Comparative Advantage and
Policy Incentives for Wheat
Production in Rainfed and Irrigated
Areas of Mexico
Derek Byerlee and Jim Longmire
CIMMYT Economics Program
Working Paper No. 01/86
This is the third study in a series that staff of the CIMMYT
Economics Program, in conjunction with national research organizations
in selected developing countries, are preparing on the comparative
advantage and policy incentives for the production of alternative crops
in particular regions. The studies will provide the experience necessary
for the preparation of a manual that can be used by colleagues to guide
similar work by national programs.
This paper addresses an issue that is important to a number of
developing countries the economics of wheat in irrigated and rainfed
or dryland areas. It presents an approach that weighs up, from both the
farmer perspective and the national perspective, the profitability of
alternative crops in an irrigated region and in a dryland region. This
enables more precise judgements to be made about the allocation of
research effort between the regions.
If extended to a wider coverage of regions, the information
provided is likely to be a very useful tool for research managers. By
simple sensitivity analysis of yields, researchers can be provided with
a more precise estimate of productivity gains needed to make crops
competitive in specific regions.
As well as providing an economic tool for advising research
managers, the approach presented in this paper provides considerable
information on how farmer incentives are affected by government
policies. A better sense of this will help agricultural researchers to
understand why adoption rates and increases in production of particular
crops are sometimes disappointing.
An initial draft of this paper was prepared in 1983. It was
subsequently updated and revised to take account of prices prevailing
for the 1984-85 crop cycles. Spreadsheets employed in the analysis, on
VISICALC for Apple micro-computers, are retained at CIMMYT's
headquarters and can be made available upon request.
This paper could not have been prepared without the active
cooperation of many people in Mexico, especially INIA, CIAMEC and CIANO,
the private and fanning sectors, and CIMMYT's staff based in Mexico.
As with all working papers, we welcome comments, criticisms or
counsel so that the paper might be improved.
Acting Director, Economics Program
'~"'~'""~;a:;" -- ; ~i"'j:,;;l: 7:l':v~r :r~r:tirrui:,:I~; i~?r ,;i nire '~'
This paper has been prepared in the CIMMYT Economics Program with
assistance from staff of INIA and regional research organizations, CIANO
in the State of Sonora and CIAMEC in the State of Tlaxcala. The
information for this study came from many sources, but we would like to
thank especially the farmers of Tlaxcala and Sonora who so willingly
provided data. The farm-level information was gathered and analyzed by
CIMMYT staff Dagoberto Flores, Laura Hernandez, Edith Hesse de Polanco
and Pedro Santamaria and to them we owe a special thanks. Other people
and organizations who provided data included Banco de Mexico, transport
operators, farm input suppliers, local banks and farm marketing
organizations to all we are vey grateful. Detailed reviews of this
paper with valuable comments were provided by Scott Pearson and Donald
Winkelmann. The typing was done by Claudia De Alba, with contributions
by Maribel Carrillo. Their excellent efforts are acknowledged. The
graphs were drawn by Miguel Mellado E., Jose Manuel Fouilloux and Rafael
de la Colina to their usual high standard.
C O N T E N T S
LIST OF FIGURES vi
LIST OF TABLES viii
EXECUTIVE SUMMARY x
1.0 INTRODUCTION 1
2.0 METHODOLOGY FOR MEASURING COMPARATIVE ADVANTAGE
AND POLICY INCENTIVES 4
2.1 The Resource Cost Ratio 4
2.1.1 Influence of Production Technique on
the Resource Cost Ratio 7
2.1.2 Influence of Transportation Costs on
Comparative Advantage 8
2.1.3 Uses of Domestic Resource Cost Analysis
in Allocating Research Resources 9
2.2 Measuring Policy Incentives 10
2.3 The Overall Measure of Subsidies 12
2.4 Data Sources and Analysis 13
3.0 THE ROLE OF WHEAT IN THE MEXICAN AGRICULTURAL
SECTOR AND MAJOR POLICY ISSUES 15
3.1 The Macro-Economic Scene 15
3.2 Recent Production Performance of Wheat in
Relation to Other Crops 19
3.3 The Demand Outlook for Wheat 21
3.4 From Food Exporter to Food Importer 21
3.5 Irrigated versus Rainfed Agriculture in
3.6 A Comparison of Wheat in Irrigated and
Rainfed Farming Systems 28
3.7 Wheat Production Techniques in Sonora and
4.0 AGRICULTURAL PRICING POLICY AND PRODUCER INCENTIVES 38
4.1 Product Prices 38
4.2 Input and Factor Prices 41
4.2.1 Seed 41
4.2.2 Fertilizer 41
4.2.3 Pesticides 45
4.2.4 Machinery 45
4.2.5 Fuel 47
4.2.6 Overall Changes in Mechanization Costs 48
4.2.7 Irrigation 50
4.2.8 Credit and Insurance 53
4.2.9 Labor 54
4.2.10 Land 56
4.2.11 Research and Extension 57
4.3 Long Distance Transportation 58
4.4 Nominal Protection Coefficients 58
4.5 Effective Protection Coefficient (EPCs) 65
4.6 Subsidy Coefficients 70
5.0 FARM BUDGETS AND THE CALCULATION OF RCRs 72
5.1 Farm Budgets for Tlaxcala 72
5.2 Resource Cost Ratios in Tlaxcala 76
5.3 Farm Budgets and Returns for Sonora 79
5.4 Resource Cost Ratios in Sonora 84
6.0 CONCLUSIONS 88
APPENDIX A. Measures of Comparative Advantage
and Policy Incentives 90
APPENDIX B. Calculation of Mechanization Costs 93
APPENDIX C. Computing the CIF Price of Safflower 98
List of Figures
Role of the Agricultural Sector in Mexican
Estimated Percentage Overvaluation of the
Mexican Peso Based on Differential
Inflation Rates in the US and Mexico
Imports of Major Categories of Crops,
Percent of Harvested Area in Irrigation
Districts in Grains, Cotton and Oil
Irrigated Wheat Area and Production as a
Percent of Total Wheat Area and Production,
Distances Between Sonora, Tlaxcala and
Wheat Area and Yields in the Yaqui Valley,
Major Cropping Systems in the Yaqui Valley,
Area and Yield.of Barley and Wheat in
Ratio of Price of Cotton to Wheat and the
Areas of Cotton and Wheat, Mexico, 1970-83
Mexican Imports and Exports of Fertilizer,
Ratio of the Price of Nitrogen to Wheat,
Mexican Price and World Price Equivalent
for Urea and Triple Super Phosphate in
Real Price of Diesel in Mexico Compared
to Estimated World Price, 1970-83
Interest Rates on Short Term Agricultural
Credit Compared to Interest on 6-Month
Savings Certificates, 1971-84
Nominal Protection Coefficients, Sonora
Nominal Protection Coefficients, Tlaxcala
NPCs for Wheat in Sonora Using the
Corrected Exchange Rate, 1970-85
Nominal and Effective Protection Coefficient
for Crops in Sonora, 1970-85
Nominal and Effective Protection
Coefficients for Crops in Tlaxcala, 1970-85
List of Tables
Resource Cost Ratios Calculated under
Varying Assumptions about Transport
Costs and the Consumption Point
Interpretation of Possible Results of the
Resource Cost Ratio and the Effective
Average Statistics for Major Agricultural
Crops in Mexico, 1980-83
Growth Rates of Area, Yield and Production
of Major Agricultural Crops in the Two
Decades of the 1960s and the 1970s
Percentage of Total Area and Production in
Irrigated Areas for Various Crops
Major Characteristics of the Yaqui Valley,
Sonora and the Altiplano of Tlaxcala/Hidalgo
Technical Parameters for the Production of
Rainfed and Irrigated Wheat
Contract Hire Rates for Machinery in
Michigan, U.S.A. and Sonora, Mexico, 1980
Indices of Prices of Water and Electricity,
Average Cost of Water in the Yaqui Valley
for Wheat Production, 1980-1985
Real Costs of Developing New Irrigated
Rates of Interests on Short-Term
Agricultural Loans, 1983
Costs of Road Transport of Grain Adjusted
for Subsidized Diesel Price, 1982 and 1983
Import Prices, Consumption Points and
Formulas Used to Calculate Nominal
Protection Coefficients in Sonora
Subsidies on Wheat Production, Tlaxcala,
Table 4.9 Subsidies on Wheat Production, Sonora,
Table 5.1 Technical Parameters for Farm Enterprises
in Tlaxcala 72
Table 5.2 Farmer Prices and World Price Equivalent
of Inputs and Outputs, Tlaxcala, 1984 73
Table 5.3 Budgets for Wheat, Barley and Maize in
Tlaxcala Using Actual Prices Paid by
Farmers, 1984 74
Table 5.4 Gross Margins and Returns on Capital for
Wheat, Barley and Maize in Selected Years 75
Table 5.5 Calculation of Resource Cost Ratios in
Tlaxcala, 1984 77
Table 5.6 Resource Cost Ratios for Wheat, Barley
and Maize, Tlaxcala 78
Table 5.7 Mechanical Operations and Inputs Used in
Constructing Farm Budgets, Sonora, 1985 80
Table 5.8 Farmer Prices and World Price Equivalent
of Inputs and Outputs, Sonora, 1985 81
Table 5.9 Farm Budgets for Wheat, Safflower, Cotton,
Maize and Soya, Sonora, 1985 82
Table 5.10 Farmers Returns by Crop, 1977 to 1985,
Table 5.11 Calculation of Resource Cost Ratios,
Sonora, 1985 85
Table 5.12 Resource Cost Ratios for Various Crops,
Sonora, 1977-1985 86
COMPARATIVE ADVANTAGE AND POLICY INCENTIVES FOR WHEAT PRODUCTION
IN RAINFED AND IRRIGATED AREAS OF MEXICO
Wheat has become an increasingly important food crop in Mexico as
consumers with rising incomes, particularly urban consumers, have
switched from maize to wheat products. Although wheat production in
Mexico has expanded rapidly in the past 25 years, imports of food
grains, including wheat, have increased since the early 1970s.
