contrast with Internet purchases of crop
inputs, farm size showed no relation to transactions for purchasing
livestock inputs and selling livestock (58 percent of online market
activity).

Agricultural commerce sites on the Internet are betting that farmers will
do business online in increasing numbers to explore the potential benefits
of e-commerce, ranging from cost savings to better and more timely
agricultural information. Assuming reliable Internet service, the Net could
also provide farmers and ranchers with the opportunity to buy and sell
commodities efficiently and conveniently.  

Mitch Morehart (202) 694-5581 and Jeffrey Hopkins (202) 694-5584
morehart@ers.usda.gov
jhopkins@ers.usda.gov



COMMODITY SPOTLIGHT
Prices for Bumper U.S. Soybean Crop Hinge on China's Imports

Early this year, very dry conditions that began in the summer of 1999
prevailed over much of the nation's major soybean producing region. Soybean
prices rallied in response to expectations of lower production in 2000. In
late April and early May, dry soils allowed a rapid planting pace for
spring crops, and partly in response to the price rise, U.S. farmers
planted 74.5 million acres of soybeans for 2000, a 1-percent rise from last
year's record. Nearly all of this year's increase was in the Northern
Plains and Lake States, where crop rotations are still adjusting to
incorporate more soybeans.

Following planting, abundant summer rains fell (though not in the
Southeast), greatly increasing soil moisture. Once the threat of widespread
drought disappeared, soybean prices fell sharply in anticipation of large
production. Based on the record acreage and a national average yield
forecast of 40.7 bushels per acre, USDA forecasts a record U.S. soybean
crop in 2000 of 2.99 billion bushels. This year's output will likely exceed
the 1998 record by nearly 250 million bushels. With beginning stocks large,
U.S. soybean supplies are expected to rise 9 percent. 

Despite this liberal supply expansion, U.S. soybean exports in 2000/01 are
projected to rise only slightly, to 1,010 million bushels from last
season's record 975 million. Primary reasons for the modest export growth
are larger soybean harvests in China and South America and shrinking
imports by the European Union (EU). In addition, the strength of the U.S.
dollar compared with currencies of major export competitors and import
buyers continues to curtail U.S. foreign trade.

With U.S. soybean demand expected to lag supply growth, ending stocks in
2000/01 are projected to swell to 465 million bushels from 280 million in
1999/2000, keeping downward pressure on soybean and soybean product prices.
The 2000/01 soybean farm price is expected to average $3.90-$4.80 per
bushel, down from the 1999/2000 average of $4.65, and to remain well below
the loan rate ($5.26 per bushel) for the third consecutive year. Thus, loan
benefits will continue to be important for soybean producers. Large
supplies of corn and other feeds will also put downward pressure on soybean
prices.

World oilseed production in 2000/01 is anticipated up 3.3 percent to 308
million metric tons. Virtually all of the increase is due to an 8-percent
rise in expected soybean production. The U.S. and China account for
three-fourths of the forecasted soybean output gain, with comparatively
modest changes for most other countries. Growth in output by China, a major
importer, is expected to contribute significantly to a projected fall in
global soybean exports, from 46.3 million tons in 1999/2000 to 45.6
million. 

China & European Union
To Curb Imports

China's domestic prices for soybeans and products, unlike those of most
nations, remain firm due to strong domestic demand and restrictive import
policies. Vegetable oil prices within China, for example, are typically
more than double world prices, because strict import quotas on vegetable
oils maintain this price wedge. For oilseeds, however, a relative absence
of import barriers provides a substantial advantage to domestic crushers in
producing the highly valued vegetable oil. With recent changes in China's
import policies for the soybean complex, including the re-imposition of a
value-added tax on soybean meal in 1998 (AO September 1999), the country
has successfully shifted toward greater reliance on domestic
oilseed-crushing capacity vs. imports of protein meal and vegetable oil. 

China's soybean imports soared 31 percent in 1998/99 and more than doubled
to a record 9 million tons in 1999/2000, accounting for nearly 80 percent
of world expansion in soybean trade last season. Soybean exports from the
U.S. to China nearly tripled in 1999/2000, up from 9 percent of total U.S.
exports to 17 percent. Conversely, China's imports of soybean meal and
soybean oil plunged in 1999/2000 from 2 years earlier. U.S. shipments of
soybean meal and soybean oil to China in 1999/2000 fell 100 percent and 80
percent, respectively.

The same factors that drove the surge in China's 1999/2000 soybean imports
also encouraged its own farmers to sow more soybeans this year instead of
corn. China's soybean area in 2000 is estimated up 10 percent, which would
push the projected crop to a relatively large 15 million tons (a serious
drought is expected to reduce yields in northeastern China). As a result,
China's soybean imports are expected to fall by one-fifth to 7.25 million
tons in 2000/01. The much larger domestic oilseed harvest would also mean
that growth in Chinese imports of protein meal and vegetable oil will be
modest next year, even if import barriers are lowered.

European Union soybean imports also are expected to decline in 2000/01,
despite smaller domestic oilseed harvests and greater availability of
global soybean supplies. The EU's recently implemented reforms (Agenda
2000) of the Common Agricultural Policy will allow minimum internal grain
prices to fall 15 percent over the next 2 years. Since WTO commitments cap
subsidized EU grain exports, much of the surplus will be fed to domestic
livestock, thereby reducing EU soybean meal consumption. The substitution
of grains will depress soybean meal prices and EU crushing margins. USDA
projects EU soybean meal consumption to decline to 27.1 million tons in
2000/01 from 27.7 million in 1999/2000. The reduction is expected to trim
EU imports of soybeans and soybean meal from 16.8 million to 16.3 million
tons and from 20 million to 19.6 million tons, respectively.

In India, large vegetable oil imports weakened demand for domestic oilseeds
earlier this year, keeping farm prices for soybeans just above the
government support level and curbing area planted. However, normal
development of India's monsoon is helping soybean yields recover from
excessive dryness last year, and India's 2000/01soybean harvest is
estimated up 10 percent to 5.7 million tons. India does not export soybeans
but processes the entire crop for the soybean oil, exporting the surplus
soybean meal produced. A larger soybean crop would boost projected Indian
soybean meal exports to 2.5 million tons from 2.3 million in 1999/2000.

Late this year, as South American soybean farmers make planting decisions,
greater U.S. and Indian competition and slower imports by China, the EU,
and Japan will dim their price outlook. As in 1999/2000, the expansion in
Brazilian soybean area should remain subdued, rising just 1 percent to 13.4
million hectares. Tight corn supplies should also encourage switching from
soybeans in southern Brazil. Parts of Brazil coped with very dry conditions
during the growing season this year, so with better weather assumed for
next year, soybean output should rise modestly to 32.8 million tons from
this year's 31.4 million. Slack world import demand may trim Brazilian
soybean exports to 9.4 million tons in 2000/01 from 10.2 million. 

A record Argentine soybean area is projected next year, a result mostly of
expanded double cropping with wheat and some switching from sunflowers. A
higher proportion of double-cropped soybeans would hold down the national
average yield, however. Consequently, Argentina's soybean production is
expected to rise only modestly from 20.7 million tons this year to 21.5
million in 2000/01. Larger competitor supplies and smaller world imports
will limit Argentine soybean exports next year to 4.1 million tons,
compared with 5.1 million this year.

China's Accession to WTO 
To Boost Soybean Product Imports

The future of Chinese agriculture, as well as world trade, will likely be
transformed once the country gains admission to the World Trade
Organization (WTO). As a prelude to getting consent from all WTO member
countries, China has signed a bilateral agreement with the U.S.

The agreement would expand market access for soybean oil by replacing
China's arbitrary, unannounced absolute quotas with a tariff-rate quota
(TRQ). Under a TRQ, a lower tariff is applied to imports within the quota,
while above the quota, no quantitative restriction exists provided the
importer pays the higher, over-quota tariff. In principle, a country could
set an over-quota tariff so high as to practically prohibit imports beyond
the quota.

Under the bilateral agreement, the quantity of soybean oil that China would
allow under the TRQ increases from 1.72 million metric tons next year to
3.26 million by 2006. The within-quota duty would be 9 percent (compared
with 13 percent currently), and the over-quota duty would gradually decline
from 74 percent to 9 percent by 2005. This reduction effectively eliminates
the TRQ, leaving just a low 9-percent tariff (equivalent to rates in other
WTO countries) on an unlimited volume of soybean oil imports. 
An increasing proportion of the quota, which is now only available to a few
state-owned importers, would be allocated to nonstate traders. China's
tariff on soybeans (3 percent) and soybean meal (5percent) would be bound
at their current low rates. Subsidized agricultural exports would be
forbidden.

Soybean oil exports from the U.S. (as well as from Argentina and Brazil)
will almost certainly expand as China's over-quota tariff declines over the
next 3 to 4 years, although the tariff-rate quota might not be entirely
filled immediately after implementation, depending on China's vegetable oil
consumption and domestic production. Competition in China's vegetable oil
market will also arise from accession agreements between China and other
countries, which increase import access for canola oil and palm oil. 

Reducing China's import barriers on vegetable oil could sharply increase
oil imports and drive down internal prices. Since many Chinese crushing
facilities are much less efficient than Western plants, the reduction in
processing margins should reverse the recent expansion in oilseed crushing
and revive imports of soybean meal as well.

Because of its size, China is already one of the world's largest consumers
of soybean meal and soybean oil. However, China's per capita use is still
relatively low compared with developed nations, and lower prices could
stimulate consumption. Following WTO accession, initial USDA estimates (AO
March 2000) of the average change in value over baseline projections of
China's imports during the next decade are: soybean oil, $352 million
higher; soybean meal, $220 million higher; and soybeans, $402 million
lower.

With concurrent changes in China's grain and livestock sectors, it is
uncertain what would be the net effect on China's domestic oilseed
production, as policies in previous years have generally been skewed toward
grain production. Allowing more meat imports into China will also affect
domestic feed consumption. Without domestic subsidies, it is likely that
Chinese oilseed farmers will switch to more profitable crops or quit
agriculture as they are exposed to more competition from more efficient
foreign producers.

The rest of the world's oilseed crushers welcome lower protection for
China's farmers and domestic processors, and greater opportunities to
export oilseed products. Both the U.S. and Argentina substantially expanded
crushing capacity in the 1990's. As China's policies promoted domestic
oilseed crushing, considerable excess capacity developed, and crush margins
throughout the world sharply narrowed. China's soybean crushing increased
by nearly one-fourth in 1999/2000, while world crush (excluding China)
declined 1 percent. The supply gluts have been most acute in the global
vegetable oil market, as robust gains in palm oil output further depressed
prices. 

