|
Items in cart [0] |
Questions? Call 888-624-8373 |
| ||||||||||||
| Copyright © 2009. National Academy of Sciences. All rights reserved. Terms of Use and Privacy Statement |
Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter.
Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 9
Food Aid Projections for the 199Os:
Workshop Proceedings
INTRODUCTION
At the request of the Agency for International Development's Bureau for Food for Peace
and Private and Voluntary Organizations, the Board on Science and Technology for Inter-
national Development (BOSTID) of the National Research Council (NRC) arranged for an
NRC-appointed pane} and a group of experts to convene for two days of discussions concern-
ing projections of needs for food aid in the decade 1990-2000. The objective of the meeting
was to examine the projections of food commodity trade and, either directly or by deduction,
food aid needs of developing countries, relying upon work of six principal groups engaged in
food commodity analysis the United Nations Food and Agriculture Organization (FAO)
Commodities and liade Division (and others), the World Bank International Commodity
Markets Division, the U.S. Department of Agriculture (USDA) Economics Research Service
Commodity Bade & Analysis Branch (and others), the International Food Policy Research
Institute (IFPRI), the Iowa State University Center for Agricultural Research and Develop-
ment (CARD), and the International Institute of Applied Systems Analysis (~lASA). The
projections of these groups were discussed from the perspective of a number of specialists
engaged in parallel types of analysis of future global economic, regional econo-political,
demographic, and climatic impact, scientific and technological research impact, and risk
forecasting.
DEFINITIONS
There are two basic approaches to defining food aid needs:
~ "Supply-stabilization~ food aid- to stabilize market prices in the recipient country
by making up the shortfall ("need gap") between production and consumption. Such
aid is thus "demand-based" or "demand-driven". Most typically it is delivered into the
local governmental or commercial food distribution system; seldom is there an attempt to
target local populations beyond existing policies. Food stabilization aid of this type helps
developing countries with economic growth, through saving foreign exchange, assisting in
generating local income, and insuring against domestic economic instability. It can help
9
OCR for page 10
to cushion their vulnerability with respect to world commodity markets, though food aid
volumes may decline when prices increase, as many donors make allocations in money terms.
· "Bunger-responsiven food aid aid sufficient to alleviate hunger through making up
nutritional deficiencies. This estimate of food needs attempts to identify populations at
nutritional risk. The goal is to provide food to affected populations, such as women and
children, suffering disproportionate deprivation, attempting to alleviate both chronic hunger
and acute periodic deprivation. This goal leads to estimates of need necessarily higher than
the minimum caloric shortfall for any specified hungry populations since some substitution
effects occur among this population. Currently, relatively little regular food aid is based on
this hunger-responsive type of estimate.
Projections for "supply stabilization" and "hunger-responsive" food aid do not include
soft credit programs of the European Community or the United States, such as the Guar-
anteed Market Supply (GMS) or Export Enhancement Program (EEP) entitlements, but
they do include sales under Title ~ of Public Law 480, as well as U.S. and other donor nation
grant aid. Past emergency aid, because of its stabilizing eject, as well as its humanitarian
motivation, is used for calculating the trade proportion of food aid in some estimates (not
those of IFPRI).
In calendar year 1987 alone, export subsidies on wheat and flour provided by the Euro-
pean Economic Community and the United States to developing countries amount to about
$1.6 billion, or more than half the value of all food aid recently provided by all countries
in the Organization for Economic Cooperation and Development (OECD). Although these
programs also provide importers an opportunity to use their foreign exchange for other
uses, perhaps for development-oriented investments or for debt repayments, they are not
appropriately defined as food aid.
METHODOLOGY
.
Projecting food commodity trade and aid has consistently been attempted since the
Second World War, thanks, in part, to the creation of the FAO, and the expansion of
trade and development concerns in industrial countries. Mostly, estimates were based
on immediate demand and supply availability rather than on detailed need estimates.
However, with improvements in the global system of reporting production and export and
import statistics, and, more recently, with the ability to store and process large amounts
of information on computers, sophistication in using agricultural statistics for forecasting
has improved. But the complex interdependence among the variables requires models that
account for many relationships. There are, therefore, no entirely satisfactory methods of
predicting food trade and food aid needs, even over periods as short as five or ten years, let
alone the fifty-year time horizon set by Brown University's Hunger Program (Kates et al.,
1988~. The most widely used studies on food aid (cereal) needs or requirements are those
prepared by the FAO and USDA. These are principally made for the coming year, or at
most, the year after.
The methods that are used to project future trade, and derivatively, aid needs, all
Involve a complex set of estimates and assumptions. These include estimates of relevant
forces affecting individual countries, for example, the expected rate of general economic
growth (which reflects purchasing power), the growth of population (which together with
income is the major determinant of demand for food), and the amount of cereal grain
production. Assumptions leading to these estimates, along with decisions as to which
countries to include in estimates (such as China and India, whose size tends to swamp
10
OCR for page 11
the data on international trends arising from smaller countries) account for the primary
differences among the main forecasts.
here is no general agreement about the validity of various assumptions. Most economists
agree that a general equilibrium model, in which as many as possible of the variables of
relevant countries' economies are allowed to interact together to assess the impact of varying
any or each of them, including the insertion of food aid, would be able to give the most
fully considered estimates. However, these general equilibrium models are so complex that
none currently exists that is agreed to be especially useful In projecting food aid needs.
The International Institute for Applied Systems Analysis QUASAR, in Laxenburg, Austria, a
center for international cooperation in developing this type of modeling, among other types
of research, has creates} one basic world food model.
Since suitable general equilibrium models are not available, many econorn~sts use partial
equilibrium models, in which a number of the variables are assumed to remain constant, or
to grow at an assumed fixed (positive or negative) rate, while other variables are projected.
Partial equilibrium analysis is widely considered to be a valid means to assess the impact of
limited changes in the dependent variables. However, the longer the period and the greater
the changes projected, the less confidence there is in the results.
For an economic factor as important to some countries as food aid, it is likely that
supplying food aid over a number of years will create economic changes that will alter
econorn~c variables assumed to remain constant. In the extreme this could mean that food
aid supplied over the first half of a projected period would cause economic growth rates to
rise, hence incomes would rise, local food prices would rise without large increases in supply,
and therefore the numbers of nutritionally disadvantaged would increase. After some lag,
however, higher food prices could stimulate both income growth among rural producers
and large increases in production, as occurred in China in the early 1980s, for example.
The implicit expectation is that eventually the effects of food aid could actually result in
a decline in its need, an agricultural growth would be adequate to cover growing domestic
demand.
The estimates presented at the workshop, with the exception of those prepared by
the IFPRI, were derived as rather simple projecl;ions or logical deductions. They were put
forward within the framework of fairly robust estimates of LDC food imports in the 1990s,
generated by models of international commodity trade in cereals and national staple foods
consumption behavior. Developing the estimates of hunger-responsive and price-stabilizing
food aid amounts involved a recognition of "normative elements in forecasting. That
is, they included those elements which, from prior experience, are taken to be normal or
reasonable amounts. In addition to the needs projected from import gaps or nutritional
requirements, there are food aid needs arising from exceptional circumstances. These may
result from political, economic or "exogenous shock" factors (such an drought, civilstrife,
or other unforeseen occurrences).
Actual food aid deliveries, as opposed to needs, may be expected to be lower than
the workshop projections of food aid needs, if we assume no changes in donor country
policies. This reflects constraints on the ability of donors to provide the estimated amounts.
As Ascher (1978) points out, "The introduction of 'normative' forecasting, which is the
use of projections to systematically explore and select goals and alternatives, is the most
prominent indication of the effort to transform the forecast from isolated information into a
decision-making process in its own right." Aside from the key assumptions of a normative
nature regarding target gods for food aid, other assumptions regarding econorn~c behavior
were involved in the estimates derived at the workshop.