Increasing demand for wheat products and Mexico's trade and
financial situation will place additional pressure on expanding domestic
wheat production. Expansion in irrigated areas will be limited to yield
increases unless wheat substitutes for competing crops. This raises the
question of whether research and policy should give more attention to
rainfed wheat production.
This study analyzes the comparative advantage of wheat in two
contrasting regions of Mexico:
S the Yaqui Valley of the State of Sonorai which is the most
important wheat growing area of Mexico. An average of over
100,000 ha of irrigated wheat are sown annually with yields
now averaging over 5 t/ha.
S the rainfed highland area of the states of Tlaxcala and
Hidalgo. Although wheat was traditionally grown in the area,
it has been largely replaced by barley. Here, while costs of
transportation to major markets are much lower than in Sonora,
prices to farmers for their grain are the same. Wheat yields
are around 2 t/ha.
This resource cost study on two of Mexico's wheat-producing areas
shows the substantial influence of the government in setting output and
input prices in the Mexican wheat industry. In general, producers have
been receiving prices below world prices for wheat. This is particular-
ly the case in Tlaxcala which is adjacent to major consuming centers
and, hence, has markedly lower transportation costs than alternative
sources of supply. Policy has also intervened to change price relation-
ships with other crops. In most cases, farmers have received prices
higher than world prices for competing crops, especially maize and
oilseeds. At the same time, cotton production was influenced as over-
valued exchange rates had the effect of reducing domestic cotton prices
in most years of the 1970s and some more recent years.
To some extent, government policy has compensated producers through
subsidies on inputs. Subsidies on fertilizer, diesel fuel, credit, seed
(in rainfed areas) and water (in irrigated areas) all exceeded 50
percent in the period 1979-82. Since then, credit and water remain the
major sources of assistance, as the direct government subsidies have
been reduced in relative terms. High levels of subsidies, however, have
encouraged both intensive use of inputs and high costs in terms of
In Sonora, cotton provides higher national economic returns in
those years when irrigation water supplies are plentiful and when land
is the limiting factor. However, when water is limiting, the case for
wheat is made more favorable. This is especially so when allowance is
made for the substantial variation in yields and international prices of
Production of the remaining crops, oilseeds and maize, seems to
have no comparative advantage in Sonora. However, in the case of soya
beans, often grown in rotation with wheat and cotton (cotton-wheat-soya
over 2 years), there is a comparative advantage when the fixed costs of
production are not attributed to soya beans as a third crop in the
Finally, for wheat in Sonora, the major research opportunity posed
by this study is to find ways to reduce production costs. With govern-
ment policy now committed to reducing subsidies there is a need to look
for ways to use more effectively water, fuel, credit, and fertilizers.
In Tlaxcala, wheat production also has a comparative advantage
relative to other crops, especially maize. This potential for wheat has
not yet been realized, partly because price policies have generally not
encouraged wheat production and partly because more profitable wheat
varieties have still to be developed and/or extended to local farmers.
Even with more appropriate varieties, wheat would still be a slightly
more risky crop (because of its longer growing cycle) relative to
The analysis of the comparative advantage of wheat production in
two regions in Mexico suggests that rainfed wheat in the central
highlands may be equally as competitive as irrigated wheat in the north
west. This is despite the fact that very little research and extension
has been devoted to wheat grown under rainfed conditions in Mexico. The
case for allocating research and extension resources to rainfed wheat in
Mexico is strengthened by this study.
COMPARATIVE ADVANTAGE AND POLICY INCENTIVES FOR WHEAT
PRODUCTION IN RAINFED AND IRRIGATED AREAS OF MEXICO
Agricultural research institutions, both international and nation-
al, are under increasing pressure to justify allocation of research re-
sources between crops and between regions. For example, in the case of
wheat many national research programs, especially in the tropical coun-
tries, are faced with the decision on whether to invest in wheat re-
search and production programs in order to introduce wheat as a new crop
(or to expand wheat production from a small base).
An agricultural research program can use many criteria for making
decisions on allocation of research resources (see Scobie, 1984). These
will often include the objectives of food self-sufficiency to avoid de-
pendence on a fluctuating world market, and more equitable income dis-
tribution. But a major criterion in allocating research resources will
and should be, the expected economic returns to the investment in re-
search. Some simple rules have been proposed to judge research resource
allocation in terms of the economic importance of the crop. The most
common is to measure research resources allocated to each crop as a per-
centage of the gross value of the crop. (See, for example, Daniels and
Nestel, 1981). Such a measure is deficient in at least two respects.
First, it will seriously underestimate resources allocated to potential
crops since their gross value of output is low at the time of the
decision. This might then lead to an under-investment in crops that
could be important in the future- and an over-investment in crops
that are stagnant or declining. Second, gross value of production,
current or potential, might be a poor guide to the contribution of that
crop to national income. This is in part because high value crops
typically also have higher costs of production. More importantly, prices
paid and received by farmers are often a poor guide to the value of the
SExamples of recent large scale expansion of new crops are soyabeans
in Brazil, sorghum in Mexico and wheat in Bangladesh.
resources when employed in alternative uses. In particular,
profitability to farmers of cereal production often reflects the effects
of policy interventions such as low producer prices to protect consumers
or subsidies on specific inputs as an incentive to producers.
Many research systems now also face conflicting signals emanating
from government policies which, on the one hand, seek to achieve food
self-sufficiency and, on the other hand, promote export crops in an ef-
fort to overcome a burdensome foreign exchange deficit. What is needed
to provide a better guide for agricultural research decisions is a means
of measuring the costs of these goals in terms of profitability to the
.Closely related to the issue of research resource allocation is the
extent to which policy incentives on the whole favor or discriminate
against a particular crop. Research decision makers and scientists often
feel that technologies emerging from the research systems are not adop-
ted because policies, especially price policies, act as a disincentive.
