<%BANNER%>
HIDE
 Front Cover
 Title Page
 Abstract
 Table of Contents
 Foreword
 Acknowledgement
 Part 1: Maize research investment...
 Part 2: The current world maize...
 Part 3: Selected maize statist...
 References
 Back Cover
 Copyright


CIMMYT PETE



CIMMYT world maize facts and trends
CITATION SEARCH THUMBNAILS PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00080052/00001
 Material Information
Title: CIMMYT world maize facts and trends
Portion of title: World maize facts and trends
Physical Description: v. : ill. ; 28 cm.
Language: English
Creator: International Maize and Wheat Improvement Center
Publisher: International Maize and Wheat Improvement Center
Place of Publication: Mexico D.F
Creation Date: 1991
Frequency: annual
regular
 Subjects
Subjects / Keywords: Corn industry -- Periodicals -- Developing countries   ( lcsh )
Corn -- Periodicals -- Developing countries   ( lcsh )
Genre: statistics   ( marcgt )
serial   ( sobekcm )
 Notes
General Note: Description based on: 1986.
 Record Information
Source Institution: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: oclc - 18149171
lccn - 88647303
System ID: UF00080052:00001
 Related Items
Related Items: CIMMYT world maize facts and trends. Spanish. CIMMYT ... hechos y tendencias mundiales relacionados con el maíz
Preceded by: CIMMYT maize facts and trends

Table of Contents
    Front Cover
        Front Cover 1
        Front Cover 2
    Title Page
        Page i
    Abstract
        Page ii
    Table of Contents
        Page iii
    Foreword
        Page iv
    Acknowledgement
        Page v
    Part 1: Maize research investment and impacts in developing countries
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
        Page 10
        Page 11
        Page 12
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
        Page 18
        Page 19
        Page 20
        Page 21
        Page 22
        Page 23
        Page 24
        Page 25
        Page 26
        Page 27
        Page 28
        Page 29
        Page 30
        Page 31
    Part 2: The current world maize situation
        Page 32
        Page 33
        Page 34
        Page 35
    Part 3: Selected maize statistics
        Page 36
        Page 37
        Page 38
        Page 39
        Page 40
        Page 41
        Page 42
        Page 43
        Page 44
        Page 45
        Page 46
        Page 47
        Page 48
        Page 49
        Page 50
        Page 51
        Page 52
        Page 53
        Page 54
    References
        Page 55
        Page 56
        Page 57
    Back Cover
        Back Cover 1
        Back Cover 2
    Copyright
        Copyright
Full Text





Rla-l








1991/92 CIMMYT

WORLD MAIZE

FACTS AND TRENDS


MAIZE RESEARCH
INVESTMENT
AND IMPACTS IN
DEVELOPING
COUNTRIES





























CIMMYT is an internationally funded, nonprofit scientific research and training organization. Headquartered in Mexico, the
Center is engaged in a research program for maize, wheat, and triticale, with emphasis on improving the productivity of
agricultural resources in developing countries. It is one of 17 nonprofit international agricultural research and training centers
supported by the Consultative Group on International Agricultural Research (CGIAR), which is sponsored by the Food and
Agriculture Organization (FAO) of the United Nations, the International Bank for Reconstruction and Development (World
Bank), and the United Nations Development Programme (UNDP). The CGIAR consists of some 40 donor countries,
international and regional organizations, and private foundations.

CIMMYT receives core support through the CGIAR from a number of sources, including the international aid agencies of
Australia, Austria, Belgium, Brazil, Canada, China, Denmark, Finland, France, India, Germany, Italy, Japan, Mexico, the
Netherlands, Norway, the Philippines, Spain, Switzerland, the United Kingdom, and the USA, and from the European
Economic Commission, Ford Foundation, Inter-American Development Bank, OPEC Fund for International Development,
UNDP, and World Bank. CIMMYT also receives non-CGIAR extra-core support from the International Development
Research Centre (IDRC) of Canada, the Rockefeller Foundation, and many of the core donors listed above.

Responsibility for this publication rests solely with CIMMYT.



Abstract: This report examines maize research in the developing world, focusing primarily on crop breeding by the public
sector (national research programs and international agricultural research centers). Based on a survey of developing country
maize research programs, the report reviews the level of human and financial resources invested in public sector maize
research in Africa, Asia, West Asia/North Africa, and Latin America. The products of that research (particularly improved
varieties and hybrids generated by public sector crop breeding programs during the past 25 years) are described, as well as
their impacts in farmers' fields. A general approach to analyzing returns to research is reviewed, and several case studies of
the returns to maize research are briefly described. The report concludes with a discussion of how such studies must be
improved to meet the increasing demand for information to justify research resource allocation decisions.


Correct citation: CIMMYT. 1992. 1991-92 CIMMYT World Maize Facts and Trends: Maize Research Investment and
Impacts in Developing Countries. Mexico, D.F.: CIMMYT.
To cite Part 1: Morris, M.L., C. Clancy, and M.A. L6pez-Pereira. 1992. Maize research investment and impacts in developing
countries. Part I of 1991-92 CIMMYT World Maize Facts and Trends: Maize Research Investment and Impacts in Developing
Countries. Mexico, D.F.: CIMMYT.

ISSN: 0257-8743
ISBN: 968-6127-71-2
AGROVOC descriptors: Zea mays, food production, economic analysis, food policies, germplasm, developing countries
AGRIS category codes: E10, E70, F01
Dewey decimal classification: 633.15


Printed in Mexico.














Contents


iv Foreword

v Acknowledgements

1 Part 1: Maize Research Investment and Impacts in
Developing Countries
2 Investment in Maize Research in Developing Countries
9 Products of the Global Maize Research Network
15 Impact of Improved Maize Germplasm
27 Returns to Investment in Maize Research
31 Summary and Conclusion

32 Part 2: The Current World Maize Situation
32 Production
33 Trade
34 Prices
35 Policy Developments Potentially Affecting Maize Production and Trade
35 Conclusion

36 Part 3: Selected Maize Statistics
37 Eastern and Southern Africa
39 Western and Central Africa
41 North Africa
42 West Asia
44 South Asia
45 Southeast Asia and Pacific
46 East Asia
47 Mexico, Central America, and the Caribbean
49 Andean Region, South America
50 Southern Cone, South America
51 Eastern Europe and Former USSR
52 Developed Market Economies
54 Regional Aggregates

55 References


57 Annex 1: Regions of the World


WIL~I







V ri I .1 ... I- *,fl~l I


Foreword


Over the years, the CIMMYT World Facts and Trends series has highlighted developments in the global
maize and wheat economies that have implications for agricultural research. This report reviews achievements
in maize research in the developing world and presents information on the human and financial resources
which have gone into producing those achievements. This study is part of a larger, forward-looking effort by
CIMMYT to gain greater insight into the level of resources devoted to maize and wheat research in the
developing world, identify the products of research, determine the magnitude of their impacts, and assess
future directions for maize and wheat research in light of what we learn.

In examining the collaborative effort to generate and disseminate improved technologies for maize production
in the developing world, we have chosen to focus primarily on the public sector. However, we also present
some complementary information on the level of private sector involvement in developing world maize
research. Our objective is to answer several fundamental questions posed by research administrators, policy
makers, and institutions that fund research: What level of resources goes into maize research? What products
come out of that research? What is the impact in the field? Are the benefits of maize research worth the costs?

Answering these questions is important for several reasons. First, the research community must be
accountable to the people it serves. Studies of returns to research such as those described in this report are
complicated to conduct, but they provide much-needed information about the effectiveness of particular
research endeavors. Second, the increasing scarcity of resources for agricultural research has heightened the
sensitivity of policy makers and funding agencies to the economic dimensions of research resource allocation
decisions. And third, the ability to choose appropriate research strategies depends to some extent on the
capacity to make informed decisions about whether and how different types of research might affect different
social groups.

The following pages lend support to a conclusion expressed in the most recent World Development Report
(World Bank 1992), which is that "experience over the past decades has demonstrated that the generation of
new knowledge is the most potent and least costly avenue to improving productivity." The evidence presented
in this report indicates that many developing countries invest significant resources in public sector maize
research, with impressive results.

However, in seeking empirical data for our study, we had to overcome more than a few difficulties.
Throughout the world research expenditures are accounted for with varying degrees of precision; in some
cases, the objectives of researchers remain undocumented; and the results of research are not consistently
monitored in the field. It is true that many programs are hard pressed merely to conduct research from one day
to the next. But clearly the research community must present stronger arguments for its support. To do so, it
must offer accurate information about its accomplishments, and that will require greater thoroughness in
documenting its activities. This study is one step in that direction. We think that our readers will find it useful.


Donald L. Winkelmann


Director General















Acknowledgements


The development of this edition of World Maize Facts and Trends was coordinated by Michael Morris. The
feature report on investment in maize research was written by Michael Morris, Colleen Clancy, and Miguel A.
L6pez-Pereira of the CIMMYT Economics Program. Several CIMMYT staff drafted original material on
special topics, much of which was eventually featured in the boxes: Pat Byrne, James Deutsch, David
Hoisington, David Jewell, and Michael Listman. Laura Saad of the CIMMYT Economics Program provided
valuable research assistance.

Part 1 of this publication could not have been completed without the assistance of maize researchers in many
developing countries. We are grateful to them for providing data: L. Adupa, A. Aguiluz, J. Alfero,
M. Alvarez, A.A. Arboleda, U. Arvidson, M. Aslam, J. Assam, A. Bejarano, L. Brizuela, A. Bueno,
A. Jim6nez, M. Caulfield, M. Caviedes, B. Clerget, N. Coulibaly, S. Dlamini, J.C. Garcia, Y. Gouro, J.E. Iken,
A.A. Ismail, S. Jacob, C. Kitbamroong, A. Koffi, K.K. Lal, N.S. Leon, L. Machinda, R. Magnavaca,
Z. Mamba, E. Maoule, J. Monteiro, A.J. Moshi, C. Mungoma, L. Narro, A. Ndiaye, Y. N'Guessan,
K. Njoroge, M. N'Kashana, M. Nxumalo, J. Pali-Shikulu, E. Rufyikiri, T. Saing, B. Sali, V. Segovia,
B. Shaner, N.N. Singh, Soumaila, D. Sousa, Subandi, A. Takem, B. Tolessa, S. Troare, C.Tseng, S. Twumasi-
Afriyie, R. Urbina, T. Vy, J. Wobil, and B. Zambezi.

Part 2 was written by Colleen Clancy.

Part 3 was compiled by Laura Saad using FAO data, as well as data provided by collaborators in national
programs throughout the world: J.J. Actis, A. Aguiluz, M. Alvarez, Z. Arafeh, M.N. Aslamy, R. Avenell,
G. Avila, A. Bejarano, A. Bekric, V. Benjasil, A. Bianchi, L. Brizuela, E.R. Capio, I. Capkova, M. Caviedes,
M. Denic, M. Doyuk, M.R. Fuentes, P. Goertz, J.A. Gonzilez, J.J. Guzmin, E.M. Hainzelin, A. Hamid,
T. Hasegawa, J. Hinterholzer, K. Lal, L. Malambo, T. Mansour, C.T. Martinez, L.M. Mazhani, G. McCaw,
D. McHugh, T. Meng, F. Mohamed, J.A. Monteiro, L. Moremoholo, M. Mudhara, C. Mungona, L. Narro,
K. Njoroge, A.O. Okoth, A. Ortega, O. Paratori, A. Pereira, C.E. Pinheiro, J. du Plessis, A. Rahman, A.B. Saz,
Z. Semgalawe, J. Sfakianakis, N.N. Singh, B. Smith, A.A. Sozinov, Subandi, T. Szundy, P. Thole, M. Tohami,
B. Tolessa, N. Tomov, C. Torres, C.T. Tseng, S. Twumasi, P.S. Ue, R. Urbina, W.H. Verburgt, A. Wahab,
and L.J. Xiong.

Many CIMMYT staff assisted in collecting data for Parts 1 and 3. We wish to thank: P. Anandajayasekeram,
H. Ceballos, H. C6rdova, A.O. Diallo, B. Gebrekidan, B. Gelaw, G. Granados, R. Hassan, L. Harrington,
Z. He, P.W. Heisey, C. de Leon, J. Longmire, W. Mwangi, A. Palmer, S. Pandey, C.Y. Tang, M. Read,
M. Saade, G. Sain, G. Srinivasan, S. Sriwatanapongse, S. Vasal, S. Waddington, and C. Wedderbur. We also
wish to thank E. Bailey of ICARDA, K.O. Makinde of IITA, and P. Moya of IRRI for their assistance on yet
another issue of Facts and Trends.

The authors are grateful to our colleagues who reviewed drafts of this report and made suggestions, including
Derek Byerlee, Michael Collins, Ren6e Lafitte, Michael Listman, Ripusudan Paliwal, Roger Rowe, Robert
Tripp, and Donald Winkelmann.

The draft was edited by Kelly Cassaday of CIMMYT. Design and production were done at CIMMYT by Eliot
Sanchez P., with the assistance of Rocio Vargas and Ma. Concepci6n Castro.












Part 1: Maize Research Investment and Impacts
in Developing Countries
Michael L. Morris, Colleen Clancy, and Miguel A. L6pez-Pereira


Of the developing world's major cereals,
maize is the third most important after
rice and wheat. Currently planted on
over 80 million hectares in developing
countries, maize is a dual-purpose crop
that is consumed directly as food and
also used extensively as feed. Driven by
population growth and rising incomes,
demand for maize keeps growing: maize
utilization rose at an average annual rate
of nearly 4% from 1961 to 1990.

This strong growth in demand for maize
has not been matched by corresponding
increases in maize supply. Although
maize production in developing coun-
tries continues to rise, the rate of
production growth has slackened in
recent decades (Figure 1). Improved
varieties' adapted to a wide range of
agroclimatic conditions have made it
possible for farmers to produce maize in
many areas where they could not grow
the crop before. But the dwindling
availability of arable land has meant that
the area planted to maize has grown
more and more slowly over the years.
Although area growth rates recovered
slightly during the 1980s after dropping
below 1% per year in the 1970s, increas-
ing population pressure on the land
makes it unlikely that past growth rates
of 2-3% per year will ever be achieved
again.

Assuming that more maize cannot be
produced by expanding the area planted,
the increased future demand for maize
will have to be met largely through yield


gains that is, by producing more
maize on the same amount of land. The
primary source of these yield gains will
be improved production technologies
generated by a global maize research
system made up of national agricultural
research systems (NARSs), international
agricultural research centers (IARCs),
and private companies.

Although this global maize research
system is crucial to securing future gains
in maize production, these production
gains will have to be achieved in a world
where resources for research are increas-
ingly scarce and challenges to research
are gaining in complexity. For that
reason, it is more important than ever to
be certain that the benefits of agricul-
tural research are worth the considerable
costs. This report contributes to the
general understanding of returns to
research by examining some of the costs
and benefits of the global maize research


Production
4 0,



3.0%


1950s


Production
5.0%

2.8%






2.2%


1960s


system dedicated to improving maize
production in the developing world. The
focus is primarily on public sector
research carried out by NARSs and
IARCs, although the role of private
companies is also discussed. Germplasm
improvement research (that is, maize
breeding) is emphasized, because the
product improved varieties and
hybrids is relatively easy to identify
and track.

The report begins by reviewing the level
of human and financial resources
invested in maize research in the
developing world. It then summarizes
the products of maize research, espe-
cially improved varieties and hybrids
generated by public sector crop breeding
programs during the past 25 years, and it
reviews the impacts of these products in
farmers' fields. Next, it addresses the
question of whether resources invested
in maize research have generated





.. Yield

SArea


Production


3.0%






0.8%

1970s


Production


1.7%



1.3%


1980s


Throughout this report, the term "improved
varieties" is used to refer to improved varieties
and hybrids.


Figure 1. Changes in sources of maize production growth in developing countries,
1950-90.







,i a rld \la,.,. .,- ,-d Trends


attractive rates of return. The report
concludes by summarizing the difficul-
ties involved in quantifying returns to
maize research and discusses implica-
tions for future maize research planning
and evaluation.



Investment in Maize Research
in Developing Countries

Relatively little formal analysis has been
done on investment in and returns to
global maize research. One reason is that
it is difficult to obtain accurate data on
the cost of maize research, particularly
research done in the developing world.
Aggregate data on research expenditures
are available for a number of countries,
but such data often cannot easily be
attributed to specific commodities
(Pardey and Roseboom 1989, Evenson
and Kislev 1971). Some analysts have
tried to estimate the proportion of
overall research resources targeted at
specific commodities, but usually they
have had to resort to indirect methods,
such as indices of scientific publications
(Judd, Boyce, and Evenson 1983).


To obtain a clearer sense of the strength
of the global maize research effort,
CIMMYT recently conducted a survey
of public institutes and selected private
companies active in maize research in
developing countries (see box, "The
CIMMYT Global Maize Research
Impacts Survey," p. 3). Much of the
information contained in this report was
collected through this survey. In addition
to the material presented in Part 1,
selected statistics generated by the
survey appear in Part 3 for individual
countries and regions.

Size of national maize research
programs-Table 1 presents data on the
size of national maize research pro-
grams, as measured by the number of
maize scientists employed in the public
sector. Two categories of researchers are
distinguished: plant breeders (including
those who directly support breeding
activities) and other researchers. Among
the 47 countries that responded to the
CIMMYT survey, the average number of
maize scientists per national program
varies considerably, ranging from 13 in
the Latin American countries to 114 in
the Asian countries.2 This variability is


largely a function of differences in the
average size of the countries in each
region, especially in Asia, where the two
largest countries surveyed (China and
India) are located. Interestingly, breeders
outnumber non-breeders in Asia and
Latin America but not in Africa.

One way to derive a better idea of the
intensity of the maize research effort in
individual countries is to relate the
number of scientists working in the
national maize program to the size of
each country's maize sector. Table 1
includes data on two measures of
research intensity: the number of public
sector researchers per million hectares
planted to maize, and the number of
public sector researchers per million tons
of maize produced. These measures of
intensity vary by region, often reflecting
the influence of one or two large






2 In interpreting the regional figures, it should be
kept in mind that a number of important
countries were not included in the survey,
notably Argentina, Chile, and Turkey.


Table 1. Numbers of public sector maize scientists by region, 1990

Number Public sector Maize Maize
of maize scientists scientists scientists
countries per million ba per million I
surveyed Breeders Others maize area maize production

Sub-Saharan Africa 21 86 188 19 15
West Asia and North Africa 3 32 59 55 16
South, East, and Southeast Asiaa 10 588 460 28 9
Latin America 13 218 106 14 7

Latin Americab 12 180 85 22 12

Source: CIMMYT survey.
a Based on data for all of China (temperate and non-temperate zones).
b Excluding Brazil.















The CIMMYT Global
Maize Research Impacts Survey


Many of the data presented in this
report were collected in 1990-91
through a global survey designed to
establish a baseline set of maize
research indicators. Two questionnaires
were used in the survey. One went to
coordinators of the national maize
programs in 47 developing countries,
most of which plant at least 100,000 ha
or more of maize, as well as a few
countries with less maize area (see
table). The questionnaire focused on
two dimensions of maize research:
research inputs (e.g., numbers of


scientists, level of research expenditures)
and research impacts (e.g., varieties and
hybrids released, area planted to im-
proved materials).

The second questionnaire was sent to 45
private maize seed companies active in
developing countries. This question-
naire, intended to complement the
questionnaire distributed to public sector
research institutions, focused primarily


Countries participating in the CIMMYT Global Maize Research Impacts Survey

Sub-Saharan West Asia and South, East, and Latin
Africa North Africa Southeast Asia America

Benin Afghanistan China Bolivia
Burkina Faso Egypt India Brazil
Burundi Morocco Indonesia Colombia
Cameroon Myanmar Costa Ricaa
C6te D'Ivoire Nepal Ecuador
Ethiopia Pakistan El Salvador
Ghana Philippines Guatemala
Kenya Taiwan-ROCa Honduras
Madagascar Thailand Mexico
Malawi Vietnam Nicaragua
Mali Paraguay
Mozambique Peru
Nigeria Venezuela
Senegal
Swazilanda
Tanzania
Togo
Uganda
Zaire
Zambia
Zimbabwe

a Less than 100,000 ha planted to maize.


Pi.- i ~ wu v. n, .;l~ ra .,'Im- ,-r 0.- opr C


on the germplasm sources private
companies used to develop varieties
and hybrids.

The survey's coverage was extensive.
Of the developing world's important
maize producers, only Argentina,
Chile, and Turkey were excluded from
the survey, since in these countries
virtually all maize is produced in
temperate environments using
germplasm quite different from that
used in tropical and subtropical
environments. The 47 participating
countries comprise approximately 93%
of the non-temperate area planted to
maize in the developing world.

Overall, the response was excellent,
both from national programs as well as
from private companies. In addition to
generating information on the level of
human and financial resources invested
in maize research, the questionnaires
generated detailed information on more
than 1,000 varieties and hybrids
released during 1966-90. This includes
virtually all varieties and hybrids
officially released by public breeding
institutions during that period in the
participating countries (with the
exception of China, where the survey
inventoried only materials released for
the non-temperate zones in Yunnan,
Guizhou, and Guangxi Provinces of
southern China).














countries. For example, the number of
maize scientists per million hectares in
West Asia and North Africa is relatively
high primarily because of Egypt, where
many public sector maize researchers
work on developing technologies for a
small but highly productive irrigated
area. In contrast, the number of maize
scientists per million hectares is rela-
tively low in Latin America because of
Brazil, where a large amount of maize
research is carried out by the private
sector.

Maize researchers' level of training-
The number of scientists working in a
national program provides some indica-
tion of the program's overall strength,
but it is an incomplete measure in that it
does not indicate their skill levels.
Scientists' skills may be assessed by
examining their level of academic
training. Table 2 presents data on
academic degrees held by public sector
maize researchers in selected developing
countries. Among the 47 countries
responding to the CIMMYT survey,
scientists holding bachelor's degrees or
lesser degrees constitute the largest
group (54%), followed by those with
master's degrees (30%) and doctorates
(16%). These aggregate figures conceal
considerable variability among indi-
vidual countries. It is also important to
note that, although many powerful maize
research programs include a high
proportion of scientists with doctoral
degrees, some very strong national
programs China's, for example -
include relatively few scientists with
doctoral degrees.

To provide some context for interpreting
these data, Table 2 also presents the
degree status of agricultural researchers


in Australia, New Zealand, and Canada
in 1981 (note that these figures refer to
all agricultural researchers, not only
maize researchers). When compared to
the figures generated by the CIMMYT
survey, a large disparity is evident at the
Ph.D. level; in general, the developing
countries responding to the CIMMYT
survey employ less than one-third the
proportion of doctoral-level scientists as
do these three industrialized countries.

Expenditure on public sector maize
research-Although skilled human
capital is obviously a prerequisite for
successful research, even the most
accomplished maize scientist is unlikely
to be effective without access to suffi-
cient resources for conducting field
trials, performing experiments, and/or
carrying out surveys. To develop a more
complete picture of the strength of the
global maize research system, it is
necessary to round out the information
on numbers of scientists with informa-
tion on the financial resources support-
ing their activities.


How much are developing countries
spending on maize research, and to what
extent does this figure vary between
countries? The CIMMYT survey asked
for information on the size of national
maize research program operating
budgets in 1990, requesting specific
estimates for five categories of research
expenditure: 1) plant breeding, 2) crop
management research, 3) staff salaries,
4) administrative costs, and 5) general
overhead.3 In most cases respondents
provided data for some, but not all, of
the five categories of expenditure; when
data were incomplete, they were
adjusted upward (using an index
calculated from the data reported by
other countries in the region, supple-
mented by data from Pardey, Roseboom,
and Anderson 1991) to arrive at an
estimated total research expenditure





3 Capital investment expenditures were excluded,
since these tend to vary greatly from year to year.


Table 2. Level of training of public sector maize scientists in selected countries

Doctoral Master's Bachelor's
degree (%) degree (%) degreea()

Egypt 52 23 25
Brazil 36 49 15
Thailand 18 55 27
Mexico 7 64 29
Kenya 3 60 37
China 3 22 75

Group of 47
developing countries 16 30 54

Canada 69 22 9
New Zealand 55 23 22
Australia 41 17 42

Source: CIMMYT survey, Pardey and Roseboom (1989).
a Includes other degrees.














figure corresponding to all five catego-
ries. In a few cases, respondents pro-
vided an estimate of the overall level of
research expenditure but were unable to
disaggregate the figure by category of
expenditure; in these cases, the expendi-
ture data were disaggregated using an
allocation index based on the Pardey,
Roseboom, and Anderson data.

Table 3 presents data on the operating
budgets of national maize research
programs in the 44 countries which
provided usable data. (Unless otherwise
noted, all research cost figures appearing
in this report are expressed in 1990 US
dollars.) In 1989-90, the average maize
program was funded at a level of
US$ 197,000 in Sub-Saharan Africa,
US$ 99,000 in West Asia and North
Africa, US$ 435,000 in Asia, and
US$ 439,000 in Latin America.4 Not
surprisingly, these regional figures
conceal considerable variability, ranging
from less than US$ 15,000 in several





4 The figure for Asia excludes China, for which
budget data were incomplete.


relatively small countries or countries in
which maize is of modest importance, to
over US$ 1,750,000 in large maize
producers such as Brazil, China, and
Mexico.