11
OCR for page 12
A second point about the estimates is that they were based on expert judgment. A con-
siderable degree of consensus among the experts encouraged a certain degree of confidence
in the average figures derived, and the boundaries around them. The ultimate test of esti-
mates which forecast future food aid flows and levels of food aid needs lies in their accuracy.
In the 1990s, actual deliveries of food aid will occur. Whether these satisfy either of the
two "needs" supply-stabilization for poor developing countries (the FAO Tow-income food
deficit countries), and hunger-responsive food aid will be difficult to assess, certainly more
difficult than whether forecasts of actual flows are accurate or not. Both needs estimates
require, in turn, projections of the demographic composition and amounts of food needed
to eliminate the caloric deficiency of hungry people in the affected countries a subject of
widely varying estimates and observations of uneven reliability, depending on the criteria
adopted.
FOOD COMMODITY TRADE AND AID FORECASTING
-
In introductory remarks, Owen Cylke, Assistant Administrator for FVA, poir~ted out
that food aid is one of the most widely shared instruments of international cooperation
among executive branch agencies of the U.S. government. With the impending change in
administration it is timely to Took at future requirements. The concern is for efficient food
systems and food security in the Third World, not just for food aid.
The following is a synopsis of the papers contributed by the six principal organizations.
The synopsis is derived from the synthesis evaluation paper presented to the meeting by
Bruno Larue of the University of Guelph, and from the ensuing discussions of participants.
Dr. Larue pointed out that it is always easier to judge forecasting models than to create
them. Forecasts are only as good as the available data, and, because they require many
compromises between what should ideally be included and what is practical to include,
they are inherently limited. There are also many uncertainties, arising from such causes
as the weather for example, Hurricane Gilbert in Jamaica and floods in Banglaclesh and
unexpected market demands arising from political changes, such as the coup d'etat in Haiti
and other so-called exogenous factors. Most of the "models" reviewed at the workshop are
designed to predict future prices and tonnages in the export-import trade market. Except
for the IFPR! process employed by Hannan Ezekiel, they are not explicitly designed to
predict food aid requirements.
Three models, the Food and Agricultural Policy Research Institute (FAPRI) model
developed by Iowa State University and the University of Missouri, the Static World Policy
Simulation Modeling (SWOPSIM) of the U.S. Department of Agriculture, and the World
Bank model, predict export-import trends in commodities, mainly cereal grains, not food
aid. The FAPR! model is based primarily on predictions of trade to and from the United
States, while the Bank and others treat U.S. policies in detail in their models because of
their importance to world food trade and aid.
The IFPRI effort, which does project food aid needs, is based on the aggregate impor-
tation of food by developing countries, translated into cereal equivalents.* The import gap
is then calculated as the difference between total domestic use of the major food crops and
production. Food and feed use projections are obtained from estimates of per capita GNP,
population growth, and income elasticities of demand. Commercial import projections are
subtracted from the import gap. These are based on growth rates of per capita income
*Using cereal equivalents for oilseeds, roots, and tubers assumes that foods are equally interchangeable;
since the bulk of food aid is in the form of cereals, there is no great error introduced by this assumption.
12
OCR for page 13
applied to base period commercial imports as a measure of the ability to pay for food. The
balance is the projected food aid needed for food price stabilization.
Some issues arise from these assumptions: there are problems in looking only at the
demand side, since if you know how much food aid you can expect to receive, it can affect
your demand for commercial imports of food and other goods. Theoretically, food aid
demand should not be limiting, since it is a free good. On the other hand, modeling the
supply side is difficult, since you are modeling political and economic behavior. For one
thing, it would have to be based on assumed relations specifying political propensity to tie
food exports to economic aid or new loans in order to reduce undesirably large domestic
stocks. However, given the data available over the short period in developing countries, the
existing LDC demand side at projected prices and availabilities remains the normal basis
for estimates. This "need", as distinct from food aid policy implementation, is affected by
many exogenous factors, and it is very difficult to predict or control.
OUTPUT OF THE MODELS
All models show better economic situations for developing countries in the high-growth
scenario; Tow growth has a negative effect. The FAPRI and World Bank models are surpris-
ingly close on production forecasts. FAPRI predicts a substantial increase in commodity
prices, whereas the World Bank and SWOPSIM models predict a decrease in prices in real
terms.
Net trade prediction figures for the period 1990-1995 show 10-15 percent growth in
exports from the producing countries, and 1~17 percent growth in LDC imports. Figures
for the Centrally Planned Economies (Figures 1 and 2) show a swing from net exporters
to net importers. Corn trade shows an increase in exports by 30-40 percent, with imports
by developing countries increasing 36 percent in response to rising incomes and demand for
animal products. Growth in soybean trade will be much smaller than wheat or corn trade.
The import gap of demand exceeding domestic supply should get wider by 2000.
An interesting implication is that high growth, stimulated through resource transfer to
developing countries from industrial countries, has a greater economic impact than freer
trade resulting from removal of restrictions on commodity trade.
Comparison of the IFPR! and FAPRI models yields different results, but not sur-
prisingly since they have different assumptions and countries included. The import gap
estimates of these two models (all figures in million metric tonnes EMMA of cereal equiva-
lents) are:
High Income E. Asia
Asia (excluding China & India)
Latin America 21
(excluding Argentina)
FAPRI
16
30
IFPRI
19.2
22.6
13.2
The proportion of food aid needed to close the import gap is anticipated to become
larger as import needs rise faster than foreign exchange earnings. One reason for this is the
expectation that the growth impact of the "green revolution" package of technologies and
support to farmers, a factor in high food production growth in the 1970s and 1980s, is largely
over and will not continue into the next decade. High-yielding varieties produced by recent
research are more likely to lead to greater stability in crop yield, rather than much higher
yields. There is, however, a prospect for greater gains from a marketing and distribution
revolution. It is increasingly possible to deliver food across national and international
13
OCR for page 14
Wheat
100
80
60
40
20
o
70
oh
O 60
11
o
oh
J
-
50
40
30
20
10
o
10
5
o
-5
-10
-15
~ . . .
.
.~;~;~;~;~;~;~;. ,, 'of,
:~:~:~:~:~:~:~:e , .....
~ ' ': ',
:~:~:~:~:~:~:~: it
2:,.:.:.:~.~
·:~:~:~:~:~:~:. / 'aim
.".'.2.2.
.. 2 ',.2.
:2:::::: ~
::. .~.:.; ;..
:::::::-:
:~:~:~:~:~:~::
·:-:- - - . .:
:-:-:-:-:-:-:-:
:::::::: :::
·:-:-:- -:::
:-:-:-:-:-:-:-:
...2............
:~:~:~:~:~:~:.
·:-:-:- - . .:
........
·:-:-:-:-:-:-:-
:-:-:-:-:-:-:-;
·:-:-:-:-:-:-:
:::: :-:-:-
.~...:...
.~
...
['2'2-L
;
20
FAPRI
World Bank 1 5
o
10 _ If,,,
.-.-.-.-......
............
·:-:-:-:-:-:-:
:-:-:-:-:-:-:-
, .~
~ _ :.:-.:.:.:.:-.:
.......
·:-:-:-:-:-:-:
:-:-:-:-:-:-:-
_ -
-5
_
Coarse Gralns
.-...-.-......
:-:- - - .::
..............