In most cases however, these policy effects are measured by very super-
ficial means such as deflation of producer prices by the consumer price
index to estimate changes in real producer prices.
The purpose of this paper is to illustrate the methodology of com-
parative advantage and policy incentives as a means of linking the re-
search decision making to the policy environment in which researchers
and farmers -make decisions. This methodology emphasizes means for val-
uing resources and outputs in terms that are meaningful for measuring
the present or potential contribution of a crop to national income. The
methodology is applied to a particular issue in Mexico: the relative
emphasis that should be given to irrigated versus rainfed wheat
production in planning future research on wheat.
The methodology presented here is not new. It has been recently
applied in a number of agricultural situations (e.g. Pearson et al,
1981). It has not, however, been utilized in research decision making.
Nor have results been analyzed and presented in a way that they can be
readily understood by the non-economist.
This paper is organized as follows. First, we present the method-
ology using a simplified example to show the basic steps. (A more formal
treatment is given in Appendix A). We then discuss the broad issues of
food policy in Mexico as it relates to agricultural research decisions
and specifically those related to wheat. This leads to a detailed
analysis of policy incentives for wheat production in Mexico that
measures the effect of government interventions in product and input
markets. Finally we estimate the profitability of wheat in an irrigated
area (the State of Sonora) and a rainfed area (the States of Tlaxcala
and Hidalgo) in relation to other crops. Two sets of calculations are
made first, using actual farm prices and, second, adjusting prices for
the effects of government taxes and subsidies. The presentation is
purposefully detailed to enable the reader (including the non-economist)
to follow the methodology and to appreciate the decisions that were
taken at each step in the analysis.
2.0 Methodology for Measuring Comparative Advantage
and Policy Incentives
2.1 The Resource Cost Ratio
Comparative advantage is an expression of the efficiency of using
resources to produce a particular product when measured against the pos-
sibilities of international trade. In a very simple example, assume that
one hectare of land and a given amount of other inputs can be used to
grow cotton or wheat. If the yield of cotton is 1 ton/ha, then at
recent international prices this cotton, when exported, will purchase
about 10 tons of wheat. Thus, the country gains relatively more by
producing and exporting cotton and importing wheat, since wheat is
unlikely to yield 10 tons/ha. Of course, this simple example ignores a
number of issues such as the fact that cotton may require more imported
inputs for its production.
A more useful measure of resource use efficiency is provided by the
resource cost ratio (RCR). Assume that country A, where wheat is grown
only on a small scale, is importing substantial quantities of wheat.
Consequently it is considering a major investment in a wheat research
and production campaign to substitute for imports. For simplicity assume
a) The government has set the price of wheat at $300/ton.
b) The major purchased input for wheat production will be
fertilizer which costs the farmer $50/ha.
c) The farmers' land, labor and capital used in wheat production
are-costed at $150/ha at market prices.
Clearly if the current yield is 1.3 tons/ha it is quite profitable
for farmers to grow wheat. Profits = (1.3x300)-150-50 = $190 or a 95%
return on an outlay of $200(=150+50).
However from the national viewpoint the resources used in growing
wheat may not be profitably employed. Assume that wheat can be imported
to the capital city, the major consumer of wheat, for $200/ton, that it
costs $20 to transport domestic wheat to the capital and that fertilizer
is subsidized by the government by 33 percent. In this case the national
value of the wheat produced is $180/ha (the cost of imports less local
transport costs) and the value of the fertilizer employed is $75/ha when
the subsidy is subtracted. Assuming as before the market value of the
farmers' land and labour resources of $150/ha, wheat production is
marginally profitable (1.3 x 180-150-75 = 9 or a return on investment of
less than 5 percent). This calculation which considers the true costs
and returns to the country we call the national profitability, as
distinct from farmer profitability.
The final step in this calculation of national profitability is to
take into account the opportunity cost of the farmers' resources of
land, labor, capital and irrigation water in other uses. In some cases
the market value of these resources may adequately reflect these costs.
In most cases, policy measures distort market values as a good measure
of their value in alternative uses. Assume that the alternative use of
the farmers' resources is in the production of cotton for the export
market. To the farmer, the value of his resources in cotton is $150,
i.e. the value at market prices. However, because of export taxes im-
posed on cotton the country actually earns a net value of $200/ha in
foreign exchange from growing cotton (i.e. the foreign exchange value of
the cotton less the cost of inputs used in producing cotton). Hence,
national profitability of wheat production is negative (1.3 x 180-200-75
= $-41/ton) and in fact there is a net loss of foreign exchange from
wheat production. This contrasts with the government's objectives to
save foreign exchange through local wheat production.
The resource cost ratio is a measure of the total cost of produc-
tion when prices are adjusted for taxes and subsidies and resources are
valued in alternative uses. It is calculated by dividing inputs and out-
puts used in production into "tradeables" and "non-tradeables" as fol-
a) Tradeables are commodities which are imported or exported,
such as wheat and fertilizer in the above example.
b) Non-tradeables are resources such as land or labor that do not
usually enter international trade.
c) All tradeable commodities are valued at their world price
equivalent. This is the price at which the commodity can be
imported (or exported), adjusted for transport costs and
exchange rate anomalies. In order to compare import prices it
is necessary to establish a common location, usually the place
of consumption of the commodity.
d) Inputs which are partly tradeable and partly domestic (e.g.
transport with tradeable fuel and spare parts, but non-
tradeable labor) are divided into their tradeable and domestic
e) Non-tradeables are valued at their returns in alternative
opportunities (again valued at world prices).
The Resource Cost Ratio (RCR) is then calculated as:
Returns to non-tradeable domestic resources in the next
best alternative use (valued at world price equivalent)
Value added to tradeables (valued at world price
equivalent) = Value of tradeable outputs value of
In the above example,
RCR = = 1.26
A ratio above one implies that the value of the domestic resources
employed is greater than the value of the foreign exchange saved or
earned. In the example, it costs $1.3 in domestic resources to save one
unit of foreign exchange. Thus, in this case the crop is unprofitable
to the nation, and no comparative advantage exists in devoting resources
This example shows a situation where policy measures make wheat
production of a crop profitable to farmers, although it is not an
efficient use of resources from the national perspective since it
decreases national income. There are, of course, many situations where
the opposite situation prevails where it would be efficient to produce a
crop, but policy measures such as low producer prices or tariffs on
inputs make it unprofitable for farmers to produce that crop.
2.1.1 Influence of Production Technique on the Resource Cost Ratio
It is a common mistake to assume that there is a unique comparative
advantage over the whole country. In fact, in most cases a country has
several different actual or potential production regions for a crop with
different technologies, yield potentials, and competing crops. Hence,
the resource cost ratio is likely to vary from region to region. This in
itself, is important information because it establishes the efficiency
of resource use between regions. If wheat can be produced in both
irrigated or rainfed areas there are clearly two quite different
strategies available for expanding domestic wheat production. Likewise
in a given region different techniques may be available such as large
mechanized farms and small farms depending on animal and manual power.
Each technique will have a different resource cost ratio.
The range of techniques included in the calculation of RCRs can
also include potential techniques. This can be used as a guide to in-
vestment in research. If in the above example researchers believe yields
can be increased to 1.6 tons/ha through additional expenditures of
$10/ha of local resources in developing an earlier variety and better
land preparation techniques, then
RCR = 99,
(1.6 x 180)-75
and wheat is now marginally efficient. If in the future there is the
possibility of a heat tolerant variety that will yield 2 t/ha then wheat
is a potentially efficient crop that justifies further investment in
research on that crop.