Price Competition 
To Remain Keen

Competitive prices are seen securing solid growth in U.S. soybean product
demand, after 2 years of poor margins and declining crush rates. Domestic
demand for both soybean meal and oil is expected to grow modestly in
2000/01, generally exceeding gains in export demand. U.S. soybean crush is
expected to rise to 1.625 billion bushels in 2000/01 from 1.57 billion last
season. 

Low feed costs and rebounding prices in the hog sector should begin to
promote herd expansion again next year. Domestic disappearance of soybean
meal is projected up 3 percent to 31.25 million short tons, compared with
an estimated 1-percent decline in 1999/2000. Soybean meal prices are
forecast at $140-$165 per short ton vs. $165 last season.

With abundant U.S. and foreign soybean and soybean oil supplies, price
competition will be keen. Projected U.S. soybean oil prices for 2000/01 are
15-18 cents per pound, little changed from the 1999/2000 average of 15.7
cents. Competitive prices and targeted foreign food aid will better
position U.S. soybean oil exporters next year, and comparatively weak
1999/2000 export shipments of 1.2 billion pounds are forecast to recover to
1.8 billion next season. Domestic disappearance of soybean oil is
anticipated up nearly 3 percent to 16.65 billion pounds.

Like their U.S. counterparts, South American oilseed processors have seen
poor margins that prevented them from operating at full capacity. In
Argentina, domestic soybean crushing is likely to remain stagnant in
2000/01, edging up just 0.1 million tons from 16.9 million in 1999/2000. In
Brazil, slightly larger domestic supplies and stronger export and domestic
demand for soybean meal and oil are boosting crush from 21.1 million tons
in 1999/2000 to 21.6 million. Brazilian soybean meal exports, particularly
to Europe, have benefited from the country's depreciated exchange rate.
Competition from Brazil sharply curtailed U.S. soybean meal exports to
Europe in 1999/2000. 

Brazil's crushers (located mostly in the south) will need larger supplies
to remain competitive. Access to soybeans grown in the expansion areas of
the center-west has been complicated by interstate value-added taxes, which
make it more profitable to export soybeans than to crush them domestically.

Despite anticipated trade liberalization, substantial growth in China's
imports of soybean products is not expected in the near term. Therefore,
the strength of Chinese import demand for soybeans will be a key
determinant in the consumption of U.S. and South American crops in 2000/01.
But given an already huge expansion in the U.S. harvest, an increase in
world soybean prices remains unlikely, even in the most optimistic analysis
of Chinese and EU demand.

Mark Ash (202) 694-5289
mash@ers.usda.gov


POLICY
Weak Prices Test
U.S. Sugar Policy

Expanding domestic sugar production and prospects for higher imports are
testing the government's ability to prevent sugar prices from dropping
below support levels. In June, USDA entered the sugar market for the first
time since 1986, purchasing 132,000 tons of refined sugar at a total cost
of $54 million. With this move, USDA projected savings of as much as $6
million in administrative costs that the government might otherwise incur
from expected loan forfeitures later in the fiscal year. The move is also
intended to support sugar growers and to help boost prices for sugar. 

The purchase announcement in May stated that at least 75 percent of an
initial (150,000-ton) purchase would be refined sugar and could be followed
by additional purchases, depending on price and market conditions. The
purchase was authorized under the cost-reduction options of the Food
Security Act (FSA) of 1985. Since June, USDA's Commodity Credit Corporation
(CCC) has been storing the purchased sugar. On August  17, USDA announced a
2-week signup period for the Sugar Payment-In-Kind (PIK) Program, which
offers sugar beet producers the option of diverting a portion of this
year's crop from production in exchange for government-held sugar.

Burgeoning Supplies 

U.S. sugar production for fiscal year 2000 (ending September 30) is
estimated at a record 9.035 million short tons (raw value)--almost 700,000
tons larger than fiscal 1999 production. One reason for this increase is
record area harvested for sugar beets and sugarcane, spurred by higher
expected returns compared with crops that normally compete with sugar for
land use, such as wheat, feed grains, hay, soybeans, and rice. Also, sugar
yields in Louisiana, which now surpasses Florida in sugarcane acreage, have
risen more than 34 percent since 1995 as more acreage has been planted to
high-yielding varieties. 

For beets, last year's generally favorable growing and harvesting
conditions permitted a clean crop, with higher sugar content than the
previous year. Beets entered storage in good condition and remained in good
shape through the winter months as sugar was extracted from them, although
winter weather conditions were less than ideal for storage.

In addition to domestic production, imports are augmenting U.S. sugar
supplies. U.S. imports are restricted by a tariff-rate quota (TRQ). Under
the raw sugar TRQ, 40 quota-holding countries are each allocated a fixed
amount which they may ship to the U.S. in a fiscal year (October-September)
at a zero or low duty. Any raw sugar that enters the U.S. above the quota
is subject to a duty of 15.36 cents per pound--high enough to be generally
prohibitive. 

As part of the Uruguay Round Agreement of the General Agreement on Tariffs
and Trade (GATT), the U.S. had agreed to bind its minimum sugar TRQ imports
at 1.256 million tons per fiscal year. When the World Trade Organization
(WTO) replaced the GATT in 1995, the U.S. minimum access commitment became
enforceable under its dispute settlement mechanisms. 

USDA establishes the total TRQ (for raw and specialty/refined sugar)
annually to control supply. If it is set too high, U.S. prices could
decline below the price support level. If it is set too low, prices could
rise to unacceptably high levels.

A year ago, USDA's fiscal 2000 projection put sugar production at record
levels, and sugar imports in excess of the minimum bound level. In November
1999, the raw sugar TRQ was established at 1.501 million tons. At USDA's
request, it was agreed that only the portion of the fiscal 2000 TRQ
corresponding to the WTO minimum access level would be imported (allocated
to quota-holding exporters), with the remainder constituting a reserve
(unallocated) that could be imported if domestic supply failed to meet
projected levels. 

In addition to domestic production and sugar imports, supply is amplified
with sugar extracted from imports of sugar syrups ("stuffed molasses")
outside the TRQ. These imports have added an estimated 125,000 tons to the
U.S. sugar supply in fiscal 2000. 

Total U.S. sugar supply (including beginning stocks) for fiscal 2000 is
currently estimated at 12.3 million tons. Total use (domestic deliveries
plus exports) is estimated at about 10.4 million tons, leaving ending
stocks at 1.91 million tons. The resulting stocks-to-use ratio is 18.4
percent, the highest level since fiscal 1986.

The abundant sugar supply relative to demand has caused U.S. sugar prices
to decrease to levels not seen in 20 years. The widely quoted No.14 New
York nearby futures price for U.S. raw sugar declined from a monthly
average of 22.61 cents per pound in July 1999 to 17.24 cents in February
2000--a 24-percent decrease. Prices rebounded to the mid-19-cent range in
mid-June, but plunged to 17-18 cents by mid-July, despite the USDA sugar
purchase.

Refined beet-sugar prices have decreased as well. Prices for spot-refined
beet sugar as quoted in the Milling and Baking News averaged only 19 cents
per pound in June and July, down more than 7 cents from a year earlier. 

Government Response Through 
The U.S. Sugar Program

The level of price support to the sugar industry is based on loan rates
legislated in the 1996 Farm Act. Sugar processors (not farmers, whose crop
can't be stored) can take out loans from the government with sugar as
collateral. The loan rate that borrowers receive for raw cane sugar is 18
cents per pound, and for refined beet sugar the rate is 22.9 cents per
pound. 

Processors take sugar program loans for a maximum term of 9 months and
repay them along with interest charges (or forfeit the collateral) before
September 30. If the TRQ is less than 1.5 million tons, sugar loans are
recourse, which like ordinary loans are repayable in cash only. Such loans
have no price-supporting effect and only serve as a mechanism for
short-term financing, with no risk of Treasury expense.

When the TRQ is higher than 1.5 million tons, loans are nonrecourse--i.e.,
the processor may forfeit the collateral in lieu of repaying the loan, and
the government has no recourse but to accept the sugar as full payment. To
the extent that processors put their sugar under loan, their return on that
sugar (minus forfeiture penalty) is protected when market prices drop below
the loan rate. Nonrecourse loans can, in theory at least, help support the
sugar price, since forfeited sugar is effectively taken off the market in
the near term. This price protection, however, incurs risk of government
Treasury expense. With the TRQ set above the trigger in fiscal 2000, loans
are nonrecourse.

Loans outstanding to the CCC as of mid-July are sizable, totaling $447
million and 1.1 million tons. Raw sugar loans made to sugarcane processors
total $183 million, with 511,164 tons under loan, or about 12.4 percent of
estimated production. Beet sugar loans total $264 million, with 620,618
tons under loan, or about 12.5 percent of estimated production. The maximum
government budget exposure if all loans are forfeited is $425 million
(taking into account offsetting revenue of $22 million from forfeiture
penalties). 

In order to discourage forfeiture, market prices must be high enough to
cover the interest expenses, transportation costs (for cane), and market
discounts (for beet sugar). Cane processors incur transportation and
distribution costs in moving sugar to the refiner, in contrast with beet
sugar which is already in refined form and requires no further processing.
Cane processors also face location discounts required by some refiners.
Sugar beet processors must recover the entire interest expense of loan
repayment in their share of the sugar's selling price (unlike cane
processors who share interest expenses with growers) as well as a 2-percent
cash discount (beet sugar is normally sold at discount to cane). 

In addition, processors must consider forfeiture penalties when deciding
whether to forfeit sugar to the CCC. The 1996 Farm Act requires that
processors who forfeit sugar pledged as collateral for a nonrecourse loan
pay a penalty of 1 cent per pound for raw cane sugar and 1.072 cents per
pound for refined beet sugar. 

Accounting for these factors, the average minimum price necessary to
discourage forfeiture in fiscal 2000 is about 19.6 cents per pound
(mainland states) for raw sugar (cane) and about 24.84 cents per pound for
beet sugar. With the mid-July New York nearby futures price between 17 and
18 cents per pound and the refined Midwest beet sugar spot price at 19
cents per pound in mid-July, forfeitures for both types of sugar seemed
likely. Most beet sugar customers had already contracted for their sugar
needs through the rest of the year by mid-July, and there was little
likelihood that spot prices would recover sufficiently to exceed the
minimum price for forfeiture. In fact, beet processors had already
forfeited 42,000 tons as of August 1.