While these budget data provide an idea
of the average total size of national
maize programs, a rather more detailed
picture of the intensity of publicly
funded maize research emerges when
budget data are related to the size of
each country's maize sector. Table 3
lists data on public sector research
expenditures per million hectares
planted to maize and per million tons of
maize produced. As can be seen,
countries in Sub-Saharan Africa spend
relatively more on maize research per
million hectares planted to maize than
countries in other regions. Although this
result may simply indicate that maize
research is inherently more costly in
Africa than elsewhere, it probably also
reflects the relative lack of private sector
research activity in many African
countries. Compared to other regions of
the developing world, Africa's relatively
high level of public sector research
expenditure per million hectares of
maize area appears even higher when


expressed in terms of expenditure per
million tons of maize produced, because
of the much lower average yields in
Africa compared to other regions of the
developing world.

In interpreting these figures, it is
important to keep in mind that differ-
ences between regions or countries in the
level of research expenditures may
reflect not only differences in the actual
level of research activity, but also
variability in the structure of research
costs. For example, the relative abun-
dance of low-priced labor in Asia
probably contributes to the region's low
level of expenditure per researcher,
whereas Africa's relatively under-
developed transportation and communi-
cation systems undoubtedly raise
research costs. Differences in research
expenditure between regions also may
result from technical factors involved in
compiling the data; for example,
uncertainty about the appropriate
exchange rate for converting budget data
from local currency to US dollars may
have introduced distortions.

When the budget data are expressed in
terms of expenditure per researcher,
average expenditures appear to be much


Table 3. Public sector investment in maize research, 1989-90

Average maize Expenditure Expenditure Expenditure
program budget" per million ha per million t per researcher
(US$ 000s) (LiS$ 000s) (LS$ 000s) I US$ 000s

Sub-Saharan Africa 197 289 228 15
West Asia and North Africa 99 178 53 3
South, East, and Southeast Asia 435 115 38 4
Latin America 439 213 114 16

Group of 44 developing countries 305 176 74 8

Source: CIMMYT survey.
a Budget data unavailable for Costa Rica, Paraguay, and Zambia.


I nrr I I .,, I ..,-J. in ,.I.r.r I anJ lffr,.h^:; .n 1-,. l i ._ l s. 'r..


Al-













higher in Latin America and Sub-
Saharan Africa than in West Asia and
North Africa or Asia. Interestingly, a
similar regional pattern was found by
Pardey, Roseboom, and Anderson (1991)
in their definitive study of aggregate
national research investment.

Level of private sector research
activity-This information on public
sector expenditures on maize research
provides an incomplete view of the
overall level of investment in maize
research, because the private sector also
conducts maize research. Unlike
publicly funded national maize pro-
grams, whose mandates typically oblige
them to address the needs of all of a
nation's maize farmers, private compa-
nies base their research investment
decisions largely on perceived profit
opportunities. Thus, the level of private
sector research activity tends to vary
considerably, being generally higher in
larger countries with well-developed
commercial maize sectors, and lower in
smaller countries where subsistence
farming is the norm.

The relationship between public and
private sector maize research is complex
and highly variable, in some cases
assuming characteristics of competition
and in others characteristics of
complementarity. In industrialized
countries, public sector researchers tend
to focus on more basic research, leaving
the development and sale of finished
varieties and hybrids to the profit-
motivated private sector. In developing
countries, that separation of activities is
often less clear, as publicly funded
researchers frequently assume responsi-
bility not only for the development and
testing of improved materials, but also
for their delivery to farmers.


It is difficult to judge the level of private
sector investment in maize research in
the developing world, because most
private companies regard this informa-
tion as confidential. While avoiding the
sensitive issue of private sector research
expenditures, the CIMMYT survey did
generate data on how many maize
scientists work in the private sector, as
well as information on the types of
activities in which they are involved.
Most private sector scientists in the 47
countries responding to the survey
concentrated on seed production and
varietal testing, activities which can only
loosely be described as research (Figure
2). Relatively few private sector re-
searchers engaged in crop improvement
activities (in other words, basic breed-
ing). This pattern is consistent with the
perception that private companies focus
on applied research (that is, the develop-
ment of finished products), while leaving
more basic breeding activities to the
public sector.


The concentration of private sector
researchers on varietal testing and seed
production is reflected by the growing
presence of private companies in the
maize seed industry. Figure 3 presents
data on the number of public and private
seed companies operating in the 47
developing countries that responded to
the CIMMYT survey. Asian countries
reported the greatest number of seed
companies, with private companies
comprising 62% of the total. Countries
in Latin America reported the second
highest number of seed companies, with
private companies comprising approxi-
mately 88% of the total. Only in Sub-
Saharan Africa did the number of
government seed companies exceed the
number of private companies, which
represent only 27% of the total.5 On the
whole, these data confirm the growing
importance of private seed companies in
the developing world. Two-thirds (67%)
of the 405 seed companies operating in
these 47 countries are private. It is


_I Seed production
I 1 \ netal tein
SCrop improvement


Sub-Saharan
Africa


West Asia and
North Africa


South, East,
and Southeast
Asia


Latin
America


Figure 2. Activities of private sector maize scientists by developing country region, 1990.


'l';llll 'l" '['rl. M ...'c F;..:r. i l ] Tre.nPd ;


H














interesting to note that domestically
owned private companies (nationals) are
either equal (or approximately equal) to
internationally owned private companies
(multi-nationals) in two of the regions,
and more numerous in the other two
regions.

Determinants of research invest-
ment-What factors determine how
much a given country invests in maize
research? Intuitively, one would expect
the level of a country's public sector
investment in maize research to depend
on the absolute importance of the
national maize crop, which would
suggest that the level of investment
increases in proportion with the area




The high proportion of private companies in
Latin America is heavily influenced by data from
Brazil, where many private companies have
regional centers for research and seed production
targeted at the Brazilian market and the markets
of smaller surrounding countries.


planted to maize and/or the value of
maize production. One might also expect
that the level of public sector investment
is influenced by the relative economic
importance of maize, measured for
example by the value of the national
maize crop expressed as a proportion of
agricultural gross domestic product
(GDP). Political considerations might
also be expected to influence the level of
investment in maize research, which
would mean that investment is higher in
countries where maize is an important
staple food. Finally, public sector
investment in maize research might be
expected to be influenced by the level of
private sector research activity, if public
and private research are regarded as
substitutes.

Analysis of the data collected through
the CIMMYT survey did not generate
clear evidence of most of these expected
relationships. Not surprisingly, the size
of national maize programs (as measured
by numbers of maize researchers and


I I Government

Private Multinational (may have branches
in several countries)
S Private National


Sub-Saharan
Africa


West Asia and
North Africa


iE
South, East,
and Southeast
Asia


Latin
America


overall level of investment) was strongly
and significantly correlated with the
importance of the national maize crop
(as measured by area planted to maize
and value of maize production). But no
significant relationships were discerned
between the intensity of the public sector
maize research effort (measured by
expenditure per unit area planted to
maize, expenditure per million tons of
maize produced, number of researchers
per unit area planted to maize, or
number of researchers per million tons
of maize produced) and any of the
hypothesized explanatory variables,
including value of the maize crop as a
percentage of agricultural GDP, percent-
age of the maize crop consumed as food,
percentage of calories in the average diet
derived from maize, and number of
private sector maize researchers. The
absence of discernible quantitative
relationships between these variables
suggests that even though public sector
investment in maize research in many
countries is roughly congruent with the
size of the crop, resource allocation
apparently is not related in any system-
atic way to other major economic and
political factors.

But even if economic and political
factors do not appear to influence the
allocation of resources to maize re-
search, other, technical factors do play a
role. The complexity of the area targeted
for maize research in individual coun-
tries can be measured as the number of
"mega-environments" per million
hectares of maize area. Mega-
environments are production zones, not
necessarily contiguous, delineated by
ecological conditions (temperature,
rainfall, soils), crop characteristics
(maturity cycle, grain color, grain


Figure 3. Number and type of maize seed companies, by region, 1990.


['P ri I %J1 i/ R l .,a.I', In%., :rnI ,:npi indJ i M r I,'r- in 1r P-ni:, U">',lrr : -- -















texture), biotic and abiotic constraints
(pests and diseases), and socioeconomic
factors (production systems, cropping
patterns, consumer preferences). The
measures of research intensity both
those based on numbers of researchers as
well as those based on research expendi-
tures were positively correlated with
the number of mega-environments. This
suggests that technical considerations do
indeed play a role in the research
resource allocation process, since policy
makers evidently allocate more re-
sources to maize when the number of
target environments adds to the com-
plexity of research.

Are there economies of scale in maize
research?-When the countries that
responded to the CIMMYT survey were
divided into five categories based on the
size of the national area planted to
maize, the measures of research intensity
were found to be significantly (and


negatively) associated with national
maize area. In other words, the number
of researchers per unit area planted to
maize and research expenditures per unit
area planted to maize both declined as
the area planted to maize increased,
indicating that the intensity of maize
research is higher in countries with
smaller maize areas (Figure 4). Expendi-
tures per researcher were found to
remain constant over the entire range of
national maize areas, with the exception
of the smallest maize area category, in
which expenditures per researcher were
found to be significantly higher (Table
4). This last pattern suggests that there is
a minimum size below which a maize
research program cannot be productive,
for below this "threshold" level the
program will lack a critical mass of
scientists and facilities. If this is true,
countries with limited maize areas may
be overinvesting in maize research
because they are not able to reduce the


size of their national program below this
threshold level.6

In contrast to the inverse correlation
found between national maize area and
both measures of public sector research
intensity, no consistent relationship was
found between the size of the national
maize area and the number of private
sector researchers. This suggests private
companies' decisions to invest in maize
research are directed not only by the size




6 Brennan (1992) has addressed the interesting
issue of whether there is a minimum threshold
size for an efficient plant breeding program.
Brennan concludes that some sort of
collaborative regional research effort would
appear to make sense from an economic point of
view whenever several small countries share a
similar production environment. Brennan's
conclusion appears to be supported by these
empirical data showing significantly higher
levels of investment intensity in countries with
small maize area.


240

200


A Taiwan (ROC)


Senegal
---- --



Cameroon

Zdmbia
Ghana

Pakistan


China


Brazil


A- Brzi


01


120 1


IndiaA Mexico


0 2 4 6 8 10 12 14
National maize area (million ha)


16 18 20 22


Figure 4. Relationship between maize researchers per unit area planted to maize and national maize area,
44 developing countries.





. ,i 1. ar.. e... .. .














of the potential commercial market in a
given country, but also by factors
influencing the companies' ability to do
business there (e.g., the legal climate,
marketing infrastructure).

Research investment in maize com-
pared to other crops-How does the
overall level of investment in maize
research compare to the level of invest-
ment in other important food crops?
Data on the level of investment in
research on rice, wheat, and maize
disaggregatedd by NARSs and IARCs)
were assembled for 25 developing
countries by Judd, Boyce, and Evenson
(1983) (Table 5). As a proportion of the
value of production, NARSs' investment
in maize research was roughly equal to
investment in rice research, although
both were considerably lower than
investment by NARSs in wheat research.


Investment by IARCs in research on all
three crops was found to be considerably
lower than that of NARSs, averaging a
miniscule 0.02-0.03% of the value of
production. Interestingly, IARCs were
found to support a larger proportion of
public sector research for maize than for
the other crops.



Products of the Global
Maize Research Network

What products are generated by the
global maize research network? Two
basic products can be distinguished:
improved germplasm (maize character-
ized by high yield potential and resis-
tance to a wide range of biotic and
abiotic stresses) and improved crop
management practices (which enable
farmers to increase yields, reduce


production costs, and/or increase
cropping intensity). Of the two, im-
proved germplasm is easier to identify
and measure, since it is a tangible
product whose origin can usually be
determined unequivocally. Improved
crop management practices are more
difficult to identify and measure, since
they are based on information whose
origin frequently cannot be established
with certainty. (Many "improved" crop
management practices cannot be
attributed to the efforts of researchers,
but rather to farmers, who developed
them through trial-and-error experimen-
tation).7 Because more data are available
on improved germplasm, the remainder



7 Traxler and Byerlee (1992) discuss the issues
involved in measuring the impacts of crop
management research.


Table 4. Relationship between maize research indicators and size of national maize area, 1989-90

Public Expenditure Expenditure Private
researchers per per researchers
per million hab researcher per
National maize area million ha (tUSS000s) iLIS$ O00s million ha

< 175 thousand ha 89 1,662 24 26
175 435 thousand ha 52 571 12 11
436 650 thousand ha 39 324 13 17
651 1,500 thousand ha 27 203 12 10
> 1,500 thousand ha 16 110 12 11

Source: CIMMYT survey.
a Size ranking based on five-year average using 1985-89 data.
b Expenditure data unavailable for Costa Rica, Paraguay, and Zambia.


Table 5. Public sector research as a proportion of value of production, average 1972-79 (25 countries)

Latin Total IARCs Ratio
Commodilt Africa Asia America NARSs spending IARCs:Total

Maize 0.44 0.21 0.18 0.23 0.03 0.13
Rice 1.05 0.21 0.41 0.25 0.02 0.08
Wheat 1.30 0.32 1.04 0.51 0.02 0.04

Source: Judd, Boyce, and Evenson (1983).


P U -,R. r-hI rr nri iI ~ ,i D I nI -














of this report focuses primarily on this
product of research. Since relatively
little information is available about
improved maize materials developed by
private companies, the emphasis is on
germplasm produced by public sector
institutions.

Number of varietal releases in
developing countries-Data on the
number of maize varieties and hybrids
released by public sector breeding
institutes in individual countries from
1966 to 1990 appear in Part 3 of this
report. The data are presented in two
forms: 1) total number of releases and
2) number of releases per million
hectares of national maize area (this
standardized measure of the rate of
varietal releases facilitates comparisons
between countries with small and large
maize areas).

Three interesting patterns appear in the
varietal release data. First, the rate of
releases per million hectares decreases


as national maize area increases (Figure
5). However, since some of the countries
that have released the fewest maize
varieties per million hectares maintain
quite accomplished national breeding
programs (Brazil, China, and India, for
example), success in maize breeding
probably should not be measured simply
by the number of varieties released.
Second, the rate of varietal releases
increases when the target environment is
more complex (as measured by the
number of mega-environments per unit
area). This finding indicates that national
breeding programs are sensitive to the
need to develop materials adapted to
distinct production conditions. Third, the
rate at which varieties are released
appears to be inversely related to their
adoption by farmers, since the percent-
age of commercially successful varieties
declines as the rate of releases increases.
(Varieties were classified as commer-
cially successful if they were adopted on
5% of the national maize area, or at least
25,000 ha.)


(35%)


Figures in parentheses denote percentage of
commercially successful releases
100


80
- 80

,U
P,







(47%)


<175 175-435 436-650 651-1500 >1500
National maize area (000 ha)

Figure 5. Rate of maize varietal releases as a function of maize area in 45 developing
countries, 1966-90.
countries, 1966-90.


This inverse relationship between the
rate of varietal releases and commercial
success lends itself to various interpreta-
tions. One interpretation is that in
countries where certification procedures
are more stringent (resulting in a lower
rate of varietal releases), released
varieties are likely to exhibit clear
superiority over currently available
materials and therefore stand a higher
chance of being adopted by farmers.
Another possibility is that when few
varieties are released farmers' choice of
varieties is limited. Therefore, more
farmers are likely to adopt a given
variety regardless of its performance,
making relatively more varieties appear
successful. A third interpretation is
related to the size of the mega-environ-
ments in individual countries. There is
some evidence that when more mega-
environments are found in a given
country each mega-environment is
smaller. In this case the target area for
each variety is limited, and so is the
likelihood that a given variety will meet
the criterion for commercial success.

Types of maize varieties released by
the public sector-Among the different
types of maize varieties released by
public sector breeding institutes between
1966 and 1990, open-pollinated materi-
als have predominated, comprising
nearly two-thirds of all materials
released (Table 6). However, public
sector institutes have also been active in
releasing conventional and non-conven-
tional hybrids, which together comprise
approximately one-third of all releases.

Table 7 presents data on the ecological
adaptation of maize materials released
by public sector breeding institutes from
1966 to 1990. Materials adapted to
lowland tropical environments have
predominated, comprising just over one-


i, -', ,"














half (52%) of all releases. Materials
adapted to subtropical/midaltitude
environments make up the next most
important group (35%), followed by
materials adapted to highland (11%)
environments. Broadly adapted materials
developed for transition zones comprise
the remaining 2%. The final line in
Table 7, which provides data on the area
planted to maize in each ecological
niche in the developing world, indicates
that the distribution of public sector
releases is highly congruent with the
relative area planted to maize within
each niche.


Historical trends in the rate and mix
of varietal releases-Figure 6 depicts
the pattern of maize varietal releases
over time. Despite showing some
variability, the rate of varietal releases
generally accelerated during the 1960s
and 1970s before leveling off in the
1980s, suggesting that the rate of
releases has stabilized now that many
national maize breeding programs have
reached maturity.

Closer examination of the varietal
release data indicates that the mix of
materials produced by national programs


has varied through time (Figure 7).
Throughout the 1960s, public sector
maize breeders in developing countries
placed greater emphasis on hybrids than
on OPVs. The emphasis on hybrids
presumably reflected the fact that maize
breeding programs in many developing
countries were modeled on breeding
programs in industrialized countries,
which at that time were generally
oriented toward hybrid development.

During the 1970s, however, the mix of
materials released by national programs
changed, with the proportion of OPVs


Table 6. Public sector maize releases in developing countries, by type of material, 1966-90 (%)a

Open-pollinated Conventional Non-conmentional
Region varieties hybrids hybrids Othersh Total

Sub-Saharan Africa 59 25 15 1 100 (295)
West Asia and North Africa 14 68 18 0 100 (28)
South, East, and Southeast Asia 62 16 8 14 100 (180)
Latin America 61 32 6 1 100 (349)

All developing countries 59 27 10 4 100 (852)
(503) (230) (85) (34)

Source: CIMMYT varietal database.
a Values in parentheses are total number of releases in each category.
b Includes synthetic varieties and other materials not classified under the previous three categories.



Table 7. Public sector maize releases in developing countries, by ecological adaptation (%)a

Tropical Subtropical/
Region lou land midaltilude Highland Otherb Total

Sub-Saharan Africa 48 36 9 7 100
West Asia and North Africa 0 100 0 0 100
South, East, and Southeast Asia 43 55 2 0 100
Latin America 64 16 18 2 100

All developing countries 52 35 10 3 100

Maize area located in this environment 60 28 11 1 100

Source: CIMMYT varietal database.
a Based on 852 varieties and hybrids released from 1966 to 1990.
b Includes materials classified in more than one of the previous three categories.


I rr I ,1 1' I, :r-' I1 1 I~ N n r.1 mi~ i.. F r rI. ..: i,,~ h : ,J,;














rising and the proportion of hybrids
declining. This development was
attributable in large part to a scarcity
during this period of improved
germplasm needed for the production of
tropical hybrids. Since many of the
hybrids released during the 1970s did




240 --
SLatin America
200- South, East, and Southeast
SWest Asia and North Afric
SSub-Saharan Africa
." 160--


d


-i0 i


1966-70 1971-75 1976-80


not significantly outperform the best
OPVs, national programs often chose to
concentrate their breeding efforts on
OPVs, which were less expensive to
develop, and seed of which was easier to
produce.








Asia
a


1981-85 1986-90


Figure 6. Rate of maize varietal releases, by developing country region, 1966-90.


100


All national programs
Large national programs
- Small national programs


2"


0


1961-65 66-70 71-75


76-80


81-85


86-9


0


In the 1980s, breeders in many develop-
ing countries began to shift back toward
an increased emphasis on hybrids. In one
sense, the shift was a natural develop-
ment which occurred as more and more
national programs had the resources to
support more sophisticated hybrid
development schemes. But the shift also
reflected the increased availability of
improved germplasm suitable for the
development of tropical hybrids. As the
result of years of sustained breeding
efforts, improved materials had become
available from which superior inbred
lines could be extracted for use in hybrid
crossing programs. As hybrids developed
from these materials began to signifi-
cantly outperform OPVs, breeders in
national programs naturally shifted their
breeding strategy. The international
centers supported the change and
stepped up their own efforts to support
the production of materials suitable for
use in hybrid breeding programs,
including inbred lines.

Interestingly, even though large (and
therefore generally more sophisticated)
national programs have consistently
released a greater proportion of hybrids
than small national programs, programs
of all sizes seem to have experienced
similar changes in breeding strategies
through time (Figure 7).

Figure 8 shows the proportion of hybrids
released by national breeding programs
during the 1960s, 1970s, and 1980s in
each region of the developing world.
Quite clearly, the mix of materials has
varied significantly between regions.
National programs in Sub-Saharan
Africa have consistently released more
OPVs than hybrids, reflecting the fact
that relatively few national programs in
Africa are actively involved in hybrid


Figure 7. Proportion of hybrids in national program releases, 1961-90 (five-year moving
average).


.. ."". ...
6(


'V. r M..rc Fair; i, I ,rrnd















development. The proportion of hybrids
released has remained relatively stable
across the past three decades, indicating
little or no change in public sector
breeding strategies and capacities.

National programs in West Asia and
North Africa have released the highest
percentage of hybrids of all four regions.
This is hardly surprising, considering the
large influence of Egypt in the regional
data. Egypt has an extremely strong
national maize program, which for many
years has emphasized the development
of hybrids.

Several countries of South, East, and
Southeast Asia support large national
maize programs, and one would expect
data for the region to reflect a high
percentage of hybrids among national
program releases. Thus, the sharp
decline in the proportion of hybrids
released during the 1970s runs contrary
to expectations. However, this pattern
resulted from the phenomenal success of


Sub-
Saharan
Africa


West Asia
and North
Africa


Suwan-1, a high-yielding OPV released
in Thailand in 1974 (see the box,
"NARSs Helping NARSs: The Story of
Suwan-l," p. 14). Suwan-1 proved to be
almost unbeatable, even by hybrids, and
throughout Asia the eventual appearance
of numerous materials with Suwan
parentage sharply increased the propor-
tion of OPVs among national program
releases during the 1970s. More re-
cently, breeding efforts in Asia have
shifted back toward increasing emphasis
on hybrids.

National programs in Latin America
have released progressively fewer
hybrids during each of the past three
decades. This trend can be attributed to
several factors. First, declining levels of
support to some national programs in
Latin America from private philan-
thropic organizations such as the
Rockefeller and Ford Foundations
curtailed expensive hybrid breeding
activities. Second, as private seed
companies have become increasingly


I 1960s

I 1J97 1(
E ay 18L


South, East,
and Southeast
Asia


Latin
America


Figure 8. Proportion of hybrids released by national programs in developing
countries, by decade.


active in the development and release of
hybrids in Latin America, some national
programs have deliberately chosen to de-
emphasize hybrids in favor of OPVs.

Global movement of maize
germplasm-Maize breeding has
become a truly global effort involving
national breeding programs, interna-
tional research centers, and private
companies. Formal and informal
networks effectively disseminate a wide
range of improved germplasm -
materials developed by national program
breeders and/or emerging from interna-
tional breeding programs as well as
seed of unimproved maize landraces and
other exotic materials maintained in
germplasm banks. (For a discussion of
how the changing economic and legal
environment of research might affect
this global germplasm network, see the
box, "Intellectual Property Rights and
the CGIAR," p. 16).

The effectiveness of CIMMYT's
germplasm distribution network (see the
box, "A Global Maize Germplasm
Distribution Network," p. 19) is re-
flected in the increasing incidence with
which materials distributed through the
network appear in varieties and hybrids
released in developing countries. Figure
9 illustrates historical trends in the
proportion of public sector releases
containing CIMMYT germplasm.s
During the years immediately following
CIMMYT's founding in 1966, relatively
few of the Center's materials appeared
in national program releases. However,





8 It is important to recognize that materials coming
out of the CIMMYT breeding program are a joint
product of the close collaboration between
CIMMYT scientists and national program
researchers. For brevity, this report refers to such
materials as "CIMMYT germplasm."


Pi'i i N I i N, ., :~ r In. mn r I -r ac r 0, r .-g C .'r. I


,-1







1.1 "1 j I rn1 r,,


NARSs Helping NARSs: The Story of Suwan-1


The spread of improved maize
germplasm from CIMMYT's breeding
program in Mexico to national breed-
ing programs throughout the develop-
ing world is a familiar story. Much less
widely known is the fact that improved
materials developed within some
national programs also have been
adopted extensively in several parts of
the developing world. A good example
of such a material is Suwan-1, a late-
maturing yellow flint variety developed
in Thailand.

In the late 1950s, the government of
Thailand began to promote maize
production as part of a national policy
designed to diversify agriculture away
from rice. As Thai farmers began to
respond to increased maize production
incentives, they were forced to rely
heavily on varieties and hybrids
introduced from the Americas (locally
adapted improved varieties were
largely unavailable). Dissatisfaction
with the performance of these exotic
materials soon led to a push for the
development of new germplasm
showing better adaptation to local
production conditions.

Germplasm improvement efforts began
in earnest in the mid-1960s, spurred by
two key institutional developments.
First, in 1966 the Rockefeller Founda-
tion moved its Inter-Asian Corn
Program headquarters from India to
Thailand, resulting in an immediate
injection of financial resources and
experienced human capital. At the
same time, Thailand's previously
fragmented national maize breeding


efforts were consolidated into a single
program run by Kasetsart University and
the Ministry of Agriculture. The
Rockefeller Foundation and Thai
researchers established a collaborative
breeding program at Farm Suwan, a
state-of-the-art research station in central
Thailand where irrigation permitted
planting up to three crops each year.
Close collaboration with CIMMYT
began after the Center's establishment in
1966. Besides sending germplasm to
Thailand for evaluation and use in their
breeding program, CIMMYT provided
training to many Thai maize researchers.