2.1.2 Influence of Transportation Costs on Comparative Advantage
The choice of consumption point and the cost of transportation are
also important when analysing comparative advantage. Returning to the
above example, assume that wheat production is possible in an inland
area which is 1,000 km from the capital city and main port a not
uncommon situation in many developing countries. If the cost of
transportation from the producing regions to the capital is $50/ton then
we can calculate the resource cost ratio under two assumptions; a) the
wheat is consumed in the producing region and hence competes with
imported wheat that must be transported from the port (i.e. capital
city) and b) the wheat is consumed in the capital city and hence the
value of domestic wheat must be adjusted for transportation costs. Table
2.1 shows the resource cost ratios for each consumption point assuming
two different transport costs. Clearly at the low transport cost, wheat
production is only marginally efficient if it is consumed at the
producing point. However, for the high transport cost wheat production
becomes efficient at the producing point but highly inefficient when
calculated for the capital city. In general resource cost ratios are
quite sensitive to the cost of both international and domestic
transport and the choice of consumption point.
It is also worthwhile noting three additional assumptions that are
implicit in the above calculations. First, if wheat is to be consumed at
the producing point some provision will usually have to be made for lo-
cal milling. Most milling facilities in wheat importing countries are
established at the port. Clearly, consumption of locally produced wheat
looses its advantage if wheat must be transported to the port for mil-
ling and the flour transported back again. Second, transport costs them-
selves are subject to considerable policy intervention. In particular,
governments often subsidize public transportation such as railways or
provide subsidies on transport inputs such as fuel. These distortions of
transport costs should be removed when calculating resource cost ratios.
Finally, input costs are also sensitive to transport costs. If
fertilizer is imported and shipped inland then its price in the
producing region should be adjusted for the higher transport cost.
Table 2.3. Resource Cost Ratios Calculated under Varying Assumptions
about Transport Costs and the Consumption Point.
Transport Cost from Producing Region
to Capital City-Port
Value of RCR
Wheat consumed in the
producing region .95 .80
Wheat consumed in the
capital city-port 1.26 1.67
a/ In this case, transport costs are added to the CIF Price. That is
for transport costs of $20/ton, RCR = 200/[(1.3 x(200+20))-75]
2.1.3 Uses of Domestic Resource Cost Analysis in Allocating Research
The resource cost ratio is a measure of comparative advantage and
can be used for a number of purposes.
a) To help decide on investment in a production program. Given
current technological coefficients and world prices the
domestic production of wheat in the above example is an
inefficient use of resources although the government might
still procede on the basis of other criteria such as food
b) To help decide on investment in research. Other things being
equal researchers will want to allocate research resources to
crops, techniques and regions which seem to be efficient users
of resources when measured from the national point of view.
Several types of decision situations are possible.
i) Allocation of research resources to a specific crop in a
given region. The resource cost ratios of potential
techniques, as in the above example, will be a basis for
such a decision.
ii) Allocation, of research resources to a specific crop
across regions. If the resource cost ratio of rainfed
wheat regions is less than that for irrigated wheat
regions, research on improved technology for rainfed
wheat will encourage expansion in regions where wheat has
its comparative advantage.
iii) Allocation of research resources across crops in a
specific region. Again a ranking of crops by their
resource cost ratio provides a measure of their
It should always be kept in mind that measures of comparative
advantage are a measure of the efficiency of resource use. Governments
have other objectives in resource allocation besides efficiency,
especially income distribution objectives. Nonetheless, the efficiency
of resource use is important and any measure which enables decision
makers to quantify the cost of pursuing other objectives will provide
considerably more information than is currently available.
2.2 Measuring Policy Incentives
Closely related to measures of comparative advantage are measures
of policy incentives. The divergence between national profitability and
private profitability is a measure of policy effects induced by a) taxes
and subsidies, b) import and exchange rate policies, c) price policies,
and d) market impefections such as monopolies.
A simple measure of policy incentives is provided by the ratio of
domestic prices to world prices (adjusted for transportation charges).
This is the "nominal protection coefficient" defined for producers as
NPC =domestic producer price
world price equivalent x official exchange rate
In the earlier example, NPC 300 1.7
where transportation charges from the producer to the city are $20-.
This rate of protection indicates that policy measures such as tariffs
or other import restrictions strongly protect local wheat producers.
Since official exchange rates are often a poor guide to the real
value of foreign exchange (i.e. the exchange rate is overvalued), it is
often useful to also calculate the nominal protection coefficient with a
"corrected" exchange rate to convert world prices to local prices. The
difficulty with this approach is of course the problem of choosing a
realistic exchange rate.
A better measure of policy incentives also takes into account ef-
fects of policies on input prices such as a subsidy on fertilizer which
increases the incentives for local production. This measure is defined
as the "effective protection coefficient", EPC, which is expressed as
Value of output Value of traded inputs per unit
Eat domestic prices of output at domestic prices
Value of output at Value of traded inputs per unit
world prices con- of output at world prices con-
verted at the offi- verted at the official exchange
cial exchange rate rate
In the example,
EPC = = 2.1
The EPC is higher than the NPC in this case, since there is a
subsidy on fertilizer. In general the EPC is a summary of the incentives
or disincentives created by government price policy interventions in
input and output markets. An EPC less than one indicates that policy is
a potential disincentive to production. However, the incentive provided
by pricing policy on a particular crop must be measured against
incentives provided to other crops. For example, if the EPC for wheat is
1.3 this might not be a particularly strong incentive to produce wheat
SNPCs can be measured at either the producing point or the consuming
point. In this paper we use the producing point so that the denomi-
nator is the farm gate equivalent of the world price. That is, in a
free trade situation, the farmer would receive the world price less
the cost of transportation. At the consuming point, the NPC is
equal to (300+20)/200 = 1.6.
if the EPC for a competing crop is 1.6.
These measures of policy incentives may be useful in understanding
trends in wheat production in a given country. For example, stagnant
production might be related to a deteriorating measure of policy
incentives provided to producers. Or measures of policy incentives might
be compared across regions to assess to what extent changes in policy
have favored particular regions.
2.3 The Overall Measure of Subsidies
The effective protection coefficient only takes account of distor-
tions in prices of outputs and inputs that are traded on the interna-
tional market. Governments commonly also influence prices for resources
used in agricultural production by providing subsidies, especially for
credit and water. In a situation where the EPC is less than one it is
useful to know if the effects of these subsidies on resources is
sufficient to compensate farmers for the tax implicit in the low
The effective subsidy coefficient is a measure of the overall
effect of government intervention in product, input and resource
markets. In the example above, let us assume that farmers receive
$150/ton for wheat. The effective protection coefficient is then:
EPC = = 0.88
That is farmers receive a price below the world price for wheat
(NPC = 150/180 = .83) and the subsidy on fertilizer is not sufficient to
compensate farmers for the low producer price. Take the case where
farmers receive subsidized credit (i.e. low interest rate loans) through
a government program and that subsidy (i.e. government costs less income
from the program) amounts to $10/ton of wheat produced. The effective
subsidy coefficient is then calculated as:
ESC (150-180) + ((75-50)/1.3) + 10 -30 + 19 + 10 0
The first term of the numerator is the difference between farmer
and world prices (at the farm gate), the second term the difference
between input costs to the farmer and equivalent world prices for inputs
and the third term the credit subsidy. The denominator is the world
price for wheat (at the farmgate). In this case the combined effect of a
government subsidy on fertilizer and credit compensates for the low
price received for wheat.
The effective subsidy coefficient, the overall measure of policy
incentives, and the resource cost ratio, the measure of comparative
advantage, are closely related. It can be shown algebraically that if
ESC is greater than zero (i.e. positive policy incentives for a crop)
and RCR is less than one (i.e. the crop has a comparative advantage),
then the crop is also profitable to farmers (and vice versa).
Table 2.2 shows four possible outcomes of the resource cost ratio
and the effective subsidy coefficient and their interpretation. The
most common situations are represented by cases where both are less than
unity or where both are greater than unity. In the first case,
governments are using the comparative advantage of the industry to keep
prices low. This sometimes happens in the case of basic food crops. In
the second case, the industry is able to survive only because of the
incentives provided by government pricing policy.