:::.:::
:-:-:-:-:::::
I::::::
:::: :-:-:
:~:~:~:~:::
·:-:- - .::
::: :-:-:-:
:-:-:-::-:-:-
::: :. .:
Soymeal
....... '.2.~.:':
·:-:-:-:-:-:::
:::: :-.-:-
:-:-:- -: :-:
·:-:-:-:-:-:::
:-:-:-:-:-:-:-
::::::::-:
·:-:-:-:-:-:::
:~:~:~:~:~:~:~:
:::::: :-:-:-
·:-:- - -: :-:
:-:-:-:-:-:-:::
:::::::::::::::
·:-:-:-:-:-:-:-
:~:~:~:~:~: :~:
~-
.
,,,, ,,~
i,.,
Hi' .
/ .
.
,."
,,
-10
70
:-:-:-:-:-:-:-
:::--:::::
a:-:-:-:-:
:-:- - - .::
·:-:-:-:-:-:--
..............
:-:-:-:-:-:-:-
...............
,.:~:~:~:~:~:~:.
:~:~:~:~:~:~:~:
, .. - . - ........
:~:~:~:~:~:::~.
:::::----::.::
·:-:-:-:-:-:-:-
:~:~:~:~:~:~:~:
. - -. -.- ~
:-:-:-:-:-:-: :1
LL
(D 50
I
z
LL
to
40
30
20
10
o
25
20
15
10
5
-10
-15
-20 L
:~:~:~:~:~:~:
·:-:-:-:-:-:-:
:-:-:-:-:-:-:-
:::: :-:-:
·:-:-:-:-:::::
:-:-:-:-:-:-:-
.:~:~:~:~:~:~:
t::::::::::::::
l:-:-:-:-:-:-:-
I-:-:-:-:-:-:-:
I:~:~:~:~:~:~:.
[:::::::::
k:-:-:-:-:-:-:
' ~ .
c.:~:~:.:.:. .:
I ~.
[:~:~:~:~:~:~:::
c .......
t.2 2.
c~ ·:
0r ~
-5 L ~ ~
, .. .
1 1
t.:~:~:.:. ;. ;~
t: :~:~:~:~:~:~:~
~...2.. ~
h:-:- - - .: :-4
:~:~:~:.:.:.. ::
k:~:~:~:~:~:~:~:d
k::::::: :-:-;
:~:~:~:.:. . :~::
:~:~:~:~:~:~:::d
;~:~:~:~:~:~:~:~:
::::::::d
:~:~:~:~:~:.
.:~:~:.:. . :~:3
:~:~:~:~:~:~:::4
E
| ~ FAPRI
1 ~
| ~ World Bank
e
1 ___L;
1-----.-.-.-.-.1
t::::::::::::::]
I:-:-:-:-:-:---]
1-:-:-:-:-:-:::]
1:-:-:-:-:-:-:-3
1::::::::::::::3
1-:-:-:-:-:-:-:4
I:-:-:-:-:-:-:-3
I: ::-. :: : : : ::- :i
7
, 1
t.:;::::~:~:~:
i:~:~:.:: a:~:
F- -
i:-:- - - :-:-:
[-:-:-:-:-:-.::
t::: :-:-:-:-
i:~:~:.:: :~:~:
1-:-:-:-:-: :::
t:~:~:~:~:~:~:~.
1::::::::::: ::
1-:-:-:. .: :-,
p-:-:-:-:-:-:::
.~:-:-:-:-
L
EXPORTS IMPORTS IMPORTS EXPORTS IMPORTS IMPORTS
DCS LDCs CPEs DCs LDCs CPEs
FIGURE 1 Projected 1995 Net Trade.
SOURCE: Adapted from Larue and
Meilke, 1988.
FIGURE 2 Percentage Change in
Net Trade 1990 to 1995. SOURCE:
Adapted from Meilke and Larue,
1988.
14
OCR for page 15
regions to feed people. More efficient means of marketing food, therefore, would have a
positive impact on food needs, even without increases in the rates of production growth to
keep up with population and economic growth.
PRESENTATION OF THE IFPRI FIGURES
Hannan Ezekiel, International Food Policy Research Institute
Dr. Ezekiel indicated that the IFPRI method estimates demand-based food aid needs.
It does not deal with the gap between need and the ability to pay. It assumes that prices
are kept reasonable through food aid, and do not cause a reduction of demand on the part
of the poor. One methodological problem is to distinguish between demand and nutritional
need; adequate data for estimating nutritional deficiencies satisfactorily are not available.
Another problem—access of poor to the food is beyond the scope of the present IFPRI
study. Increasing the incomes of the poor (or the average incomes of the country as a whole)
to the levels at which the poor are able to buy the food they require is a gigantic problem,
one far beyond the solution of food aid by itself.
Food import gaps cannot always be filled commercially because of countries' foreign
exchange shortages; food aid fills this unfilled gap. There is no such thing as food aid
"needs" per se, however. "Needs" derive from assumptions about stable per capita food
availability and commercial import capacity. Ultimately they are determined by donor
policies, which result in suitable criteria being framed for the purpose. Seeing food aid as a
tool for development helps to set these criteria properly. In addition, there is an important
need for equity which, because of the nature of food, must be estimated independently, and
addressing it by, say, feeding the urban poor, must be done without leading to neglect of
agriculture. If used properly, food aid should increase commercial demand faster because
it supports good policies, thereby increasing incomes. However, if it supports bad policies,
decreasing incomes may lead to more need for food aid to avoid starvation. Thus future
food aid needs are in part a product of its effectiveness in earlier periods. In every case,
food aid must be seen in a specific context, or else it is only a mechanical response and is
not likely to be used as an important development resource. Specific IFPRI projections for
market stabilizing program food aid are given in tables in the paper presented (Appendix
A; Ezekiel, 1988b.)
These IFPRI estimates are based on trends. Income trends combined with a population
growth factor are used to estimate demand, which, in turn, is the basis for import gap
estimates. In principle, food aid should fluctuate in response to production in recipient
countries, though the volume of food aid exports may lag behind production changes,
based on food stock adjustments taking place first (Johnson, 1979~. Upper and lower
limits for need can be projected around a central trend, assuming average variability. Data
are aggregated over a large number of countries from FAO supply utilization accounts,
without modification. The period 1972-1983 is used to estimate trends because using
long base period (the 23 years from 1961-1983) to make projections misses significant recent
alterations in the trends in some countries, both positive and negative. The future is likely
to be more like the immediate past 10-12 years than like the last 23.
7
PRESENTATION OF FAO FIGURES
Richard Perkins, Commodities and Made Division
Dr. Perkins pointed out that food aid is already included in historic trade data, and
15
OCR for page 16
therefore appears in the estimates of countries' future import needs, which assume that
supplies of donated food will continue. Using country import statements may give rise
to anomalies, he noted. Factors that cause this may be upward revisions of production
estimates without revisions in income or population estimates. This could lead, for example,
to Ethiopia being seen as requiring no food aid, and/or, as in the Sudan, refugees not being
counted since they are absorbed into the population without distinction for their special
food aid needs. Needs estimates based on national averages do not account for distribution
inequities and are therefore very inaccurate.
The FAO gathers a great deal of data, and much of it is used in studies of future
world food trends. It has not, however, made projections for food aid requirements recently.
The most recent FAO food aid requirements estimates are based on production, demand,
and trade Projections for the 1985-1990 neric~c] mind m?~.tlL`~mn.t.ic~.~1 meanly similar to
~ ~ r ~ O _ ~ A ~ ~~ _~~ V _
~ ~ T ~ ~ ~ ~ / ~ ~ _ ~ ~ ~ · . ~ ~ ~ . . ~ . ~
the Ub~A/~UL Strain, Unseen, and Livestock' model and plann~ng-type models with
a fairly optimistic production and trade scenario. For the 1985 estimates of food aid
requirements, a revised approach was adopted to include separate components reflecting
balance of payments support and project food aid (for example, for Food for Work and food
security reserves) plus an emergency component based on shortfalls in cereal production.