Dealing with Surplus Sugar

Results from the June 2000 USDA sugar purchase are unclear. On the positive
side, the purchase seems to have reduced the costs USDA would have incurred
through defaults on nonrecourse loans. Purchase prices averaged 20.5 cents,
which is less than the sum of the loan rate (22.9 cents per pound) and
accumulated interest (about 1.16 cents per pound) minus the forfeiture
penalty (1.072 cents per pound). 

However, the purchase does not seem to have had any noticeable effect on
sugar prices. Industry observers, including independent analysts, have
suggested that USDA should have offered to purchase much more sugar in
order to affect the market price. Some sugar processors and growers had
initially suggested a larger purchase, in the neighborhood of 250,000 to
370,000 tons. They now argue that USDA should consider a second purchase
offer, acting quickly to restore market confidence.

As of August 1, CCC is holding an inventory of 174,000 tons of refined
sugar, an amount equivalent to 2 percent of the total sugar production
forecast for 2000. The inventory includes sugar that was recently
forfeited, as well as the sugar purchased in June. Additional sugar
forfeitures, which can take place on September 1 and on October 1, will
likely boost government-owned stocks further. Processors intending to
forfeit are required to file a 30-day notice with the CCC, but they are not
bound to forfeit once they have filed notice. 

What will USDA do with the sugar that is forfeited, in addition to the
sugar purchased in June? On August 1, USDA announced a Payment-In-Kind
(PIK) program, offering sugar beet farmers the option of foregoing harvest
in exchange for sugar held by the CCC. On August 17, USDA announced a
2-week signup period beginning August 21 for the PIK Program. Farmers are
limited to $20,000 in PIK payments. By reducing this year's harvest, the
PIK program will help alleviate sugar overproduction, reduce federal
expenditures by reducing probable crop loan forfeitures in fiscal 2001, and
reduce government storage expenditures. The amount of sugar available for
the PIK program is likely to increase in the coming months as sugar pledged
as collateral for CCC loans is forfeited.

Another potential policy option was selling sugar for the manufacture of
ethanol, but the corn industry indicated strong opposition because of
adverse impacts on the corn market. And disposal in the international
market (at a loss, which would violate World Trade Organization export
subsidy commitments) or as emergency food aid was not widely viewed as an
appropriate option.

Low Price Outlook 
For Fiscal 2001 

The market situation may not improve in the coming year. Large predicted
sugar supplies with only modest demand growth indicate continued economic
distress for the industry.

On the supply side, USDA is projecting fiscal 2001 cane and beet sugar
production at 8.973 million tons, slightly below the current year's
estimated record level. Cane sugar production is expected to be higher in
2001 because of an expected record year in Louisiana as well as more
production in Florida and Texas compared with fiscal 2000. In contrast,
beet sugar production is expected to be down, with a return to more normal
crop yield patterns and the possible closing of two processing plants in
California.

In addition, imported sugar is expected to add substantially to U.S. sugar
supplies in 2001. Combined allocated portions of the TRQ will likely be
close to the minimum WTO access of 1.256 million tons, although the raw and
refined sugar TRQ's for fiscal 2001 have not yet been announced.

Non-TRQ imports are projected at 448,000 tons. These include sugar for the
Refined Sugar and Sugar-Containing Products Re-export Programs and the
Polyhydric Alcohol Program (315,000 tons), high-tier tariff sugar (8,000
tons), and sugar extracted from sugar syrups entering outside the sugar TRQ
(125,000 tons).

Imports from Mexico could be as high as 250,000 tons. According to the
North American Free Trade Agreement (NAFTA) side-letter agreement, Mexico's
duty-free access to the U.S. market will increase from 25,000 metric tons
(raw value) to the smaller of  250,000 metric tons or Mexico's "net surplus
production." (Net surplus production is the difference between Mexico's
projected production in metric tons, raw value, less the sum of projected
domestic consumption of sugar in metric tons, raw value, and high-fructose
corn syrup in metric tons.)

Total U.S. sugar supply could reach 13 million tons, 4 percent over the
estimated supply for fiscal 2000. As for sugar demand, use is forecast up
only 1.3 percent to 10.56 million tons, including exports under the Refined
Sugar Re-export Program (175,000 tons). Projected ending stocks for fiscal
2001 could therefore be as high as 2.44 million tons, implying an ending
stocks-to-use ratio of over 23 percent. 

An open question for the industry is whether the fiscal 2001 U.S. sugar
loan program will again be set as nonrecourse. If the raw sugar tariff-rate
quota is 1.5 million tons or below, the loan program will be recourse. USDA
would lose its authority to purchase sugar under the cost reduction options
of the 1985 Food Security Act, and the price-supporting feature of U.S.
sugar policy (nonrecourse loans) would evaporate for the fiscal year. 

When the sugar industry faced the recourse/nonrecourse issue in fiscal
2000, USDA established a raw sugar TRQ above the 1.5-million-ton trigger.
Last year's announcement of the raw sugar TRQ was delayed 6 weeks while
debate of the plan proceeded within the Administration. The "reserve"
portion (249,000 tons) made available for allocation at the discretion of
USDA remains untapped as the fiscal year draws to close. 

The decision on the size of the fiscal 2001 raw sugar TRQ is normally
announced before October 1. For fiscal 2001, the WTO minimum access for the
U.S. sugar TRQ is 1.256 million tons. Assuming duty-free sugar imports from
Mexico under the NAFTA side-letter are not counted as part of the WTO
minimum access, Mexico's projected duty-free shipments will be added to the
minimum access to determine the size of the TRQ. If Mexico's net surplus
production is at least 244,035 metric tons (269,000 short tons), the U.S.
raw sugar TRQ will be above the 1.5-million-ton trigger, and loans for
fiscal 2001 will be nonrecourse. On the other hand, if Mexico's net surplus
production is lower than 244,035 metric tons, then USDA may choose to have
a portion of the raw sugar TRQ remain unallocated as in fiscal 2000, so
that loans remain nonrecourse and price support remains intact.

If loans are nonrecourse in fiscal 2001, possible U.S. government budget
exposure from loan forfeitures is very likely to be much higher next year.
With domestic sugar consumption failing to keep pace with growth in
domestic production plus imports in the foreseeable future, the sugar
market will remain under pressure, making it difficult to keep prices above
support levels without continuing to reduce output through a PIK program or
incurring large Treasury costs.  

Stephen Haley (202) 694-5247 and Nydia Suarez (202) 694-5259
shaley@ers.usda.gov
nrsuarez@ers.usda.gov


RESOURCES & ENVIRONMENT

Confined Animal Production Poses Problems of Manure Management 

Livestock and poultry manure applied to farmland provides a valuable source
of organic nutrients. On many operations, careful nutrient management,
including use of manure, can reduce or eliminate the use of commercial
fertilizers. But nitrogen and phosphorus from manure can cause quality
problems when they enter water systems. Reducing flows of excess nutrients
from the application of animal waste to cropland has become a growing
challenge to confined animal operations.

Nitrogen is easily soluble and is transported in surface runoff, tile
drainage, and water leaching through soil (AO May 2000). Phosphorus is only
moderately soluble, and relative to nitrogen, not very mobile in soil. But
sediment-adsorbed phosphorus can transport considerable amounts of
phosphorus to surface waters through erosion, and the potential for
dissolved phosphorus loss to surface and groundwater increases with buildup
of phosphorus in the soil. 

The opportunity to jointly manage animal waste and crop nutrients as part
of a single operation has decreased with the trend toward fewer, larger,
and more specialized animal production operations, which have inadequate
land available for utilizing manure.

According to the 1997 Census of Agriculture, sales of confined animal
species (feedlot beef cattle, dairy, swine, and poultry) totaled over $75.4
billion, more than 45 percent of total farm sales. Federal policies that
affect the industry's manure management costs--e.g., through the Clean
Water Act (CWA) and farm legislation--can have significant economic effects
on the livestock and poultry sectors. In addition, a growing number of
states are implementing regulations directed specifically at confined
livestock and poultry operations (see article on page 20).

This article presents national and county-level estimates of numbers of
animals and quantity of manure nitrogen produced on confined animal
operations (feedlot beef, dairy, swine, and poultry), as well as farmland
acreage available for nitrogen application. The estimates are a joint
effort of three USDA agencies-the Economic Research Service (ERS), Natural
Resources Conservation Service (NRCS), and National Agricultural Statistics
Service (NASS). 

The study examines national data on farms that could be regulated under the
CWA as point-source discharge sites, and on farms that may be eligible for
assistance under the Environmental Quality Incentives Program (EQIP) of the
1996 Farm Act. Estimates of manure production and of land available for
application are based on data from the four most recent Censuses of
Agriculture (1982, 1987, 1992, and 1997). The question addressed is: If a
livestock or poultry operation applies its manure to the available farmland
(cropland and pasture) under its control at an optimal rate to meet the
nutrient needs of crops grown, how much excess nitrogen would require
disposal?

If the operator controls adequate land for manure application, the focus of
manure disposal should be on farm-level solutions. For producers who can
gain access to land off their farms, manure disposal involves additional
considerations such as timing of transfer and applications, liability for
improper application, and transportation costs. Areas that have
insufficient cropland for spreading manure at optimal rates will need other
manure disposal strategies, with manure management costs depending on the
manure management strategy employed and the extent of potential
problems--e.g., variable nutrient content in the manure, establishing
markets for excess manure nutrients, and manure storage constraints that
necessitate coordination of production flows and manure nutrient usage. 

Concentration in Animal Production 
& Manure Output

The number of farms with confined animals has declined dramatically and
steadily from 435,000 farms in 1982 to 213,000 in 1997. The number of
animals on these farms  is measured based on an animal unit (AU), which
allows multi-species comparisons relative to some specific standard--e.g.,
1,000 pounds of live animal weight. Using the 1000-pound definition in this
analysis means an AU is equivalent to 1.14 head of feedlot beef, 0.74 dairy
cow, 2.67 swine for breeding, 9.09 swine for slaughter, 250 laying hens and
pullets greater than 3 months old, 455 broiler chickens or pullets less
than 3 months old, 50 turkeys for breeding, or 67 turkeys for slaughter. 

All the decline in numbers of confined animal farms occurred in the
smallest size groups--i.e., very small operations with fewer than 50 animal
units (AU), and small operations with 50 to 300 AU. In contrast, the number
of medium-size operations (300-999 AU) grew by 4,400 farms, and large farms
(at least 1,000 AU) more than doubled to almost 4,000 farms. However, in
1997, medium-size farms accounted for only about 6 percent of all confined
animal farms and large farms almost 2 percent, so that very small and small
farms still dominate the number of confined animal farms by a wide margin. 