2.4 Data Sources and Analysis
It is clear that data will be needed from a wide range of sources
in order to calculate measures of comparative advantage and policy
incentives. One of the most important needs is reliable information at
the farm level on technical coefficients such as input levels and yields
as well as prices paid and received by farmers. In our case we used data
from farm level surveys which had been conducted over a number of years
in the two selected wheat growing areas. This was supplemented by a
mini-survey of machinery owners to provide a more detailed breakdown of
machinery costs such as depreciation, fuel and operator labor.
Table 2.2 Interpretation of Possible Results of the Resource Cost Ratio
and the Effective Protection Coefficient
Effective Protection Resource Cost Ratio
Coefficient Less than one Greater than one
Industry is efficient and Industry is not effi-
is not protected. Govern- cient and is not pro-
Less than 1 ment policy exploits in- tected.
dustry comparative advan- Likely to be a stagnant
tage by keeping prices or declining industry.
Industry is efficient and Industry is not effi-
at the same time protec- cient but favorable
Greater than 1 ted. Will usually have price policy allows
strong incentives to pro- domestic production.
Data on transportation costs were obtained through phone interviews
with the major trucking companies and through published annual reports
of the railways. The border prices of outputs and inputs were mostly
estimated from published sources. Two very useful publications in this
regard are the Monthly Bulletin of Agricultural Statistics published by
the FAO and which gives FOB and CIF prices for many commodities, and
Agricultural Statistics published annually by the USDA which provides
data on farm prices for inputs in the USA.
Data were analyzed on an Apple II micro-computer using the VISICALC
software package. This is a particularly useful package for this type of
work since arrays of data can be managed and budgets constructed. Once
constructed, sensitivity analysis can be performed or data from another
year or region can be analyzed very rapidly since the methods of cal-
culation for variables such as RCRs are recorded in the program.
In order to provide a longer term perspective on policy incentives
we analyzed data for several years. This is particularly important in
Mexico given that the recent economic situation is not representative of
the past, nor, necessarily of the future.
3.0 The Role of Wheat in the Mexican Agricultural Sector
and Major Policy Issues
The section provides a brief overview of some of the major trends
in production, consumption and imports of major food and feed crops in
Mexico, with particular emphasis on the role of wheat. The objective is
to provide a background in which to view the analysis of major policy
issues, especially those that relate to food self-sufficiency versus
export crop promotion.
3.1 The Macro-Economic Scene
The decade of the 1970s was one of important structural adjustment
in the Mexican economy. While the rate of growth of GNP was high, aver-
aging 5.2 percent per year between 1970 and 1980, growth of the agri-
cultural sector averaged only 2.3 percent per year, substantially below
the population growth rate.
The greatest shift occurred in the role of agriculture in external
trade. In 1970, agricultural exports accounted for over 50 percent of
total exports earnings from goods and services. By the period 1980-82,
the share of agricultural exports in trade had declined to 9 percent
(see Figure 3.1), reflecting in large part, the growth of oil revenues.
Cotton was until recently the single largest source of agricultural ex-
port earnings- In the early 1960s cotton accounted for 50 percent of
agricultural exports. By 1980-82 the share of cotton had declined to
less than 20 percent.
Overall, agricultural imports as a proportion of total imports have
tended to rise (Figure 3.1) and in 1981, Mexico registered for the first
time a net deficit in the balance of trade of the agricultural sector.
Nonetheless agricultural imports as a proportion of total imports are
lower than the average of 14 percent for all middle income developing
- Coffee is now Mexico's most important agricultural export.
Role of the Agricultural Sector in Mexican External Trade.
Agricultural Exports as a
Percent of Total Exports
Cotton as a Percent of
Agricultural Imports as a
Percent of Total Imports
Source: Informe Anual del Banco de Mexico, S.A.
An important aspect of macro-economic policy during the 1970s has
been the management of inflation and foreign exchange rate adjustments.
Domestic inflation has generally run ahead of that in Mexico's main
trading partners. Exchange rates have been fixed and maintained by
import controls and foreign borrowing until a sharp devaluation was
forced. As a result, the exchange rate was overvalued for much of the
decade. A rough measure of exchange rate overvaluation is provided by
adjusting the exchange rate using 1954 (a year of devaluation) as a
base, by the differential between Mexican and U.S. inflation rates.
Using this measure the exchange rate was overvalued by about 20 percent
at the beginning of the decade (see Figure 3.2). This increased to 45
percent in 1975, a difference that was largely eliminated by the 1976
devaluation. Continuing internal inflation led to a return to an over-
valuation of 50 percent in 1981. Since then the peso has been sharply
devalued and by 1982 was probably undervalued. The devaluation has been
accompanied by high inflation -rates approaching 100 percent per year.
In 1983 a two-tiered exchange system was in operation an official
rate for exports, essential imports and debt service repayment and a
market rate for other foreign exchange transactions. This complicates
the present analysis since agricultural exports and food imports were
subject to the lower exchange rate, while many inputs such as
agricultural chemicals and spare parts were imported at the higher rate.
To facilitate analysis we used the average of the two exchange rates or
about 130 Pesos/US dollar as the exchange rate for 1983.
The high inflation rate of 1982 and 1983 was also a problem in the
analysis. Prices changed drastically between the beginning of the crop
cycle and the end. (For example, diesel prices for the 1982/83 wheat
cycle were 4 Pesos/lt for land preparation but 14 Pesos/lt for
harvesting and transport.) It was also difficult to compare
profitability between crops because of different cycles. At the same
time, government policy toward the agricultural sector has changed in
the 1980's because of the need to adjust to new economic realities. For
this reason we use results over a period of years to obtain a longer
term perspective on policy effects on the agricultural sector.
FIGURE 3.2 Estimated Percentage Overvaluation of the Mexican Peso Based
on Differential Inflation Rates in the US and Mexico.
3.2 Recent Production Performance of Wheat in Relation to OLher Crops
In Tables 3.1 and 3.2 we summarize production statistics for wheat
in relation to other major crops. Crops are grouped into food grains,
feed grains, oil seeds and export crops. All these crops except sorghum,
rice and beans, compete with wheat in one of the areas chosen for this
study. (Maize competes in both areas).
Maize is the basic food grain in Mexico. However, with rising
incomes and urbanization, wheat tends to substitute for maize in the
diet, and per capital maize consumption declines (increasing amounts of
maize, however, are probably used for animal feed).
Wheat production increased rapidly in both the decade of the 1960s
and 1970s due largely to yield increases. It should be noted, however,
that the figures for the period 1970-72 to 1980-82 are heavily influ-
enced by the excellent wheat year in 1982 when both area and yields were
exceptionally high compared to previous years. Maize production has in-
creased nearly as rapidly as wheat with the source of growth largely due
to yield in the 1970s.
Over the last two decades the mix of oil seed crops has shifted
markedly from cotton seed to safflower and soya beans. The latter two
crops were introduced in the early 1960s and area expanded rapidly. How-
ever, yields have not increased significantly in either case.
Among feed grains the rapid expansion of sorghum is well known.
However, although barley is much less important, its yield has increased
consistently throughout the last two decades. In fact yields of barley -
a rainfed crop have increased more rapidly than wheat which is largely
an irrigated crop.
Finally, production of cotton, an export crop, has consistently
declined due to a decline in area. This largely reflects replacement by
import substituting crops, such as oil seeds and sorghum.
Table 3.1 Average Statistics
for Major Agricultural Crops in Mexico
Imports as Per-
Area Yield Production Net Imports cent of Appar-
(000 ha) (ton/ha) (000 ton) (000 ton) rent Consumption
Maize 6938 1.82 12615 1706 12
Wheat 892 4.08 3642 567 13
Rice 159 3.44 547 60 10
Beans 1868 0.71 1323 257 16
Safflower 314 1.06 334 0 0
Soya beans 336 1.79 603 966 62
Cotton seed 272 1.48 403 60 a- 13
Sorghum 1518 3.43 5204 1725 25
Barley 267 2.04 545 68 11
Cotton lint 272 0.94 256 -154 -151
a/ 1982-83 not included
Table 3.2 Growth Rates of Area, Yield and Production of Major Agricul-
tural Crops in the Two Decades of the 1960s and the 1970s.