The total requirement was estimated at 20.2 MMT of cereals. The projection models are
usually neutral with respect to income distribution and food price policies. They are also
neutral about targeting populations or leakage limits.
Another approach adopted by FAO has been to base food aid needs on nutritional
considerations—to increase food intake to "acceptable levels." The guidelines for food aid
adopted by FAO in 1981 give priority to low income food deficit countries. Out of 12 MMT
of food aid in 1987, 10 MMT went to low income food deficit countries, an allocation trend
to be welcomed. The definition of a nutritionally "minimal level of intake" was established
by the Fifth World Food Survey of 1985, based on the distribution of food energy intake
measured through recent household consumption surveys. Very conservative low limits were
set of 1.2-1.4 BMR units per capita (corresponding to 145~1610 KCal/day in India and
1550-1720 KCal/day in Egypt) as minimum maintenance requirements for energy, with
no allowance for movement or work energy for adults. It then calculated the minimum
additional amounts of energy (as cereal equivalents) required to bring the average daily
intake up to these minimum levels for the people in developing countries which at the time
were below this level. These estimates, without allowing for leakage or targeting difficulties,
amount to 8-14 MMT respectively for reaching the 1.2-1.4 BMR limits.
The energy distribution curves used in the survey in 1979-1981 implicitly included in
their established base the approximately 9 MMT of food aid actually provided during that
period. Therefore the 8-14 MMT plus any further increase required for the portion of food
aid that "leaks" (that is, fails to reach the targeted malnourished population) should be
added to the existing food aid levels in any projection of extra needs of that period. As
a basis for extrapolating such figures, by the year 2000 FAO projections for the absolute
number of malnourished people in LDCs indicate an increase over 1979-1981 figures of
roughly 10 percent. (While the fraction of the worId's population receiving lower than a
minimum calorie need will drop, the absolute numbers are not expected to do so.) A further
10 percent allowance for leakage would give a total estimate of 20.6-27.8 MMT additional
food required.*
*The calculations are: 9 MMT + ~ MMT + 10% of 17 = 18.7, + 10% of 18.7 = 20.6 MMT; 9 MMT
+ 14 MMT + 10% of 23 = 25.3, + 10% of 25.3 = 27.8 MMT.
16
OCR for page 17
Such figures are based, however, on the assumptions that 1) all countries would be able
to purchase their net import requirements commercially by 2000; and 2) real world market
prices equal to those in 1983-1985 would prevail. It may be noted, however, that a rather
large decline is projected in the agricultural trade surplus of the developing countries as a
whole, which could undermine such assumptions.
PRESENTATION OF WORLD BANK FIGURES
Ronald Duncan, International Commodity Markets Division
Dr. Duncan indicated that World Bank figures are not for food aid, but projections of
_ 1 ~ · te · ~ ~ ~ ~ ~ . ~ ~ ~ ~ · . · ~ .
lV-U~IVll ~t 6~O mu my ~~. iiie shorts DanK proJectlons see snort-run price increases
through 1989, as the last year of a 3-year cycle, then falling prices in 1990-1991. In the Tong
run, to 2000, they see prices in real terms declining as a continuation of the post-Second
World War trend. Growth in LDCs' per capita GNP will increase food demand. Their
ability to meet this demand will depend on policy responses within the LDCs to restraints
on production, and on import policies. Areas of particular concern for food imports are
sub-Saharan Africa, Bangladesh, and the small southeast Asian countries (excluding China
and India). The World Bank focuses on LDCs in its projections. Other world bodies are
taking care of projections for the industrial countries. Major countries within the World
Bank's area of interest, however, and which the World Bank is assisting with policy reforms,
include countries other than those included above, such as India, Indonesia, Thailand and
Argentina, and also China.
Macroeconomic assumptions indicate a 1989-1990 slowdown in industrial country eco-
nomic growth, and then a rather optimistic period 1990-200~during which econorn~c
growth in OECD countries should be greater than 3 percent, with real interest rates falling,
boosting investments compared with 197~1985 a period when investments fell as real
interest rates rose. LDCs' growth averages are predicted to be 4.5 percent, but this includes
China, with 6-7 percent, and India with 4.5 percent, as well as African and Latin American
growth, which is much lower. The Centrally Planned Economies (excluding China) are
projected at a somewhat pessimistic 2.3 percent. If the CPEs grow faster, the grain trade
market could be much larger.
The outlook for grains trade is the subject of some dispute within the World Bank.
When growth in grain yields slows, some tend to assume that more investment in irrigation
will lead to higher rice yields. However, with increasing incomes in LDCs, there will be
a further shift in relative demand toward coarse grains and from rice to wheat (Japan's
market for grains in 1955 was 95 percent rice, but fell to less than 70 percent in 1985~.
Bennet's Law suggests the inevitability of the growth in demand for coarse grains for feeding
livestock. There should thus be less alarm at falling rice consumption. Africa is different,
showing increased rice consumption with rising incomes. There will therefore be a projected
annual import growth in wheat and coarse grains to both LDCs and CPEs of 5-8 percent.
The actual rate of growth will, it is believed, depend on the rate of implementation of
price/policy reform.
The emphasis on the economic environment is justified because, as in the case of demand
for rice. food consumption r~fl~rt~ ~~`r~rn1 f~f~t.~r~ for inet~n'*= a'7he;~;~ .. ~~;_A~
HAT ~~ ~ ~ ~ ~ ~ ~ ~ ~ At. ~ ~ ~ _ ~ ' ~ ' _ ~ ~ A I _ _ ~ _ ~ ~ _ I ~ 1 _ _ _ 1 ~
__¢ ~ ~ ^_~ w_^ ~ ^~^ ^~—US—~ ~ ~—~1~ ~1 ~11 PI 1~.
~1~1_ _ ~ · 1 · · 1 1 . ~ · ~ · · . ~
~ nag ~r Is possible to remove suns~ct~es without economic chaos was shown in one province
in China where the price was tripled, and consumers were given a lump sum to assist their
adjustment to the new prices. There was no upheaval; a 25 percent decline in rice purchases
reflected mainly an end to feeding cheap rice to chickens.
17
OCR for page 18
There is a major problem in using aggregated commodity trade figures because of the
huge influence of the large countries such as India and China.
PRESENTATION OF THE FAPRI FIGURES
W. Meyers, Center for Agriculture and Rural Development,
Iowa State University
Dr. Meyers observed that the meeting was an important step in bringing together the
commodity trade modelers with those making food aid projections to compare the output
of the models and understand why differences might occur. Key assumptions used in the
FAPRI model are macroeconomic, based on Wharton/Chase econometrics and Project Link
(the University of Pennsylvania macroeconomic modeling project) data rather than World
Bank projections. Its forecast is for an average GNP growth rate of less than 4.5 percent.
The baseline used in the paper circulated to participants was that presented at the Buenos
Aires XX International Conference of Agricultural Economists in August 1988, but prepared
before the drought in the United States, and does not include changes in U.S. policies in
response to drought.
The model's major conclusions were that U.S. acreage planted, as opposed to idled
through set-aside programs, will increase towards the minimum limit of 40 million acres
conservation reserve mandated in the 1985 Food Security Act, leaving the United States
with reduced flexibility to respond to future stresses. The effect of the drought on prices has
driven up estimates for 1988-1989, but these should return to the trend as Tong as policies
remain the same. A slight increase in prices for 1986-1996 is foreseen, returning to resume
the long-term decline thereafter (Figure 3~. Cereal stocks will also be drawn down to the
pre-1985-1986 level by the drought, and there will be much less criticism of the reserve
policy than there was in 1987 (Figure 4~.
FAPRT trade trends project growth of net imports by LDCs, less by the Centrally
Planned Economies (CPEs) and industrial countries. There will be greater pressure on
foreign exchange in developing countries. This will reflect factors such as the growing
impact of debt servicing.