In a departure from previous breeding
strategies, which had stressed screening
of imported varieties and hybrids, the
collaborative research team at Farm
Suwan decided to concentrate on
developing a genetically heterogeneous
composite material that would incor-
porate germplasm from a wide range of
sources. Eventually, 36 germplasm
sources from many parts of the world -
which were mostly received through
Mexico and India were selected to
form Thai Composite #1. Work on this
population continued throughout the late
1960s and early 1970s, leading to steady
improvements in grain yield and other
agronomic characteristics. Beginning in
1971, after it had became apparent that
downy mildew caused by Sclerospora
sorghi posed an important new con-
straint to maize production in Southeast
Asia, downy mildew resistant sources
from the Philippines were crossed in.
Following several additional cycles of
full-sib recurrent selection, Suwan-1 was
released in Thailand in 1974.


In the years immediately following its
release, Suwan-1 was rapidly adopted
throughout the main maize production
zones in central Thailand. The variety's
success could be attributed in large part
to its superior performance compared to
previously available materials, espe-
cially its resistance to downy mildew.
Favorable institutional conditions also
facilitated the spread of Suwan-1, in
particular the existence of a well-
organized private sector input supply
industry, which was able to deliver seed
and fertilizer efficiently to most farmers.

The obvious success of Suwan-1 soon
generated interest outside Thailand.
Requests for seed came first from
neighboring countries in Southeast Asia,
many of which soon released the variety.
Eventually, these initial requests for seed
were followed by requests emanating
from further away. As more maize
breeders got a chance to evaluate
Suwan-1, the variety's superior combin-
ing ability became apparent, leading to
its increasing use as a breeding material.
Gradually, Suwan-1 established itself as
one of the most widely used breeding
materials in the tropical world. By 1990,
Suwan-1 was present in 22 varieties and
hybrids released by national breeding
programs in 13 countries; these varieties
and hybrids were being grown on
approximately 1.4 million hectares (see
figure). To a considerable extent, this
wide-scale use of Suwan-1 by national
programs was made possible through
CIMMYT's global germlasm exchange
system.

But the story of Suwan-1 does not end
here; like many good stories, it has a few
sequels. Suwan-2, an earlier maturing







I In i R, rh r.~ r.r,. ,1 fir IRa..


version of Suwan-1, was developed by
the national program for farmers who
could not quite fit Suwan-1 into their
cropping systems. And Suwan-3, which


yields 10-15% more than Suwan-1 and
contains additionally improved
germplasm from CIMMYT, is now
rapidly replacing Suwan-1 in Thailand.


Origins of germplasm used in developing Suwan-1.
ANTIGUA, BARBADOS, BRAZIL, COLOMBIA, CUBA, EL SALVADOR, GUATEMALA, GUAYANA, INDIA, MEXICO,
PERU, PHILIPPINES, PUERTO RICO, USA.


Countries in which Suwan-1 has been released or used as a breeding material.
BRAZIL, CAMEROON, CHINA, COLOMBIA, INDIA, INDONESIA, NEPAL, PHILIPPINES, SENEGAL, THAILAND,
VENEZUELA, VIETNAM, ZAIRE.


as progress was made in improving
CIMMYT's pools and populations,
NARS breeders began to adopt them
with increasing regularity, and by the
late 1980s over 75% of all NARS
releases contained at least some
CIMMYT germplasm. Many of the
releases containing such germplasm are
lowland tropical materials, since that is
where CIMMYT has historically placed
greatest emphasis.

While the IARCs have served as an
important source of improved
germplasm for national breeding
programs, NARS breeders also take
advantage of materials developed by
other national programs (these materials
may be distributed directly from one
national program to another, or through
the IARCs). Several germplasm com-
plexes developed by NARS breeders
have been used widely throughout the
developing world as parents for locally
developed materials; Suwan-1 is an
outstanding example.



Impact of Improved
Maize Germplasm

The global maize research system of
NARSs and IARCs has clearly been
successful in developing and releasing
improved maize varieties and hybrids.
However, as indicated above, the
number of varietal releases is an imper-
fect measure of research impacts,
because many varieties released for
commercial distribution are never
adopted widely by farmers. A better
measure of the impact of breeding
research is the area planted to improved
materials, since this provides an indica-
tor that can be used in directly estimat-
ing the additional maize production
attributable to improved germplasm.






LV ri N. j .,J r-rd


Intellectual Property Rights and the CGIAR


Over the past decade, a large amount
of agricultural research activity has
shifted from the public to the private
sector, particularly in the industrialized
countries. As privately funded agricul-
tural research organizations have
assumed an increasingly important role
in the development of new technolo-
gies, their willingness to afford free
access to the fruits of their research has
frequently conflicted with the need to
generate financial returns. At the same
time, public sector research budgets in
many countries have been declining in
real terms, causing some research
administrators to view research
products as a possible source of new
funding which will enable them to
continue to operate for the benefit of
the public. These two trends the rise
of private sector research and the
decline of public sector research -
have led to increased calls in certain
sectors for stronger laws governing the
protection of intellectual property
rights (IPR), so that those who develop
new technologies can generate finan-
cial returns from their sale, or merely
prevent the expropriation of these
technologies by others.

The Consultative Group on Interna-
tional Agricultural Research (CGIAR)
now finds itself caught up in a far-
ranging, emotional debate about
whether and how IPR legislation
should be applied in the developing
world to the products and processes of
agricultural research. At one extreme


of this debate lie biotechnology firms,
private seed companies, and some public
research institutes, all pushing for IPR
legislation to strengthen the ability of
research organizations to exercise
control over the technologies they
develop. Aside from seeking to make
existing IPR legislation stronger, these
institutions favor extending coverage
into new areas. At the opposite extreme
lie advocacy groups which argue that
improved agricultural production
technologies should be made freely
available to all, particularly improved
plant varieties developed at least in part
using germplasm originally collected
from the fields of Third World farmers.
A number of intermediate points of view
are put forward by groups who recognize
the desirability of strengthening incen-
tives for investment in research, but who
also recognize the need to maintain
access to the fruits of research by those
who may be unable to exercise effective
demand in the market.

How do these developments affect the
international agricultural research
centers of the CGIAR? The basic goal of
the CGIAR is to increase the productiv-
ity of resources devoted to agriculture,
with special emphasis on resource-poor
farmers in developing countries.
Achievement of this goal is generally
assisted by free and open exchange of
technology, including products (e.g.,
improved germplasm, laboratory
enzymes, machinery), processes (e.g.,
laboratory techniques) and information
(e.g., breeding strategies, information on
genetic sequences and gene functions).


The CGIAR centers facilitate flows of
technology and scientific information
through a spectrum of research organiza-
tions ranging from national agricultural
research systems, universities, and
private companies. However, the
CGIAR recognizes that policies de-
signed to facilitate free and open
exchange of technology are often
disadvantageous for private companies,
which by their very nature depend on the
ability to capture economic benefits
from the technologies they generate. To
the extent that private companies are
discouraged from investing in research,
farmers and consumers everywhere
lose out.

The financial incentives provided by IPR
legislation can work to the benefit of
farmers in developing countries, as
evidenced by the success of private
maize seed companies in countries such
as Guatemala, India, Kenya, Mexico,
Thailand, and Zimbabwe. But while IPR
legislation has clear advantages in that it
can stimulate increased investment in
private sector research, at the same time
it could potentially create hazards for
publicly funded research organizations
by compromising their ability to work
for the benefit of their target clientele.
For example, with the strengthening of
IPR legislation, CGIAR centers might be
required to restrict the use of products
and processes developed using inputs or
procedures available only under com-
mercial licensing agreements.

Recognizing the importance of IPR
issues, the CGIAR is currently involved
in a process of information gathering,
reflection, and debate designed to lead to
a CGIAR policy or at least a set of







i'.ri I ki i.- Re-e 'v, i -.IrrnIn ,,id I rr P- dJ. nn ..u.,I r- : A M


guidelines on IPR. This process has
explicitly sought to incorporate the
widest range of views, not only from
within the CGIAR system but also
from NARSs, farmer groups, the donor
community, and representatives from
the private sector. While details have
yet to be worked out, a broad consen-
sus is emerging that the CGIAR policy
should carry the strongest possible
commitment to maintain free access to
products and process of research, both
for the CGIAR centers and for their
clients in the developing countries.

The CGIAR cannot simply ignore these
issues, because to serve its clients
effectively it must maintain access to
products and processes developed by
private companies. At the same time,
the CGIAR clearly has a responsibility
to protect the interests of its clients,
particularly NARSs and farmers in
developing countries. Thus, CGIAR
centers have an important role to play
in supporting development of "rules of
the game" which will allow both public
and private sector research organiza-
tions to operate efficiently and equita-
bly, both in countries which choose to
implement IPR legislation as well as in
those which do not. The challenge will
be to create sufficient economic
incentives to ensure socially optimal
levels of private sector investment, yet
at the same time to protect the interests
of those who may lack the economic or
political power to compete effectively
in the marketplace.


Table 8 presents data for 1990 on the
area planted to improved maize varieties
and hybrids in 45 important maize-
producing countries. Since the genetic
composition of open-pollinated maize
varieties can change rapidly in farmers'
fields through natural crossing with other
maize cultivars, these estimates refer
only to area planted to seed of improved







240


varieties and hybrids purchased within
the previous three years. Overall, 50% of
the total maize area in these countries
was planted to improved materials. Use
of improved germplasm is proportionally
highest in South, East, and Southeast
Asia (57%) and West Asia and North
Africa (57%), followed by Sub-Saharan
Africa (47%) and Latin America (46%).


SContaining no CIMMYT germplasm

JA Ccontanimng CI Nl Nl gcmlplani


121)








1966-70 1971-75


1976-80 1981-85


1986-90


Figure 9. Incidence of CIMMYT germplasm in maize varieties and hybrids released by
public sector institutions in developing countries, by five-year period, 1966-90.




Table 8. Area planted to improved maize varieties and hybrids in the tropics and
subtropics, by region (1990)

National Area under Area under
maize area improved improved varieties
in 1990 varieties (% of national
Region (million ha) (million ha) maize area)

South, East, and Southeast Asia 19.0 10.8 57
West Asia and North Africa 1.2 0.7 57
Sub-Saharan Africa 14.4 6.8 47
Latin America 23.8 10.9 46

All developing countries 58.5 29.2 50

Source: CIMMYT survey.
Note: Argentina, Chile, Turkey, and temperate maize area of China excluded.






V. ] P.., I j- j I J


Highland
I (5.9Et


Figure 10. Distribution of area planted to improved maize germplasm in developing
countries (non-temperate production environments). 1990.


Disaggregated b) Region:


Subiropical jnd
midaliitude
I 3.8





QJ


Highland
(I 14.0 r)


Distribution orarea planted to improved
maiie germplasm. Sub-Saharan Africa.


Highland

/'Subtropical and 0 4%"
middhitude
L~ (26.8%)


Distribution of area planted to improved
maize germplasm. South. East. and
Southeast Asia.


0
Subtropical and
midaltirude
(100 )


Distribution of area planted to improved
maize germplasm. West Asia and North
Africa.


Subtropical and
midalitude
3.4%) r-


Distribution of
maize germplas


While the figure of 29.2 million hectares
planted to improved varieties and
hybrids represents a significant achieve-
ment for the global maize research
network, the proportion of the develop-
ing world's maize area planted to
improved materials is nevertheless lower
than the equivalent figures for rice and
wheat. The discrepancy can be attributed
to two principal factors. First, many of
the developing world's maize farmers
plant local varieties selected over
hundreds of years to satisfy complex sets
of production and consumption require-
ments; until considerable location-
specific adaptive breeding is done,
improved varieties and hybrids rarely
meet these requirements. In contrast,
wheat and rice tend to be grown over
large, relatively homogeneous produc-
tion environments (including irrigated
zones) which can be more easily
targeted by breeders. Second, farmers'
adoption of improved maize generally
depends on a well-functioning seed
industry, which is still lacking in many
developing countries. Once again, maize
differs in this respect from self-pollinat-
ing species such as rice and wheat, seed
of which can be passed directly from
farmer to farmer without much loss of
genetic purity.

Figure 10 shows how the area planted to
improved maize varieties and hybrids is
distributed across ecological niches.
Improved germplasm has had its greatest
impact in lowland tropical zones, since
fully two-thirds of the area planted to
improved materials is located in this
environment. This result is to be
expected, given that 60% of the develop-
ing world's non-temperate maize area
lies in lowland tropical environments


area planted to improved
m. Latin America.















A Global Maize Germplasm Distribution Network


The world's largest formal maize
germplasm distribution network is
administered by the International
Testing Unit of the CIMMYT Maize
Program in collaboration with scien-
tists in over 80 countries. The main
objectives of this network are to
facilitate the evaluation of families of a
given maize population through
standard trials in a range of environ-
ments and to distribute experimental
materials for use in local breeding
programs.

The standard trials distributed by
CIMMYT consist of maize germplasm
at an advanced stage of improvement.
These trials are organized according to
ecological adaptation (lowland tropi-
cal, subtropical/midaltitude, highland),
maturity (early, intermediate, late), and
grain color (white, yellow). Special
purpose trials aimed at evaluating
specific characteristics are offered on
an irregular basis. Each year a letter
listing the various trials available is
sent to approximately 300 maize
researchers in both developing and
industrialized countries. Although the
majority of these cooperators work in
national breeding programs, CIMMYT
trials are also grown by researchers at
universities, non-profit organizations,
and private companies.'


In response to a request for a trial,
CIMMYT provides the cooperator with
seed, instructions for planting and
managing the trial, forms for recording
data, and documentation for moving the
seed through local phytosanitary and
customs channels. In return, the coopera-
tor agrees to grow the trial according to
the instructions, to collect data on the
performance of trial materials, and to
report the data back to CIMMYT in a
timely manner. At a minimum, research-
ers must report data on the number of
days to flowering, grain yield, and grain
moisture content. Other information
typically reported includes plant and ear
height, degree of lodging, and resistance
to major maize pathogens.

Once the data are received at CIMMYT
headquarters, they are analyzed and
returned to the cooperator. Results are
published twice yearly in the Interna-
tional Maize Testing Reports, which
include summaries of trial results for
individual locations, as well as com-
bined analysis across all sites in which a
particular trial was grown. The com-
bined analysis is a useful indicator of the
stability of a material's performance
under a variety of en ironmental
conditions.


After examining results of a trial for
individual locations and across loca-
tions, a cooperator maj decide that one
or more materials merit further
investigation and request seed from
CIMMYT. Following additional
testing, national programs sometimes
decide to recommend a CIMMYT line
for immediate release and use by
farmers, w without additional selection.
More commonly, however. additional
selection is done prior to release in
order to improve local adaptation.
Other uses of CIMN1YT experimental
materials include their incorporation
into population improvement pro-
grams, employ meant a, a parent in a
non-conventional hybrid, and use as a
source material for a specific trait or
for the extraction of inbred lines. The
demand for CIMMYT inbred lines has
become particularly strong as public
sector interest in hybrid de-velopment
has increased (see p. 12); these
materials are expected to make an
important contribution to h brids in
developing countries in the near future


In processing requests for trials, the
Iniemationai Teling Unit gi.e, preference
to requevis from nallondl program,


a1















The Tuxpefio Germplasm Complex


Tuxpefio, one of approximately 250
maize landraces found in the New
World, was developed along the
lowland tropical coast of eastern
Mexico over many hundreds of years
of selection in farmers' fields. Al-
though its origins cannot be known
with certainty, it is believed that
Tuxpefio was formed by hybridization
between two other landraces, Olotillo
and Tepecintle, which in turn may have
resulted from hybridization between
floury materials from South America
and teosinte, a wild relative of maize
(CIMMYT 1986). Tuxpefio, a white
dent material, is distinguished by its
tallness (averaging 3-4 m in its native
habitat), good resistance to ear rots and
foliar diseases, and an inherent
disposition to respond to conditions of
high fertility. Although most heavily
concentrated around the port city of
Tuxpan in the state of Veracruz, the
landrace is widely distributed from the
northeast of Mexico to the Yucatan and
beyond.

During the 1940s, while collecting and
classifying germplasm for use in the
world's first global maize improvement
program, Rockefeller Foundation
scientists and their colleagues in the
Mexican Ministry of Agriculture were
quick to recognize the superior
performance of Tuxpefio materials.
However, they realized that the
original landrace collection had
considerable drawbacks and would
have to undergo extensive refinement
before it could serve as a successful
breeding material. Major problems


included the excessive plant height, the
relatively high ear placement, and the
low harvest index (the plant's ratio of
grain to dry matter). The Rockefeller
Foundation and Mexican Ministry of
Agriculture scientists, and later their
successors in the CIMMYT Maize
Program, therefore started an intensive
selection program to overcome these
undesirable characteristics.

Early attempts to reduce the height of
Tuxpefio materials focused on the
introduction of dwarfing genes. How-
ever, this strategy which in rice and
wheat led to the well-known "Green
Revolutions" proved unsuccessful in
the case of maize. Plants carrying
dwarfing genes suffered from greatly
shortened lower intemodes and conse-
quently proved highly susceptible to
lodging. CIMMYT breeders therefore
opted for the slower but eventually more
successful strategy of bringing down
plant height through recurrent selection
methods. Over a 10-year period, a
dramatic 37% reduction in plant height
was achieved in 15 cycles of recurrent
selection. Meanwhile, by selecting for
such traits as reduced tassel and leaf
size, CIMMYT breeders raised the
harvest index from 0.30 to 0.45, which
had the effect of increasing yield
potential at a rate of 4.4% per cycle. The
time needed for the plant to reach
maturity was reduced by nearly 0.5% per
cycle, significantly shortening the
plant's growth cycle and thereby
improving its ability to escape drought
late in the growing season.


As the superior performance of Tuxpefio
germplasm became apparent, a number
of distinct breeding populations were
developed by breaking off fractions of
the original population and applying
increased selection pressure for different
traits (see figure). Five current
CIMMYT breeding populations claim
direct Tuxpefio ancestry: Population 21
(Tuxpeiio 1), Population 43 (La Posta),
Population 49 (Blanco Dentado-2),
Population 63 (Tuxpefio o2), and
Tuxpefio Sequfa. Many other CIMMYT
breeding populations contain lesser
amounts of Tuxpefio germplasm,
including Population 22 (Mezcla
Tropical Blanco), Population 24
(Antigua-Veracruz 181), Population 26
(Mezcla Amarilla), Population 27
(Amarillo Cristalino 1), Population 28
(Amarillo Dentado), Population 29
(Tuxpefio Caribe), Population 44
(Tuxpeiio AED), Population 45 (Ama-
rillo Bajio), and Population 46
(Templado Amarillo Cristalino).
Tuxpefio germplasm has also been used
extensively by the IITA Maize Program,
whose popular TZPB and TZPB-SR
breeding populations were directly
derived from Tuxpeiio sources.

During the past 25 years, Tuxpefio
germplasm has been widely distributed
in the developing world, primarily
through CIMMYT's international trials
(which have included numerous experi-
mental varieties developed using
Tuxpefo materials), as well as through
the Center's Maize Germplasm Bank
(which contains over 850 separate
Tuxpefio landrace accessions). In
addition, significant amounts of
Tuxpefio seed have been distributed
through the Mexican national maize


'"







Part 1: Maize Research Investment and Impacts in Developing Countries


program. Tuxpefio's superior perfor-
mance, as well as its excellent combin-
ing ability, have made it a favorite
source material for many national
breeding programs. Between 1966 and
1990, NARSs released 147 varieties and
hybrids developed from CIMMYT's and
IITA's "mainly" Tuxpefio populations;
in 1990, these varieties and hybrids were
planted on approximately 3.8 million
hectares. Private companies have also
made extensive use of Tuxpeio
germplasm, although few data are
available on the number of private sector
releases claiming Tuxpeio parentage,
nor on the area planted to these releases.
Tuxpefio germplasm has even made its
way into temperate production zones; for
example, Tuxpefio materials are widely
recognized for their strong influence on
the dent maizes of the southern USA, as
well as on materials of the North
American Corn Belt.


The extraordinary popularity of
Tuxpefio germplasm testifies to the
ability of the global maize research
system to identify, improve, and
distribute superior germplasm. Al-
though the extensive use of Tuxpefio
materials might raise questions about a
narrowing in genetic diversity at the
global level, maize breeders point out
that the original Tuxpefio landrace
collections contained a wide range of
genetically distinct materials. Conse-
quently, varieties and hybrids devel-
oped from Tuxpefio sources retain
tremendous genetic diversity despite
their common heritage. This diversity
is further extended by the tendency of
open-pollinating varieties to outcross in
farmers' fields, which ensures that their
genetic composition is continually
changing.


Tuxpeno race
reservoir

Family tree of the Tuxpenio germplasm complex.


(CIMMYT 1990). Many of the improved
varieties and hybrids developed for these
production environments were derived
from a single source material, the
Tuxpeiio complex originating in Mexico
(see box, "The Tuxpefio Germplasm
Complex," opposite).

Benefits of improved germplasm-
What benefits do farmers expect to
receive when they adopt an improved
maize variety or hybrid? Farmers are
usually interested first and foremost in
increased grain yield, which may be
associated with many different charac-
teristics depending on local production
constraints. Specific yield-enhancing
traits typically include reduced plant
height, increased rooting depth, in-
creased stalk strength, improved husk
cover, insect resistance, disease resis-
tance, tolerance of too little or too much
moisture, tolerance of temperature
extremes, tolerance of unbalanced soil
conditions (e.g., acid soils), and/or
responsiveness to fertilizer. In addition
to traits associated with increased grain
yield, other characteristics may also be
influential in determining adoption, such
as grain quality (appearance, taste,
smell, cooking characteristics,
storability) or fodder quantity and
quality.

Formal efforts to compare the perfor-
mance of older and newer materials have
tended to concentrate on grain yield.
Most standard varietal trials rank
materials according to grain yield,
although other characteristics are also
recorded (e.g., days to 50% silking, plant
height, ear height, leaf area, ears per
plant). Although varietal trials usually
focus on yield factors, it is important to







World Maize Facts and Trends


remember that yield increases do not
necessarily constitute the only advantage
offered by a new variety or hybrid. If
farmers require an early maturing
material to enable more timely planting
of the crop following maize in the
rotation, they may readily adopt an
improved variety or hybrid which yields
the same as their current material but
which can be harvested two weeks
earlier. Conversely, if farmers growing
maize for home consumption have
strong preferences for a certain grain
type, they may fail to adopt an improved
variety or hybrid that yields significantly


more than their current material but
whose grain quality is considered
unacceptable. Thus, even though yield
gains provide a convenient measure for
quantifying the economic benefits
associated with the adoption of an
improved variety or hybrid, they remain
an incomplete measure.


There are compelling reasons for
breeders to evaluate germplasm under
tightly controlled experimental condi-
tions, but varietal trials done on experi-
ment stations can be misleading because
they are typically conducted using much
higher levels of inputs and management


than farmers use. Since there may be
strong effects resulting from the interac-
tion between a variety and the level of
management, trials on the experiment
station often are complemented by on-
farm trials designed to estimate the yield
gains farmers themselves are likely to
realize. For this reason, maize varieties
and hybrids submitted for certification
and eventual commercial release
undergo a series of evaluations which
include at least one cycle of on-farm
testing. On-farm varietal trials come in
two forms: researcher-managed trials
and farmer-managed trials.


Table 9. Performance of improved maize materials vs. local checks in selected countries, researcher-managed trials

Mean yield
ol improved Mean yield Yield
Improved material of local check advantage
Country material Local check (kg/ha) Ikg/ha. 1%

Indonesiaa Arjuna Harapan 4,450 4,322 3
Moroccob Guich 29 Bercheed 2,190 1,990 10
C6te d'Ivoirec Ferke 7928 CJB 2,130 1,920 11
Thailandd Suwan-1 Local 2,870 2,550 13
Mexicoe Across 8243 Criollo 3,990 3,480 15
Haiti' La Maquina Chicken corn 1,300 1,110 18
Malawig MH 16 NCSM 41 1,758 1,456 21
Paraguayh Guarani V-312 Avati moroti 3,350 2,720 23
Philippinesd IPB Var 2 Tiniguib 1,565 1,255 25
Benin' Sekou 81 Local 2,742 2,142 28
CameroonJ CMS 8501 Local 2,611 1,994 31
Zairek Babungo Shaba-1 2,330 1,780 31
Nepal' Rampur Composite Local 3,261 2,422 35
Burkina Fasom SR 22 Local 3,400 2,500 36
Guatemalan HB-83 Local 4,227 3,075 37
Ghanao Dobidi Local 4,045 2,543 59
BurundiP Mugamba 1 Local 2,973 1,721 72
Togoq Pirsabak 7930 Local 1,918 1,080 78


Dahlan et al. (1987).
B. Sali, personal communication.
I. Ndabalishye, personal communication.
Satorre et al. (1991).
R.H. Lafitte, personal communication.
Yates and Martinez (1985).
Smale, Heisey, and Leathers (1991).
Pham, Waddington, and Crossa (1989).
Dossou (1989).


Woldetatios et al. (1991).
Berhe (1989).
Lal (1991).
Sanou (1991).
C6rdova (1984).
Edmeades, Dankyi, and Marfo (1991).
E. Rufyikiri, personal communication.
Togo Ministry of Agriculture (1988).


Sources: a
b
c
d
e
f
g
h
I














Evidence from on-farm trials man-
aged by researchers-In researcher-
managed on-farm trials, researchers
control the non-experimental variables
(e.g., land preparation method, planting
date, planting method, fertilization,
weed control, insect control, harvesting).
The main advantage of researcher
management is a much higher degree of
uniformity in the trials, resulting in
lower variability due to non-experimen-
tal factors and facilitating analysis of
results. The disadvantage of researcher
management is that non-experimental
variables may not reflect farmers' actual
practices, which means yield gains may
be exaggerated if there are significant
germplasm x environment interactions.