1961-63 to 1970-72 1970-72 to 1980-82
Area Yield Production Area Yield Production
( percent/year ) ( percent/year )
Maize 1.5 2.5 4.0 -0.5 4.0 3.5
Wheat -1.0 4.6 3.6 1.8 3.2 5.0
Rice 1.1 1.6 2.8 0.6 3.1 3.7
Beans .8 2.4 3.2 0.4 2.0 2.4
Safflower 20.0 2.0 21.9 4.4 -3.0 1.2
Soya 22.0 -1.1 20.6 6.9 .0 6.9
Cotton Seed -6.2 2.8 -3.4 -4.2 1.0 -3.2
Sorghum 21.4 1.5 23.0 4.6 2.5 7.1
Barley .1 5.2 5.2 2.9 4.2 7.1
Cotton -6.2 3.9 -2.8 -4.2 1.1 -2.9
3.3 The Demand Outlook for Wheat
The consumption of wheat in -Mexico has grown at 4.5 percent over
the last decade. This reflects rapid population growth, growing incomes,
urbanization and declining real prices for bread. Projections of the
demand for wheat suggest a minimum growth rate of 3.5 percent annually.
Population will grow at an annual rate of 2.5 percent from 1980-2000.
The income elasticity of demand for wheat products is generally assumed
to be around 0.4-0.5 (Lamartine-Yates, 1981) although some estimates
place the elasticity somewhat higher (e.g. 0.6 by Bredahl (1981) and
0.6-1.0 by Lustig (1980)). Assuming a growth in per capital income of
2.0-2.5 percent, wheat consumption per capital is estimated to grow by
1.0 percent per year due to income gains.
At a growth rate of 3.5 percent per year, wheat consumption will
reach a minimum of 5.1 million tons in 1990 and 7.2 million tons in
2000. Other projections place consumption at higher levels. For example,
Lamartine-Yates (1981) forecasts consumption to grow at 5.5 percent per
year reaching 6.3-6.8 million tons by 1990. Much also depends on future
bread pricing policy. If consumer subsidies on bread are reduced then
there will be a slowdown in growth of bread consumption.
3.4 From Food Exporter to Food Importer
Despite an i--ressive growth in the production of most of the major
crops, there has been a sharp increase in dependence on imported food in
the 1970s. Mexico became a net exporter of food grains in the 1960s in
large part due to increased wheat production. However, by 1970-72, this
surplus had been converted into a deficit which has steadily increased.
Figure 3.3 shows these imports by three categories. Food grains (wheat,
maize, rice and beans) have been imported to the extent of over 3
million tons in recent years or about 15 percent of national production.
Feed grain imports have increased even more rapidly despite the rapid
growth in domestic production. Finally imports of oilseeds have also
jumped sharply in recent years, again despite a very high growth rate of
Figure 3.3 Imports of Major Categories of Crops, Mexico, 1965-82.
(Maize, Wheat, Rice, Beans)
74-75 77-79 80-82
W 1000" 604
- 1965-67 s8 -7, 7I-7 7- 7-
S-61 68-70 71-73 74-75 77-79 80-82
1965-67 68-70 71-73 74-75 77-79 80-82
This increased dependency on imported food has led naturally to
considerable debate on the need to achieve self-sufficiency in basic
foodstuffs. While self-sufficiency has been an important policy goal
during much of the 1970s, major efforts were not made to reverse the
trend toward increased food imports until the end of the 1970s when the
SAM program (Sistema Alimentario Mexicano) provided special incentives
for increasing production of basic foods. Although these incentives were
largely eliminated in 1983, food self-sufficiency remains a stated goal
of the Plan Nacional de Desarrollo (De la Madrid, 1983).
At the same time, with the current chronic foreign exchange con-
straints there is again talk of increasing production of export crops.
However, given that export crops, especially cotton, compete with some
basic food and oilseed crops such as wheat, safflower and soya, for land
and water resources, there is obviously a conflict between the goals of
food self-sufficiency and earnings of foreign exchange through export
3.5 Irrigated versus Rainfed Agriculture in Mexico
The debate on comparative advantage in food crops versus export
crops in Mexico is also closely tied to the debate on irrigated versus
rainfed farming. Historically, most investment spent on agriculture has
been directed toward irrigated agriculture. Even as late as 1971-74 the
proportion of agricultural investment in irrigation was over 70 percent.
The proportion of total cultivated area under irrigation has risen to
over 23 percent in the 1970s and early 1980s, compared with about 18
percent in 1965. Because of low yields in much of the rainfed area, the
proportion of the value of agricultural output produced in irrigated
areas was nearly half.
There is a sharp division between crops grown under irrigation and
those grown under rainfed conditions. As shown in Table 3.3 wheat, cot-
ton, safflower and soya are largely grown under irrigated conditions.
Maize, beans and barley are largely rainfed crops often produced under
difficult climatic conditions. Sorghum is the only crop that is grown
extensively under both irrigated and rainfed conditions.
Table 3.3 Percentage of Total Area and Production in Irrigated Areas
Percent of Percent Total Percent of Percent Total
Area Under Production from Area Under Production from
Irrigation Irrigation Area Irrigation Irrigation Area
Wheat 50 73 91 b 97
Maize 6 a/ 12a/ 18 b/ 29
Rice 47 60 46 65
Beans 3 9 15 25
Barley 2 7 22 46
Sorghum 61 70 44 53
Safflower 80 82 80 87
Soya Beans 50 44 78 85
Cotton 65 na 79 95
a/ Calculated for 1960-62.
/ Calculated for 1975-77.
na = Not available
There is naturally a debate as to whether a larger proportion of
the irrigated areas should be devoted to basic food crops such as maize
and beans. In fact, irrigated maize is quite important accounting for
some 20 percent of irrigated area in 1981. Given considerably higher
yields than in rainfed areas this is equivalent to about one third of
national maize production and perhaps almost one half of marketed sur-
plus since most maize in irrigated areas is commercially produced .
The share of irrigated land devoted to export crops, especially
cotton, has fallen while the production of oilseeds and other crops has
SDuring 1981 and 1982 the area under maize in irrigated areas was
increased due to special incentives of SAM.
increased (Figure 3.4). Grains account for over 50% of harvested area
and wheat has consistently accounted for around 20 percent of the
harvested area under irrigation.
Looking to the future it seems that the area under irrigation will
expand more slowly in the future because it has become increasingly
costly to develop new irrigated areas. Lamartine-Yates (1981) estimates
that a maximum of 1 million ha of new land could be brought under irri-
gated between 1975 and 2000- This places a special premium on the
efficient use of existing irrigated areas.
Wheat, an irrigated crop, offers a special opportunity to diversify
from the irrigated districts without sacrificing the goal of self-suf-
ficiency. Wheat has, of course, been a major success story with rapid
expansion of production through the 1960s and 1970s. Wheat production
increased from 1.35mt in 1960-62 to 3.48mt in 1980-82, while area re-
mained relatively constant. Wheat yields increased because of rapid in-
creases in yields in irrigated areas but also because of a switch from
rainfed to irrigated areas. Historically, wheat has been produced under
rainfed conditions in the summer cycle. The Mexican altiplano (including
the States of Tlaxcala, Mexico, Puebla and Hidalgo) in colonial times
produced a wheat surplus for export to the Caribbean colonies. As late
as 1962 half of Mexican wheat area was sown in rainfed areas. Since that
time however the importance of rainfed wheat has declined to reach only
8 percent of area in 1974 (Figure 3.5). In the altiplano where over
100,000 ha of wheat were grown in the 1950s, wheat practically disap-
peared. This decline was in part due to the emphasis in government pol-
icy on promoting wheat research and production in irrigated areas as
well as to the rise of competing crops in rainfed areas, especially
malting quality barley.