Real income growth in Africa (ranging from 0-3 percent per year) is expected to result
in a widening trade gap in wheat and coarse grains (Figures 5 and 6~. A gap requiring
~ ~ ~ v ~ cat
imports is broadly typical of all developing country areas. Even though Asian economic
growth will stay at about a constant 5 percent per year, well above African growth, the
higher African growth in its import gap probably reflects the Tow base starting point for
African predictions and higher marginal propensity to import food as incomes rise.
Though the improved economic trends are encouraging, in many countries growth in
production cannot keep pace with growth in demand from population and income increases.
While countries in east Asia with high growth can cover their increased grain imports using
foreign exchange generated by exports, many other LDCs have heavy debt burdens and
foreign exchange constraints that will limit their ability to increase imports substantially.
Importers' costs will also depend heavily on whether the local currency is likely to appreciate
or depreciate with respect to the U.S. dollar, and it is the same countries with foreign
exchange shortages whose currencies are likely to depreciate, causing the costs of imported
commodities to rise.
The region under the greatest pressure, in terms of potential reductions in per capita
consumption of commodities, is Africa, followed by the Middle East and Latin America,
excluding Argentina. The Asian region is likely to do better, because production growth is
18
OCR for page 19
500
450
400
O 350
-
ire
To
~ a.
\
/ \ / \ Soybean
1 \ / \
-1 \ I '
300 / l\\
250
200
150
100
I \"\
\
~ _ ~
| ~ Wheat \ ''`
1/: \\~~~~\ '`
// ~ ~
~ Corn g /\\
/
'A \
_ ~x
_'
50
71/72 76/77 81/82 86/87 91/92
YEAR
FIGURE 3 Major Cereals: U.S. Gulf Port Real Prices FAPRI Model
0.29
0.23
O 0.21
0.19
0.17
0.15
0.13
0.11
1~
j L\
. _\\ ~
J
i ~
I ~ ~
~ / I I
HI I
\ / ~ I
/ t1
~ ~ ma/ 1 1 I/ 1
71/72 76/77 81/82 86/87
YEAR
FIGURE 4 World Cereal Levels: Stocks-to-Use Ratio FAPRI Model
19
96/97
/ The at
/ / \\
/ 1 \ \
I \ \
I \
1 \
1 \
1 \
11 \
1
l
l
Feed Grains
-
1 1
91/91 96/97
OCR for page 36
GFDL
it'
~0
{4>D--\
NCAR
~ x
4~J
FIGURE 15 Temperature (June, July, August). The distribution of surface air temperature change (°C)
for a doubling of atmospheric carbon dioxide concentration for June, July, and August simulated by the
global climatic models of GFDL (left), GISS (center), and NCAR (right). Stipple indicates temperature
increases greater than 4°C.
GFDL
q~7
o: go
~?
OWEN
GISS
~1~
FIGURE 16 Precipitation (June, July, August). The distribution of precipitation rate change (mm/day)
for a doubling of atmospheric carbon dioxide concentration for June, July, and August simulated by the
global climatic models of GFDL (left), GISS (center), and NCAR (right). Stipple indicates a decrease in
precipitation rate.
36
OCR for page 37
GFDL
I...
. l . ~ \
- 1~1.~1~1141.1.\
~2'gIl./
o
~7
GISS
_~N
KG O \
~7
oN
NCAR
of
FIGURE 17 Soil Water Content (June, July, August). The distribution of soil water change (cm) for a
doubling of atmospheric carbon dioxide concentration for June, July, and August simulated by the global
climatic models of GFDL (left), GISS (center), and NCAR (right). Stipple indicates a decrease in soil
water.
yields have been generally high and stable since the m~d-1970s in the Great Plains region
because the weather has been generally good until, of course, spring and summer of 1988.
Some have attributed the 1988 drought to the long-awaited appearance of the greenhouse
effect. CO2 concentration has been increasing throughout the 20th century. How then do
we explain the generally benign weather of the last decade? Hot years, cold years, wet
years, and drought years are all well known in the Great Plains. One need not invoke the
greenhouse effect to explain the most recent drought.
Jones and Wigley in the United Kingdom and Hansen and Lebedeff in the United
States have shown an increase in global mean temperature of about 0.5°C since 1900. Their
records show that the warming trend was interrupted by cooling for about 20 years between
1940-1960. Since 1960, warming has resumed. Karl et al. (1988) have computed the average
temperature for the 48 contiguous states during the same period. They observe a warming
trend until the mid-1930s. From then on, temperatures have decreased not only until 1960
but, for all intents and purposes, to this time, although there has been a slight upswing
since the mid-1970s. Although the Karl et al. data show that the mean temperature in the
48 states is now about 0.4°C warmer than it was in 1905, it is 0.5°C cooler than it was in
1935.
The impact of increased temperature and consequent changes in climate could change
the comparative agricultural advantage many countries now enjoy. Canada and Japan and
parts of the Soviet Union might benefit with yields increasing in a warmer world. Other
countries such as the United States might suffer losses particularly if the scenarios of reduced
precipitation come to pass. However, some adjustments are possible, for example, changes in
planting dates, changes in varieties, introduction of new crops. Most agricultural scientists
hold that there are many tactics and strategies available to help agriculture adapt to the
37
OCR for page 38
kinds of temperature change that are predicted by the global climate models. Major changes
in precipitation might be more difficult to deal with. However, the certainty assignable to
model predictions of regional precipitation changes is much Tower than that assignable to
predictions of temperature change.
Another factor to consider is that plants will respond to the higher concentration of CO2
In the air with increased rates of photosynthesis and with improved water use efficiency,
that is photosynthesis per unit of water consumed. These responses, often termed the
"CO2 fertilization effect" should act to moderate detrimental climate changes and perhaps
augment the effects of beneficial climate changes.
Dr. Hutchinson discussed the ability to predict emergency food aid requirements, and
concluded that in the 1990s our ability to predict drought or famine will not improve enough
to make any effective difference to the farmer. He pointed out that we must be able to
monitor better, and though remote sensing is very useful for observing changes over time in
soil moisture and crop acreage, there is a need to improve ground truth with better data.
We really do not know people's needs, market situations, patterns of drought and famine,
and the ability of people to respond to them. The 1974/1975 famine-affected population in
the Sahel should not have survived, but they did. What did they do? What institutional
responses are needed? Drought will continue, and will be incorporated in some models,
so that there should be plans for drought in national planning, as we are admonished in
biblical tradition.
.
AFRICAN REGIONAL CONTEXT
Christopher DeIgado, IFPRI
Dr. DeIgado observed that world food price projections can be useful to the policy
debates concerning the particular problems of Africa. He underlined the African situation:
on the demand side it is useful to distinguish three types of commodities rice and wheat,
other cereals, and non-cereals. The main factors influencing demand are income and
population growth in the urban and rural areas.
He agreed that world commodity prices will decline. The composition of cereal import
needs of LDCs are driven by GNP per capita. There is a shift in commodity composition
which is not linked to price, and usually not much discussion of non-cereal food aid, which
is relatively import ant in some parts of Africa.
There axe five issues touched on by the conference discussion that are especially relevant
to food aid in Africa:
1. The projected continued decline in world food prices. This is very important,
because it is acting in opposition to the strategy for increasing African food production
through price incentives to farmers.