Data from on-farm trials appear mostly
in the "informal" literature (e.g., annual
reports of national programs, results of
verification trials published by varietal
certification authorities, seed company
bulletins). It is difficult to generalize
about the performance of improved
materials compared to farmers' materials
because so much depends on the
materials and the level of management.
Table 9 summarizes results of several
sets of researcher-managed on-farm
varietal trials. Many of these trials
evaluate levels of fertilizer; the data in
Table 9 correspond to the treatment most
representative of farmers' practice
(usually no fertilizer or very little
fertilizer).

Evidence from trials managed by
farmers-In farmer-managed trials, all
variables other than variety are con-
trolled by the farmer (e.g., land prepara-
tion, planting date, planting method,
fertilization, weed control, insect


control, harvesting). The main advantage
of farmer management is that the trials
reflect the level of management actually
provided by farmers. This can be
extremely important in evaluating the
performance of improved maize
germplasm, because farmers often
manage their maize in ways that may be
suboptimal for the maize enterprise
considered in isolation but which
maximize returns from the overall
farming system. On the other hand, in
evaluating the performance of improved
materials under farmer management,
researchers must remain aware that the
evaluation may be biased if the im-
proved materials being tested require
changes in management practices with
which farmers are still unfamiliar. For
example, in many cases farmers initially
plant improved materials at very low
densities because they do not realize that
the materials have improved resistance


to lodging and reduced height; at low
densities, the materials may yield far
less than they would at higher densities.
Farmer-managed trials thus require
careful monitoring by researchers.

Table 10 summarizes results of a
sampling of farmer-managed on-farm
varietal trials.

Adoption case studies-Previous
sections of this report have shown that
even though the global maize research
network has been successful in develop-
ing improved varieties and hybrids, only
about 50% of the developing world's
maize area is planted to these materials.
Evidence from on-farm varietal trials
suggests one reason for the relatively
slow rate of adoption: when grown under
low levels of management (such as those
typically provided by resource-poor
farmers in developing countries),
improved materials often fail to yield


Table 10. Performance of improved maize materials vs. local checks in selected
countries, farmer-managed trials

Mean fieldd Mean yield
of improved of local Yield
Improved Local material check gain
Country material check (kg/ha) (kg/ha) (%)

Nepal" Ganesh-2 Madi 1,349 2,722 -50
Malawib CCA Local 1,300 1,400 -8
Mexicoc H-430 H-422 8,400 8,000 5
Nigeriad TZSR Local 1,720 1,590 8
Pakistane Azam Local 5,580 5,050 10
Ghana' Aburotia Local 2,460 2,180 13
Cameroong Kasai I Local 2,980 2,620 14
Beninh TZSR Local 1,880 1,620 16
Zaire' Babungo Shaba-1 1,710 1,360 26
NigeriaJ TZSR-W Local 1,990 1,560 28


Sources: a
b
c
d
e


Chand et al. (1991).
Hildebrand (1984).
Cota et al. (1991).
Mutsaers (1991).
Byerlee, Khan, and Saleem (1991).


' Tripp et al. (1987).
g McHugh (1991).
h Versteeg and Huijsman (1991).
Vogel, Berhe, and Hennesey (1991).
j Mutsaers and Walker (1991).


Ic.n~i...,,.iril. '1;
Iii







I I FA .. I -', J


better than traditional varieties by a
significant margin. Once again, this
serves to emphasize the critical impor-
tance of proper management in allowing
the superior yield potential of improved
germplasm to be expressed.

While evidence from varietal yield trials
is important for understanding farmers'
adoption behavior, yield differentials
alone rarely explain adoption patterns.
Recent empirical work suggests that a
farmer's decision whether or not to
adopt an improved variety or hybrid is
tempered by a complex set of technical,
economic, and institutional factors. The
studies described below illustrate how
some of these factors work, both when
improved maize varieties are widely
adopted and when adoption is less
widespread.

Kenya-Gerhart (1975) has described
the remarkable success of hybrid maize
in Kenya. During a 10-year period
beginning in the early 1960s, hybrids
developed by the national maize
breeding program spread rapidly
throughout the country, adopted by the
overwhelming majority of both large-
scale commercial farmers and small-
scale subsistence farmers, and approach-
ing 100% coverage in high-potential
production environments.

Hybrids were clearly a superior technol-
ogy with the capacity to boost maize
yields significantly, particularly in
favorable production zones. Gerhart
recognized the fundamental importance
of that fact in analyzing the reasons for
the successful adoption of hybrids.
However, according to Gerhart, at least
four other factors were critically


important in ensuring the success of
hybrid maize in Kenya. First, the
commercial seed production and
distribution system (originally developed
to serve the needs of large-scale com-
mercial farmers) effectively delivered
the technology initially to large-scale
farmers and, eventually, to small-scale
farmers. Second, farmers could obtain
subsidized fertilizer along with the
improved seed. Third, a well-developed
system of rural infrastructure provided
both large-scale and small-scale farmers
with ready access to markets, thus
ensuring the profitability of adopting the
new technology. Fourth, extension and
training services provided by the Kenyan
Ministry of Agriculture were instrumen-
tal in familiarizing farmers with the
technology and ensuring its correct
application.

Ghana-The Ghana Grains Develop-
ment Project, a collaborative effort
involving Ghana's Crops Research
Institute, the Grains Development Board,
CIMMYT, IITA, and the Canadian
International Development Agency, was
established in 1979 with the goal of
improving grain and legume production
in Ghana through the development and
use of appropriate grower recommenda-
tions. One major achievement of the
project was the development and
promotion of improved maize varieties,
along with improved management
practices.

An extensive survey conducted 11 years
after the inception of the project indi-
cated that considerable success had been
achieved in promoting the use of
improved maize varieties (Ghana Grains
Development Project 1991). By early


1990, nearly half of Ghana's maize area
was planted to improved varieties, and
approximately 58% of all maize farmers
were planting an improved variety in at
least one of their fields. Three factors
contributed to the adoption of improved
varieties. First, extensive on-farm
varietal trials had led to the selection of
improved varieties that yielded signifi-
cantly more than farmers' traditional
varieties, even under traditional manage-
ment strategies. Second, an extensive
on-farm agronomic testing program
identified changes in crop management
practices that allowed the improved
materials to express their higher yield
potential (e.g., increased fertilizer doses,
increased plant density). Third, close
links between researchers and extension
agents ensured the widespread dissemi-
nation of accurate information about the
improved seed-fertilizer technology.

Interestingly, the survey also identified
several factors which have worked
against adoption of improved maize in
Ghana. Two in particular were singled
out. First, many survey respondents
reported that the grain of improved
varieties is inferior for processing into
local foods, and that the inferior grain
quality of improved varieties is some-
times reflected in a price differential in
local markets averaging 10-15%; thus,
even when improved varieties yield
more than local varieties, the additional
yield is not always sufficient to compen-
sate farmers for the lower price they
receive. Second, even when a farmer
decides to grow an improved variety,
seed is not always readily available
through official distribution channels
(only 20% of the survey respondents had
purchased seed from official sources).














Malawi-Despite extensive efforts to
develop maize hybrids for Malawi,
hybrids have not been as widely adopted
there as in some other African countries.
However, recent work by Smale (1992)
and Smale et al. (1991) suggests that
hybrids are beginning to make modest
inroads: 28% of the farmers surveyed in
three districts during the 1989-90
growing season reported using hybrid
materials. Nevertheless, one noteworthy
feature of the Malawian experience is
the limited extent of adoption; even
farmers who grow hybrids continue to
devote over 60% of their total maize
area to local varieties.

Smale et al. examined the factors
motivating individual farmers' decisions
to grow hybrids. Their analysis corrobo-
rated evidence that Malawians strongly
prefer flint-type maizes, thus making
many of the first hybrids released in
Malawi unacceptable for reasons of
quality (many early releases were dent-





100-

90




S60 Unfertilized local maize

50 Fertilized hybnd maize

8 40 Fertilized local maize
& -,.-


type materials subject to high processing
and storage losses). Smale et al. demon-
strated how this preference has led to
incomplete adoption of hybrid maize in
many regions in Malawi. When local
maize is assigned a value premium to
reflect its better processing and storage
characteristics, the expected net returns
to fertilized local maize are higher than
the expected net returns to hybrids. For
example, Figure 11 shows that only 35%
of the fertilized local maize plots give
net returns of less than MK 200/ha,
whereas 45% of the fertilized hybrid
plots and 55% of the unfertilized hybrid
plots give net returns lower than
MK 200/ha. As long as hybrid maize
fails to meet consumers' quality require-
ments, it will not be grown in preference
to local maize.

Paraguay-In view of the relatively
limited commercial importance of maize
in Paraguay, resources devoted to maize
research by the national program have


been modest, and germplasm improve-
ment has centered on screening and
evaluating imported materials rather
than developing varieties and hybrids
locally. Since the germplasm require-
ments of Paraguay's large-scale com-
mercial farmers are largely met by
private seed companies based in neigh-
boring Argentina and Brazil, the
Paraguayan national maize program has
concentrated on developing materials
suitable for use by the small-scale, semi-
subsistence farmers who produce most
of the country's maize.

In February 1990, researchers from
CIMMYT and the Paraguayan national
maize program carried out an extensive
survey of maize producers in the
country's major maize-producing region.
The survey revealed that use of im-
proved maize germplasm varies consid-
erably in Paraguay depending on the
category of farmer and the eventual use
of the maize. Large-scale commercial
farmers who grow maize to feed
livestock rely heavily on improved
materials; virtually 100% of these
farmers reported planting improved
open-pollinated varieties or hybrids. The
main reasons these farmers gave for
using improved materials include their
high yield potential and responsiveness
to fertilizer, as well as their suitability
for livestock feed. In contrast, small-
scale farmers who grow maize mainly
for home consumption and to feed
livestock on the farm reported planting
mostly local materials. A modest 20% of
small-scale farmers did acknowledge
planting limited areas to improved
varieties to produce feed for their own
chickens and pigs, but most said they
were uninterested in more extensive


-400 -200 0 200 400 600 800 1,000
Net returns (MK/ha)

Figure 11. Cumulative net returns distributions for three maize technologies in Malawi,
1989-90.


i K, P. 1 0.- ii,` PiV







lr 1


adoption, stating that the grain of the
improved materials is unsuitable for
preparing traditional dishes. In a few
cases, however, farmers stated that they
would be interested in planting improved
materials, but that seed was not available
through formal procurement channels.

Nicaragua and El Salvador-Since the
mid-1970s, CIMMYT breeders and
pathologists have worked closely with
colleagues from Nicaragua and El
Salvador in developing resistance to
corn stunt virus, a potentially devastat-
ing disease of maize in Central America
and the Caribbean.

As a result of that work, two stunt-
resistant varieties were released in
Nicaragua as NB-6 (released in 1985)
and NB-12 (released in 1986). In 1991,
researchers from the Nicaraguan national
program and CIMMYT conducted a
survey in Nicaragua's Region IV to
assess the adoption of NB-6 and NB-12
(Borb6n and Sain 1992). At the time the
survey was conducted, maize production
was well on the way to recovery from
the disruptive effects of the recent civil
disturbances, although area planted had
still not returned to the all-time high
levels achieved in the late 1970s. Borb6n
and Sain concluded that 75% of all
farmers had planted NB-6 or NB-12 in
1991, and that nearly 90% of the
Region's total maize area had been
planted to these two varieties. Farmers
growing the varieties attributed their
preference for NB-6 and NB-12 to four
main factors: high yield potential,
resistance to stunt, resistance to abiotic
stresses (principally drought), and good
eating quality. Among those who had
not planted NB-6 or NB-12, many cited
the lack of seed as the main impediment
to adoption.


In El Salvador, two stunt-resistant maize
hybrids have been released: H-53 in
1988 and H-57 in 1991. These hybrids
yield 20-25% higher than the popular but
susceptible hybrids H-5 and H-3
(Aguiluz et al. 1991 and C6rdova 1991),
which are grown extensively in El
Salvador (covering 50% and 12% of the
1990 national maize area, respectively).
The new stunt-resistant hybrids are
expected to make a similar strong impact
in farmers' fields in the near future.

All four of these hybrids are examples of
the positive results of collaboration
between the Salvadoran national
program and CIMMYT. Throughout El
Salvador the current rate of adoption of
improved maize varieties and hybrids is
estimated to be 70%, one of the highest
in Latin America. El Salvador has also
registered one of the highest national
average maize yields in the region:
2.2 t/ha in 1990. These are quite signifi-
cant achievements, considering that most
adopters are small-scale farmers growing
maize in challenging climatic and
topographic conditions, often in crop-
ping systems with sorghum and/or dry
beans. The sources of this success are a
well-defined research program with
realistic, clear objectives; exacting
attention to good basic seed production
and seed quality control measures
oriented toward farmers; a well-
organized and aggressive extension
program; and an effective private
industry that combines the use of its own
materials with those of the public sector
to develop outstanding hybrids
(Manzano and C6rdova 1986).

Pakistan-In the 1970s, the Pakistan
Agricultural Research Council (PARC)
and CIMMYT conducted a research and
extension program in the Swat Valley of


Northwest Pakistan, with the objective
of developing and promoting a package
of improved maize production practices
(Byerlee, Khan, and Saleem 1991).
Following 10 years of work, results were
disappointing; only fertilizer and, to
some extent, an improved maize variety
had been widely adopted in the project
area. An extensive diagnostic survey was
therefore done to analyze the reasons
behind this relative lack of success. The
survey findings helped re-orient the
research design, resulting in much higher
rates of adoption of a revised package of
improved technologies.

One crucial finding of the diagnostic
survey was the economic importance of
green maize fodder, which accounts for
approximately 50% of the total value of
production from maize fields. Farmers in
Swat have developed management
practices which allow them to maximize
the revenue from combined fodder and
grain production; these practices include
seeding maize at extremely high
densities and then systematically
removing plants throughout the growing
season for use as fodder. An important
shortcoming of the original research
program was the failure to recognize the
economic value of fodder in Swat, which
led to the promotion of an inappropriate
variety. Once the value of maize as
fodder and grain was recognized,
researchers were able to convey this
information to maize breeders, who
could then identify a high-yielding
variety that tolerates high planting
densities. The improved variety eventu-
ally adopted by the majority of maize
farmers, Azam, differed from earlier
recommended varieties in this critical
aspect.







Srr1 K I RE K;..-3rC I, In% m, .1 .5,,J inln.a..I .11 0I. II.'pr! 'rN I


Summary: the prerequisites for
adoption-This brief review of findings
from selected case studies cannot
provide a complete picture of the
complex forces driving the varietal
adoption process. However, a common
theme emerging from these examples is
that adoption is not inevitable simply
because an improved variety or hybrid
yields more than a farmer's current
material. Superior yield performance is
frequently a necessary condition for
success (although not always), but it is
rarely a sufficient one. As these case
studies indicate, regardless of the
potential yield gain offered by an
improved variety or hybrid, widespread
farmer adoption is unlikely unless one or
more additional conditions are met.
Typically these include a well-function-
ing seed industry, ready availability of
complementary inputs such as fertilizer,
farmers' knowledge of how to use the
technology, and compatibility with the
farming system. If farmers produce
maize for sale, successful adoption may
also depend on economic incentives in
the form of favorable producer prices, as
well as ready access to output markets
(which in turn requires rural infrastruc-
ture, especially roads). If farmers
produce maize for home consumption,
successful adoption frequently will
depend also on the new variety's
suitability for local processing, storage,
and consumption.



Returns to Investment in
Maize Research

Previous sections of this report have
presented evidence on the human and
financial resources devoted to maize
research at the global level; reviewed the
products of the global research effort


(especially breeding); described the
impacts of these products at the farm
level; and discussed some of the techni-
cal, economic, and institutional factors
influencing the impact of research. This
final section addresses the question, Are
the benefits of maize research worth the
considerable costs? The question is
important, because as increasing
demands are made on the resources
available for agricultural research,
policy makers (and funding agencies)
have become extremely sensitive to the
economic dimensions of research
resource allocation decisions.

How are research investments evalu-
ated?-Investments in agricultural
research are usually evaluated in terms
of the economic rate of return, which
can be thought of as the rate of interest a
bank would have to pay to generate the
same return as the research project. In
most studies, this rate of return is
calculated using the economic surplus
approach.9 The economic surplus
approach views research expenditures as
the source of technological change that
shifts the supply function for a commod-
ity down (in the case of a cost-saving
technology, which makes it possible to
produce the same amount of a commod-
ity with fewer inputs) or to the right
(in the case of an output-increasing
technology, which makes it possible to
produce more of the commodity with the
same amount of inputs). Gains in
economic surplus attributable to the
research are calculated by estimating
changes in the quantity of the commod-
ity that is produced and consumed, as
well as any changes in the prices
received by producers and paid by
consumers.


Although early work using the economic
surplus approach was based on simplistic
assumptions about the nature of the
supply and demand curves, over time the
procedures used in research evaluation
have been refined substantially.10 It is
now clear that the distribution of
research benefits among producers and
consumers depends upon the shapes of
the supply and demand curves, as well as
the nature of the supply curve shift
resulting from technological change.
Government price policies are also
important in determining how the
benefits of research are distributed
among groups of producers and
consumers.

To determine rates of return using the
economic surplus approach, it is neces-
sary to calculate streams of costs and
benefits generated by the research
investment. Figure 12 illustrates the
stream of costs and benefits typically
associated with a plant breeding pro-
gram. During an initial period, net
benefits remain negative because
research costs are being incurred (for
example, in the crossing, selection, and
evaluation of experimental materials)
without any benefits being realized.
Eventually the research produces an
improved variety, which is approved for
release after undergoing a certification
process. Following a lag necessary for
the production and distribution of seed,





9 For useful reviews of the literature on evaluating
returns to investment in agricultural research, see
Schuh and Tollini (1978), Scobie (1979), and
Norton and Davis (1981).

10 Progressive refinements of the basic method
appear in Griliches (1958), Peterson (1967),
Schmitz and Seckler (1970), Fishel (1971),
Akino and Hayami (1975), Hayami and Herdt
(1977), Lindner and Jarrett (1978), Norton and
Davis (1981), and Rose (1980)







c, \,,rl,'I .l i/c Fair- nj Trends


600-
6 Research costs Pe
Research benefits


ak adoption


0 5 10 15 20 25
Years


30 35 40 45


Figure 12. Undiscounted stream of costs and benefits associated with a maize
breeding program.


Research costs
Research benefits


(Discontinued at 8% per year)


300


200--
E Certification Year of
begins release Peak adoption
1 Research
begins

0 nnO


0 5 10 15 20 25 30 35 40
Years


45 50


Figure 13. Effects of discounting the stream of costs and benefits associated with
a maize breeding program.


500-


400

300-

200R
R

100


0


Certification
begins
researchh Year of
begins \release


the variety is taken up by farmers, with
the rate of adoption typically following
an S-shaped (or logistic) curve. Assum-
ing the variety leads to improved yields
in farmers' fields, the original research
investment (made years earlier in most
cases) now begins to generate benefits in
the form of increased production. The
stream of net benefits consequently turns
positive, increasing as the area planted
to the new variety expands, reaching a
maximum at peak adoption, and then
declining as the variety is gradually
replaced by another, newer variety.

While the relative sizes of costs and
benefits are obviously important in
evaluating a research investment, their
distribution through time is also impor-
tant, because benefits realized far off in
the future are considered less valuable
than benefits realized in the short term.
This phenomenon is referred to by
economists as the time value of money.
To accommodate the time value of
money, research costs and benefits are
discounted. Figure 13 illustrates how
discounting depresses the value of net
benefits realized near the end of the
period of analysis relative to those
realized near the beginning. Discounting
is an important concept used in analyz-
ing returns to investments, because the
productivity effects of research can
persist for up to 30 years or more
(Pardey and Craig 1989).

Once the stream of net benefits gener-
ated by a plant breeding program has
been estimated, it is possible to deter-
mine a unique discount rate that equates
the discounted streams of costs and
benefits, thus driving total net benefits to
zero. This discount rate is known as the


(111..-___














internal rate of return (IRR). The IRR
tends to be extremely sensitive to the
temporal distribution of costs and
benefits. As a practical matter, calculat-
ing the IRR for a plant breeding program
therefore requires knowledge not only of
levels of costs and benefits, but also of:
the lag involved in the development of a
new variety; the lag involved in its
certification and release; the lag in-
volved in the production and distribution
of seed; and the lags involved in
adoption (and eventual disadoption) of
the variety by farmers.

An idea of how several researchers have
determined rates of return to maize
research is given by the studies
described below.

Selected studies of returns to maize
research-Beginning in the mid-1950s,
numerous studies have examined the
impacts of research on agricultural
productivity and output growth. A
number of these studies have estimated
rates of return to maize research. It is
worthwhile reviewing some results of
this work to gain insight into the rates of
return associated with several well-
known maize research programs.

Pioneering work on the returns to
investment in maize research was done
in the USA by Griliches (1957).
Griliches focused on the development of
hybrid maize, an achievement he
referred to as "one of the outstanding
technological successes of the century."
Griliches estimated private and public
expenditures on hybrid maize in the
USA between 1910 and 1955. He then
estimated annual gross social returns as
the value of the increase in maize
production attributable to the use of


hybrids, netting out the additional costs
involved in producing hybrid seed and
making an adjustment for the price
change associated with the production
increase. On the basis of the resulting
flow of net benefits, Griliches estimated
that the investment made from 1910 to
1955 in hybrid maize research in the
USA had an IRR of 35-40%.

Ardito-Barletta (1971) followed the
approach proposed by Griliches to
estimate returns to the Rockefeller
Foundation's investment in agricultural
research in Mexico from 1942 to 1960.
Maize was one of four commodities
included in the analysis, along with
wheat, sorghum, and potatoes. Ardito-
Barletta estimated the stream of net
benefits attributable to the Rockefeller
Foundation's maize research program as
the difference between the program's
annual costs and benefits (the latter
calculated based on the area planted to
improved maize varieties, average yield
gains attributable to use of improved
varieties and associated inputs, and the
price of maize). Using a range of likely
values for key parameters to reflect
uncertainty in some of his empirical
estimations, Ardito-Barletta concluded
that the IRR to the Rockefeller
Foundation's investment in maize
research in Mexico during 1942-60
ranged from 26% to 59%.

Yrarrazaval, Navarrete, and Valdivia
(1979) examined the returns to maize
varietal improvement research and
extension in Chile during 1949-77.
Using the conservative assumption to
value research benefits as a proportion of
actual historical production (which the
authors argue was severely depressed by
artificially low domestic producer


prices), rather than as a proportion of
potential production (which would have
been higher had high international parity
prices been allowed to prevail in Chile),
Yrarrazaval et al. calculated an IRR of
21-27%.

Nagy (1985) examined the returns to
maize varietal improvement research in
Pakistan from 1964-65 to 1980-81.
Using an economic surplus framework
amended to explicitly incorporate the
fact that most of the maize produced in
Pakistan is retained for home consump-
tion, he estimated internal rates of return
of 15-23%. Nagy conceded that these
figures probably overstated the returns to
total investment in maize research, since
no cost was assigned to the contribution
of IARCs (which have been significant,
considering CIMMYT has supported
maize research in Pakistan from the
early 1970s to the late 1980s).

Librero and Perez (1987) used a produc-
tion function approach to estimate the
returns to maize research and extension
in the Philippines during 1956-83." A
major problem encountered by Librero
and Perez was how to distinguish
between the benefits attributable to
research and those attributable to
extension, since in the Philippines these
activities must be considered comple-
mentary in some respects and substitutes
in others. Experimentation with a variety
of functional forms and lag structures
generated internal rates of return ranging
from 27% to 48%.





The production function approach differs
somewhat from the economic surplus approach,
but it can also be used to calculate the internal
rate of return to investment in research. For a
description of the basic methodology, see Schuh
and Tollini (1978), Scobie (1978), Norton and
Davis (1981).


Pri i %J,, -J imp-i'l. ... I":







korlds i ru M .c ,;*:[ and Trr.jd


Norton, Ganoza, and Pomareda (1987)
estimated the returns to publicly funded
research and extension activities for
maize in Peru during 1981-86. Their
approach was innovative in several
respects. In addition to considering
extension activities as complementary to
(and hence analytically inseparable
from) research, these researchers
extended the basic economic surplus
framework to consider the effects of
possible shifts in demand over time,
pivotal vs. parallel supply shifts, and
alternative government pricing policies.
Based on conservative assumptions
about the productivity increases attribut-
able to maize research and extension,
they estimated returns to maize research
and extension in Peru ranging from 10%
to 23%.

Karanja (1990) used a production
function approach to examine the returns
to investment in maize research,
extension, and seed development in
Kenya. Regression analysis revealed that
between 1955 and 1988 an increase in
maize research expenditure by 1%
increased maize production by 0.4% a
decade later, representing a marginal
rate of return of around 40% and an
average rate of return of 68%. Like
Gerhart 15 years earlier, Karanja
stressed the crucial complementary roles
of the private maize seed industry, the
public sector extension service, and
favorable government policies in
increasing maize production and leading
to the overall success of the maize
research program.