At the same time that the number of confined animal farms was falling, the
number of confined animal units rose 10 percent. On very small farms, AU's
dropped 64 percent overall to 1.6 million, while on small farms, AU's fell
74 percent to 11.1 million. Meanwhile, AU's on medium-size farms grew by
more than half-from 4 million to 6.4--and almost doubled on large farms to
reach 14.5 million.

Average AU per farm increased 6-17 percent for the lower three size classes
between 1982 and 1997, but dropped 10 percent--from 4,019 AU, on average,
to 3,643 AU-for large confined animal operations. Large swine feeding
operations proliferated during the period, and large swine operations
generally have fewer AU than other types of confined animal operations.

Quantities of nutrients produced by confined AU's rose about 20 percent in
1982-97, while acreage on livestock and poultry farms declined. The result
is a 20-percent increase in estimated excess manure nutrients during a
15-year period, because of inadequate cropland for utilizing manure on the
producing farms. For example, confined animals produced an estimated 1.23
million tons of recoverable manure nitrogen (collectible for spreading) in
1997, but 73 million acres of cropland and permanent pasture controlled by
operators of confined livestock and poultry operations is estimated to have
assimilative capacity for only 38 percent of the calculated nitrogen
available. This is one reason for increased policy attention focused on
confined livestock operations. 

Inability to assimilate all manure nutrients produced on the farm occurs on
operations of all sizes, but not equally. In 1997, about 15 percent of very
small farms and 72 percent of large operations had inadequate capacity to
utilize all the nitrogen produced onfarm. Very small farms produce only
about 2 percent of the national total of excess nutrients, while small
farms (50-299 AU) produced more recoverable manure nitrogen than any other
size class-almost 500,000 tons--and about 30 percent of total excess
nitrogen, primarily accounted for by poultry production.

Nutrient production from medium- and large-size confined animal operations
increased significantly during 1982-97, and quantities of total recoverable
manure nitrogen and excess nitrogen almost doubled. Recoverable manure
nitrogen production on medium-size operations increased 68 percent, and
excess nitrogen by 83 percent; on large farms the corresponding increases
were 102 percent and 104 percent. Medium-size farms accounted for 6 percent
of confined animal operations but for 20 percent of 1997 excess nitrogen
from confined animal production, while large farms accounted for 2 percent
of confined animal farms and almost half of excess nitrogen.

Farms subject to regulation under current CWA rules are designated
concentrated animal feeding operations (CAFO's) based on number of animal
units and amount of point-source discharge from the operation. CAFO's are
not directly identified in Census of Agriculture data. Because the
regulatory impact of the CWA on CAFO's is of interest to policymakers, ERS
has constructed a category of farms--"potential" CAFO's--that would likely
be considered CAFO's under EPA rules. Farms are designated as potential
CAFO's from estimates of annual average numbers of animals on the farms,
derived from data on annual number of animals sold and year-end
inventories. Potential CAFO's--5 percent of all confined animal
farms--include all farms in the large-size category and most in the
medium-size.

Potential CAFO's more than doubled from 1982 to 1997, increasing from about
5,000 farms to 11,200, while the number of AU's on these farms increased
from 9.1 million (30 percent of total confined AU's) to 18 million (54
percent of total confined AU's). Nationally, the average number of AU's on
each potential CAFO has remained stable, so the gain in AU's on potential
CAFO farms was due simply to the increase in number of potential CAFO's.
Potential CAFO's could be the source of over half of estimated excess
nitrogen from all confined animal operations. 

High Excess Nitrogen from Poultry 

Confined animal operations and resulting manure nitrogen are not evenly
distributed across the nation. In 1997, the Southern Seaboard region--a
major poultry- and swine-producing area-generated the largest quantity of
recoverable manure nitrogen (256,000 tons, or over 20 percent of the
nation's confined animal total). The region also has farms with among the
fewest acres per AU on which to apply manure, so it accounts for the
largest quantity of excess nitrogen (200,000 tons, or over 27 percent of
the national confined animal total).

The Southern Seaboard leads in production of recoverable manure nitrogen
despite having about half the AU of the Heartland region. Nutrient
production differs by species, with some types of poultry producing up to 5
times as much nitrogen per AU as feedlot beef cattle. While both the
Heartland and Southern Seaboard regions produce significant numbers of
swine, the Southern Seaboard region has more poultry and fewer cattle,
resulting in greater recoverable manure nutrients from fewer AU. 

Total recoverable manure nitrogen declined from 1982 to 1997 in both the
Northern Crescent and Basin and Range regions, but increased in all other
regions. The Southern Seaboard showed the greatest increase in both
absolute and relative terms--95,000 tons, an increase of almost 60 percent. 

About three-fourths of U.S. counties contain farms that have to dispose of
recoverable manure nitrogen in excess of onfarm crop and pastureland needs.
While production of excess manure nitrogen does not always contribute to
water quality and other environmental problems, manure movement off
confined livestock farms is necessary to avoid excess nitrogen
accumulation. Areas with excess manure may need mechanisms to encourage
land application on other farms, or to provide incentives for alternative
manure treatment strategies. 

Generally, excess manure nitrogen is greatest in counties with the largest
concentration of confined animals, although AU numbers and excess manure
nitrogen are not perfectly correlated. For example, northern Alabama and
Georgia, where poultry is dominant, have high calculated levels of excess
nitrogen because poultry manure has a high nitrogen content per AU and land
available for spreading is limited. Northeastern Iowa and southern
Wisconsin have a relatively high concentration of animals but lower excess
nitrogen than might be expected, because there is more available land per
farm and lower nitrogen production per AU.

The concentration of excess manure nutrients on small-size poultry farms
and on all larger sized operations may provide an opportunity to
effectively target policies to reduce excess manure nutrients. The
potential exists to develop and utilize economical, effective off-farm
technologies, since operations are geographically concentrated (minimizing
manure transportation costs) and are species-dominant (producing relatively
homogeneous manure for processing).

Future of National Policies 
Affecting Animal Operations

Federal policies related to regulation of manure produced on confined
animal operations are still evolving. The Clean Water Act (CWA)--passed in
1972 and administered by the Environmental Protection Agency (EPA)--is the
major piece of Federal legislation affecting animal operations. The CWA
defines water quality in terms of designated beneficial uses (e.g.,
drinking water, recreational use, and aquatic life support) and establishes
criteria to support each use. USDA's Environmental Quality Incentive
Program (EQIP)-authorized by the 1996 Farm Act-replaces most previous
financial assistance programs and better targets assistance to areas most
needing actions to improve or preserve environmental quality. 

Under the CWA, National Pollutant Discharge Elimination System (NPDES)
permits are required for point sources (facilities that discharge directly
into a discrete ditch or pipe) that will empty into navigable waters. NPDES
permits for animal feeding operations currently focus solely on developing
engineering (technology-based) solutions to reduce runoff and spills from
manure storage and treatment structures. 

Under 1974 NPDES regulations, several criteria may be used to designate an
animal feeding operation (AFO) as a concentrated animal feeding operation
(CAFO), thereby labeling it a point source. The criteria may include number
of animals, days in confinement, lack of vegetation in the confinement
area, and potential for waste runoff into waterways. For example, an AFO
could be designated a CAFO if the farm confines 1,000 or more slaughter or
feeder cattle for a total of 45 days annually, or if the farm confines 300
head of slaughter or feeder cattle for 45 days annually and discharges
directly into a waterway. Threshold animal numbers are specified for
slaughter and feeder cattle, dairy cows, swine, laying hens, broilers,
chickens, turkeys, horses, sheep, ducks, or may be a combination of
animals.

EQIP is a voluntary agricultural program that can improve water quality
through changes in farm nutrient management practices. EQIP provides
technical, educational, and financial assistance to farmers and ranchers
for adopting structural, vegetative, and management practices that protect
or enhance environmental quality. By statute, half the program's available
funding is targeted to conservation problems of livestock and poultry
producers.

All 213,000 confined livestock and poultry farms are eligible for nutrient
management technical assistance under EQIP. Operations with fewer than
1,000 AU are also eligible for financial assistance with manure storage or
treatment facilities. Operations with more than 1,000 AU-the 2 percent that
produce 35 percent of excess nitrogen-are not eligible for government
financial assistance to design and build manure management facilities.

Limited funds may lessen the effectiveness of EQIP. Funds allocated by EQIP
were near $200 million for 1997 and 1998, but declined to around $175
million in 1999 and 2000. Even if total annual EQIP funding were devoted
solely to manure management planning, average spending would be only $820
per confined livestock or poultry farm. 

USDA and EPA announced a new initiative in 1999--the Unified National
Strategy for Animal Feeding Operations--that will set minimum standards for
all state water quality protection programs. Regulations to implement the
Unified Strategy are currently under review.

Under the Unified Strategy, all animal feeding operation (AFO) owners and
operators would be expected to develop and implement site-specific
comprehensive nutrient management plans (CNMP), including onfarm
application and off-farm disposal. The strategy will revise the criteria
that identify operations requiring an NPDES permit. The largest operations
will still require a permit, but NPDES permits will also be required of
operations with unacceptable conditions, regardless of size, that pose a
significant risk of water pollution or public health problem or that are
concentrated in a watershed designated as impaired because of nutrient
discharge from AFO's. For example, many poultry farms in the small-size
category that are not currently required to obtain NPDES permits might be
required to have them in the future, if their concentration in the
watershed makes a significant contribution to water quality problems. 

Under current EPA proposals for future NPDES permits, development of a CNMP
will be a required part of the permit process. Permit applications will
include management strategies for manure collection, storage, and
disposal--including use of manure nutrients in crop production. 

The CNMP requirement brings land application of manure into the Federal
NPDES permitting process for the first time. The costs of implementing
off-farm manure management strategies are still to be determined. But more
stringent application of the CNMP requirement on potential CAFO's could
significantly reduce the possibility of excess nutrients entering water
sources.

Noel Gollehon (202) 694-5539 and Margriet Caswell (202) 694-5540
gollehon@ers.usda.gov
mcaswell@ers.usda.gov


SIDEBAR - Confined Animal Production
Estimating Excess Manure Nitrogen

Farm-level "excess" of manure nitrogen on a confined livestock farm is
manure nutrient production less crop assimilative capacity. Manure nitrogen
production is estimated using the number of animals by species, standard
manure production per animal unit, and nutrient composition of each type of
manure. Recoverable manure nitrogen is the amount that can be collected and
disposed of by spreading on fields or transporting off the producing farm.