The scope for further expansion of wheat in irrigated areas is
limited. Significant area expansion is only possible if wheat is
There is also scope for more efficient water use in existing irri-
gated areas. Water use efficiency is estimated to be less than 50
percent (Palacios, 1975).
Figure 3.4 Percent of Harvested Area in Irrigation Districts in Grains,
Cotton and Oil Seeds, 1958-81.
Source: Informe Estodistico, SARH.
Figure 3.5 Irrigated Wheat Area and Production as a Percent of Total
Wheat Area and Production, Mexico, 1960-81.
substituted for other crops. Likewise, it is expected that the yield of
irrigated wheat, which has exceeded 5.0 t/ha in recent years, will
expand more slowly in the future. Recognizing this, the Mexican
Government through SAM gave special subsidies for rainfed wheat
production and this together with the recent release of improved
varieties for the rainfed areas led to some reversal of the decline in
rainfed wheat. In 1981 the area under rainfed wheat reached 154,000 ha
or 18 percent of the total wheat area.
With the rapid expansion of wheat demand and limited potential for
expanding irrigated wheat, the government is projecting an increased
role for rainfed wheat in the future with the possibility of as much as
1.5 m ha of rainfed wheat by the year 2,000 (or double the area of irri-
gated wheat) (Rodriguez Vallejo, 1982). An additional advantage of rain-
fed wheat is that much of it could be produced closer to consuming
points, thus substantially reducing the need for long distance trans-
portation from the Northwest.
3.6 A Comparison of Wheat in Irrigated and Rainfed Farming Systems
This study analyzes the role of wheat in two contrasting regions of
Mexico. The first region is the Yaqui Valley of the State of Sonora
which is the most important wheat growing area of Mexico. An average of
over 100,000 ha of irrigated wheat are sown annually with yields varying
from 4.5 to 5.1 t/ha. The second region is the rainfed area of the
States of Tlaxcala and Hidalgo, centered on the valleys of Calpulalpan
and Apan and the surrounding low lying hills. Although wheat was
traditionally grown in the area, it was largely replaced by barley
during recent decades. Barley production was stimulated by the demand
for malting quality barley for the nearby Mexico City breweries.
These regions were chosen because farm level surveys have been un-
dertaken as part of on-farm research and training in the area. They re-
present an established irrigated wheat area but with a serious disad-
vantage of transportation costs because of the long distance to main
consuming points, and an area where wheat is a minor crop but has
considerable potential (Figure 3.(,). General details of each area are
given in Table 3.4.
The Yaqui Valley: Wheat production in the Yaqui Valley has
increased rapidly as a result of yield increasing technology (Figure
3.7). Yields increased most rapidly during the 1960s with the use of
semidwarf wheat varieties and improved cultural practices. Yields have
risen less in the 1970s although continued release of new varieties, a
narrowing of the performance gap between small ejido farmers and large
private farmers and improvements in land quality through levelling,
drainage and salinity control have led to average yields close to 5 t/ha
in recent years.
Wheat in this area largely competes with cotton and safflower for
available land and irrigation water, which is in short supply in many
years. Cotton was the major crop in the valley but after the 1950s has
given way to wheat and oil seeds. This reflects a drop in world cotton
prices in the 1960s, rapid technological change in wheat production and
government policy to encourage wheat production. Cropping patterns are
set by water allocations and in most years water allocations favor wheat
over cotton. This reflects a policy to encourage food crops but also the
fact that cotton requires at least 50 percent more water than wheat.
Farmers perceive cotton to be a risky crop because of yield variability
due to insect attack and weather. Price risk is also high since there is
no guaranteed price for cotton and prices vary according to interna-
tional prices and exchange rates, as well as discounts charged by local
Safflower is a relatively new crop whose area jumped sharply in the
1970s, although yields have changed little. It is particularly suited to
the lighter alluvial soils. It also has the advantage that its water
requirements are relatively low about two-thirds those of wheat.
Both cotton and safflower are commonly grown in rotation with
wheat. Cotton and safflower require row cultivation during the cooler
winter months which helps eradicated the grassy weeds, phalaris and wild
Figure 3.6 Distances Between Sonora, Tlaxcala and Veracruz, Mexico.
Major Characteristics of the Yaqui Valley, Sonora and the
Altiplano of Tlaxcala/Hidalgo
Yaqui Valley, Sonora
Altiplano of Tlaxcala/Hidalgo
Soya beans, Maize
Private farmers (20-100+ha)
Collective Ejidos (5ha/
for many years. Well
Active labor market and
Nearly all farmers work
with banks and credit
Well developed input
Well developed markets.
Wheat sold directly to
government buying agents,
killers or co-operatives.
Barley, Maize, Wheat, Maguey
None. Animal Grazing
Private farmers (1-100/ha)
Mechanization of major oper-
ations in recent years. Well
established machinery rental
Active labor market with com-
peting non-farm jobs
Most farmers now use bank
Some problems in input
No wheat purchased until 1981.
Most farmers sell to private
oats, which are major problems in wheat.
In fact, a wheat-cotton rota-
tion was found to be an extremely effective means of weed control in
one recent field survey (Byerlee, 1981). Likewise, such a rotation is
beneficial for cotton which is subject to serious insect problems.
Hence, there are substantial advantages to maintaining a crop rotation
rather than dependence on a single crop such as wheat or cotton.
Maize and soyabeans are grown as second crops. Maize is commonly
planted in August and harvested in January to March. It is often grown
FIGURE 3.7 Wheat Area and Yields in the Yaqui Valley, 1960-82.
(Doq/) PIl!A 4.Dq4M
as a "fill in" crop between wheat and cotton (see Figure 3.8). Soyabeans
are a popular second crop in the rotation. Most farmers prefer a
wheat-soyabean-wheat-soyabean rotation. However, high water requirements
(about 50 percent above those of wheat) restrict the area of soyabeans
in most years.
Agriculture in the Yaqui Valley is highly commercialized. Almost
all operations are mechanized with most labor employed in irrigation and
weed control in row crops. The labor requirements of cotton depend on
whether it is harvested manually or mechanically. In recent years
mechanical harvesting has become more common. Market systems are well
developed for both inputs and products. Farmers generally received the
guaranteed price for their sales. Most farmers also receive short-term
credit from either official sources or private bank or credit unions.
Extension is still deficient although new varieties are adopted quickly,
largely through an efficient seed distribution program.
Tlaxcalz,'Aidalgo- Unlike Sonora, Tlaxcala and Hidalgo are areas
of traditional agriculture with large numbers of small farmers where
technological change had little impact until recently. A major influence
on choice of crops and technologies in this area is the incidence of
climatic hazards, especially drought and frosts. Rainfall is most
reliable and frost incidence least in the period June to August. This
especially favors short season crops such as barley. Barley was
traditionally grown for animal feed but is now largely produced for
Barley production received a major stimulus in the late 1960s with
the release of improved varieties with good malting qualities. Improved
varieties, herbicides for broad leaf weeds and fertilizers were adopted
very rapidly and by the end of the 1970s nearly all farmers were using
these practices (see Byerlee and Hesse de Polanco, 1982). Barley yields
almost tripled in this period (see Figure 3.9) although-the sharp jump
in 1976 may reflect a revision of the statistical estimation procedures.
1/ For more details on cropping systems and production practices in
the area, see Byerlee, Harrington and Marko (1981).
FIGURE 3.8 Major Cropping Systems in the Yaqui Valey, onora.
FIGURE 3.8 Major Cropping Systems in the Yaqui Valley, Sonora.-
oiv. May Sep+t.
4. Whoat-Syobeon-Soff lower
/ Maize and Soyobeans in These Rotations are Second Crops Which are Subject to
the Availability of Water.