Summary: the evidence on returns to
research-What general conclusions
can be drawn from these case studies?
Perhaps the most striking result from the
country studies is that rates of return to
investment in maize research have
consistently been positive and, in most
cases, substantial. Studies done in a wide
range of geographical and historical
settings from large industrialized
countries such as the USA to much
smaller developing countries such as
Pakistan and Peru have frequently
estimated rates of return averaging 10-
40%. It is interesting to note that results
on this order of magnitude continue to
be reported in spite of increasing
methodological refinements introduced
since Griliches' pioneering study. In
interpreting these results, it must of
course be recognized that there is almost
undoubtedly some degree of bias in the
way cases are selected for study; returns
to research are more likely to be calcu-
lated in instances in which research is
thought to have been successful.

But if the studies provide evidence that
investment in maize research has often
generated attractive rates of return, they
also demonstrate the tremendous
practical problems involved in attempt-
ing to calculate returns to research
empirically. One recurring problem
arises in attempting to distinguish
between benefits attributable to research
and to extension. Because of the
difficulty in distinguishing between the
two, many analysts have chosen to
include extension expenditures as a
research cost, arguing that extension
activities are instrumental in the dis-
semination of improved technologies
and consequently must be viewed as a
sort of complementary input which
cannot be omitted from the cost side of
the analysis.


Another recurring problem involves the
attribution of benefits to different types
of research, an issue of considerable
interest to efficiency-minded research
managers charged with allocating scarce
research resources among competing
uses (e.g., plant breeding vs. crop
management research). Although a few
analysts have attempted to measure
returns to plant breeding and crop
management research separately, in
doing so they have had to confront the
thorny problem of partitioning yield
gains achieved in farmers' fields into the
portion attributable to adoption of
improved germplasm and the portion
attributable to use of improved manage-
ment practices. But these two are
complementary inputs, because the size
of the yield gain attributable to adoption
of an improved variety depends on the
extent to which the farmer also adopts
improved management practices, and
vice versa. Under these circumstances,
estimating separate returns to plant
breeding and crop management research
becomes somewhat arbitrary (which is
why many analysts choose not to
distinguish between them).

Finally, the case studies also support the
view that research impacts depend to a
great extent on factors unrelated to the
improved technology itself. For this
reason, many analysts report rates of
return calculated under a range of
assumptions about government policies,
recognizing that the impact of improved
technologies (and hence the size of
research benefits) depends partly on
economic and institutional factors which
essentially have little to do with the
technology.







P, R,.-j-rch In,, .r,,. I n.j imp.: -.' in 1,, k-i.rir C-i~re


Summary and Conclusion

Based on the evidence presented in this
report, what conclusions can be drawn
about the level of investment in maize
research in developing countries, as well
as the impacts of this research?

First, it is clear that reliable data on the
level of maize research investment and
research impacts are not readily avail-
able. Many of the national maize
program administrators surveyed for this
report were unable to estimate with
precision the total amount spent on
public sector maize research, and they
frequently were unable to identify and
quantify the products of their own
research programs. Data on private
sector research costs and returns also
proved difficult to obtain, for they were
considered confidential. Without reliable
data on research costs and returns, the
payoffs to research are difficult to
determine, and an assessment of the
cost-effectiveness of research resource
allocation becomes impossible. This
points to the need for more careful
monitoring and recording of research
expenditures within national programs,
as well as more systematic documenta-
tion of research impacts.

Yet despite some questions about the
quality of the available data, the evi-
dence presented in this report makes
clear that many developing countries
invest significant resources in public
sector maize research. A large propor-
tion of these resources has been concen-
trated in traditional breeding programs,
whose output can be measured in terms
of improved varieties and hybrids. With
strong support from the IARCs, which
themselves improve germplasm and
facilitate its distribution throughout the


developing world, many national
breeding programs have released
varieties and hybrids at an impressive
rate during the past three decades. The
rate of varietal releases has if anything
accelerated in recent years, suggesting
that traditional breeding methods will
continue to deliver substantial benefits
during the foreseeable future as many
national programs continue to mature.
At the same time, the obvious benefits
generated by traditional breeding
methods may explain the reluctance of
some NARSs to cut back their existing
breeding programs in order to move
more rapidly into biotechnology re-
search, whose costs tend to be high and
whose benefits remain somewhat
speculative.

The widespread dissemination of
improved materials and their regular use
by many national breeding programs
attests to the effectiveness of the
collaborative global maize research
system. Yet one must be careful not to
equate a high rate of varietal releases
with success in maize breeding; after all,
a high rate of releases means little if
relatively few of the releases are adopted
by farmers and planted on a large scale.
The obvious question arises: just how
well have improved maize materials
been accepted by farmers?

At present, approximately 50% of the
developing world's maize area is planted
to improved materials, compared to over
80% each for wheat and rice. While
some would say that global breeding
efforts therefore have been relatively
less effective for maize than for other
crops, others would contend that maize
breeding is arguably more difficult,
given the tremendous diversity of the
mostly rainfed production environments
in which maize is grown, the unusual
complexity of many maize-based


farming systems, the relatively greater
importance of grain quality factors
(including processing, storage, and
consumption characteristics), and the
open-pollinating nature of the crop itself.
No attempt will be made here to resolve
the longstanding debate over whether
maize breeding is inherently more
difficult than breeding for other crops.
However, judging from numerous case
studies, it is clear that successful
adoption of improved maize germplasm
depends not only on the performance of
the germplasm per se but also on
additional factors, many of them beyond
the control of breeders. For example,
adoption of improved maize materials
often requires a well-functioning seed
industry, which is not nearly as impor-
tant in the case of self-pollinating crops
such as wheat and rice.

Despite the fact that success depends on
a multitude of often unpredictable
factors, maize research merits continued
government support. Numerous empiri-
cal studies have revealed that returns to
public sector investment in maize
research, while variable, tend to be
attractive by conventional investment
criteria. At the same time, the variability
in empirical results underlines the need
for more and better analysis of research
resource allocation issues. This will
require not only refinements to existing
analytical methods, but also improve-
ments in the data relating to research
costs and returns. In an era when
resources for agricultural research are
increasingly scarce, and in which
research funding decisions are made
with close attention to the bottom line,
maize researchers will have to take
greater care in monitoring the cost of
their activities and documenting the
impacts of their work.











Part 2: The Current World Maize Situation


World maize production and consump-
tion continue to increase at a steady
pace. Production remains higher in
industrialized countries, but the gap has
narrowed during the past decade due to
faster production growth in developing
countries. In addition to expanding their
levels of maize production, developing
countries have also boosted maize
imports in the face of population growth
and increases in livestock production.



Production

During the past four decades, world
maize production increased steadily,
experiencing only minor fluctuations
while rising to nearly 500 million metric
tons (MT). However, while production
in developing countries showed steady
growth, the 1980s were a volatile decade
for maize production in industrialized
countries (Figure 14). In 1983, severe
drought sharply reduced the industrial-
ized nations' maize output. The return of
more normal weather patterns in 1984
and 1985 helped production recover, but
poor weather in 1986, 1987, and 1988
again reduced production. By the end of
the decade, output was once more on the
rise, reaching almost 475 MT in 1990.

Developing countries now contribute
over 40% of world maize production.
Production gains achieved in developing
countries during the past decade are
attributable to increases in yields and in
area planted, both of which grew at
approximately 1.5% per year (Figure
15). The growth in maize area in


developing countries reflects a reversal
of the pattern observed in previous
decades, when production growth
resulted primarily from yield increases.
Much of the recent area growth can be
attributed to the expansion of maize into
non-traditional areas or crop rotations.
For example, maize is replacing sor-
ghum and millet in certain dryland areas
in Sub-Saharan Africa; in parts of India,
maize is now grown as a winter crop. In
contrast to the experience of developing
countries, during the 1980s production
growth in industrialized countries
followed the pattern established in
previous decades: maize area declined
slightly, and production growth resulted
entirely from yield increases.

In coming years, world maize stocks and
production will be taxed to meet
dramatically increased demand from two
regions Eastern and Southern Africa






500 -


400--


S300 --


a0
200 .. .


1950 1955 1960


(where production has been devastated
by drought) and the countries of Eastern
Europe and the former Soviet Union
(where political and economic turmoil
have severely disrupted production).

Eastern and Southern Africa-
Favorable harvests in Eastern and
Southern Africa in 1989 and 1990 were
followed in 1991 by the worst drought of
the 20th century, afflicting even food-
exporting countries which have usually
escaped severe droughts in the past. The
drought is thought to be an effect of El
Nifio, a meteorological phenomenon
which influences global atmospheric
circulation.

The effects of the drought have been
particularly severe in Southern Africa.
South Africa and Zimbabwe, which
normally export maize to neighboring
countries, suffered 75-80% declines in


1965 1970 1975 1980 1985 1990


Figure 14. World maize production, 1951-90.







r i Th,..L r -nr %'..rld N ,. cii to


maize production in 1992. As a result,'
the region may have to import more than
7 MT of food in 1992, three times the
normal amount. With maize accounting
for 80% of the region's total grain
output, outright famine threatens some
areas. Maize shortages, accompanied by
soaring prices, will have the greatest
effect on the poorest segments of
society, which rely on maize as a daily
staple.

Although Southern Africa was hardest
hit by the drought, Eastern Africa was
also affected, with Kenya, Ethiopia,
Somalia, and Tanzania all suffering
production declines. Poor rains in
Ethiopia and Sudan also have implica-
tions for irrigated production in Egypt,
because the major source of the Nile
River is in western Ethiopia and eastern
Sudan.


Eastern Europe and the former Soviet
Union-Many of the new republics
formed in Eastern Europe and the former
Soviet Union have suffered severe grain
shortages as a result of recent political
and economic upheavals. Implications
for these nations' cereal economies vary,
depending on the time frame being
considered. In the near term, increased
grain imports will be needed. Many of
these imports will have to come in the
form of food aid, since most of the new
republics lack foreign reserves with
which to purchase grain. Western
countries have begun to send humanitar-
ian aid, and some have begun to arrange
credit to assist the new republics in
purchasing grain. Over the longer term,
assuming the republics can make a
successful transition to market econo-
mies, they should be able to reduce
cereal imports by increasing the effi-
ciency of production and distribution.


i Yield
Area


6 --


4 4


2


0

De
-2
1950s


veloping countries

1960s 1970s 1980s


-







Industrialized countries
1950s 1960s 1970s 1980s
1950s 1960s 1970s 1980s


Trade

During the 1950s, 1960s, and 1970s,
while world maize production was
increasing by 250%, maize trade rose by
an astonishing 400%. World trade in
maize peaked in 1980, when over 80 MT
were exported worldwide.

After three decades of steady growth,
during the 1980s global trade patterns
were determined by production fluctua-
tions. The relationship between produc-
tion and trade was generally inverse:
favorable production years usually
resulted in decreased trade due to
increased domestic supplies in importing
countries, while unfavorable production
years generally resulted in increased
trade to meet production shortfalls in
importing countries. Thus from 1981 to
1986 maize was traded in world markets
in decreasing quantities. The following
three years saw moderate growth in
trade, although in 1990 world trade
experienced another moderate decline.

Exporters-The USA remains by far
the world's largest exporter of maize,
contributing almost three-quarters of the
71 MT of maize exported worldwide in
1990. United States exports nearly
doubled between 1986 and 1990 as
producers responded to strong world
prices. Aggressive export-promotion
policies also helped US producers regain
a large part of the world market share it
had lost in previous years.

Throughout the 1980s, France, Argen-
tina, China, and South Africa shifted
positions behind the USA in maize
exports. Aided by high domestic


Figure 15. Sources of growth in maize production in developing and industrialized
countries, 1950-90.


i,;







'* ~ **


producer prices coupled with an aggres-
sive export-promotion program, France
supplanted Argentina as the second
largest exporter, accounting for 10% of
world exports in 1990. After dramati-
cally entering the world market in 1985,
China is now the third largest exporter of
maize, accounting for 5% of the world
exports. China sells most of its maize in
East Asian markets, where it can
compete successfully with North
American and European producers due
to substantially reduced transport costs.
Argentina, the fourth largest exporter,
regained some of the market share it
recently lost and now supplies 4% of the
world total. South Africa, plagued by
bad weather, saw its world market share
fall from 7% in 1989 to only 2% in
1990. The continuing drought will
curtail South Africa's capacity to export
maize.

Importers-The major importers of
maize continue to be Japan, the former
Soviet Union, South Korea, Taiwan, and
Mexico, which together purchased
nearly two-thirds of the 74 MT of maize
imported in 1990. Japanese demand for
maize remained steady, accounting for
more than one-fifth of the total world
maize imports in 1990. The former
Soviet Union has been the second largest
importer of maize, occasionally surpass-
ing Japan when domestic production has
run low. In 1990, the former Soviet
Union accounted for 20% of total world
maize imports, a figure which is ex-
pected to rise. South Korea was the third
largest importer of maize in 1990,
accounting for 8% of total maize
imports. Taiwan, the fourth largest
maize importer in 1990, accounted for


6% of total maize imports, purchasing
94% of its maize from the USA and the
remainder from South Africa. Mexico is
currently the fifth largest maize im-
porter. Unlike the other large importing
countries, which use maize as a feed
grain, the Mexican government restricts
the domestic consumption of maize to
food. Record maize crops in 1990 and
1991 prompted the Mexican government
to allow food maize carried over from
the previous year to be reclassified as
feed, thus decreasing the demand for
feed imports. Given current high stock
levels, maize imports to Mexico next
year are expected to be at their lowest
level in 10 years.



Prices

After strengthening sharply in 1988 in
the wake of global production shortfalls,
world maize prices (adjusted for infla-
tion) fell slightly in 1989 and again in


1990 as production rebounded with the
help of favorable weather. By early
1992, the main international reference
price (US Yellow No. 2, FOB Gulf
Ports) stood slightly below the long term
trend (Figure 16). However, world maize
prices are expected to rise from current
levels in the face of increased demand
for imports from Eastern Europe and the
former Soviet Union. The USA, Canada,
and the European Community (EC) have
made available millions of dollars in
credit for these countries, much of it
specifically designated for wheat and
coarse grain imports. In addition, severe
production shortfalls in Southern and
Eastern Africa are likely to be translated
in increased demand for imports, which
could exert additional upward pressure
on global prices.


S200


1 150



100


1960 1965 1970 1975 1980 1985


1990


Figure 16. Evolution of the real price of maize in international markets, 1961-90.







Part 2: The Current \ .rI,.1 P.i ,. ', I'a. ri, i


Policy Developments
Potentially Affecting Maize
Production and Trade

The proposed EC policy reform-
Increased EC expenditures on agricul-
ture and large accumulations of grain
stocks, beef, and dairy products have
spurred proposals that would radically
reform the Common Agricultural Policy
(CAP). While these proposals are under
debate, the EC has adopted a one-year
set-aside program to supplement the
current five-year program designed to
reduce area planted to surplus grains.
The EC Commission would like to
reduce cereal production by 5%, but
most analysts believe that output will
fall by only 1 or 2% because of low
participation in the program. Since the
five-year set-aside program began in
1988, 1.9 million hectares of land have
been removed from production. This
area represents less than 3% of the
arable land in the EC.

North American Free Trade Agree-
ment (NAFTA)-The NAFTA would
reduce trade tariffs and quotas between
the US, Canada, and Mexico over 10
years. Because of already relatively low
bilateral tariffs and non-tariff barriers
between USA and Mexico, and the fact
that the Mexican economy is only 3.6%
of the size of the US economy, overall
net benefits to US industries would
initially be relatively small. However,


agricultural trade opportunities between
the two countries are enormous. Since
the trade liberalizations of 1986, the
value for all US agricultural products
exported to Mexico has more than
doubled. With growth in Mexico's
population (expected to reach 100
million by 2000), maize imports will
increase significantly. Under NAFTA,
trade barriers would be lifted further and
agricultural trade should have even
greater potential for increased growth.



Conclusion

Throughout the 1980s, world maize
production experienced severe fluctua-
tions, mostly as a result of poor weather
in many important producing countries,
but also because of policy changes in
North America and Europe. Despite
production instability, maize consump-
tion continued to rise steadily. With
demand outstripping supply in many
countries, the volume of world trade


rose. Export patterns meanwhile under-
went notable changes, with traditionally
prominent exporters such as South
Africa and Argentina losing market
share to France and China, the new
player in the export market.

Future prospects for global maize
markets are clouded by uncertainty.
Demand for maize will almost undoubt-
edly increase during the next few years
as the countries of Eastern Europe and
the former Soviet Union struggle to
become democracies with market
economies. Increased demand will also
come from Southern and Eastern Africa,
which are experiencing their worst
drought of the century. But even if
global demand for maize is likely to
increase for structural reasons, supply is
more difficult to predict. Depending on
whether and how the Uruguay Round of
the General Agreement on Tariffs and
Trade and the NAFTA are negotiated
and implemented, significant changes
could occur in world maize production
patterns.











Part 3: Selected Maize Statistics


The tables that follow present 34
statistics related to maize production,
trade, consumption, research, and prices,
as well as some basic economic indica-
tors. The statistics were selected to
provide the latest available information.

Countries listed in the tables are classi-
fied either as maize consumers or
producers. Maize consumers include
developing countries consuming over
100,000 t of maize per year, and
developed countries consuming more
than 1 million tons of maize per year.
Maize producers include developing
countries in which maize production
exceeded 100,000 t/yr or accounted for
at least 50% of total maize consumption,
and developed countries in which maize
production exceeded 1 million tons per
year or accounted for 50% of total maize
consumption. Average 1988-90 data
were used in the classification.

Unless otherwise indicated, the regional
aggregates include all of the countries of
a particular region, even countries for
which data have not been reported indi-
vidually (for a list of countries belonging
to each region see Annex 1, p. 57).

All prices reported in the tables were
converted to US dollars at official
exchange rates.


Notes on the Variables

Variable 1: The source of this informa-
tion was the FAO diskette of population
statistics (1991).

Variables 2-3: These data were obtained
from the World Bank World Develop-
ment Report (1992).

Variables 4, 5, 10-14, 17-18: The source
of these variables was the FAO diskettes
of production statistics (1991). Growth
rates were calculated using the semilog
model:

In Y = a + bX + u,


where:


In Y = the natural logarithm of
variable Y,
X = time period (year), and
b = growth rate of Y.

The function describes a variable, Y,
which displays a constant proportional
rate of growth (b > 0) or decay (b < 0).

Variables 6-8, 15, 16: These variables
were obtained from the FAO Agrostat/
PC diskettes (1992)

Variables 19-23: These data were
obtained from the FAO diskettes of trade
statistics (1991). Net imports were
calculated as imports minus exports.
Negative numbers indicate that a country
is a net exporter. Consumption was


calculated as production plus net
imports. Growth rates were calculated
using the formula given above.

Variables 24-29: These data (which are
for 1966-90) were collected through a
general country survey of knowledge-
able maize scientists. Data on the
number of researchers refer to 1990,
although in some cases 1989 is the
reference year. Some data were esti-
mated by CIMMYT staff.

Variables 30-31: The source of this
information was the FAO Fertilizer
Yearbook (various issues). Growth rates
were calculated using the formula given
above.

Variables 32-34: These data were
collected through a general country
survey of maize scientists and econo-
mists. Data for the majority of the
countries refer to the maize crop
harvested in 1990-91, although in some
cases 1991-92 is the reference year. The
maize price is the average post-harvest
price received by farmers. The nitrogen
price is usually the price paid by farmers
for the most common nitrogenous fer-
tilizer (usually urea). In some countries,
only the price of compound fertilizer
was available; in these cases the variable
refers to the price of active nutrients
only, whether N, P20,5 and/or K20.








Eastern and Southern Africa


3r


Producers


IAngola Burundi Ethiopia Kenya Lesotho Madagascar Malawi Mozambique


I Estimated population. 1991 million)
2 Estimated growth rate of population,
1989-2000 c .r l
3. Percapita income 1990 1 LSSi
4. Per capital cereal production. 1988-90 (kg/yr)
5. Growth rate of per capital cereal production,
1981-90 1It/lyr


6. Maize area harvested. 19 9-91 1000ha)
7 Maize yield. I19-91 It/hal
8 Maize production. 1989-91 ill(0 t)
9. Gro\w h rate of maize area. 1971-80(%)
10. Growth rate of maize area. 1981-90 (%)
I I. Growth rate of maize lield. 1971-80(%)
12 Growth rate of maize yield. 1981-90(%)
13. Growth rate ol maize production,
1971-80 (1 )
14. Growth rate of maize production,
1981-90 1 It
15. Maize area as percent of total cereal area,
1989.91 ( ii
16 Aserage yield of all cereals. 1989-91 (t/ha)
17 Growth rate of eld of all cereals,
1971-80 (1,/yvr
18. Growth rate of yield of all cereals,
1981-90 ti'l r


19 Net imports of maze. 1989-90 1000 t)
20 Net imponr of maize per capital.
1988.90 (kg/lri
21 Per capital total maize consumption,
1988-90 ikg/yri
22. Growth rate of per capita maize consumption,
1971 -1 () ic./ r
21 Growth rate of per capital maize consumption,
1981-90 1i /yr

24 Public sector maize \arietal releases,
1966-90
25. Public sector maize \anetal releases
per million ha maize area. 1966-90
26 Number of public sector maize
researchers. 1090
27 Number of public sector maize researchers
per million ha maize area. 19(90
28 Number of prit ale sector maize
researcher,. 1990
29. Number of pri% ate sector maize researchers
per million ha maize area. 1990

30. Fertilizer applied per heciare of arable land,
1987-88 ikg nuirlents/hai
.31. Growth rate of NPK consumption perhaof
arable land. 197(1-71 to 198889 (%/yr)

32 Farm price of maize. 19901-91 IUS$/t)
33 Ratio of farm-leel nnrogen pnce
to maize price. 1990-91
3-1 Farm wage in kg of maize per day, 1990-91


I


5.6 51.2 24.8 1.8 12.3


9.0 16.2


3.1 3.4
210 120
56 127


-3.9 -8.0 -1.2 0.7 -0.7 -1.1 -4.0 -2.1


1,070
1.7
1,791
-0.6
4.5
4.2
0.2


1,535
1.7
2,572
0.0
3.6
-1.8
0.9


1,332
1.1
1,481
0.3
1.7
0.1
-1.2


1,008
0.4
370
4.3
1.2
-4.0
-0.8


-2.3 1.6 3.6 -1.9 5.0 1.4 0.4 0.3

-2.4 2.3 4.7 4.6 3.2 4.1 0.5 0.4

86 57 21 79 61 12 94 65
0.3 1.4 1.2 1.6 0.7 1.9 1.1 0.4

-3.8 1.1 4.6 -1.6 8.4 -1.0 0.4 -3.9

-6.2 2.0 0.6 0.4 -0.2 2.2 -1.4 -2.9


71 0 8 -122 66 -23 85 248

7 0 0 -5 38 -2 10 16

30 32 38 115 107 12 179 40

2.2 -0.1 1.3 -4.6 5.4 -1.2 -2.3 2.0


-5.8 -0.7 2.2 -0.5 0.7 -1.0


11 16 20

89 14 13

2 12 58

16 10 39

0 0 2

0 0 1


3 2 5 51 14


12.7


152 188 167

4.7 4.0 3.7
6 5 8


4.3 14.0 -1.0


-2.3 0.9


6 17 14

40 13 19


2

13

5

33


8 10

6 13

2 11

1 15


22 1

7.8 -2.6


86 395

6.9 1.9
.4 3


) ------------~-----


--


I


I


i~iij~














Producers I


Rwanda Somalia Swaziland Tanzania Uganda


Zambia Zimbabwe


Regional
total or
average


1. Estimated population, 1991 (million) 7.5 7.7 0.8 28.0 19.3 8.7 9.9 244.4
2. Estimated growth rate of population,
1989-2000 (%/yr) 3.9 3.1 .. 3.1 3.3 3.1 2.4 3.1
3. Per capital income 1990 (US$) 310 120 810 110 220 420 640 248
4. Per capital cereal production, 1988-90 (kg/yr) 38 84 192 156 82 214 285 127
5. Growth rate of per capital cereal production,
1981-90 (%/yr) -4.8 3.1 6.7 1.1 -1.0 3.3 -1.3 -1.1

6. Maize area harvested, 1989-91 (000 ha) 85 213 83 1,820 406 788 1,148 10,851
7. Maize yield, 1989-91 (t/ha) 1.2 1.1 1.4 1.4 1.5 1.9 1.6 1.3
8. Maize production, 1989-91 (000 t) 103 238 115 2,634 603 1,462 1,836 13,953
9. Growth rate of maize area, 1971-80 (%) 4.2 0.3 -1.2 4.2 -0.9 -6.8 0.1 0.4
10. Growth rate of maize area, 1981-90 (%) 1.1 5.0 4.2 4.2 3.2 70 -1.9 2.8
11. Growth rate of maize yield, 1971-80 (%) 1.6 1.0 -2.5 5.4 0.0 7.8 -3.1 0.4
12. Growth rate of maize yield, 1981-90 (%) -0.5 5.2 6.0 3.1 0.8 -0.1 2.9 1.0
13. Growth rate of maize production,
1971-80 (%) 5.8 0.4 -3.7 9.6 -0.9 1.0 -3.0 0.8
14. Growth rate of maize production,
1981-90 (%) 0.5 10.2 10.3 7.2 4.0 6.9 0.9 3.8
15. Maize area as percent of total cereal area,
1989-91 (%) 34 32 96 61 38 86 71 42
16. Average yield of all cereals, 1989-91 (t/ha) 1.2 0.7 1.4 1.4 1.5 1.7 1.5 1.1
17. Growth rate of yield of all cereals,
1971-80 (%/yr) 1.2 0.3 -2.3 3.9 2.9 7.5 -1.4 1.0
18. Growth rate of yield of all cereals,
1981-90 (%/yr) -0.2 3.0 6.1 3.7 0.1 -0.3 3.4 0.6