Each farm's nitrogen assimilative capacity (amount of nitrogen taken up by
plants that are removed from the field at harvest) is based on onfarm
production (acreage multiplied by yield) of 24 major field crops and
pasture recorded by the Census of Agriculture. County, regional, and
national estimates of excess nitrogen levels are aggregated from farm-level
excess estimates (these meet all Census of Agriculture confidentiality
requirements for publication).

The calculation process has the potential to overstate excess nitrogen on
some farms--because many production farms move manure off the farm instead
of utilizing it on land they control--or to understate because it ignores
commercial fertilizer applications. Nevertheless, the excess values
calculated here represent a consistent, national estimate of manure
nitrogen that would need to leave producers' farms in order to be managed
in a manner that reduces the potential for undesirable nutrient flows into
the environment. 


RESOURCES & ENVIRONMENT
Environmental Regulation & Location of Hog Production

Environmental regulation, and the added costs generally associated with
compliance, are considerations often factored into the choice of a business
location. It has been hypothesized that geographic variation in
environmental regulations and enforcement can induce a migration of
industries across state or country boundaries to "pollution havens" where
compliance costs associated with environmental regulations are lower.

Analysis of how environmental regulation and enforcement at the state and
county level (instead of at the Federal level) have affected location
decisions by industrial agriculture can provide some insight into whether
the pollution haven phenomenon applies to agriculture. In addition, it may
help explain why efforts to regain some national control of the regulatory
process by implementing national standards have engendered negative
reactions. For example, local pressures could cause Congress to balk at
appropriating funds for enforcement if the U.S. Environmental Protection
Agency (EPA) tightens existing Federal water quality laws through
regulations proposed for confined animal feeding operations.

Study of whether environmental regulation causes agricultural businesses to
relocate may also shed some light on effects of environmental regulation in
the international arena. Proposals to harmonize (reconcile) environmental
standards across international boundaries add to the urgency of the
question because of concerns raised that trade liberalization could induce
increased investment in agricultural production in countries with lower
environmental standards.

Two emerging issues addressed by USDA's Economic Research Service (ERS)
are: 1) the relationship between stringency of regulation and location of
animal production, and 2) environmental implications of confined animal
production (see article on page 13). This article discusses some of the
reasons for heightened interest in the links between stringency of
environmental regulation and location of the U.S. swine industry. ERS
analyzes the impacts of environmental regulation on the location of animal
production using information from studies presented at an ERS-Farm
Foundation workshop on industry location analysis, as well as extensive
review of recently published analyses. 

Hog Industry Relocation 
& Concentration

Regulations to protect the environment have historically addressed concerns
about environmental pollution from identifiable "point" sources in the
manufacturing sector. But advances in understanding the potentially
damaging effects of pollutants in runoff from agricultural production
sites--i.e., point- and nonpoint-source pollution--have led to efforts to
extend environmental regulation to agricultural activities as well.

A report by the EPA published in the Federal Register concludes that
agriculture is the leading source of pollutants in assessed rivers and
streams, contributing to 59 percent of reported water quality problems and
affecting about 170,000 river miles of the assessed waterways. Unlike
manufacturing, however, it is difficult to correlate damage to the
environment with production activities at a specific farm or animal
production operation. Nevertheless, concern about the environmental effects
of agricultural production is becoming more widespread, exacerbated by the
proliferation of large animal production facilities, particularly those
concentrated in certain geographic areas.

Recently released data from the 1997 Census of Agriculture indicate the
number of hog operations in the U.S. has decreased by half in 10 years, but
total inventory has remained relatively constant as smaller operations exit
and the average operation gets larger. Swine production is more mobile than
other livestock sectors. Hogs can be transported more easily than other
livestock, and are not tied to the land, as are cattle. Also, contract
operations account for a large share of hog production, and when a
contractor moves or expands into a new region, new contracts can be
negotiated in the new location.

Hog production has expanded in recent years in areas in the South and in
nontraditional areas of the West, and a number of counties that were only
minimally involved in the hog industry as of 1992 now have significant
numbers of hogs. This has prompted speculation that large operations moved
to those areas because of possibly less stringent environmental
regulations.

Some high-profile environmental accidents have pointed to the risk
potential of concentrated animal production. For example, the problem of
leakage from large waste lagoons attracted public attention when millions
of gallons of manure overflowed in North Carolina in the aftermath of
Hurricane Floyd in 1999.

Implementation of environmental regulations can impose compliance costs on
producers and reduce profits. Estimates from one study of hog producers in
the U.S. and the European Union (EU) put U.S. waste management costs at
$0.40 to $3.20 per hog, which represents 1-8 percent of total hog
production costs for the operations studied, higher than in previous years
because of added costs of regulatory compliance. Because of the stringency
of the EU Nitrate Directive, estimated costs of compliance for hog
operations there are higher than in the U.S., raising concerns about EU
export competitiveness.

Producers may respond to existing or impending costs of regulation by
exiting the industry or by changing the scale and/or location of
production. Moving to a different state or country might mitigate or bypass
the costs of local or domestic environmental regulations altogether, but
adding new capacity at the same site might enable economies of scale that
offset additional costs of compliance. However, responses that promote
larger hog operations create potential for greater volumes of hog manure to
adversely affect water quality in a local area.

State-level estimates in December 1999 indicate that 17 states account for
the vast majority of very large hog and pig operations (inventory exceeding
5,000 head). North Carolina, Iowa, and Minnesota stand out in number of
very large operations. Perhaps even more significantly, however, very large
operations in Colorado, Oklahoma, and Texas, while much fewer in number,
account for almost all hog production in those states.

EPA requires operations with an inventory of more than 1,000 animal units
to have National Pollution Discharge Elimination System (NPDES) permits for
manure storage or to demonstrate that there is no runoff from the farm (EPA
defines 1,000 animal units for hogs as 2,500 head). However, interpretation
of the regulation varies from state to state, and many states pursue
enforcement only in response to citizen complaints. According to EPA, a
very small proportion of operations with more than 2,500 hogs had acquired
the appropriate manure storage permits.

Type of ownership of hog producing and packing operations appears to play a
role in the locational response to environmental regulation. Individual
producers with family-owned operations are not likely to move operations to
different locations as a result of regulatory changes. Instead, they  are
more likely to continue operating, perhaps at a different scale, or shut
down the enterprise. In addition, as the hog industry moves toward more
production under contract, contractees who grow hogs for larger operations
may have limited ability to adapt if they incur additional costs from
regulations and get no financial assistance from contractors. In the past,
production contracts allowed for specific returns on the finished product,
but have left the costs of manure management to the producer.

Most large corporate production companies already operate facilities in
multiple states, easing the shift of production between states in response
to changes in business conditions. For example, Purina has production
facilities in seven states. Similarly, many top packers also operate
multiple plants across states, so the economic benefits of clustering
production and packing facilities together could be maintained even as
production capacity shifts. Given advances in litter production technology
(i.e., more litters per sow and more pigs per litter), businesses that own
over 100,000 sows could produce 2 million pigs a year for slaughter,
promising large potential savings on transportation costs from clustering
facilities in fewer, more hospitable locations.


Analyses of business location decisions often focus on four factors:
natural endowments, economic costs, business climate, and public policies
(including environmental regulation). International location studies based
on interviews with business executives have rated political stability,
taxes, exchange rate convertibility, and repatriation of profits as key
factors in foreign investment decisions. Environmental regulations were
ranked much lower on the list of considerations.

Studies of the hog industry in particular indicate that significant
variables (factors) in location decisions for hog farms are precipitation,
existing percentage of large hog farms in the state, feed costs, and
density of production. Evidence indicates that the recent shift in hog
operations to western states (primarily Colorado, Oklahoma, and Texas)
resulted in part from savings in transportation costs, because the move
puts exportable products one day closer to the Japanese market compared
with producers in the Midwest and South. In addition, the West offers a
relatively disease-free environment for raising animals. Nevertheless,
production shifts to these more sparsely populated regions highlights the
relationship between location, concentration, and environmental impact.

As animal operations become larger, more states are looking at ways to
protect environmental quality from excess animal waste. Large confined
animal operations can present major problems at the local level. Part of
the potential environmental impact lies in the assimilative capacity of
soil and crops to prevent nitrogen and phosphorous from reaching local
surface water and groundwater resources. The National Pollution Discharge
Elimination System point-source permit system-part of the Clean Water
Act-addresses on-site storage of manure, but not disposal.

Regulatory Stringency & 
Enforcement Vary

States' policies regulating nonpoint-source pollution may vary because of 

*   the design of Federal water policy laws,

*   characteristics of the nonpoint-source pollution, and 

*   characteristics of the states that have to deal with water quality
issues.

Federal water quality laws reflect both the nation's desire to address
existing environmental problems, and the conviction that states should have
sufficient authority and flexibility to design and implement their own
environmental laws. States also have the option to provide funding for
voluntary programs to address the environmental needs of local areas.

When the Clean Water Act was passed in 1972, point sources were seen as the
primary culprits in water and air pollution, so the discharge permit
program was designed to limit emissions by known polluters. Nonpoint-source
pollution was considered a lesser problem that could be left to the states
to manage. In fact, there is some benefit to relegating nonpoint-source
pollution law to state or local level jurisdictions that are closer to the
problem-e.g., more detailed knowledge of the problem and more sensitivity
to impacts of the solution.

A possible drawback to locally developed policies is that local
jurisdictions sometimes have insufficient resources to develop and enforce
regulatory programs. In addition, regulations at a local level may not
effectively address transboundary issues, which may lead to an increase in
frequency of pollutant flows from one jurisdiction to another. If there is
a solution to a transboundary issue, it often comes from the coordination
of activities of local jurisdictions by a Federal government agency like
the EPA. 

Nonpoint-source pollution is characterized by difficulty in observing
runoff and by natural variability of pollution flows with changes in
weather, so linking observations of particular management practices
associated with confined hog feeding operations to changes in water quality
is problematic. And predicting how changes in management practices will
affect water quality presents challenges.

Differences within states in farming practices, land forms, climate, and
hydrologic characteristics is another complication in policy design.
Variation in the environmental impact of agricultural production can occur
even within relatively small geographic areas. Transboundary effects,
uncertainty in measuring actual water quality damage, and time lags in the
movement of a pollutant into a water system also factor into policy design.

Forty-four states have passed laws or instituted programs that either
protect water quality directly by curbing point-source pollution, or
protect it indirectly by regulating an agricultural production practice
associated with generation of nonpoint-source pollution. Some state laws
are follow-ons to Federal clean water laws, while others respond to chronic
local problems such as nitrates or pesticides in groundwater. To help
improve water quality, states may institute controls on inputs or practices
and land use, offer economic incentives, and provide for educational
programs.