2. The continued rising share of income in urban areas. This is more important in
determining the commodity composition of demand than any other factor; the latter is a
prime determinant of the demand for imported wheat and rice there has been a shift in
commodity consumption in West Africa since the 1960s, according to FAO figures, where
rice and wheat consumption rose by 16 kg per head while sorghum fell by 23 kg per head;
3. IMP structural adjustment policies are working in Africa, but the debt situation is
grim and demand on foreign exchange for debt servicing is a major constraint on growth;
this will tend to reduce the ability to import food commercially;
4. Hard times ahead suggest that political stability will be a major issue: food aid is
an important tool for reducing pressure on governments;
38
OCR for page 39
PRICES
Rice price = Rice import
parity price
Sorghum import
parity price
Sorghum price
Rice export
parity price
Sorguhum export
parity price
~ \ Rice supply \ - / ~
___ ~~ ~ ——————,
I /~ \ / Sorghum
1~1 1 ~ 1 1\1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 ~ 1\1 1 1 1/1 1 supply
/ \ \\ / curve beck
/ \ V weather
I ~ \ A:
Rice demand Sorghum
t I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I demand
I _
QUANTITI ES
FIGURE 18 Commodity Domestic Supply and Demand in Africa
5. The non-cereal food aid picture is confused by subsidies to farmers in the industrial
countries: milk is cheaper in the Sahel than it is in Europe, where there is a large surplus;
the impact of this on production incentives in Africa will be the hot political issue in food
aid in the l990s.
Food aid in Africa in the 1980s is very controversial, because it is linked to structural
adjustment, price policies, devaluation, and the commodity demand for rice and wheat. The
commodity composition in consumption is not synchronous with production and leads to
very important equilibrium issues. Some donors want African governments to resist using
food aid. However, rice and wheat play an important role in assuring the food security of the
poor under increasing urbanization in Africa, particularly West Africa, which is particularly
ill-suited to produce these crops locally. Food aid is thus very important in stabilizing prices
of these commodities and keeping governments in power.
The main problem is that governments have little control over relative prices among
cereals because of the large gap between export and import parity prices for major cereals
(due to high internal transport and marketing costs see Figure 18~. However, food aid
can play a role in helping governments to stabilize relative cereals prices. Political stability
requires rice and wheat imports which governments cannot buy, hence the need for food aid.
Biangular arrangements offer the potential for even relatively rich countries to use food
aid productively, for example, for Zimbabwe to import wheat and export maize to countries
like Mozambique.
Structural adjustment to get agriculture moving will increase the benefit/cost ratio of
domestic production, but prices are beyond the ability of governments to control. Therefore,
the only way to get benefits to farmers is to decrease unit production costs. This requires a
massive increase in capital investment. Public investment in infrastructure and institutions
can be a major catalyst to help price incentives encourage the massive private investment
(by farmers) to do the job. Delgado suggested food aid could be used to allow farmers not
to plant for one season and instead use their labor for capital improvement.
In the ensuing discussion, the dilemma was underlined in which African countries
typically base food security policies on increasing food prices. Governments cannot bank
on a long-term downward trend in prices, yet this is what has happened over the past
39
OCR for page 40
100 years. They may be right, since it is possible that the agricultural policy dilemmas
of overproduction in the industrial countries may not continue. Demand for food marches
on at an average 3 percent increase per year, while supply fluctuates. In the Tong run, a
change will occur to the benefit of producers when livestock demand increases in the LDCs,
as incomes rise, perhaps in 20 years. This implies that the concentration on coarse grains
is correct, since in addition to human consumption they will be required to feed livestock,
and must therefore be produced in Africa on a large and growing scale.
ESTIMATING FOOD AID
The Chairman pointed out that earlier projections of needs were made by a few people
on the basis of very primitive figures and guesswork. Now complex methodologies are
available for processing data, and we have 25 years experience; what have we been able to
learn, and can we forecast food aid with any degree of accuracy?
Projections are complicated when they are made for longer than short run: 1990
projections should be reliable, 1995 fairly good, 2000 less reliable. Complications arise from
regional breakdowns for different commodity demand, exogenous shocks, and other factors.
They are based on demand-driven analysis, projections of trade numbers, and supply-side
forecasting.
The latter—a supply-side forecast is relatively simple:
FAt=FAt_n b
(where Food Aid, FA, at time t, is estimated from a previous food aid time t minus n
years ago—times a weighting factor b reflecting institutional inertia in the budget. Other
factors such as price effect and growth forces, either high or low, might also be added to
this equation.)
The AID representatives observed that they not only need numbers, but must be able
to explain the numbers to Congress. How do the numbers relate to commercial markets,
non-commercial supplies and other subsidy programs? Are we looking for a single figure
for food aid, or trends? There is now a very large export subsidy program, the Export
Enhancement Program (EEP), amounting in 1987 to $2 billion. How do we account for a
need that is partially satisfied by these concessional but non-aid programs?
Since actual food aid allocations are made in a highly political environment which is
largely divorced from "need criteria", and since most t50 percent or more] food aid in fact
~ ,
replaces commercial imports (even though by international convention it is supposed to
be "in addition" to commercial food trade), would it not be valid to first estimate how
much food will likely flow through commeTciai trade channels (as the BES/~lASA, World
Bank and FAPR! models do), and then separately estimate how many people would still be
hungry or severely malnourished given these food trade levels, and finally derive the food
aid need estimate from these trade and malnourished projections?
There followed a discussion concerning the ability to model food aid estimates. There
is a methodological dilemma inherent in the validity of the modeling process. On the one
hand it is argued that a sound methodological approach to medium-term food aid estimates
(5-10 years) must be based on a general equilibrium model. (A general equilibrium model
requires that the interrelationships among the many variables of the entire economy can
only be described validly by including all the variables together in the analysis.) The IlASA
model described by Dr. Frohberg is such a general equilibrium model, but because of the
complexity of the economy and data constraints, it is still fairly primitive and it may be
some time before an adequate general equilibrium mode] is available to assess food aid
40
OCR for page 41
1
needs and their impact, if supplied, on the economies into which they are inserted. On
the other hand, many economists argue that it is not necessary to include all the variables
simultaneously, and that a partial equilibrium model, in which the outcome of changing key
factors such as world cereal prices, or supplying food aid at varying levels to a country or
region, can be modeled independently of other factors (such as GNP growth rates) that can
essentially be treated as constants. The IFPRI, World Bank and USDA estimates assume
that this partial equilibrium modeling is valid.
As there was no way to resolve this methodological dilemma in the time available for
discussion, the modelers agreed that the only way to arrive at estimates was to assume that
many of the factors could be held constant and see what food aid projections would result.
Dr. Kates also requested the modelers to indicate both high and low levels of food aid that
would surprise them if they were the actual levels in 1995 or 2000.
In exchanges that followed, various approximations and extrapolations from trade
projections using assumptions about the future proportion of food aid, allowed various
participants to propose food aid need estimates as starting points for refinement, but not
as conclusions of their research.
Estimates
USDA Projection to 2000
56 MMT represents annual nutritional needs of 69 countries (1988/89 level of 37.5 MMT
at compounded 3.8 percent growth to 2000)
29.4 MMT status quo current annual amount at 3.8 percent growth to 2000
21.1 MMT annual food aid shipments 1976-85 trend extrapolated to 2000 (excluding
export subsidies)
17.6 MMT lean year trend extrapolated to 2000.
56 MMT is the high side surprise limit; 21.1 the Tow side.
World Bank Projection
By the year 2000, 229.3 MMT is the projected net industrial country commodity export
to LDCs. In 1988, 10 percent of industrial country commodity exports to LDCs are food
aid; if this level is maintained to 2000, the Bank estimate of food aid will thus be 23 MMT.
IlASA/:B[S Estimate
Projection of the mode} to 2000 shows 165 MMT net imports annually by LDCs
(excluding China). Food aid or other subsidized imports estimates for 2000 are around 30
MMT/year.
"Food gap" (nutritional) deficit is estimated at 50 MMT, calculated from 400 million
hungry people.