19. Net imports of maize, 1989-90 (000 t) 0 10 10 -34 2 93 -402 80
20. Net imports of maize per capital,
1988-90 (kg/yr) 0 1 13 -1 0 11 -43 0.35
21. Per capital total maize consumption,
1988-90 (kg/yr) 14 46 194 99 28 211 176 64
22. Growth rate of per capital maize consumption,
1971-80 (%/yr) 2.5 -1.1 -6.7 5.7 -3.3 -1.4 -4.8 -1.0
23. Growth rate of per capital maize consumption,
1981-90 (%/yr) -2.8 -0.4 4.2 2.0 0.1 2.3 -2.7 0.0

24. Public sector maize varietal releases,
1966-90 .. .. 13 4 16 27
25. Public sector maize varietal releases
per million ha maize area, 1966-90 .. .. 8 11 21 23
26. Number of public sector maize
researchers, 1990 .. 5 14 7 14 6
27. Number of public sector maize researchers
per million ha maize area, 1990 .. .. 67 8 18 18 5
28. Number of private sector maize
researchers, 1990 .. .. 4 2 0 0 19
29. Number of private sector maize researchers
per million ha maize area, 1990 .. .. 53 1 0 0 17

30. Fertilizer applied per hectare of arable land,
1987-88 (kg nutrients/ha) 1 2 46 8 .. 16 54 11
31. Growth rate of NPK consumption per ha of
arable land, 1970-71 to 1988-89 (%/yr) .. -0.1 2.5 6.4 .. 3.9 0.6 3.8

32. Farm price of maize, 1990-91 (US$/t) .. .. .. 51 108 68 72
33. Ratio of farm-level nitrogen price
to maize price, 1990-91 .. 8.2 3.7 6.4 6.4
34. Farm wage in kg of maize per day, 1990-91 .. .. .. 3 4 7 17


Eastern and Southern Africa (continued)







Western and Central Africa


Producers


Benin Burkina Faso Cameroon C6oe dI'oire Ghana Guinea

1. Estimated population, 1991 (million) 4.8 9.3 12.1 12.4 15.4 5.9
2. Estimated growth rate of population,
1989-2000 (%/yr) 2.9 2.9 2.9 3.5 3.0 2.8
3. Per capital income 1990 (US$) 360 330 960 750 390 440
4. Per capital cereal production, 1988-90 (kg/yr) 127 222 80 103 72 141
5. Growth rate of per capital cereal production,
1981-90 (%/yr) 2.8 4.1 -3.9 0.8 4.5 5.0

6. Maize area harvested, 1989-91 (000 ha) 460 207 217 689 547 91
7. Maize yield, 1989-91 (t/ha) 0.9 1.3 1.8 0.7 1.4 1.0
8. Maize production, 1989-91 (000 t) 411 270 400 491 745 96
9. Growth rate of maize area, 1971-80 (%) 1.5 3.0 0.5 6.3 -3.3 -3.7
10. Growth rate of maize area, 1981-90 (%) 0.7 7.5 -8.4 4.1 3.3 11.2
11. Growth rate of maize yield, 1971-80 (%) 3.5 3.3 -2.0 -3.8 -1.2 0.0
12. Growth rate of maize yield, 1981-90 (%) 4.0 3.9 6.0 -2.3 6.0 -0.3
13. Growth rate of maize production,
1971-80 (%) 5.0 6.4 -1.4 2.5 -4.5 -3.8
14. Growth rate of maize production,
1981-90 (%) 4.6 11.4 -2.5 1.8 9.3 10.9
15. Maize area as percent of total cereal area,
1989-91 (%) 71 8 27 49 51 11
16. Average yield of all cereals, 1989-91 (t/ha) 0.8 0.7 1.1 0.9 1.1 0.9
17. Growth rate of yield of all cereals,
1971-80 (%/yr) 3.0 3.2 0.4 -1.8 -1.5 0.0
18. Growth rate of yield of all cereals,
1981-90 (%/yr) 3.7 2.8 4.2 -0.6 6.1 2.7

19. Net imports of maize, 1989-90 (000 t) 2 2 7 12 9 0
20. Net imports of maize per capital,
1988-90 (kg/yr) 0 0 1 1 1 0
21. Per capital total maize consumption,
1988-90 (kg/yr) 96 27 35 42 47 17
22. Growth rate of per capital maize consumption,
1971-80 (%/yr) 2.6 3.1 -3.9 -1.1 -4.4 -5.0
23. Growth rate of per capital maize consumption,
1981-90 (%/yr) 1.7 7.6 -5.5 -1.5 3.6 11.3

24. Public sector maize varietal releases,
1966-90 9 18 22 12 12
25. Public sector maize varietal releases
per million ha maize area, 1966-90 20 83 92 17 26
26. Number of public sector maize
researchers, 1990 6 5 35 8 16
27. Number of public sector maize researchers
per million ha maize area, 1990 14 22 146 12 34
28. Number of private sector maize
researchers, 1990 0 0 0 5 0
29. Number of private sector maize researchers
per million ha maize area, 1990 0 0 0 7 0

30. Fertilizer applied per hectare of arable land,
1987-88 (kg nutrients/ha) 3 4 5 11 4 1
31. Growth rate of NPK consumption per ha of
arable land, 1970-71 to 1988-89 (%/yr) 4.3 19.7 7.4 3.1 5.5 -4.9

32. Farm price of maize, 1990-91 (US$/t) .... 120 143 92
33. Ratio of farm-level nitrogen price
to maize price, 1990-91 .... 7.2 7.6 5.9
34. Farm wage in kg of maize per day, 1990-91 .... 33 17 15







Western and Central Africa (continued)


0.


Producers


Nigeria Senegal


Togo


Zaire


Regional
total or
average


1. Estimated population, 1991 (million) 9.5 111.0 7.6 3.6 36.5 260.8
i 2. Estimated growth rate of population,
S 1989-2000 (%/yr) 3.0 2.8 3.1 3.2 3.0 2.9
- 3. Per capital income 1990 (US$) 270 290 710 410 220 388
2 4. Per capital cereal production, 1988-90 (kg/yr) 239 123 135 144 37 115
5. Growth rate of per capital cereal production,
1981-90 (%/yr) 4.2 2.9 0.9 2.3 0.7 2.0

6. Maize area harvested, 1989-91 (000 ha) 172 1,550 99 261 1,211 5,723
7. Maize yield, 1989-91 (t/ha) 1.3 1.3 1.3 1.0 0.7 1.1
8. Maize production, 1989-91 (000 t) 216 1,955 123 269 874 6,083
9. Growth rate of maize area, 1971-80 (%) -7.1 -12.0 6.9 1.9 2.3 -1.0
S 10. Growth rate of maize area, 1981-90 (%) 10.4 15.1 4.6 4.3 6.1 5.5
S11. Growth rate of maize yield, 1971-80 (%) 8.3 4.8 0.7 -1.7 1.0 0.0
-a 12. Growth rate of maize yield, 1981-90 (%) 2.9 -0.5 1.8 2.0 -2.6 1.7
S 13. Growth rate of maize production,
. 1971-80 (%) 1.1 -7.1 7.6 0.2 3.2 -1.0
S 14. Growth rate of maize production,
0 1981-90 (%) 13.2 14.6 6.4 6.3 3.5 7.2
15. Maize area as percent of total cereal area,
1989-91 (%) 7 13 8 43 70 17
S 16. Average yield of all cereals, 1989-91 (t/ha) 0.9 1.2 0.8 0.8 0.8 0.9
17. Growth rate of yield of all cereals,
1971-80 (%/yr) 0.6 5.9 2.0 -3.3 0.7 2.1
18. Growth rate of yield of all cereals,
1981-90 (%/yr) 2.6 0.9 2.5 1.6 -1.6 1.6

19. Net imports of maize, 1989-90 (000 t) 2 0 9 2 58 227
o 20. Net imports of maize per capital,
1988-90 (kg/yr) 0 0 1 1 2 1
21. Per capital total maize consumption,
" 1988-90 (kg/yr) 25 19 19 79 26 25
S 22. Growth rate of per capital maize consumption,
1971-80 (%/yr) -2.4 -8.1 -1.2 -2.1 0.4 -3.1
23. Growth rate of per capital maize consumption,
1981-90 (%/yr) 5.0 7.8 3.7 2.9 -1.8 2.5

24. Public sector maize varietal releases,
1966-90 10 37 12 4 15
25. Public sector maize varietal releases
S per million ha maize area, 1966-90 71 23 103 18 13
26. Number of public sector maize
researchers, 1990 4 28 12 19 4
27. Number of public sector maize researchers
per million ha maize area, 1990 30 17 98 86 3
S28. Number of private sector maize
S researchers, 1990 0 11 0 3 0
S 29. Number of private sector maize researchers
per million ha maize area, 1990 0 7 0 14 0

S30. Fertilizer applied per hectare of arable land,
1 1987-88 (kg nutrients/ha) 6 10 5 7 0 6
S 31. Growth rate of NPK consumption per ha of
arable land, 1970-71 to 1988-89 (%/yr) 12.4 21.0 -0.8 18.8 -0.7 9.7

32. Farm price of maize, 1990-91 (US$/t) .. 201
S33. Ratio of farm-level nitrogen price
C to maize price, 1990-91 .. 1.1
34. Farm wage in kg of maize per day, 1990-91 .. 11







North Africa


Producers


Egypt


Consumers


Morocco


Algeria


Libya


Tunisia


Regional
total or
average


1. Estimated population, 1991 (million) 53.1 25.6 25.7 4.7 8.3 117.4
2. Estimated growth rate of population,
1989-2000 (%/yr) 1,8 2.4 2.8 3.6 1.9 2.2
3. Per capital income 1990 (US$) 600 950 2,060 .. 1,440 1,071
4. Per capital cereal production, 1988-90 (kg/yr) 215 297 57 81 107 186
5. Growth rate of per capital cereal production,
1981-90 (%/yr) 1.6 8.1 -4.8 -1.0 -7.3 2.2

6. Maize area harvested, 1989-91 (000 ha) 861 389 1 1 .. 1,252
7. Maize yield, 1989-91 (t/ha) 5.7 1.0 ..... 4.2
8. Maize production, 1989-91 (000 t) 4,866 391 ..... 5,260
9. Growth rate of maize area, 1971-80 (%) 2.5 -1.7 ..... 0.9
10. Growth rate of maize area, 1981-90 (%) 0.0 -0.1 ..... -0.1
11. Growth rate of maize yield, 1971-80 (%) 0.8 0.7 ..... 2.0
12. Growth rate of maize yield, 1981-90 (%) 3.3 11.6 ..... 3.8
13. Growth rate of maize production,
1971-80 (%) 3.3 -1.0 ..... 2.8
14. Growth rate of maize production,
1981-90 (%) 3.2 11.5 .... 3.7
15. Maize area as percent of total cereal area,
1989-91 (%) 38 7 0 0 .. 10
16. Average yield of all cereals, 1989-91 (t/ha) 5.5 1.3 0.8 0.7 1.2 2.0
17. Growth rate of yield of all cereals,
1971-80 (%/yr) 0.3 -1.2 0.7 -4.3 -1.9 -0.3
18. Growth rate of yield of all cereals,
1981-90 (%/yr) 3.2 7.6 0.0 2.2 -1.1 3.9

19. Net imports of maize, 1989-90 (000 t) 1,585 112 1,116 258 257 3,327
20. Net imports of maize per capital,
1988-90 (kg/yr) 31 5 46 59 32 30
21. Per capital total maize consumption,
1988-90 (kg/yr) 116 21 46 59 32 72
22. Growth rate of per capital maize consumption,
1971-80 (%/yr) 3.0 -0.4 22.2 29.6 26.3 3.2
23. Growth rate of per capital maize consumption,
1981-90 (%/yr) 1.1 3.1 14.1 14.8 -3.6 2.3

24. Public sector maize varietal releases,
1966-90 13 15
25. Public sector maize varietal releases
per million ha maize area, 1966-90 16 40
26. Number of public sector maize
researchers, 1990 69 18
27. Number of public sector maize researchers
per million ha maize area, 1990 83 48
28. Number of private sector maize
researchers, 1990 23 0
29. Number of private sector maize researchers
per million ha maize area, 1990 28 0

30. Fertilizer applied per hectare of arable land,
1987-88 (kg nutrients/ha) 400 37 23 41 21 66
31. Growth rate of NPK consumption per ha of
arable land, 1970-71 to 1988-89 (%/yr) 6.7 5.6 1.9 10.4 6.2 5.8

32. Farm price of maize, 1990-91 (US$/t) 127
33. Ratio of farm-level nitrogen price
to maize price, 1990-91 2.38
34. Farm wage in kg of maize per day, 1990-91 11.9


Producers Consumers__







I ~


Afghanistan


Turkey I


Jordan


1. Estimated population, 1991 (million) .. 56.9 57.0 19.5 3.4
2. Estimated growth rate of population,
1989-2000 (%/yr) .. 1.9 3.4 3.4 3.8
3. Per capital income 1990 (US$) .. 1,630 2,490 .. 1,240
4. Per capital cereal production, 1988-90 (kg/yr) 218 514 209 113 35
5. Growth rate of per capital cereal production,
1981-90 (%/yr) -4.5 -1.0 -1.7 -2.9 -1.5

6. Maize area harvested, 1989-91 (000 ha) 264 513 3 42 <1
7. Maize yield, 1989-91 (t/ha) 1.6 4.0 .. 2.9
8. Maize production, 1989-91 (000 t) 436 2,067 .. 121
9. Growth rate of maize area, 1971-80 (%) 0.0 -1.0 .. 14.7
10. Growth rate of maize area, 1981-90 (%) -0.9 -1.6 .. 16.1
11. Growth rate of maize yield, 1971-80 (%) 1.3 3.3 .. 5.4
12. Growth rate of maize yield, 1981-90 (%) 0.3 7.9 .. -0.5
13. Growth rate of maize production,
1971-80 (%) 1.2 2.3 .. 20.1
14. Growth rate of maize production,
1981-90 (%) -0.6 6.3 .. 15.6
15. Maize area as percent of total cereal area,
1989-91 (%) 11 4 0 1 0
16. Average yield of all cereals, 1989-91 (t/ha) 1.2 2.1 1.3 0.6 0.8
17. Growth rate of yield of all cereals,
1971-80 (%/yr) 2.9 4.0 3.9 -4.5 -7.9
18. Growth rate of yield of all cereals,
1981-90 (%/yr) -0.1 1.1 0.5 1.1 3.8

19. Net imports of maize, 1989-90 (000 t) 0 404 648 545 323
20. Net imports of maize per capital,
1988-90 (kg/yr) 0 7 12 30 102
21. Per capital total maize consumption,
1988-90 (kg/yr) 50 44 12 35 102
22. Growth rate of per capital maize consumption,
1971-80 (%/yr) -0.3 0.0 20.8 25.7 13.6
23. Growth rate of per capital maize consumption,
1981-90 (%/yr) -1.0 6.3 -6.2 14.4 9.2

24. Public sector maize varietal releases,
1966-90
25. Public sector maize varietal releases
per million ha maize area, 1966-90
26. Number of public sector maize
researchers, 1990 4 19
27. Number of public sector maize researchers
per million ha maize area, 1990 9 38
28. Number of private sector maize
researchers, 1990 0 24
29. Number of private sector maize researchers
per million ha maize area, 1990 0 48

30. Fertilizer applied per hectare of arable land,
1987-88 (kg nutrients/ha) 6 58 70 45 73
31. Growth rate of NPK consumption per ha of
arable land, 1970-71 to 1988-89 (%/yr) 6.7 7.3 12.2 13.7 17.6

32. Farm price of maize, 1990-91 (US$/t) .. 196 86
33. Ratio of farm-level nitrogen price
to maize price, 1990-91 .. 2.0 0.2
34. Farm wage in kg of maize per day, 1990-91 .. 36 21


West Asia






Producers Consumers







West Asia (continued)


Consumers


Lebanon


Saudi
Arabia


Syria


Regional
total or
average


1. Estimated population, 1991 (million) .. 14.6 13.0 200.9
2. Estimated growth rate of population,
1989-2000 (%/yr) .. 3.7 3.6 2.7
3. Per capital income 1990 (US$) .. 7,050 1,000 2,762
4. Per capital cereal production, 1988-90 (kg/yr) 28 264 261 277
5. Growth rate of per capital cereal production,
1981-90 (%/yr) 14.0 23.7 -3.6 -1.0

6. Maize area harvested, 1989-91 (000 ha) 2 2 59 886
7. Maize yield, 1989-91 (t/ha) .... 2.7 3.2
8. Maize production, 1989-91 (000 t) .... 158 2,802
9. Growth rate of maize area, 1971-80 (%) .... 12.6 0.2
10. Growth rate of maize area, 1981-90 (%) .... 13.6 -1.4
11. Growth rate of maize yield, 1971-80 (%) .... 6.6 2.5
12. Growth rate of maize yield, 1981-90 (%) .... -0.9 5.3
13. Growth rate of maize production,
1971-80 (%) .... 19.3 2.7
14. Growth rate of maize production,
1981-90 (%) .... 12.6 3.9
15. Maize area as percent of total cereal area,
1989-91 (%) 5 0 2 3
16. Average yield of all cereals, 1989-91 (t/ha) 1.9 4.7 0.8 1.6
17. Growth rate of yield of all cereals,
1971-80 (%/yr) 1.8 0.0 4.8 3.2
18. Growth rate of yield of all cereals,
1981-90 (%/yr) 6.2 13.0 0.8 1.7

19. Net imports of maize, 1989-90 (000 t) 111 402 152 2,904
20. Net imports of maize per capital,
1988-90 (kg/yr) 41 30 13 16
21. Per capital total maize consumption,
1988-90 (kg/yr) 42 30 22 32
22. Growth rate of per capital maize consumption,
1971-80 (%/yr) 4.8 24.1 29.0 5.0
23. Growth rate of per capital maize consumption,
1981-90 (%/yr) -4.0 -13.2 -2.7 1.0

24. Public sector maize varietal releases,
1966-90
25. Public sector maize varietal releases
per million ha maize area, 1966-90
26. Number of public sector maize
researchers, 1990
27. Number of public sector maize researchers
per million ha maize area, 1990
28. Number of private sector maize
researchers, 1990
29. Number of private sector maize researchers
per million ha maize area, 1990

30. Fertilizer applied per hectare of arable land,
1987-88 (kg nutrients/ha) 75 463 52 60
31. Growth rate of NPK consumption per ha of
arable land, 1970-71 to 1988-89 (%/yr) 2.9 30.0 11.6 9.5

32. Farm price of maize, 1990-91 (US$/t)
33. Ratio of farm-level nitrogen price
to maize price, 1990-91 ... 1.3
34. Farm wage in kg of maize per day, 1990-91 ... 7







South Asia


44


Producers


Myanmar


Nepal


Pakistan


Regional
total or
average


I. Estimated population. 1991 (million) 864.4 42.5 19.6 125.2 1,187.5
g 2 Estimated growth rate of population,
I 99 2000 ityr 1.7 2.0 2.5 2.7 1.8
5 3. Per capital income 1990 I LS$) 350 ... 170 380 338
4. Per capital cereal production, 1988-90 (kg/yr) 232 346 292 172 230
Q 5 Growth rate of per Lapita cereal production,
19S 1 -90 i'O ri 1.2 -2.7 2.3 -2.1 0.6

6. Maize area haresied. 1989-91 (000 ha) 5,856 124 760 856 7,674
7 Maize l field. 1989-91 i/hai 1.5 1.5 1.6 1.4 1.5
8 Maize production. 1989-91 (000 t) 8,975 190 1,222 1,185 11,645
'U 9 Growth rate of maize area. 1971-80 (%) 0.0 4.5 0.0 1.6 0.3
10 Growth rate of maize area. 1981-90 (%) 0.0 -3.0 4.9 1.5 0.5
SII Growth rate of maize sield, 1971-80 (%) 1.3 5.7 -2.1 1.3 1.0
S 12. Growth rate ol maize field, 1981-90 (%) 2.7 -0.5 -0.3 1.6 2.3
I3. Groiwh rate ol maize production,
S 1971-80 i1 1.4 10.2 -2.1 2.9 1.3
14. Grovih rate of maize production,
S 1981--90 i- 2.8 -3.5 4.7 3.2 2.8
S1 Maize area as percent ol total cereal area,
1989-91 1 6 2 25 7 6
16 Average fieldd of all cereals, 1989-91 (t/ha) 1.9 2.7 1.9 1.8 2.0
17 Growth rate of yield of all cereals,
1971-80 1%/ ri 2.2 4.7 -1.0 2.9 2.4
18 Growth rate of field of all cereals,
1981 -901 1/ /r i 3.5 -0.5 1.7 0.8 2.9

19. Net imports of maize. 1989-90 (000 t) 94 -5 1 0 119
a. 20 Net impons of maize per capital,
1988.-00kg/ ri <1 <-1 0 0 <1
21. Per capital tloal maize consumption,
1988-90 ikg/r 11 5 58 10 10
22 Growth rate of per capital maize consumption,
1 47I -80 oi/Nri -0.7 9.2 -4.7 0.3 -0.9
23. Grow th rate of per capiia maize consumption,
19 81 -9.0 11'-rt i 0.7 -4.6 2.4 -0.6 0.6

2-1 Public sectorr maize arielal releases,
S 1966-90 43 15 9 21
2 Public ,ecior maize .areilal releases
per million ha maize area. 1966-90 7 122 13 25
26 Number of public sector maize
researchers. 1990 168 11 26 26
27. Number of public sector maize researchers
S per million ha maize area. 1990 28 89 39 30
S 28 Number of pnalie sector maize
researchers. 199t1. 50 0 0 13
29. Number of private sector maize researchers
per million ha maize area. 1990 8 0 0 15

., 30. Fertilizer applied per hectare of arable land,
I 1987.88ikg nutrients.hai 65 11 24 83 65
S31 Growth rate of NPK consumption per ha of
arhble land. 1970.71 to 1988-89 (%/yr) 9.2 10.4 11.0 9.6 9.2

32 Farm price of maize. 1090-91 (US$/t) 113 .. 146
S33. Raiio of farm-le el nitrogen price
to maize price. 1000-91 2.6 .. 2.5
3-1 Farm wage in kg ol maize per day, 1990-91 9 .. 7


1







Southeast Asia and Pacific


Producers


Indonesia Kampuchea Philippines Thailand Vietnam


Consumers


Malay.sia Singapore


Regional
total or
average


1. Estimated population, 1991 (million) 186.7 8.3 63.1 56.5 68.0 18.2 2.7 421.5
2. Estimated growth rate of population,
1989-2000 (%/yr) 1.6 1.9 1.8 1.4 2.1 2.3 1.2 1.7
3. Per capital income 1990 (US$) 570 .. 730 1,420 .. 2,320 11,160 1,146
4. Per capital cereal production, 1988-90 (kg/yr) 277 309 227 449 291 101 .. 278
5. Growth rate of per capital cereal production,
1981-90 (%/yr) 1.7 4.1 0.5 -0.4 1.7 -3.8 .. 1.0

6. Maize area harvested, 1989-91 (000 ha) 3,037 40 3,699 1,644 484 20 .. 8,961
7. Maize yield, 1989-91 (t/ha) 2.1 1.3 1.3 2.5 1.5 .... 1.8
8. Maize production, 1989-91 (000 t) 6,445 52 4,677 4,035 720 .... 16,026
9. Growth rate of maize area, 1971-80 (%) 0.5 1.7 3.5 4.1 7.2 .... 2.5
10. Growth rate of maize area, 1981-90 (%) 2.1 -10.3 2.0 1.9 3.9 .... 2.0
11. Growth rate of maize yield, 1971-80 (%) 4.1 -2.7 2.3 0.4 0.5 .... 2.4
12. Growth rate of maize yield, 1981-90 (%) 3.8 4.3 3.0 0.0 4.6 .... 2.6
13. Growth rate of maize production,
1971-80 (%) 4.6 -1.0 5.7 4.5 7.7 .... 4.9
14. Growth rate of maize production,
1981-90 (%) 5.9 -6.0 5.1 2.0 8.5 .... 4.6
15. Maize area as percent of total cereal area,
1989-91 (%) 23 2 52 14 7 3 .. 21
16. Average yield of all cereals, 1989-91 (t/ha) 3.8 1.4 2.0 2.0 3.0 2.6 .. 2.7
17. Growth rate of yield of all cereals,
1971-80 (%/yr) 3.3 -3.5 3.5 -0.3 -1.5 2.5 .. 1.3
18. Growth rate of yield of all cereals,
1981-90 (%/yr) 2.1 5.4 1.8 0.5 3.3 -0.8 .. 2.0

19. Net imports of maize, 1989-90 (000 t) -100 0 174 -1,208 -41 1,339 143 400
20. Net imports of maize per capital,
1988-90 (kg/yr) -1 0 3 -22 -1 77 53 1
21. Per capital total maize consumption,
1988-90 (kg/yr) 36 6.3 78 55 12 6 53 41
22. Growth rate of per capital maize consumption,
1971-80 (%/yr) 3.4 1.0 3.2 0.2 2.6 11.2 14.9 4.6
23. Growth rate of per capital maize consumption,
1981-90 (%/yr) 3.7 -8.7 1.6 18.4 6.5 4.0 -14.3 4.0

24. Public sector maize varietal releases,
1966-90 18 .. 14 7 22
25. Public sector maize varietal releases
per million ha maize area, 1966-90 6 .. 4 4 43
26. Number of public sector maize
researchers, 1990 37 .. 43 22 60
27. Number of public sector maize researchers
per million ha maize area, 1990 12 .. 11 13 118
28. Number of private sector maize
researchers, 1990 7 .. 117 65 0
29. Number of private sector maize researchers
per million ha maize area, 1990 2 .. 31 38 0

30. Fertilizer applied per hectare of arable land,
1987-88 (kg nutrients/ha) 113 .. 63 39 81 151 .. 80
31. Growth rate of NPK consumption per ha of
arable land, 1970-71 to 1988-89 (%/yr) 12.1 .. 4.4 7.6 0.7 3.0 .. 7.6

32. Farm price of maize, 1990-91 (US$/t) 119 49 177 96 120 167
33. Ratio of farm-level nitrogen price
to maize price, 1990-91 2.1 .. 2.6 4.4 4.0
34. Farm wage in kg of maize per day, 1990-91 14 9 .. 24 4 31










46 East Asia


I Eliimated population. 1991 million
2 Eimated gro%%th rate ol population.
1989.2000 1 i/\,rI
3 Per Lapla income 1990 1 LIS5S
4 Per Iapita cereal production. 198.X-90 Ikgl/)r
5. Growth rate of per capital cereal production.
1l9 .1-90ic./iri

6 Maize area harnesied. 1989.9 1 I(110 hai
7 Maize Nield. 19~l -Q1 tl/hai
X Maize production. 1989-91 111)( lO
Q Gromilh rate of ma.ze area. 19 1 -80 -I I
10 Grow ih rate of maize area. 1981 -90 1C
I I Gro\ th rate of maize N eld. li7 I -,s, I i
12 Gro, lh rate of maze field. I191 -1.1 I I
13 Groth rate of maize production.
1971 -S0) it,
11 Grolth rate of maize production.
19 .1-90 i -
I Maize area a, percent of 1otal cereal area.
1989-91 1, Ci
16 A. erjge 1ield ol all cereal. 1989-91 litha
I Groul h rate of N field of all cereal,.
1971-80 11 r/,ri
IK. Gro th rate of field of all cereal,.
19 1.-90 iC rl

19 Nel import, of maize. 1989-90t.l 11110
20 Nel imporcn of maie per capital.
1988-90 kp/' r
2I Per capita total maize consumption.
19h8-90 I kg/. rl
22. Gro th rate of per capital maize conumption.
1971 -80i 1 / r
23 Gro wlh rate of per capita maize comnumption.
198 1-90 'N\r1

24. Public sector maize %arielal releases.
196hh-90
25 Public sector maize \n.'etal release,
per million ha maize area. l966-90
26. Number nl public sectorr maize
researcher 19941
27 Number ol public .eclor maize researcher,
per million ha maize area. 1990
28 Number ol pri% ate eclor maize
researcher,. 1990
29 Number of prit ae secior maize researchers
per million ha maize area. 1990

30 Fenilizer applied per hectare of arable land.
1987- 8 ikg nuinentlshai
31 Growth rate ot NPK consumption per ha of
arable land. 1970-'l to 19i,-s-89 I,/ri'.