Difficulty in measuring the stringency of environmental regulations is a
limitation for analysis of whether state environmental regulations affect
the location or expansion decisions of hog producers. Environmental indices
that rank states on level of environmental protection are of limited use
for agricultural analysis, particularly indices that predate rapid growth
in an industry like swine production. The components underlying the indices
do not relate specifically to agricultural industries or to environmental
problems spawned by concentration in livestock production. For example, one
index assigns states to four categories of environmental
protection--environmentally progressive, struggler, delayer, or
environmentally regressive--in 1990 and 1994. While this ranking highlights
the potential for states to move up or down in environmental protection, it
does not take into account environmental problems that did not even exist a
few years ago. Recent research has started improving these indexes.

Specificity can add stringency to regulation. For example, states may
develop regulations specific to an industry to give more regulatory
attention to a perceived problem. However, specific regulation can also
reflect efforts to stave off even more stringent regulation--known as a "no
more stringent than" law. By enacting a legislative prohibition on future,
more stringent, environmental regulations, states may be seeking to
encourage facilities to locate there. 

Regulations that include reporting requirements and that indicate some
accountability for firms' actions have greater stringency than those that
simply recommend best management practices. The number of permit bars or
blocks that preclude violators from obtaining new permits until problems
have been addressed is a better indicator of regulatory stringency than the
number of penalties, since penalties may or may not be imposed for
environmental infractions due to lack of enforcement capability or funding.

Another indicator of stringency is sufficient resources and staff allocated
to enforcement by state agencies. Rational enforcement agents should be
optimizing some weighted function of their agency's political interests and
the general social welfare. Level of enforcement may not significantly
affect firms' locational response to regulatory restrictions if expected
costs of noncompliance are less than expected costs of compliance. In fact,
very few operations in any state have been penalized in the past, and the
penalties were generally small compared with overall costs of the
operation.

Even with Federal laws like the Clean Water Act and the Clean Air Act,
enforcement is normally delegated to state agencies. However, government
agencies don't usually take on the task of regulation in advance of a
problem, so regulation generally lags the appearance of environmental
damage. Areas that develop the most stringent regulations will tend to be
those that already have environmental problems, that have the most
production with potential to cause environmental problems, or that have
production close to population centers where citizens are concerned about
potential problems.

No matter how stringent, sometimes state laws are ineffective because they
are applied unevenly. For example, a study commissioned by the Indiana
legislature reveals that many of the state's environmental regulations only
apply to new operations, because older operations are "grandfathered
in"--i.e., not subject to the new rules. However, grandfathering may be
politically necessary to get environmental legislation passed. 

Does Environmental Regulation 
Influence Location?

Conjecture is that animal industries tend to move to areas with a lax
environmental regulatory structure. Lax structure can mean either no effort
to enforce, or lack of institutional capabilities or financial resources to
enforce. It may also mean an absence of perceived need for environmental
regulation or enforcement. Locational shifts may involve moves between
geographic areas, or clustering within a given area.

Clustering may occur in areas where existing climatic and geologic factors
such as slope or rainfall make it less costly to comply with standardized
regulations. For example, protecting a lagoon from overflowing is easier
and is lower cost on land that is not a floodplain or where the
distribution of rainfall is not problematic. Clustering has a cumulative
effect in lowering costs, with processing facilities drawing in more
production facilities that may in turn draw in more processing, allied
agribusiness, and input suppliers.

Studies examined indicate that hog operations locate wherever they can
function on a large scale and realize unit-cost savings. Compliance costs
for environmental regulations were only a minor consideration in the past,
but this could change with likely stricter future regulations governing
larger producers. Mitigating environmental problems in areas of expanding
hog production can nevertheless be consistent with profitable operations.

Producers can lower compliance costs  by altering practices. For example,
modifying the cropping system can increase the capacity of farmland to
absorb nitrates and phosphorous from manure, and feed supplementation with
phytase reduces the amount of phosphorous excreted by hogs. Since much of
the best technology for dealing with pollution from hogs is expensive,
clustering many large operations in an area can make use of the technology
more cost-effective. For example, a custom applicator for manure
facilitates injecting manure into the soil locally rather than transporting
it long distances. Joint ownership and use of such machines increases
cost-effectiveness and reduces compliance costs for all.

One somewhat surprising finding is that stringent regulation--which doesn't
necessarily imply stringent enforcement--may actually attract industries to
states. Since specificity in regulations makes the rules clear for
industries planning for future operations, the uncertainty of having to
deal with regulations as they develop is reduced. However, the more a state
spends on environmental enforcement, the less likely a given firm will
locate in that state. Differences in level of enforcement among nearby
states, especially if competitors already operate in the area, may also
affect location decisions. For example, new operations might be
disadvantaged if they incur costs not imposed on existing businesses. 

Additional research is needed to estimate the potential impacts of new
state and Federal water quality regulations on the animal production
sector. For example, compliance costs for the Unified National Strategy for
Animal Feeding Operations-an initiative announced by USDA and EPA-will be
one subject for future research. Research in the future also will explore
the relationship between type, size, and location of operation, and unit
costs for compliance with particular environmental laws.

Location decisions, while important at the state level, also have an
international context, with concerns about large production companies
shifting investment outside the U.S. Production in other countries would
still face variations in environmental regulations. The European Union
experience with its Nitrate Directive is instructive, demonstrating that
limiting producers' options with strict regulation of nitrate levels in an
area with a limited land base has the potential to greatly reduce the scale
and to influence the location of animal production. For example, an EU hog
producer has built production facilities in five U.S states, in part
because of EU environmental constraints.

Harmonization of environmental standards across international boundaries is
a contentious topic in World Trade Organization (WTO) discussions, because
of possible effects on the location of agricultural businesses, as well as
geographic dispersion of the emissions. If uniform environmental
regulations were to raise costs of production in some countries so high
that they could no longer be competitive in export markets, producers in
those countries would likely appeal for an exemption, and some countries
might be willing to enhance their export competitiveness at the expense of
the environment.

With its abundant land base, the U.S. is generally better able to
accommodate compliance with environmental regulations. However, certain
localities within the U.S.--e.g., where manure disposal is a problem (see
map on page 18)--could have difficulty complying with stricter
environmental regulations.

John Sullivan (202) 694-5493, Utpal Vasavada (202) 694-5610, and Mark Smith
(202) 694-5490
johnp@ers.usda.gov
vasavada@ers.usda.gov
mesmith@ers.usda.gov
A list of references is available from the authors.


SPECIAL ARTICLE
Transportation Bottlenecks Shape U.S.-Mexico Food & Agricultural Trade

Anyone visiting Laredo, Texas quickly notices that this bustling city is a
major gateway for trade between the U.S. and Mexico. In fact, it is the
busiest of all ports of entry for commercial trade along the more-than
2,000-mile U.S.-Mexico border. Delays are common, with tractor-trailers
lined up waiting to carry cargo across the border. South of the border,
queues are several miles long with Mexican trucks waiting to cross into the
U.S.

The high volume of traffic at Laredo and other border crossings symbolizes
the dynamic and fast-growing trade relationship between the U.S. and
Mexico, spurred by economic growth on both sides of the Rio Grande and,
beginning in 1994, by the progressive elimination of numerous tariff and
quota barriers as part of the North America Free Trade Agreement (NAFTA). 

Food and agricultural trade between the U.S. and Mexico has been a part of
this growth, more than doubling in the last 10 years to a forecast $10.9
billion in fiscal 2000. Mexico is now the fourth-largest U.S. export market
for farm products ($5.9 billion) and ranks third as a source of farm
imports ($5 billion). The trade is driven by three factors, each associated
with a distinctive transportation pattern:

*   Income growth in Mexico, with the exception of the 1995 recession, has
been significant, averaging about 5 percent per year since implementation
of NAFTA. More and more of Mexico's 98 million people have moved into the
middle to upper classes, now estimated at about 30 million. Many of them
reside in the industrial heartland, the "golden triangle"-an area outlined
by Mexico's three largest cities: Monterrey, Mexico City, and Guadalajara.
This is also where a good deal of value-adding activity takes place, such
as transforming raw agricultural imports like feedgrains and oilseeds into
meat products targeted for domestic consumption. U.S. products destined for
the region move primarily by truck and, to a much lesser extent, by rail
along the Laredo-Mexico City corridor. Some bulk commodities move through
the U.S. seaports of Galveston and New Orleans to Veracruz and other
Mexican seaports, then to interior locations. 

*   Income growth in the U.S. has led to dietary diversification and to
demand for a stable year-round supply of certain foods. Mexico's climate
and investment in irrigation have enabled an export-oriented industry in
its northwest to develop to meet U.S. demand for off-season fruits and
vegetables. A large share of Mexico's horticultural product exports moves
northward by truck primarily through Nogales, Arizona, and to a lesser
extent through the Rio Grande Valley, including the Texas towns of Hidalgo,
McAllen, and Mercedes. The products are stored in warehouses and
distributed to grocery chains and markets throughout the U.S. This trade is
seasonal, peaking in November-March. 

Development of the maquiladora system--assembly of foreign component parts
for re-export--is based on inexpensive Mexican labor, plentiful U.S. and
other foreign capital, and a policy environment encouraging investment and
trade. The system, a result of comparative advantage and government policy,
employs 1.2 million workers in 3,521 plants, according to a March 2000
report by U.S.-Mexico Chamber of Commerce, and it accounts for about 40
percent of Mexico's total exports. Three-quarters of the plants are in
Mexico's six border states with the U.S.

About 30 percent of the maquiladora factories are engaged in textile and
apparel manufacturing and are important buyers of U.S. cotton, textiles,
and yarn. These border areas, where population and income growth has been
faster than in other parts of the country, in turn provide markets for U.S.
food and agricultural products. Because much of the output is exported to
the U.S., the maquiladora system is closely linked to the performance of
the U.S. economy. The system is a large contributor to traffic congestion
along the border close to where many of the plants are located,
particularly at El Paso-Ciudad Juarez, Laredo-Nuevo Laredo, San
Diego-Tijuana, and Brownsville-Matamoros.

Rising trade has led not only to congestion but also, in some instances, to
costly delays at the border and elsewhere. Particularly vulnerable are
time-sensitive perishable products that make up a sizable share of both
south- and northbound food and agricultural trade. 