IFPR] Estimate
Based on 85 Tow-income LDCs, excluding China, India, Nigeria, and Brazil, and at 2
levels of per capita GNP cutoff:
41
OCR for page 42
Program Food Aid Requirements in MMT Cereal Equivalents (CE) per year
By 1990 By 2000 By 1995
All countries 37 74 55
Countries with < $800 average GNP/head 19 39 29
FAPR] Estimate
Net average imports by LDCs (MMT CE/yr) 1990 2000 1995
World Bank estimate 103 177 129
FAPRI 95 117
IIASA/BLS — 165+11
Net exports from industrial countries
World Bank 134 229 171
FAPRI 129 151
World Bank shortfall 31 52 42
FAPRI shortfall 36 34
Nutritional Need Estimates
Dr. Pinstrup-Andersen projected three levels (A, B. and C) with slightly different
assumptions, based on Dr. Ezekiel's IFPRI figures, as shown in Table 7.
FAO Estimate
The 1979 projection of food aid need in 1985 was 17-18 MMT, reasonably close to the
actual estimate of need in 1985 of 22 MMT. Actual food aid provided was 10 MMT. In 1988,
the actual food aid supplied is projected to approximate 9 MMT. The unofficial estimate
of need for 1990 is 18-20 MMT, and 8-10 MMT is likely to be a "politically acceptable
estimate" of actual amount to be supplied.
Food aid approximates 10 percent of total overseas development assistance (ODA).
OECD countries are unlikely to increase the total ODA level in 2000.
FAO estimates of net cereal import needs of all countries in 2000 are 115 MMT, and
the gross needs (including China and India) 160 MMT. These figures do not include an
estimated additional 10 percent for dairy, fish and fruit components.
For the Tow-income food deficit countries, based on requests and percentage of food aid
in imports 1984-86 (with very little for feed), food aid needs are 19 MMT minimum and 38
MMT maximum. The average is thus about 30 MMT. The lower end of the range assumes
an optimistic domestic cereal production scenario for developing countries.
POINTS OF CONSENSUS REACHED BY THE WORKSHOP
1. There will be an increase in trade of cereals with LDCs, especially imports.
2. There will be, on average, a decline in real-term commodity prices.
3. The shortfall between domestic production and requirements will increase.
4. More ambitious domestic policies on nutrition using food targeting would increase
the estimates of food aid needed.
5. African needs have the widest boundaries, showing the greatest uncertainty. Ten
years is very short to expect dramatic change in growth rates in agricultural production.
42
OCR for page 43
TABLE 7 Pinstrup-Andersen Estimate of Nutritional Food Aid Needs
A B C
Estimated numbers of food deficient people 400
(in millions)
Assumed calorie deficiency/person/day
500 400
300 300 300
Numbers and calorie' expressed as 15 18 15
MMT grain per year
Substitution (percent)3 50 50 67
Food aid total (from need/~ubetitution)4 30
36 45
Net addition to food supply5 15 18 30
1Two alternative scenarios
2Additional grain consumption by calorie deficient population to
eliminate deficits
3Percent of food aid used for replacing current consumption. Two
alternative scenarios.
4Amount of food aid targeted on deficient households needed to eliminate
deficiencies.
5Since the net addition to market supply is less than what is projected
to be needed to keep prices stable (IFPRI estimates 37 MMT), no
price-depressing effect will occur.
These often require investment in economic infrastructure, as well an development of agricul-
tural technology. It takes 20 years to introduce a new improved crop variety widely enough
to farmers to make an impact on national production. If production grows, however, income
grows (the rule of thumb is: one job created in agriculture creates five jobs in the general
economy) and demand grows, both production and demand changes, affecting "need". Asia
may be a larger arena of growth in need by the year 2000.
How much food aid is possible? Just to achieve a minimum estimate of 19 MMT by 2000
requires doubling current levels, and this would mean a 10 percent per year increase, whereas
current budget plans do not anticipate such a rise. A growing "needs" gap, especially in
Africa, is possible, therefore, although not certain, given the uncertainty already noted.
6. Exogenous shocks create the need for systematic projection planning under uncer-
tainty. Food aid projections are not good at including exogenous shocks.Negative exogenous
shocks include:
.
Climatic effects
- increased variability
increased UV affecting photosynthesis
change in ozone layer
volcanism, and "volcanic winter" effect
occurrence of worst weather events together in the eastern and western or
northern and southern hemispheres.
Economic effects
- debt renunciation
43
OCR for page 44
recurrence of rapidly rising energy costs
recession due to industrial countries' policies.
Positive exogenous shocks include:
Climatic effects
climatic improvements in short-growing areas of the USSR, Canada, etc.
· .
Economic effects
diversion of armament expenditures to development funding.
O Technology effects
unexpected research breakthroughs in plant breeding, production, storage or
marketing technologies.
There is little likely impact of research on productivity during the next decade, because
of the length of time it takes for research to be translated into technology available widely to
farmers. Research is able to respond more quickly to outbreaks of disease than previously,
because of increased sophistication in both biotechnology and communications among re-
searchers. Short-term research contributions may be anticipated in improving postharvest
technology.
The only way of handling exogenous shocks is to recalculate the estimates to take into
account their projected effects, because including exogenous shocks in forecasts would make
modelling impossible because of the number of variables. The only way they can be handled
is analogous to the disclaimer included in insurance policies: "in the event of exogenous
shocks, the estimates have to be revised". This is therefore a potentially fruitful area for
research, including the stochastic modeling suggested by Dr. Klein.
7. Numbers of hungry people.
Dr. Kates pointed out that the estimates of hungry people vary by a factor of 6, from
Dr. Duncan's conservative 100 million to Dr. Pinstrup-Andersen's 400-500 million, with
FAO's "Food-Poor Population" of 350 million close to the mean. This is partly as a result
of differences in definition of "hunger", partly because of the difficulty in obtaining reliable
figures. In fact, there has been remarkable progress in reducing the percentage of hungry in
the population, from close to 35 percent in 1950 to 17 percent in 1988 (Figure 19~. There is
some evidence that the rate of improvement slowed over the past 5-6 years, and the curve
may not continue downward. It needs to be underscored that the program to reduce hunger
is stagnating and must be improved over the next decade to overcome a serious hiatus. Dr.
Rogers pointed out the need to revise the definition of hunger, to take into account the
different thresholds of need for women, children and growth and activity.
Dr. Perkins noted that in 1988 10 MMT is the total amount of food aid estimated to be
delivered to Tow-income food deficit countries, while 20 MMT is the minimum actual level
of need. Doubling the present level is the maximum that could be realistically expected
by 2000 from the standpoint of political acceptance, since it involves a 10 percent annual
increment.
Dr. Perkins indicated the actual commitments of food aid in
1967 4.7 MMT committed by the food aid convention
1980 7.6 MMT
2000 16 MMT projected.
The range of figures derives from a number of different sources. Dr. Andersen suggested
that the most useful references include the FAO Fifth World Food Survey (1985) and the
Agriculture: toward 2000 (1987 revision), the Kates et al. Hunger Report 1988, and the
Sub-Committee on Nutrition report (ACC/SCN, 1987~.
It was agreed there is need to revise upward the nutritional need thresholds to take
44
OCR for page 45
- -
30
z
I 20
at
LL
to
10
_
—_
_,
-
~.
_ _
_
—_
1 950 1 960 1 970 1 980 1 990 2000
,`
YEAR
FIGURE 19 Change in Percentage of Hungry People with Time
into account the special needs of women and children, and to stress that the demand-
based estimates of need do not represent nutritional needs estimates these must be more
precisely targetted by country and by needy groups within countries. Nutritional needs
must be met by methods that go beyond food aid, through entitlement, food for work, food
stamp programs, or other mechanisms.