32. Famn price of maize. 1990-041 iLS$.'It
33. Raino of farm-level nitrogen price
to maize price. 1990-91
31 Farm \age in kg of maize per da\. 10009(.91


Producers


China


1.131 0

I .




13
3"O0
333

I 1

21.80-14
4.3
89.922
7i
12
44
2h


Consumers


Korea
D.P.R


Korea
Republic


-13 2

11 9
5.300
205

-0

214
4.3
105
-18
.29
I27
20

1i19


17

6.10(1


3.4- 24


Taiwan


*II 3



I -I('
140

-36

8-1
4'

4 7
109
1 0
39

5 7


(I I

4 627


I I

6 2


262 33, 4-!10

1I.. 4 _2.0


*Data for 1988-90.
** World Bank (1991).


Regional
total or
average


1.218.5

S3
555


199

21.48
-1
94.478
2 7
1.-
4.3
2.7

7.0

SI


22
4.2


2 5

7.501

6

79

59

2.5







Mexico, Central America, and the Caribbean


Producers


El Salvador Guatemala Haiti Honduras Mexico Nicaragua Panama

1. Estimated population, 1991 (million) 5.3 9.4 6.6 5.3 89.9 4.0 2.4
2. Estimated growth rate of population,
1989-2000 (%/yr) 1.8 2.8 1.9 2.9 1.8 3.0 1.6
3. Per capital income 1990 (US$) 1,110 900 370 590 2,490 .. 1,830
4. Per capital cereal production, 1988-90 (kg/yr) 157 170 63 125 262 129 133
5. Growth rate of per capital cereal production,
1981-90 (%/yr) 1.5 0.3 -2.5 -1.4 -2.3 -2.3 0.1

6. Maize area harvested, 1989-91 (000 ha) 288 611 208 384 6,953 199 72
7. Maize yield, 1989-91 (t/ha) 2.0 2.0 0.8 1.4 1.9 1.3 1.3
8. Maize production, 1989-91 (000 t) 565 1,230 168 551 13,036 251 93
9. Growth rate of maize area, 1971-80 (%) 4.1 -0.1 -0.5 4.3 -1.8 -3.7 -0.2
10. Growth rate of maize area, 1981-90 (%) 1.4 1.3 3.0 2.2 -0.5 2.3 2.7
11. Growth rate of maize yield, 1971-80 (%) 1.8 2.2 -3.9 -2.0 3.9 4.0 2.9
12. Growth rate of maize yield, 1981-90 (%) 2.1 1.8 -2.6 0.0 0.0 3.3 3.8
13. Growth rate of maize production,
1971-80 (%) 5.9 2.1 -4.3 2.2 2.1 0.3 2.7
14. Growth rate of maize production,
1981-90 (%) 3.5 3.1 0.3 2.1 -0.6 5.6 6.4
15. Maize area as percent of total cereal area,
1989-91 (%) 67 87 55 80 69 69 41
16. Average yield of all cereals, 1989-91 (t/ha) 1.8 2.0 1.0 1.4 2.3 1.5 1.8
17. Growth rate of yield of all cereals,
1971-80 (%/yr) 1.3 2.2 -2.5 -2.3 4.1 4.5 3.0
18. Growth rate of yield of all cereals,
1981-90 (%/yr) 1.7 1.5 -1.3 0.3 0.0 -0.9 2.5

19. Net imports of maize, 1989-90 (000 t) 56 62 4 41 3,684 32 32
20. Net imports of maize per capital,
1988-90 (kg/yr) 11 7 1 8 42 9 14
21. Per capital total maize consumption,
1988-90 (kg/yr) 127 150 29 112 182 83 55
22. Growth rate of per capital maize consumption,
1971-80 (%/yr) 4.8 -0.2 -5.2 0.3 1.2 -2.5 -0.3
23. Growth rate of per capital maize consumption,
1981-90 (%/yr) 2.2 0.9 -2.0 -0.3 -1.4 0.8 3.1

24. Public sector maize varietal releases,
1966-90 12 45 .. 23 83 15
25. Public sector maize varietal releases
per million ha maize area, 1966-90 43 62 .. 64 11 72
26. Number of public sector maize
researchers, 1990 10 16 .. 17 85 9
27. Number of public sector maize researchers
per million ha maize area, 1990 35 22 .. 47 11 43
28. Number of private sector maize
researchers, 1990 11 22 .. 6 55 12
29. Number of private sector maize researchers
per million ha maize area, 1990 39 30 .. 17 7 58

30. Fertilizer applied per hectare of arable land,
1987-88 (kg nutrients/ha) 111 69 2 21 71 57 67
31. Growth rate of NPK consumption per ha of
arable land, 1970-71 to 1988-89 (%/yr) -1.6 5.1 8.9 0.1 6.3 4.6 1.9

32. Farm price of maize, 1990-91 (US$/t) 181 176 .. 125 212 176 220
33. Ratio of farm-level nitrogen price
to maize price, 1990-91 3.1 3.8 .. 4.4 1.7 4.1 5.9
34. Farm wage in kgof maize per day, 1990-91 12 17 .. 12 27 16 27








48 Mexico, Central America, and the Caribbean (cont'd)


Consumers


Costa Rica


Cuba


Dominican
Republic


Jamaica


4 1


1 Estimated population, 1991 (million)
2. Estimated growth rate of population,
1989-2000 (%/yr)
3 Per capital income 1990 (US$)
4 Per capital cereal production, 1988-90 (kg/yr)
5 Growth rate of per capital cereal production,
1981-90 (%/yr)


6. Maize area harvested, 1989-91 (000 ha)
7 Maize yield, 1989-91 (t/ha)
r Maize production, 1989-91 (000 t)
9 Growth rate of maize area, 1971-80 (%)
10 Growth rate of maize area, 1981-90 (%)
II Growth rate of maize yield, 1971-80 (%)
12 Growth rate of maize yield, 1981-90 (%)
I 3 Growth rate of maize production,
1971-80(%)
14 Growth rate of maize production,
1981-90 (%)
15. Maize area as percent of total cereal area,
1989-91 (%)
Ih. Average yield of all cereals, 1989-91 (t/ha)
I 7. Growth rate of yield of all cereals,
1971-80 (%/yr)
I ? Growth rate of yield of all cereals,
1981-90 (%/yr)


3.1

1.9
1,900
105

-3.6


2.5

0.7
1,500
1

-10.5


I I -- -


Regional
total or
average


150.1

1.8
2,144
196

-1.9


8,886
1.8
16,144
-1.3
-0.1
3.4
0.2


19 Net imports of maize, 1989-90 (000 t) 168 687 397 155 5481
S 20. Net imports of maize per capital,
S 1988-90 (kg/yr) 57 65 57 64 38
2I Per capital total maize consumption,
S 1988-90 (kg/yr) 86 75 65 65 144
S 22. Growth rate of per capital maize consumption,
S 1971-80 (%/yr) -3.3 9.4 10.9 2.4 1.6
23. Growth rate of per capital maize consumption,
1981-90 (%/yr) 6.2 4.3 8.6 -4.4 -0.6

24 Public sector maize varietal releases,
1966-90 13
25 Public sector maize varietal releases
per million ha maize area, 1966-90 317
S 2b. Number of public sector maize
researchers, 1990 9
27 Number of public sector maize researchers
S per million ha maize area, 1990 220
28. Number of private sector maize
researchers, 1990 7
29 Number of private sector maize researchers
per million ha maize area, 1990 171

z 30 Fertilizer applied per hectare of arable land,
a 1987-88 (kg nutrients/ha) 191 180 52 109 79
S I Growth rate of NPK consumption per ha of
S arable land, 1970-71 to 1988-89 (%/yr) 2.2 4.8 -2.2 -1.3 4.3

12 Farm price of maize, 1990-91 (US$/t) 178 196
v 33 Ratio of farm-level nitrogen price
S to maize price, 1990-91 1.9 3.6
34. Farm wage in kg of maize per day, 1990-91 26 36







Andean Region, South America 4






Producers
Regional
total or
Bolivia Colombia Ecuador Peru Venezuela average

4 I Etimated population. 1991 million 5 33 3 Ii. 220 21. I 9)4
2 Esnimaied gro.ih rare of population.
198 l 0-' I0O ic irl 2.5 1.5 2 i 2 2.1 I 9
- 3 Per capital income 1990i tiS11 630 1.260 49S 1.160 2.500 1.433
4. Per caplla cereal production. 1988-90 ikg.r.') r !1 122 I \4 10- li19
5 Gro.lh rate of per capita cereal production.
1 9 l-90 'I ,'rliI 6 -05 63 I1 2 I.f1

6 Mlaiwe area harcied. l189L91 1000 hai 269 806 -1I 3X6 41-0 2.39)8
7 Maize ield. 1989-91 it/hja 1 6 I 5 1.2 2 I) 21 I
8 Maize production. 1989-91 10001) 439 I 177 531 7"i I.11li. 3.92S
9 Groth riteofmnaize area. 1971 -80 1 3.2 00 -57 I -10 .1.4
10 Growth rate of maize area. 1981-90 1 -10 29 92 32 79 4-
II. Gro.ilh r.le of maize field. 1'-7801 1 1.5 I 10 5 0 4 I 9
12 Groth rae ol'maize \%eld. 1981-1 I i0 -I4 .1 3 -34I 10 4.- 06
13 Grotuh rare of maize production.
1971-980t 3.7 0 2 1 0 ) 0.5
S 1-1 Grow th rale of maize production.
19 1 -90ti1 -2.4 25 5 8 4 1 12.3 5I0
15 Maize area aw percent of total cereal area.
198991 I I 44 (I 56 48 57 5u
16 A\erage fieldd of l ll cereal. 1989-ol t1 lha 1 I 2 5 2.3 2 2
17 Growth rdic of cld of all cereals.
1971-80 ( /\rl 0 2 2 5 5 4 1.0 5 5 3 2
Ix Growth rate of eld of all cereal.
1981 90 lCr/nrl .12 (1 -0 7 1.9 2.0 0 1

19 Net imports of maize. 1989.- 0 iO0 II -I 25 -2 377 198 616
. 20 Net import of miize per capia.
S 98890 kg/r -I I -. I I 10 7
S21 Per capital lolal maize conumplion.
1988-9(0 Ikg ) r 53 33 -12 58 68 9
22. Grow ih rate of per capita maize consumpnon.
1971.80i1!' rri I I 00 -4 8 0.1 5 I
S23. Grciu th rate of per capital maize consumption.
1981 -901 IOcr -5 .0.9 3 0.6 -59 22

2- Public sector maize anrieal releases.
1966-90 16 50 10 17 24
25 Public sector maize arietal releases
S per million ha maize area. 1966-90 61 60 29 53 19
'6. Number of public wecior maize
researcher-. 1990 22 11 23 3 24 4
27. Number of public sector maize researcher,
S per million ha maize area. 199(1 4 12 66 106 49
28 Number of prl\ate sector maize
researchers. 1990 6 19 6 I 25
20. Number of prT.ate sector maize researchers
per million ha maize area. 1990 23 23 17 56 51

S3.1 Fertilizer applied per heciare of arable land.
1987-8 Ikg nutrienis/ha, 2 98 27 58 1'8 7s
31 Gro\ih rate ol NPK consumption per ha of
arable land. 1970.71 t1 1988 -9iir/\rl 3 1 4 7 6." 0 II 5 6 1

32 Farm price of maize. 1990-91 itiS/til 24-1 159 162 1[11 216
33 Ratio of farm-le\el nirrogen price
to maize price. 1990-91 5 0 26 3 2 8 1.5
34. Farm age in kg of maize per da\. 199.0-1 14 17 14








0 Southern Cone, South America
5O


Producers


Argentina


Brazil


Chile Paraguay Uruguay


-, r


I. Estimated population. It01 million
2 Estimated growth rate of population.
1I%9 2 -,1011 'e % r i
3. Per tapita income 1940 I .iS1
-1 Per capilt cereal production. Il S-.90 ikg.,rir
5 Gro',th rale of per Lapita cereal production.
19 l-.90 i c,' r

6 Maize area har\teed. Ir.-)l I1100 ha
SMaize Nleld. lu~ )l I i /hal
8 Maize production. '189-91 i)11)il t
9 Growth rate of mjaie area. IL71 -80 r',
10 Gro.th rate of maze area. cI )I 90 i' I
II Gro th rate ol maie ,eld. I4-1- li I I
12 Gro th rate of mnaie Meld. I9l-9(i', ",
I 3 Groth rate of majle production.
197 1 -SO I'; I
14 Gro rth rate of irmize production
19S .1 t I
15. Malire area a, percent of total cereal area.
19h9-91 1' I
Ib .A.erage wield ol all cereal.. 199-91 it/hai
17 Grom th rate oif field of all cereal,.
1971 -,i ll i ', / r i
I Grow th rate of field ol all cereal,.
I u8 1 -90 c" / r i

I Net imporn of majie. l19.I-90 1001) 11
20. Net imlport o nmaize per capital
1989-90 ikg/r./i
2I Per capia total mjize con'umption.
1988-90, 1kg.,r
22 Grow, ih rate of per capital maize conunipilon.

23 Growth rale of per capila maize conunmplion.
19.81-,0 ct/N r i

21 Public ectLor maize \arielal release .
1066-9(1
25 Public 'ector maize \ Jreeal release
per million ha male area 1966.00
2o Number of public ,ector mdize
reearcher,. 1990
2" Number of public sector maize researcher.
per million ha maize are.. 194(0
28 Number of prit ate ector maize
researcher'.. 1O90
20 Number of pri ale ecror maize researcher,
per million ha maize area. 1990(

.10. Fernihzer applied per heclare of .rabl e land
19S7-.s ikg nurlientl/hai
31 GCro, th rate o NPK consumption per ha of
arable land. 1970 -71 o10 ) .,*9 (4'l/,ri

32 Farm price ol maize. 1)4l-90)91 ILiSh/l
33 Ratio oI larm-le\el nitrogen price
tr mail e price. I 00.1.91
34 Farm % age in kg of maize per da,. 19 0-.9 I


32.6 152.5


I I)

hl69

-* 2

1. 17
1.3
"'.692
--1.7
7 8
35
0 7

-12


1.7
2,680
270

0.8


12,071
1.9
23,505
1.3
0.9
1.3
1.2

2.6

2.1


1.9 5.6

0.3 0.9


13.3

1.3
1,940
230

7.9


4.4

2.8
1,110
390

7.8


0.6
2,560
446

1.7


516
2.0
1,040
8.4
3.6
2.7
5.3

11.0

8.9

64
1.9

2.3

4.7


-1-


8.3 -4.6


I


Regional
total or
average


205.9

1.6
2,548
328

-2.5


14,472
2.2
31,202
0.0
-0.5
1.4
0.2

1.5

-0.2

48
2.1

1.2

0.7


-2,582

-13

149

0.8

0.0


5 48 74 4 51 35

44 2.6 2.6 2.5 0.5 3.3

58 123 137 183


___________________________________________ .1. ______________________________________________________ a








Eastern Europe and Former USSR*


Producers


Hungary Romania


Consumers


German
Former Democralic
USSR* Yugoslavia Republic


Regional
tolal or
average


1. Estimated population, 1991 (million) 9.0 15.7 10.5 23.3 291.0 24.0 Note* 414.9
2. Estimated growth rate of population,
1989-2000 (%/yr) -0.2 0.3 -0.4 0.4 0.6 0.6 Note* 0.5
3. Per capital income 1990 (US$) 2,250 3,140 2,780 1,640 .. 3,060
4. Per capital cereal production, 1988-90 (kg/yr) 935 777 1275 792 715 630 665 729
5. Growth rate of per capital cereal production,
1981-90 (%/yr) -0.9 2.1 -0.5 -2.6 2.3 -2.2 2.2 1.5

6. Maize area harvested, 1989-91 (000 ha) 516 164 1,104 2,588 4,168 2,269 7 10,929
7. Maize yield, 1989-91 (t/ha) 4.0 4.7 5.7 3.1 2.7 3.7 .. 3.4
8. Maize production, 1989-91 (000 t) 2,068 777 6,335 8,022 11,235 8,313 .. 37,310
9. Growth rate of maize area, 1971-80 (%) -0.3 2.2 -1.1 0.9 -3.9 -1.0 .. -1.2
10. Growth rate of maize area, 1981-90 (%) -2.5 -0.4 -0.5 -2.2 1.8 0.0 .. -0.1
11. Growth rate of maize yield, 1971-80 (%) 0.4 1.6 4.2 3.7 2.6 3.4 .. 3.2
12. Growth rate of maize yield, 1981-90 (%) -5.3 -0.3 -4.2 -5.9 1.9 -4.4 .. -2.5
13. Growth rate of maize production,
1971-80 (%) 0.2 3.9 3.1 4.6 -1.3 2.3 .. 2.0
14. Growth rate of maize production,
1981-90 (%) -7.9 -0.7 -4.7 -8.0 3.7 -4.5 .. -2.5
15. Maize area as percent of total cereal area,
1989-91 (%) 24 7 40 44 4 54 0 8
16. Average yield of all cereals, 1989-91 (t/ha) 4.1 5.0 5.2 3.1 1.9 3.7 4.8 2.3
17. Growth rate of yield of all cereals,
1971-80 (%/yr) 1.1 2.3 4.0 3.4 0.3 3.0 0.5 0.7
18. Growth rate of yield of all cereals,
1981-90 (%/yr) -0.7 2.9 -0.5 -1.4 4.5 -1.2 2.4 3.3

19. Net imports of maize, 1989-90 (000 t) 655 160 -79 103 14,269 265 1,159 16,897
20. Net imports of maize per capital,
1988-90 (kg/yr) 73 10 -7 4 50 11 70 40
21. Per capital total maize consumption,
1988-90 (kg/yr) 260 63 475 305 105 340 70 130
22. Growth rate of per capital maize consumption,
1971-80 (%/yr) 2.4 7.4 3.0 4.1 6.1 1.3 9.2 4.6
23. Growth rate of per capital maize consumption,
1981-90 (%/yr) -6.4 -7.4 -4.2 -8.0 2.0 -4.1 -1.6 -2.2

24. Public sector maize varietal releases,
1966-90
25. Public sector maize varietal releases
per million ha maize area, 1966-90
26. Number of public sector maize
researchers, 1990
27. Number of public sector maize researchers
per million ha maize area, 1990
28. Number of private sector maize
researchers, 1990
29. Number of private sector maize researchers
per million ha maize area, 1990

30. Fertilizer applied per hectare of arable land,
1987-88 (kg nutrients/ha) 218 311 277 127 117 131 367 137
31. Growth rate of NPK consumption per ha of
arable land, 1970-71 to 1988-89 (%/yr) 3.1 1.4 0.7 4.9 5.2 3.2 -0.3 3.8

32. Farm price of maize, 1990-91 (US$/t) .. 148 124 .. 177
33. Ratio of farm-level nitrogen price
to maize price, 1990-91 .. 2.1 2.9 .. 3.3 2.4
34. Farm wage in kg of maize per day, 1990-91 .. 38 31 .. 39 42

Data cover the same area previously designated as the USSR.


Bulgaria


Czecho-
sloakia


I osmr


'"
:""'







Developed Market Economies


5,, ,


Producers


Federal
Republic
Canada France or Germany


Greece Italy


South
Africa


Spain


1. Estimated population, 1991 (million) 7.6 26.7 56.6 78.8* 10.1 57.7 36.0 39.2
2. Estimated growth rate of population,
1989-2000 (%/yr) 0.2 0.8 0.4 0.1* 0.2 0.1 2.2 0.2
3. Per capital income 1990 (US$) 19,060 20,470 19,490 22,320 5,990 16,830 2,530 11,020
4. Per capital cereal production, 1988-90 (kg/yr) 672 1805 999 425 468 301 370 531
5. Growth rate of per capital cereal production,
1981-90 (%/yr) 0.6 -1.0 1.6 1.1 -2.4 -1.1 -0.5 5.5

6. Maize area harvested, 1989-91 (000 ha) 195 1,041 1,756 228 213 810 3,426 493
7. Maize yield, 1989-91 (t/ha) 7.9 6.7 6.7 7.2 9.4 7.6 2.8 6.5
8. Maize production, 1989-91 (000 t) 1,545 6,952 11,804 1,642 2,006 6,144 9,720 3,188
9. Growth rate of maize area, 1971-80 (%) 4.7 7.5 -0.1 0.2 -2.3 0.5 -0.3 -2.4
10. Growth rate of maize area, 1981-90 (%) -0.2 -1.6 1.1 4.9 2.0 -3.2 -2.7 3.3
11. Growth rate of maize yield, 1971-80 (%) 2.2 1.6 0.7 2.0 7.2 3.5 3.0 3.4
12. Growth rate of maize yield, 1981-90 (%) 0.7 1.5 0.9 2.1 -0.2 1.4 4.6 2.8
13. Growth rate of maize production,
1971-80 (%) 6.9 9.1 0.6 2.2 5.0 4.1 2.7 1.0
14. Growth rate of maize production,
1981-90 (%) 0.6 -0.1 2.0 6.9 1.8 -1.8 1.9 6.1
15. Maize area as percent of total cereal area,
1989-91 (%) 21 5 19 5 14 18 56 6
16. Average yield of all cereals, 1989-91 (t/ha) 5.4 2.5 6.2 5.9 3.5 4.0 2.0 2.5
17. Growth rate of yield of all cereals,
1971-80 (%/yr) 2.0 1.1 1.6 1.3 3.6 2.1 1.8 3.4
18. Growth rate of yield of all cereals,
1981-90 (%/yr) 2.2 -0.2 2.8 2.7 -0.3 0.8 2.7 5.2

19. Net imports of maize, 1989-90 (000 t) -185 531 -6,833 1,033 -18 1,012 -1,578 1,362
20. Net imports of maize per capital,
1988-90 (kg/yr) -24 20 -122 17 -2 18 -46 35
21. Per capital total maize consumption,
1988-90 (kg/yr) 177 259 92 42 187 125 232 120
22. Growth rate of per capital maize consumption,
1971-80 (%/yr) 5.9 6.5 3.7 -2.2 9.3 0.0 0.0 4.3
23. Growth rate of per capital maize consumption,
1981-90 (%/yr) -1.4 0.1 -8.9 -0.8 -2.1 -3.2 0.4 -6.4

24. Public sector maize varietal releases,
1966-90
25. Public sector maize varietal releases
per million ha maize area, 1966-90
26. Number of public sector maize
researchers, 1990
27. Number of public sector maize researchers
per million ha maize area, 1990
28. Number of private sector maize
researchers, 1990
29. Number of private sector maize researchers
per million ha maize area, 1990

30. Fenilizer applied per hectare of arable land,
1987-88 Ikg nurrients/ha) 212 47 312 411 165 172 63 98
31. Grow th rate of NPK consumption per ha of
arable land, 1970-71 to 1988-89 (%/yr) -0.4 5.1 1.1 0.0 3.8 3.4 2.3 1.8

32. Farm price of maize, 1990-91 (US$/t) 294 85 204 194 250 222 81
33. Ratio of farm-level nitrogen price
to maize price, 1990-91 5.3 7.1 .. 2.9 1.4 1.9
34. Farm wage in kg of maize per day, 1990-91 182 .. 170 206 104 200

Includes data from the former German Democratic Republic.