Food and agricultural trade between the U.S. and Mexico grew briskly in the
1990's despite border and infrastructure constraints. During this time,
growth in U.S. agricultural exports to Mexico outstripped growth in
shipments to almost all other major U.S. foreign markets, some of which are
more developed than Mexico, including the European Union (EU), Japan,
Taiwan, and South Korea. While growth in exports to China, Southeast Asia,
and South America was more rapid in the mid-1990's, it was not sustained in
these markets because of the effects of financial crises, recession, and
supply-side factors on import demand (e.g., record crops in China). Growth
in food and agricultural exports to Canada and Mexico was roughly equal;
growth in total U.S. exports to Mexico was significantly faster than to
Canada in spite of a much more integrated and seamless U.S.-Canada
transportation system.

More than 45 percent of the food and agricultural products now crossing the
U.S.-Mexico border is perishable--about three-quarters of northbound and
one-fifth of southbound food and agricultural trade. This trade includes
fresh and frozen fruits and vegetables as well as chilled and frozen dairy,
livestock, and poultry products. U.S.-Mexico two-way perishable trade in
1998-99 was larger than with any other U.S. trading partner, slightly more
than U.S.-Canada and more than double the volume with the EU and with
Japan.

Extensive trade in perishables is a sign of a sophisticated transportation
system. The refrigeration requirements and, in some cases, the short
shelf-life of perishable products demands more intensive management,
greater speed in marketing, and an unbroken cold chain from the point of
production to the point of consumption.

Behind Border Congestion 

Under NAFTA, trucks were to eventually be able to travel freely throughout
member countries, as regulations that limited truck movement were
eliminated. But the prohibition of reciprocal truck access continues
because of U.S. concerns about safety shortcomings in Mexican trucking
(overweight trucks, lack of operational logs, and no limits on number of
hours driven per shift). Long-haul U.S. and Mexican trucks whose cargo is
destined for locations deep within each country cannot simply drive across
the border to destinations beyond the commercial zone (typically 20 miles
beyond the border or covering several counties). Instead, U.S. and Mexican
truckers must deliver their trailers to the border, and hire short-haul
"drayage" tractors to pull their trailers across the border. Long-haul
trucks on the other side then pick up the trailers and take them to their
destination. Since about 80 percent of the value of U.S.-Mexico trade moves
by truck, continuation of the complicated three-step transfer system is
probably the main contributor to border congestion.

Such a system, particularly along the Texas-Mexico border where much of the
long-haul transferring takes place, increases cross-border traffic. For
example, almost half of the 127,863 trucks crossing at Laredo and nearby
Colombia Solidarity International Bridge in June 1999 pulled empty trailers
or none at all. 

The truck-crossing system also substantially increases the time needed to
cross. For example, delays at the Laredo border range from 4 to 23 hours,
according to analysis by USDA's Foreign Agricultural Service in May 1999.
Removing border bottlenecks would reduce travel time between Chicago and
Monterrey, Mexico by as much as 40 percent, according to estimates by Texas
A&M International University. 

Some north-bound delays result from efforts to interdict drugs and
undocumented immigrants. U.S. drug officials estimate more than 60 percent
of cocaine entering the U.S. comes through Mexico. And according to the
Immigration and Naturalization Service, more than half of the estimated
275,000 annual illegal immigrants to the U.S. come from Mexico. Other
delays arise from inspections by USDA's Agricultural Marketing Service, its
Animal and Plant Health Inspection Service, and the U.S. Food and Drug
Administration. U.S.-Mexico food and agricultural trade is among the most
inspected because of the high volume of food and agricultural trade,
especially perishable products.

Inadequate infrastructure is also a factor at the border, as well as in
some parts of Mexico, increasing transit times and shipping costs. Based on
World Bank data, Mexico's roads and rail system are less developed than
those in the U.S. and Canada. Mexico's road system is not nearly as
comprehensive as the U.S. system, as measured by roads per square
kilometer, and provides less service as measured by road length per capita. 

Overcoming Obstacles

Bottlenecks at the border and inadequate infrastructure are, in effect, a
tax on trade, raising the cost of doing business with Mexico through delays
and through degradation of fresh products. However, a broad spectrum of
incremental measures is expanding the capacity and efficiency of the
increasingly integrated U.S.-Mexico transportation system, reducing the
effects of constraints, and allowing the system to accommodate trade
growth.

Increasing the throughput of trucks at the border can be accomplished in a
number of ways: through expansion of physical facilities at crossing
points, deployment of more customs personnel, expansion of operating hours,
application of new technologies for checking cargo, and automation of
paperwork required for exports and imports.

The speed of processing and inspection is particularly important in
Nogales, since a large share (60 percent) of U.S. fresh fruits and
vegetables from Mexico crosses at this point. Recent investments have
expanded parking capacity at the Customs compound and reorganized the flow
of trucks to help handle heavier traffic. The Customs compound, originally
designed to handle 400 northbound trucks per day, now handles 1,000 to
1,400 daily.

To alleviate growing congestion on both sides of the border at Laredo,
Texas, a fourth bridge was recently completed within the city limits, and
is used exclusively for commercial traffic. This new bridge has
significantly reduced the long lines of tractor-trailers, sometimes
stretching back as far as 4 or 5 miles along Interstate 35. There is
already discussion of a fifth Laredo bridge. But congestion may be due more
to inefficient use of bridges and failure to utilize them for much of the
day. And questions have arisen over the rationale for building new bridges
in Laredo, when the problem is actually that nearly half of all crossings
there involve trucks pulling empty trailers or no trailers at all.

Mexico's infrastructure has improved somewhat in recent years, with
substantial public investment in highway construction and development of
strategic nodes and feeders to connect regional and state road networks.
But recently developed modern toll roads in Mexico are underutilized
because the tolls are too expensive for widespread commercial use. 

In addition to toll roads, much work has been done to modernize North-South
highway corridors by widening roads to include safe shoulders.
Nevertheless, some sections have minimal or non-existent shoulders and are
in poor repair. These highway sections are scheduled to be modernized by
2001. While rail track is generally in good condition, Mexico's railroads
are undercapitalized due to being state run for many years. The situation
is changing since privatization was initiated in 1995.

New technology is reducing inspection times at the border. In 1998, the
U.S. Customs Service began using a fixed X-ray unit that allows agents to
scan an entire truck, reducing the need to unload suspicious cargo. In
1999, officials at Nogales started using a hand-held system which performs
about 200 X-ray inspections a day, compared with 60 for fixed-location
machines.

Mexico is also upgrading inspection procedures through its Customs
Modernization Program to reduce the time for full inspections of southbound
trucks from 90 minutes to 10 minutes or less. This program includes
enhancement of inspection equipment at major points of entry; overhauling
customs computers; and simplifying customs clearance, including the use of
a single NAFTA customs document.

Mexico is installing X-ray equipment, both mobile and fixed units like
those on the U.S. side. Top priority is being given to crossings at Nuevo
Laredo and Colombia across from Laredo, Texas, and at Ciudad Ju rez across
from El Paso, Texas. The gulf coast port of Veracruz, the Pacific coast
port of Manzanillo, and the Mexico City airport are also priorities.

A computerized trade data system is being developed that will permit
simultaneous filing of import data at multiple government agencies. This
will reduce the redundancy of paperwork required by Customs, USDA, and
other government agencies with jurisdiction over imports or exports. 

Mexican and U.S. customs authorities have harmonized hours of operation at
some crossings, but operating practices vary widely from one crossing
location to another. Some high-volume crossings such as Laredo and Otay
Mesa in California operate every day. Others are closed on Sunday and have
reduced hours on Saturdays and holidays. Part of this variability arises
because certain ports specialize as crossings for certain cargoes, such as
fruit and vegetable imports (Nogales), Maquiladora trade (Otay Mesa, El
Paso, Laredo, Brownsville), and long-haul trade (Laredo). 

Some observers assert that operating hours for border crossings need to
expand to a 24-hour-a-day, seven-day-a-week schedule. In Laredo, hours were
extended until midnight as an attempt to ease peak-hour congestion, but few
trucks took advantage of the later hours because warehouses and freight
forwarders were not operating at those hours.

Developing Free Trade Zones &
Alternative Routes

Development of free trade zones on both sides of the border helps
circumvent congested border crossings. Instead of being inspected and
stored at the border, goods proceed to a bonded warehouse at a free trade
zone site where products are cleared by customs and other agencies. Duties
are deferred until imported goods are assembled or leave the site. 

In the U.S., the San Antonio "Kelly USA" Intermodal Facility--already a
global transportation hub--is slated to become a free-trade zone. The
facility has potential to relieve congestion at the border ports of entry
at Laredo, Eagle Pass, and El Paso, because of its location at the
intersection of highways I-35 (north-south) and I-10 (east-west).

Another such site, ADNPlus Industrial Multiport, is being developed as a
free trade zone in Monterrey; it allows for the shifting of customs
clearance for some southbound freight from the congested Laredo crossing to
a location 140 miles south. The site covers 44 million square feet and is
adjacent to the Monterrey airport.

The Multiport park will have terminals for agricultural products, including
a grain elevator, as well as for a range of other freight, including cars,
chemicals, and steel products. It also will provide intermodal services for
railroad, truck, and air cargo carriers. Other free trade zones include the
Alliance Airport in Fort Worth, Texas.

Development of alternative land and sea routes is yet another way to reduce
delays and costs. Shippers of food and agricultural products are already
shifting away from Laredo, the busiest port. The share of major categories
of U.S. food and agricultural exports going through the Laredo Customs
District has declined from 63 percent in 1993-94 to 55 percent in 1998-99.
The largest declines were for cereals (exports of $1.3 billion in 1999),
fresh fruit ($208 million), oilseeds ($819 million), and vegetable oils
($356 million).

For bulk commodities, like cereals and oilseeds, the shift has been away
from truck and rail shipment through Laredo to ocean shipment through New
Orleans and Galveston to Veracruz and other Mexican ports. For higher-value
products primarily shipped by truck, the shift is to other land ports, like
Brownsville and Eagle Pass (within the Laredo District), El Paso, or
Nogales. Many of the ports around the Gulf of Mexico are expanding and
upgrading facilities in anticipation of growing ocean trade between Mexico
and the U.S.

Development of the Port of Manzanillo on Mexico's Pacific Coast has allowed
for more Mexican food and agricultural exports to Los Angeles-Long Beach,
bypassing land routes to the U.S. west coast. It has also bolstered direct
shipments to Japan and other Asian destinations. 

The option to adjust shipping routes depends on product perishability and
the availability of lower cost alternatives. Time-sensitive products
require prompt delivery, and shipping choices are limited by the urgency of
reaching the final destination quickly. More storable products, like
cotton, grain, and oilseeds, afford shippers more alternatives because time
is usually not as critical. 