REFERENCES AND BIBLIOGRAPHY
ACC/SCN. 1987. First Report on the World Nutrition Status. Food and Agriculture Organization of the
United Nations, Rome.
Ascher, W. 1978. Foreca~tir~g: An Appraisal for Policy-Maker and Planners. The Johns Hopkins University
Press, Baltimore.
Brown, L. 1988. '`The Changing World Food Prospect: the Nineties and Beyond." Washington, D.C.:
Worldwatch Institute, Worldwatch Paper 85.
Ezekiel, H. 1988a. "Medium-term Estimates of Demand-based Food Aid Needs and their Variability: A
Note on Methodology.n International Food Policy Research Institute, Washington D.C. 9pp.
Ezekiel, H. 1988b. "Medium-term Estimates of Demand-based Food Aid Needs and their Variability.n
International Food Policy Research Institute, Washington D.C. 13pp.
FAO. undated. aFAO Agricultural Commodity Projections to 1ggo.n FAO Economic and Social Develop-
ment Paper No. 62, pp. 201-212.
FAO. 1984. "Assessing Food Aid Requirements: A Revised Approach." FAO Economic and Social
Development Paper No. 39.
FAO. 1985. "The Fifth World Food Survey." Rome.
FAO. 1987. "Agriculture: Toward 2000." Conference of FAO, Twenty-fourth Session, Rome, 7-26
November 1987. Document C 87/27, Revised Version 1987. Food and Agriculture Organization of
the United Nations, Rome.
FAPRI. 1988. "FAPRI Ten-Year International Agricultural Outlook. Summary and Tables," and "FAPRI
Staff Report #1-88.~ Ames, Iowa: Center for Agricultural and Rural Development, Iowa State
University.
Frohberg, K. 1988. "Food Aid Requirements of Developing Countries.n Organization for Economic
Cooperation and Development, Paris, France. Xerox, llpp.
Hanrahan, C.E. 1988. "The Effectiveness of Food Aid: Implications of Changes in Farm, Food Aid, and
Trade Legislation." Proceedings of a CRS Workshop held on April 25, 1988. Washington, D.C.:
Congressional Research Service, The Library of Congress, Document 88-493 ENR, 155pp.
Johnson, D. Gale. 1979. PA Liberal View." Chapter 9 in Raymond F. Hopkins and Donald J. Puchala,
"The Global Political Economy of Food." 1979. The University of Wisconsin Press.
45
OCR for page 46
Johnson, S.R., W.H. Meyers, P. Westhoff and A. Womack. 1988. "Agricultural Market Outlook and
Sensitivity to Macroeconomic, Productivity and Policy Changes." Paper prepared for the Twentieth
International Conference of Agricultural Economists, Buenos Aires, Argentina, August 23-31, 1988.
Karl, T.R., H.F. Diaz and G. Kukla. 1988. "Urbanization: Its Detection and Effect in the United States
Climate Record. Journal of Climate. (in press).
Kates, R.W., R.S. Chen, T.E. Downing, J.X. Kasperson, E. Messer and S.R. Millman. 1988. The Hunger
Report: 1988 The Alan Shawn Feinstein World Hunger Program, Brown University, Providence, Rhode
Island, 45pp.
Larue, B. and K.D. Meilke. 1988. A Di~cwsiorz of Lor~g-Term Agr~c?tltural Commodity Forecasts and Food Aid
Needs. Department of Agricultural Economics and Business, University of Guelph, Ontario, Canada.
Xerox 15pp.
Meyers, W.H., S. Devadoss, and B. Angel. 1988. "Commodity Market Outlook and Trade Implications
Indicated by the FAPRI Analysis." Center for Agricultural and Rural Development, Iowa State
University, Ames. Xerox, 34pp.
Meilke, K.D. and B. Larue. 1988. "A Critique of Agricultural Trade Models. Paper prepared for the
Twentieth International Conference of Agricultural Economists, Buenos Aires, Argentina, August
23-31, 1988.
Mitchell, D.O. 1988. "Outlook for Grains and Soybeans to 2000." Paper prepared for the Twentieth
International Conference of Agricultural Economists, Buenos Aires, Argentina, August 23-31, 1988.
Washington, D.C.: International Commodity Markets Division, The World Bank. Xerox, 21pp.
Parikh, K.S. and W. Tims. 1986. 'From Hunger Amidst Abundance to Abundance Without Hunger." (An
overview of the policy findings of the Food and Agriculture Program of the International Institute
for Applied Systems Analysis). Laxenburg, Austria: International Institute for Applied Systems
Analysis. Executive Report 13. 37pp.
Paulino, L.A. 1986. "Food in the Third World: Past Trends and Projections to 2000.~ Washington, D.C.:
International Food Policy Research Institute, Research Report No. 52.
Perkins, R.J. 1988. "Statement for the NRC Workshop on Food Aid Requirements and Food Trade in the
1990s.~ Rome: U.N. Food and Agriculture Organization, Commodities and Trade Division. Xerox,
9pp.
Pinstrup-Andersen, P. 1988. "Changing Patterns of Consumption Underlying Changes in Trade and
Agricultural Development. (Publication from the December 1986 Trade Consortium Meeting in
Mexico, pp 69-90.) Washington, D.C.: U.S. Department of Agriculture.
Pinstrup-Andersen, P. 1988. Prevalence of Malnutrition and Dietary Deficiencies as they Relate to
Health and Performance Problems (Paper Prepared for the 1988 World Food Conference, Des
Moines, Iowa, June 5-9, 1988. Ithaca, New York: Cornell Food and Nutrition Policy Program,
Division of Nutritional Sciences, Cornell University. Xerox, 30pp.
Roningen, V.O., P.M. Dixit and R. Seeley. (undated) "Agricultural Outlook for the Year 2000: Some Alter-
natives~ Washington, D.C. U.S. Department of Agriculture, Economic Research Service. Unpublished
draft. Xerox, 10pp.
Rosenberg, N.J. 1988. "Global Climate Change Holds Problems and Uncertainties for Agricultures, in
Ann Tutwiler, (ed.) U.S. Agriculture in a Global Setting: Art Agenda for the Future. Washington, D.C.:
Resources for the Future. Chapter 13.
Rosenberg, N.J. 1988. "Some Highlights of a Resources for the Future Workshop: Controlling and
Adapting to Greenhouse Warming (held at the National Academy of Sciences, Washington, D.C.,
June 14-15, 1988.) Xerox l9pp.
Sanderson, F.H. 1984. World Food Prospects to the Year 2000.~ Washington, D.C.: Resources for the
Future, Reprint 219.
Sarma, J.S. and P. Yeung. 1985. "Livestock Products in the Third World: Past Trends and Projections
to 1990 and 2000~. Washington, D.C.: International Food Policy Research Institute, Research report
No. 49.
Schlesinger, M.E. and J.F.B. Mitchell. 1985. "Model Projections of the Equilibrium Climatic Response to
Increased Carbon Dioxide," in M.C. MacCracken and F.M. Luther, eds., Projecting the Climatic Effects
of Increa~r~g Carbon Dioxide. DOE/ER-0237. Washington, D.C.: U.S. Department of Energy, Carbon
Dioxide Research Division.
Trostle, R.G. 1988. "Food Aid Needs During the 1990s." Washington, D.C.: U.S. Department of
Agriculture, Commodity Economics Division, Economic Research Service. Xerox, 10pp.
USDA. 1988. "World Food Needs and Availabilities, 1988/89: Summer." Washington, D.C.: Economic
Research Service, U.S. Department of Agriculture.
Young, V., D. Bier, and P. Pellett. 1989. A Theoretical Basis for increasing current estimates of amino
acid requirements in adult man, with experimental support. American Journal of Clinical Nutrition
50 :80-92.
46
Representative terms from entire chapter:
world food