Austria







Developed Market Economies (continued) a


Producers I


United Belgium-
Switzerland States Luxumbourg


Consumers


Japan Netherlands Portugal


United
Kingdom


1*


Regional
total or
average


1. Estimated population, 1991 (million) 6.7 252.1 10.3 123.9 15.0 10.4 57.7 841.5
2. E~timaled growth rate of population,
1989-2000 1l r i 0.4 0.8 0.1 0.3 0.5 0.4 0.2 0.6
3. Percapita income 1990(US$1 32,680 21,790 15,540 25,430 17,320 4,900 16,100 19.382
4. Per capital cereal production, 1988-90 (kg/yr) 197 1,081 225 115 89 149 385 67()
5. Growth rate of per capital cereal production,
1981-90 (%/yr) 4.0 -2.2 1.1 -0.4 -0.8 3.0 0.3 -0.9

6. Maize area harvested, 1989-91 (000 ha) 27 27,046 8 0 0 264 0 35.574
7. Maize yield, 1989-91 (t/ha) 8.7 7.2 .. .. 2.6 .. 6.7
8. Maize production, 1989-91 (000 t) 238 194,177 .. .. 674 .. 238,519
9. Growth rate of maize area. 1971-80 (%) -1.5 2.3 .. .. -1.7 .. 1.7
10. Growth rate of maize area, 1981-90 (%) 4.7 -1.1 .. .. -1.7 .. -1.1
11. Growth rate of maize yield, 1971-80 (%) 3.1 1.4 .. .. 0.3 .. 1.9
12. Growth rate of maize yield, 1981-90 (%) 1.7 0.9 .. .. 7.1 .. 1.2
13. Growth rate of maize production,
1971-80(% 1.5 3.8 .. .. -1.4 .. 3.6
14. Growth rate of maize production,
1981-90 (%) 6.4 -0.2 .. .. 5.4 .. 0.1
15. Maize area as percent of total cereal area,
1989-91 (%) 13 42 2 0 0 28 0 24
16. Average yield of all cereals, 1989-91 (t/ha) 6.3 4.6 6.4 5.7 6.9 1.6 6.2 4.0
17. Growth rate of yield of all cereals,
1971-80 (%/yr) 1.3 1.3 1.0 0.2 3.4 -2.0 2.0 1.3
18. Growth rate of yield of all cereals,
1981-90 (%/fr) 2.3 1.0 2.3 0.9 0.7 5.6 1.2 1.3

19. Net import of maize, 1989-90 (000 t) 129 -51,680 1,023 16,125 1,893 725 1,457 -34,427
20. Net imports of maize per capital,
1988-90 (kg/Ar) 19 -209 100 131 127 70 25 -42
21. Per capital total maize consumption,
1988-90 (kg/yr) 56 488 105 131 128 134 25 223
22. Growth rate of per capital maize consumption,
1971-80 (%t~rlN 1.4 -0.2 1.8 7.9 0.3 9.3 0.1 1.0
23. Growth rate of per capital maize consumption,
1981-90 (%./yr -2.3 -0.9 -2.8 1.6 -2.5 -11.4 -4.6 -1.2

24. Public sector maize varietal releases,
1966-90
25. Public sector maize varietal releases
per million ha maize area, 1966-90
26. Number of public sector maize
researchers. 1990
27. Number of public sector maize researchers
per million ha maize area, 1990
28. Number of private sector maize
researchers. 1990
29. Number of pri ate sector maize researchers
per million ha maize area, 1990

30. Fertilizer applied per hectare of arable land,
1987-88 (kg nutrients/ha) 431 94 507 415 650 78 346 116
31. Growtlh rate of NPK consumption per ha of
arable land, 1970-71 to 1988-89 (%/yr.) 0.9 0.8 -0.7 0.8 -0.1 3.3 2.5 0.7

32. Farm price of maize. 1990-91 (US$t) .. 91
33. Ratio of farm-level nitrogen price
to maize price. 1990-91 .. 4
34. Farm wage in kg of maize per day, 1990-91 .. 478


I







Regional Aggregates


Less
developed
countries


Developed
market
economies


Eastern
Europe and
former USSR*


World


1. Estimated population, 1991 (million) 4,103.0 841.5 414.9 5,360.9
2. Estimated growth rate of population,
1989-2000 (%/yr) 1.9 0.6 0.5 1.6
3. Per capital income 1990 (US$) 801 19,382 .. 4,075
4. Per capital cereal production, 1988-90 (kg/yr) 253 670 729 358
5. Growth rate of per capital cereal production,
1981-90 (%/yr) 0.3 -0.9 1.5 -0.2

6. Maize area harvested, 1989-91 (000 ha) 82,643 35,574 10,929 129,146
7. Maize yield, 1989-91 (t/ha) 2.4 6.7 3.4 3.7
8. Maize production, 1989-91 (000 t) 201,520 238,519 37,310 477,349
9. Growth rate of maize area, 1971-80 (%) 0.8 1.7 -1.2 0.9
10. Growth rate of maize area, 1981-90 (%) 1.3 -1.1 -0.1 0.5
11. Growth rate of maize yield, 1971-80 (%) 3.0 1.9 3.2 2.6
12. Growth rate of maize yield, 1981-90 (%) 1.7 1.2 -2.5 0.4
13. Growth rate of maize production,
1971-80 (%) 3.8 3.6 2.0 3.5
14. Growth rate of maize production,
1981-90 (%) 3.0 0.1 -2.5 0.9
15. Maize area as percent of total cereal area,
1989-91 (%) 19 24 8 18
16. Average yield of all cereals, 1989-91 (t/ha) 2.4 4.0 2.3 2.7
17. Growth rate of yield of all cereals,
1971-80 (%/yr) 2.7 1.3 0.7 1.9
18. Growth rate of yield of all cereals,
1981-90 (%/yr) 2.0 1.3 3.3 1.9

19. Net imports of maize, 1989-90 (000 t) 18,073 -34,427 16,897
20. Net imports of maize per capital,
1988-90 (kg/yr) 5 -42 40
21. Per capital total maize consumption,
1988-90 (kg/yr) 54 223 130 86
22. Growth rate of per capital maize consumption,
1971-80 (%/yr) 3.1 1.0 4.6 1.7
23. Growth rate of per capital maize consumption,
1981-90 (%/yr) 1.3 -1.2 -2.2 -0.8

24. Public sector maize varietal releases,
1966-90
25. Public sector maize varietal releases
per million ha maize area, 1966-90
26. Number of public sector maize
researchers, 1990
27. Number of public sector maize researchers
per million ha maize area, 1990
28. Number of private sector maize
researchers, 1990
29. Number of private sector maize researchers
per million ha maize area, 1990

30. Fertilizer applied per hectare of arable land,
1987-88 (kg nutrients/ha) 78 116 137 100
31. Growth rate of NPK consumption per ha of
arable land, 1970-71 to 1988-89 (%/yr) 8.0 0.7 3.8 3.6

32. Farm price of maize, 1990-91 (US$/t)
33. Ratio of farm-level nitrogen price
to maize price, 1990-0 [
34. Farm wage in kg of maize per day, 1990-91

Data for former USSR cover the same area previously designated as the USSR.












References


Aguiluz, A., R. Urbina, R. Celado, and H.
C6rdova. 1991. Efecto del mejoramiento para
resistencia al achaparramiento sobre
rendimientos de cultivares de maiz evaluados
en siete ambientes de Centro Amdrica y el
Caribe. In Desarrollo y mejoramiento de
germoplasma para resistencia afactores
adversos bi6ticos y abi6ticos 1990.
Guatemala, Guatemala: CIMMYT Regional
Maize Program for Central America and the
Caribbean. Pp. 67-72.

Akino, M. and Y. Hayami. 1975. Efficiency and
equity in public research: Rice breeding in
Japan's economic development. American
Journal ofAgricultural Economics 57(1): 1-
10.

Ardito-Barletta, N. 1971. Costs and Benefits of
Agricultural Research in Mexico. Ph.D.
thesis. Chicago, Illinois: University of
Chicago.

Berhe, T. 1989. 1987-88 Annual Report:
Agronomy/FSR. Report prepared for USAID/
International Institute of Tropical
Agriculture/Zaire R.A.V. Project. Mimeo.

Borb6n, C., and G. Sain. 1992. El uso de
variedades resistentes al achaparramiento en
la region IV de Nicaragua. Managua: Centro
de Investigation en Granos Basicos,
Ministerio de Agricultura y de Ganaderia.
Mimeo.

Brennan, J.P. 1992. Economic Criteria for
Establishing Plant Breeding Programs.
CIMMYT Economics Working Paper 92-01.
Mexico, D.F.: CIMMYT.

Byerlee, D., K. Khan, and M. Saleem. 1991.
Revealing farmer rationality: On-farm maize
research in Swat Valley, Northern Pakistan.
In R. Tripp (ed.), Planned Change in Farming
Systems: Progress in On-Farm Research.
Chichester, UK: John Wiley and Sons. Pp.
169-190.

Chand, S.P., D.P. Sherchan, T.P. Tiwari, D.K.
Rijal, G.B. Gurung, B.D. Gurung, M.B.
During, Y.B. Thapa, R.K. Neupane, and B.
Bajracharya. 1991. Maize Research Report
1989. PAC Working Paper No. 17.
Kathmandu, Nepal: Pakhribas Agricultural
Centre.

CIMMYT. 1990. CIMMYT's Five-year Budget:
1990-1994. Mexico, D.F.: CIMMYT.

CIMMYT Maize Staff. Improving on
Excellence: Achievements in Breeding with
the Maize Race Tuxpeio. CIMMYT: Mexico,
D.F. 1986.


C6rdova, H. 1984. Formulaci6n de hibridos de
maiz en base a families de hermanos
completes y sus implicaciones en la
producci6n de semilla commercial. In Anais XV
Congress Nacional de Milho e Sorgho, 2-6
July, 1984, Maceio, Brazil. Brasilia: Empresa
Brasileira de Pesquisa AgropecuBria. Pp. 377-
404.

C6rdova, H. 1991. Desarrollo y mejoramiento
de germoplasma para resistencia a factors
adversos bi6ticos y abi6ticos y producci6n de
semilla: estrategias y logros 1986-91. In
Desarrollo y mejoramiento de germoplasma
para resistencia a factors adversos bi6ticos
y abi6ticos 1990. Guatemala, Guatemala:
CIMMYT Regional Maize Program for
Central America and the Caribbean. Pp. 1-14.

Cota, O., J.M. Valenzuela, A. Ortega, and A.A.
Soqui. 1991. CIANO H-430: Nuevo hibrido
de maiz blanco. Folleto Tecnico No. 17.
Sonora, Mexico: Secretaria de Agricultura y
Recursos Hidraulicos.

Dahlan, M., Heriyanto, Sunarsedyono, Sri
Wahyuni, C.E. Van Santen, J.P. Van
Staveren, and L.W. Harrington. 1987. Maize
On-Farm Research in the District of Malang.
MARIF Monograph No.3. Malang, Indonesia:
Malang Research Institute for Food Crops.

Dossou, R.A. 1989. Amelioration du mais au
Benin: Points saillants des r6sultats de 1988.
Paper presented at the IITA/SAFGRAD
Maize Network Workshop, 20-24 March,
1989, Lomd, Togo.

Edmeades, G.O., A.A. Dankyi, and K. Marfo.
1991. On-farm research in the transition zone
of Ghana. In R. Tripp (ed.), Planned Change
in Farming Systems: Progress in On-Farm
Research. Chichester, UK: John Wiley and
Sons. Pp. 63-84.

Evenson, R.E., and Y. Kislev. 1971. Investment
in Agricultural Research and Extension: A
Survey of International Data. Center
Discussion Paper No. 124. New Haven,
Connecticut: Yale University, Economic
Growth Center.

FAO. (various years.) Fertilizer Yearbook.
Rome, Italy: Food and Agricultural
Organization of the United Nations.

1991. Diskette of population
statistics. Rome, Italy: Food and Agricultural
Organization of the United Nations.

1991. Diskettes of production
statistics. Rome, Italy: Food and Agricultural
Organization of the United Nations.


1991. Diskettes of trade statistics.
Rome, Italy: Food and Agricultural
Organization of the United Nations.

1992. Agrostat/PC (diskettes).
Rome, Italy: Food and Agricultural
Organization of the United Nations.

Gerhart, J. 1975. The diffusion of hybrid maize
in Western Kenya. Mexico, D.F.: CIMMYT.
Mimeo (abridged by CIMMYT).

Ghana Grains Development Project. 1991. A
Study of Maize Technology Adoption in
Ghana. Mexico, D.F.: GGDP.

Griliches, Z. 1957. Hybrid corn: An exploration
in the economics of technical change.
Econometrica 25(4): 501-522.

Griliches, Z. 1958. Research costs and social
returns: Hybrid corn and related innovations.
Journal of Political Economy 66: 419-431.

Harrington, L.W., R. Krisdiana, and Herianto.
1991. On-farm research and farmer adoption
of new maize technology in Malang District,
Indonesia: A logit analysis. Mimeo.

Hayami, Y., and R.W. Herdt. 1977. The impact
of technological change in subsistence
agriculture on income distribution. American
Journal ofAgricultural Economics 59(1):
245-256.

Hildebrand, P. 1984. Modified stability analysis
of farmer managed, on-farm trials. Agronomy
Journal 76: 271-274.

Judd, M.A., J.K. Boyce, and R.E. Evenson.
1983. Investing in Agricultural Supply.
Discussion Paper No. 442. New Haven: Yale
University, Economic Growth Center.

Karanja, D. 1990. The Rate of Return to Maize
Research in Kenya: 1955-88. M.S. thesis.
East Lansing, Michigan: Michigan State
University.

Lal, K.K. 1991. Status of maize crop in Nepal.
In Proceedings of the Fourth Regional Asian
Maize Workshop, Islamabad, Pakistan,
September 21-28, 1990. Mexico, D.F.:
CIMMYT and the Pakistan Agricultural
Research Council. Pp. 117-128.

Librero, A.R., and M.L. Perez. 1987. Estimating
Returns to Research Investment in Corn in
the Philippines. Los Bafios, Phillippines:
Laguna Philippine Council for Agricultural
Resources, Research, and Development.








-S ,.


Lindner, R.K., and F.G. Jarrett. 1978. Supply
shifts and the size of research benefits.
American Journal of Agricultural Economics
60(4): 976-980.

Manzano, M., and H. C6rdova. 1986. Quince
afios de investigaci6n en maiz en El Salvador.
Unpublished report. Mexico, D.F.: CIMMYT.

McHugh, D. 1991. A maize and groundnut
stepwise trial in the Ndop Plain, Cameroon
(1987). In H.J.W. Mutsaers and P. Walker
(eds.), On-Farm Research in Theory and
Practice. Ibadan, Nigeria: International
Institute of Tropical Agriculture. Pp. 122-
127.

Mutsaers, H.J.W. 1991. Technology evaluation
and diagnosis: The potential for improvement
of maize-cassava cropping systems in South-
Western Nigeria. Paper presented at the
RCMP Review Meeting, 28 August 1991,
International Institute of Tropical
Agriculture, Ibadan, Nigeria.

Mutsaers, H.J.W., and P. Walker. 1990.
Farmers' maize yields in S.W. Nigeria and
the effect of variety and fertilizer: An
analysis of variability in on-farm trials. Field
Crops Research 23: 265-278.

Nagy, J.G. 1985. The overall rate of return to
agricultural research and extension
investments in Pakistan. Pakistan Journal of
Applied Economics 4(1): 17-28.

Norton, G.W., V.G. Ganoza, and C. Pomareda.
1987. Potential benefits to agricultural
research and extension in Peru. American
Journal ofAgricultural Economics 69(2):
247-257.

Norton, G.W., and J.S. Davis. 1981. Evaluating
returns to agricultural research: A review.
American Journal of Agricultural Economics
63(4): 685-699.

Pardey, P., and B.J. Craig. 1989. Causal
relationships between public sector
agricultural research expenditures and output.
American Journal of Agricultural Economics
71(1): 9-19.

Pardey, P., and J. Roseboom. 1989. A Global
Data Base on National Agricultural Research
Systems. International Service for National
Agricultural Research (ISNAR) Agricultural
Research Indicator Series. Cambridge, UK:
Cambridge University Press.

Pardey, P., J. Roseboom, and J.R. Anderson
(eds.). 1991. Agricultural Research Policy:
International Quantitative Perspectives.
Cambridge, UK: Cambridge University Press.


Peterson, W. 1967. Returns to poultry research
in the United States. Journal of Farm
Economics 49: 656-659.

Pham, H., S.R. Waddington, and J. Crossa.
1989. Yield stability of CIMMYT maize
germplasm in international and on-farm trials.
In P.B.R. Hazell and J.R. Anderson (eds.),
Variability in Grain Yields: Implications for
Agricultural Research and Policy in
Developing Countries. Baltimore, Maryland:
Johns Hopkins. Pp. 185-205.

Rose, F. 1980. Supply shifts and the size of
research benefits: Comment. American
Journal ofAgricultural Economics 62(4):
834-837.

Sanou, J. 1991. Rapport technique sur la
recherche adaptive du mais en milieu rdel.
Project CEE/IITA/MAIS. Ougadougou,
Burkina Faso: INERA. Mimeo.

Satorre, C.E., A. Mulato, B. Pantinople, and L.
Harrington. 1991. On-farm variety trial of
maize under coconut in Davao del Sur,
Philippines. Paper presented at the 11th
Annual AFSRE Symposium, October 1991,
East Lansing, Michigan State University.

Schmitz, A., and G. Seckler. 1970. Mechanical
agriculture and social welfare: The case of
the tomato harvester. American Journal of
Agricultural Economics 52(3): 569-578.

Schuh, G.E., and H. Tollini. 1978. Costs and
Benefits of Agricultural Research: State of
the Art and Implications for CGIAR.
Washington, D.C.: Consultative Group for
International Agricultural Research.

Scobie, G. 1978. The impact of technical change
on income distribution: The case of rice in
Columbia. American Journal of Agricultural
Economics 60(1): 85-91.

Smale, M., with Z.H.W. Kaunda, H.L. Makina,
M.M.M.K. Mkandawire, M.N.S. Msowoya,
D.J.E.K. Mwale, and P.W. Heisey. 1991.
Chimanga Cha Makolo, Hybrids, and
Composites: Farmer Adoption
Characteristics in Kasungu, Blantyre, and
Mzuzu Agricultural Development Divisions,
1989-90. CIMMYT Economics Program
Working Paper 91-04. Mexico, D.F.:
CIMMYT.

Smale, M. 1992. Risk, Disaster Avoidance, and
Farmer Experimentation: The
Microeconomics of HYV Adoption in
Malawi. Ph.D. dissertation. College Park,
Maryland: University of Maryland.


Smale, M., P.W. Heisey, and H.D. Leathers.
1991. Maize of the ancestors and modem
varieties: The microeconomics of HYV
adoption in Malawi. Mimeo.

Togo Ministry of Agriculture. 1989. 1988
annual report. Mimeo.

Traxler, G., and D. Byerlee. 1992. Crop
Management Research and Extension: The
Products and Their Impact on Productivity.
CIMMYT Economics Paper No. 5. Mexico,
D.F.: CIMMYT (forthcoming).

Tripp, R., K. Marfo, A.A. Dankyi, and M. Read.
1987. Changing Maize Production Practices
of Small-Scale Farmers in the Brong-Ahafo
Region, Ghana. Mexico, D.F.: Ghana Grains
Development Project.

Versteeg, M.N., and A. Huijsman 1991. Trial
design and analysis for on-farm adaptive
research: A 1988 maize trial in the Mono
Province of Benin. In H.J.W. Mutsaers and P.
Walker (eds.), On-Farm Research in Theory
and Practice. Ibadan, Nigeria: International
Institute of Tropical Agriculture. Pp. 111-
121.

Vogel, W.O., Berhe, T, and R.D. Hennessey.
1991. On-farm trials comparing the local
streak-disease susceptible maize with a
streak-disease resistant variety in southern
Zaire. In H.J.W. Mutsaers and P. Walker
(eds.), On-Farm Research in Theory and
Practice. Ibadan, Nigeria: International
Institute of Tropical Agriculture. Pp. 176-
187.

World Bank. 1991. World Bank Development
Report 1991. New York. Oxford University
Press.

1992. World Development Report.
New York: Oxford University Press

Woldetatios, T., C.N. Pouboum, M.A. Messia,
M.T. Beson, S.A. Almy, and C.M. Ateh.
1991. Adaptive research on improved maize
varieties to the diversified agro-ecological
zones of South West Province, Cameroon.
Agricultural Systems in Africa 2(1): 53-63.

Yates, M., and J.C. Martinez. 1985. On-farm
research methodologies at work: Progress
report from Les Cayes, Haiti. Mexico, D.F.:
CIMMYT. Mimeo.

Yrarrazaval, R., R. Navarrete, and V. Valdivia.
1982. Costos e beneficios sociales de los
programs de mejoramiento varietal de trigo
y maiz en Chile. In M. Elgueta and E.
Venezian (eds.), Economia y organizaci6n de
la investigacidn agropecuaria. Santiago,
Chile: INIA.


tvorld N13w Ficin 3ndn Tr~rnd.









Annex 1: Regions of the World


Annex 1: Regions of the World


Eastern and Southern Africa
Angola
Botswana
Burundi
Comoros
Djibouti
Ethiopia
Kenya
Lesotho
Madagascar
Malawi
Mauritius
Mozambique
Namibia
Rwanda
Seychelles
Somalia
Sudan
Swaziland
Tanzania
Uganda
Zambia
Zimbabwe

Western and Central
Africa
Benin
Burkina Faso
Cameroon
Cape Verde
Central African Republic
Chad
Congo
C6te d'Ivoire
Equatorial Guinea
Gabon
Gambia
Ghana
Guinea
Guinea Bissau
Liberia
Mali
Mauritania
Niger
Nigeria
Reunion
Sao Tome
Senegal
Sierra Leone
St. Helena
Togo
Zaire


North Africa
Algeria
Egypt
Libya
Morocco
Tunisia

West Asia
Afghanistan
Bahrain
Cyprus
Iran
Iraq
Jordan
Kuwait
Lebanon
Oman
Qatar
Saudi Arabia
Syria
Turkey
United Arab Emirates
Yemen Arab Republic
Yemen Democratic Republic

South Asia
Bangladesh
Bhutan
India
Maldives
Myanmar
Nepal
Pakistan
Sri Lanka

Southeast Asia and the Pacific
American Samoa
Brunei
Cook Islands
East Timor
Fiji
French Polynesia
Guam
Hong Kong
Indonesia
Kampuchea Republic
Kiribati
Laos
Macau
Malaysia
Nauru
New Caledonia
Niue
Norfolk Islands
Pacific Islands
Papua New Guinea
Philippines


Samoa
Singapore
Solomon Islands
Thailand
Tokelau
Tonga
Tuvalu
Vanuatu
Vietnam
Wallis and Futuna Island

East Asia
China
Korea, D.P.R.
Korea, Republic
Mongolia
Taiwan

Mexico, Central America, and
the Caribbean
Antigua
Bahamas
Barbados
Belize
Bermuda
Cayman Islands
Costa Rica
Cuba
Dominica
Dominican Republic
El Salvador
Grenada
Guadeloupe
Guatemala
Haiti
Honduras
Jamaica
Martinique
Mexico
Montserrat
Netherlands Antilles
Nicaragua
Panama
St. Christopher and Nevis
St. Lucia
St. Pierre and Miquelon
St. Vincent Granadines
Trinidad and Tobago
UK Virgin Islands
US Virgin Islands

Andean Region
Bolivia
Colombia
Ecuador
French Guyana


Guyana
Peru
Surinam
Venezuela

Southern Cone, South America
Argentina
Brazil
Chile
Paraguay
Uruguay
Falkland Islands

Eastern Europe and
Former USSR
Albania
Bulgaria
Czechoslovakia
Germany, East
Hungary
Poland
Rumania
Former USSR
Yugoslavia

Developed Market Economies
Australia
Austria
Belgium-Luxembourg
Canada
Denmark
Faeroe Islands
Finland
France
Germany, West
Greece
Greenland
Iceland
Ireland
Israel
Italy
Japan
Malta
Netherlands
New Zealand
Norway
Portugal
South Africa
Spain
Sweden
Switzerland
United Kingdom
United States










































































































Lisbo 27 Aprtd Pota 6-41 0660 V1S.co D F. S
A *t











COPYRIGHT NOTICE

CIMMYT International Maize and Wheat Improvement Center

The CIMMYT International Maize and Wheat Improvement
Center holds all rights to the source text and shall be
considered the copyright holder for the text and images of
these publications.

The CIMMYT International Maize and Wheat Improvement
Center has made this publication available to the University of
Florida, for purposes of digitization and Internet distribution.

The CIMMYT International Maize and Wheat Improvement
Center reserves all rights to this publication. All uses,
excluding those made under "fair use" provisions of U.S.
Code, Title 17, Section 107 are restricted.

Contact the CIMMYT International Maize and Wheat
Improvement Center for additional information and
permissions.