Except for air transport, trucks are the most expensive mode of
transportation, but they are the most flexible and better able to guarantee
delivery at a particular time and place. Rail and ocean shipping are
cheaper, but their dependence on links with other modes of transportation
for final delivery can cause uncertainty. 

Making Rail More Competitive

Rail transportation in Mexico is becoming more competitive vis a vis
trucking, according to the proceedings of the fifth Agricultural Food
Policy Systems Information Workshop (Feb. 2000). In recent decades,
Mexico's national railroad, Ferrocarriles Nacionales de Mexico, experienced
chronic operating losses and poor productivity. Its share of the nation's
cargo traffic was about 20 percent in 1980 but dropped to about 10 percent
by 1995. A constitutional amendment in 1995 paved the way for privatization
of the system, which divided the railroads into five concessions, including
three main lines: the northeast corridor from Laredo to Mexico City; the
northwest corridor through Hermosillo and Nogales and Saltillo to Eagle
Pass; and the ports of Veracruz and Coatzacoalcos to Mexico City. Other
concessions were a Mexico City terminal and a number of shorter lines. 

The report also indicates that improved management and upgraded equipment
are reducing transit times and costs. Between 1994 and 1998, for example,
rail transit times over the 1200-kilometer Laredo-Mexico City corridor
declined from 67 to 50 hours, which reduces costs for U.S. grain and
soybean rail shipments to Mexico City. The overall level of rail traffic
between the U.S. and Mexico almost doubled between 1992 and 1998.

Greater integration of Mexico's rail system with that of the U.S., and
investments in warehousing and intermodal facilities, are helping to make
shipping by rail a more attractive alternative than trucking. Pre-clearance
by Customs of rail traffic avoids trains having to stop at the border,
which formerly was the procedure. 

In 1999, the Kansas City Southern Railway Company (KCSR) formed an alliance
with the Canadian National Railway, which already had merged with Illinois
Central, to form the "NAFTA Railway," linking Canadian, Mexican, and U.S.
shippers through the heart of the U.S. Corn Belt. KCSR was also part of the
successful consortium obtaining the northeast concession, thus facilitating
the interchange of freight into Mexico through Laredo. A loaded railcar in
the interior of the U.S. can go directly to Mexico City, compared with the
three handlings needed when going by barge, ship, and truck via New
Orleans, Veracruz, and finally Mexico City. 

Outlook for Reducing 
Transport Costs 

While incremental measures-streamlining and automating customs clearance,
expanding border facilities, and improving infrastructure-will continue to
reduce the effects of transportation bottlenecks, two factors will affect
the next generation of growth in U.S.-Mexico food and agricultural trade. 

One is continued development of Mexico's rail system that has been spurred
on by privatization in the second half of the 1990's and by greater
integration with the U.S. and Canadian rail systems. This low-cost mode of
transportation, currently with a small share of the Mexican freight market,
has significant potential to become more competitive with trucking,
primarily for dry cargo, but also for refrigerated products. Critical to
the future of rail in Mexico is investment in intermodal connections with
trucking and ocean shipping services to fully realize its low-cost,
long-haul advantage. 

The second factor is liberalization of truck access, as agreed under NAFTA,
which could challenge the rail system's competitive potential. Free truck
access would dramatically increase the capacity of certain border points to
process and clear cargo, thus lowering transaction costs and possibly
raising trucking's already dominant share of U.S.-Mexico trade.  

William T. Coyle (202) 694-5216 
wcoyle@ers.usda.gov

Contributors: John Link, Bill Kost, Julieta Ugaz, and Constanza Valdes
(Economic Research Service); Todd Drennan (Foreign Agricultural Service);
Keith Klindworth (Agricultural Marketing Service). Valuable comments were
provided by Ed Wueste and Henry Nevares (Texas Department of
Transportation), Sylvia Grijalva (Federal Highway Administration), and Lee
Frankel (Fresh Producers Association of the Americas).

SIDEBAR - Special Article
The Maquiladora System

Much of the traffic across the U.S.-Mexico border is generated by the
maquiladora system. Maquiladora activities largely involve manufacturing
plants in Mexico, which assemble products using U.S. or other foreign
components. Many of the products of these factories are destined for
consumption in the U.S. market and, therefore, become U.S. imports. The
system began in 1965 when Mexico relaxed strict controls on foreign
investment, customs, and immigration. It was formalized into law in 1971
under the Border Industrialization Program (BIP).

A large share of maquiladora trade is automobiles and parts, electrical
components, and other consumer goods. U.S.-Mexico maquiladora trade is
primarily between states on either side of the U.S.-Mexico border, and
between the Mexican border states and the northeastern U.S. (industrial
sector). Traditional U.S.-Mexico trade, by contrast, is more diverse in
terms of product origins and destinations, and is usually shipped further
into the interior of Mexico or the U.S. Traditional trade consists of
products destined for consumption or use as input components for
manufacturers of locally consumed products within either Mexico or the U.S.

More than three-quarters of maquiladora plants are located in the six
Mexican states along the U.S. border. This tends to concentrate maquiladora
system shipping within the border region. Some maquiladora factories
produce partial assemblies in Mexico and final product assembly is
performed in the cross-border U.S. city. This commonly occurs along the
Texas-Mexico border, for example, between the cities of El Paso, Texas, and
Ciudad Juarez, Chihuahua. There are also situations where partial
assemblies are prepared in Mexico and shipped to a corresponding U.S.
production plant in interior states such as Michigan or Illinois.

One of the fastest growing maquiladora sectors is textile and apparel
manufacturing. In the last decade, soaring bilateral trade has positioned
both Mexico and the U.S. among the world's largest exporters of processed
cotton products, creating the world's largest cotton textile trade
relationship. Trade between the two countries now accounts for almost 10
percent of all world trade in cotton textiles. Mexico replaced China in
1995 as the largest source of imported cotton textiles for the U.S., and by
1999 Mexico's share of U.S. imports reached 20 percent. During this period,
Asia's share dropped from about 60 to 45 percent. 

A relatively recent trend is the establishment of maquiladora factories
within the interior states of Mexico. As a result, more and more
maquiladora trade is shifting from along the border to interior locations.
Facilities located in coastal areas like the Yucatan are more accessible by
water-borne transportation than over land.

Binational Border Transportation Planning and Programming Study, Task Force
8 Report, Current Trade and Passenger Flow Data, Final Report, La Empresa
Barton-Aschman, May 8, 1997; Steve MacDonald (ERS).

SIDEBAR - Special Article
U.S.-Mexico Trucking Provisions under NAFTA

Transportation issues were a minor section in the North American Free Trade
Agreement (NAFTA) described in Chapter 12 dealing with cross-border trade
in services. Maritime services were not addressed because of prohibitions
in the Mexican and Canadian constitutions. Since there were few
restrictions on trucking between the U.S. and Canada, the main NAFTA
trucking issue was access of U.S. and Canadian truckers to Mexico's
interior and vice versa. At the time of the agreement, access by Mexican
carriers to the U.S. and by U.S. truckers to Mexico was limited to
commercial zones about 20 miles inside the border (sometimes more-up to 100
miles). All other shipments crossing the border had to be transferred to
local drayage firms, for movement across the border, and then to domestic
trucking companies for movement into the interior.

Provisions of NAFTA allowed investment in trucking firms in other NAFTA
countries as long as those firms were engaged in intra NAFTA trade. U.S.
and Mexican trucking firms were to be allowed to enter freely into the
border states of the other country in December 1995. And by January 1,
2000, Mexican and U.S. trucking firms were to be allowed free access to any
part of the other country. Trucks were to meet height and width, safety and
driver licensing requirements of the other country. Nevertheless, truck
access currently is not allowed.

Chronology of events:

Dec. 14, 1995-Letter from U.S. trucking interests to President Clinton
requests delay in opening U.S. border to Mexican truckers because of safety
concerns:

*   Mexican trucks are too heavy (120,000 gross vehicle pounds, compared
with 80,000 in the U.S.);

*   Mexican trucks are too old (15 years, compared with 5 years in the
U.S.);

*   Mexican trucks are not required to have front brakes and anti-lock
systems;

*   Mexican truck drivers are not required to keep logbooks and are not
restricted to 10 hours of driving per day.

December 15, 1995--Teamsters Union files suit to delay opening of the
border.

December 18, 1995--Federico Pena, then U.S. Transportation Secretary,
announces an indefinite delay in opening the border while safety issues are
addressed.

February 15, 1996--President Clinton announces a 1-year ban on
implementation of free trucking access in border states between the two
countries.

September 1998--Mexico requests a binding arbitration panel from the NAFTA
Commission to push the U.S. to open its border to Mexican trucks.

October 1999--President Clinton repeats opposition to open access for
trucks because of unresolved safety issues.

SIDEBAR - Special Article
Mexico's President-Elect Supports Trade 

On July 2, 2000, Mexican voters elected Vicente Fox of the center-right
National Action Party (PAN) to succeed Ernesto Zedillo, as president of
Mexico. Fox takes office on December 1, ending seven decades of rule by the
Institutional Revolutionary Party (PRI).

Fox, a former governor of the Mexican state of Guanajuato, is a businessman
whose career includes running Coca-Cola's Mexico operation. In his
campaign, he made a strong commitment to fiscal discipline, stronger trade
ties with the U.S., and a more secure climate for foreign investment. He
also promoted changes to the constitution that would allow competition in
the electrical and petrochemical sectors, including privatization of Pemex,
Mexico's petroleum monopoly. His support came disproportionately from the
young, urban, and better-educated population, many of whom have benefited
the most from the North American Free Trade Agreement. 

Fox is a strong supporter of free trade and envisions the free movement of
labor throughout North America by 2010. In his view, investment in
education and raising labor productivity in Mexico will reduce illegal
immigration to the U.S., and lay the groundwork for a free labor market
throughout North America. His support for free markets in North America
suggests likely support for modernizing infrastructure and facilitating
trade, which could translate into reducing bottlenecks in the U.S.-Mexico
transportation system.

This article was developed under the auspices of USDA's Mexican Emerging
Markets Project. 

For more information, plan on attending:

Transportation Bottlenecks in the U.S.-Mexico 
Food System Workshop  
January 24-25, 2001
Location to be announced

Sponsored by USDA's Emerging Markets Program, the Economic Research
Service, and Mexico's Agricultural Secretariat (SAGAR). Contact: Bill Coyle
(202) 694-5216, wcoyle@ers.usda.gov

END_OF_FILE