FIGURE 3.9 Area and Yield of Barley and Wheat in Tlaxcala, 1960-80.
Wheat-Rainfed Tloxcalo 2.0
Wheat production declined until recently (see Figure 3.9) reflec-
ting a) lack of a wheat marketing outlet, b) unsuitable varieties and
c) lack of active promotion in contrast to barley which was promoted by
a private association of brewers. By 1980 most of these obstacles had
been removed. INIA had released two wheat varieties, Cleopatra and
Zacatecas, for dryland conditions. The government marketing agency,
CONASUPO, began to receive wheat and under SAM, credit and special
incentives became available for wheat production. As a result, wheat
area increased from 2900 ha in 1980 to over 13000 ha in 1982. However
problems remain before wheat can replace barley on a major scale. Ear-
lier, more disease resistant varieties are still required- Seed is a
problem: farmers have often been provided with seed of older varieties
quite unsuitable to the area. Marketing still needs to be improved since
farmers receive a price well below the guaranteed price, in part because
of problems of grain quality and impurities but largely because of
corrupt practices of the marketing agency.
The other major crops, maguey and maize, are declining or stagnant
crops. The declining demand for pulque, the major product of maguey, and
high labor requirements have influenced the reduction in maguey area.
Maize remains largely a subsistence crop. The long growing cycle exposes
maize to a high incidence of climatic risks. Technological change and
yield increases in maize have been modest and are largely due to adop-
tion of intermediate levels of fertilizer use. There is substantial
potential to increase maize yields but climatic risks will be a serious
disincentive to increased expenditures on maize production. Also the
labor intensity of maize production, particularly for harvesting,
discourages area expansion.
3.7 Wheat Production Techniques in Sonora and Tlaxcala
Table 3.5 shows technical parameters used as the basis for the cal-
culation of Resource Cost Ratios for each production technique. These
1/ Two new varieties were released by INIA in 1982 which are earlier
parameters were derived from field surveys in the area and represent the
most common production technique. As expected, input use per hectare is
substantially higher in irrigated wheat. However, when converted to in-
puts used per ton of wheat produced the input-output coefficients are
not very different. Irrigated wheat uses significantly more labor but
less mechanical inputs and seed.
Technical change may alter these possibilities. In rainfed areas,
availability of improved varieties and use of improved timing of opera-
tions should enable average yields of 2.5 ton/ha which is close to the
yields obtained in on-farm experiments in the area over a five year
period. In irrigated areas, improved varieties released in 1981, im-
proved weed control and irrigation and possible change of sowing prac-
tices should bring average yields to 5.5 t/ha by 1990. In both cases and
especially in Sonora, there are possibilities of reducing costs of oper-
ations, through reduced tillage, lower seed rates and improved fertil-
izer management which should improve the efficiency of wheat production.
Table 3.5 Technical Parameters for the Production
of Rainfed and Irri-
Rainfed Wheat Irrigated Wheat
Yield 2.0 ton/ha 4.7 ton/ha
Per ton Per ton
Inputs Per ha of wheat Per ha of wheat
Tractor (Hrs.) 7.5 3.75 12.5 2.66
Combine (Hrs.) 1.0 .50 1.25 .27
Labor (person days) 2.6 1.30 9.5 2.02
Seed (kg) 120 60 170 36
Nitrogen (kg) 70 35 190 40
Phosphorous (kg) 30 15 30 6.4
Herbicide for broadleaf
weeds (It) .75 .38 2 .43
Insecticide (No. of applic.) 0 0 1 0
Irrigation water (cm) 0 0 85 18
4.0 Agricultural Pricing Policy and Producer Incentives
4.1 Product Prices
Mexico has had a system of guaranteed prices for most food, feed
and oil seed crops. As we have seen these guaranteed prices acted as
effective farmer prices for the irrigated areas but are subject to
discounts in the rainfed areas. Guaranteed prices are the same through-
out the country so that there is no allowance made for differential
transportation costs between different regions of the country.
Cotton is the major crop analyzed in this study that does not have
a guaranteed price. Cotton prices are fixed at world prices, converted
at the official exchange rate and then increased by a factor (about 5
pesos/1001b lint in 1982/83) to obtain the local price in Sonora. The
price of cotton seed is, however, set by the guaranteed price.
Much of the earlier analysis of producer prices in Mexico has
focused on declining real prices, using price indices as deflators.
However, real prices are not a good guide for either farmers or policy
makers. Comparison of wheat's relative price with respect to other
crops, to input prices and world wheat prices are better indicators.
For those crops covered by guaranteed prices, prices maintain a
fairly close relationship to one another. Maize and wheat prices gener-
ally moved together with a slight margin in favor of maize. Soyabean and
safflower prices also move together with soyabean prices generally 10-20
percent above safflower prices. Wheat prices have generally been 45 to
50 percent of safflower prices.
The ratio of the wheat price, which is fixed by the government, to
the cotton price, which is determined by the prevailing export price,
fluctuates quite sharply from year to year as shown in Figure 4.1. In
1974 and 1975, the relative price of wheat increased sharply due to the
increase in the guaranteed price of wheat. However, in 1976 the increase
in international prices of cotton combined with a devaluation of the
Ratio of Price of Cotton to Wheat and the Areas of Cotton
and Wheat, Mexico, 1970-83.
Source: Econotecnia Agricola VII Num.9 Sept. 83.
Agenda Economica Agricolo 1983.
Mexican peso led to a relative price of wheat only 40 percent of a year
earlier. Thereafter the relative price of wheat increased in large part
because of overvaluation of the exchange rate during the period, which
kept increases in domestic cotton prices well below the inflation rate.
In 1982, two large devaluations restored the price of cotton relative to
wheat. Since then, the ratio has declined as the official peso revalued
in real terms against the US dollar, at least until the current
arrangements were altered in July, 1985.
These variations in relative prices of cotton and wheat often lead
to large shifts in area between cotton and wheat. For example cotton
area fell to a record low in 1976 in response to low prices relative to
wheat in previous years. Likewise, in 1982, the wheat area in irrigated
areas was well above normal because of a reduction in cotton area (see
Figure 4.1). The possibilities of increasing cotton area with favorable
prices (as, for example, in 1983) are limited by water restrictions as
well as farmers' risk aversion.
In rainfed areas, the main competing crop to wheat is barley.
Barley prices received by farmers have generally been slightly higher
than the guaranteed price for wheat, but both prices move together. In
the period 1980-83, wheat prices tended to be equal to or slightly
higher than barley. Nonetheless, other factors such as availability of
suitable varieties and an adequate marketing system are probably more
important in farmers' decisions on wheat versus barley than these small
variations in relative prices.
The effect of these various price interventions will depend on the
response of wheat production to changes in the price of wheat and com-
peting crops. Recent econometric analyses of supply response suggest
that wheat production is only moderately responsive to price. An
increase in the real price of wheat of one percent will lead to an in-
crease in the production of wheat by 0.2-0.4 percent (Bredahl, 1981 and
Rosales, 1982). There is evidence that wheat production is quite re-
sponsive to a change in safflower prices (Rosales, 1982). Although none
of the studies analysed the response of wheat to cotton prices, farmers
in the Sonora have in recent years clearly made large changes in their
allocation of land between cotton and wheat, depending on their relative
4.2 Input and Factor Prices
Input pricing and distribution is in the hands of both the public
and private sector. Seed is multiplied by a public company, PRONASE,
and certified seed is sold at a fixed price in all producing regions.
Wheat seed costs have been generally double the commercial price of
We have no information on whether PRONASE is subsidized, but the
relationship of the price of commercial seed to grain would suggest that
there is a sufficient margin to cover seed processing and distribution
costs. In the case of wheat, seed is purchased from growers at 10 per-
cent above the guaranteed price. It is then treated, bagged, and graded
and resold to farmers the following cycle at about 50-100 percent above
this price. Seed can be regarded as a tradeable item since'considerable
quantities of wheat seed have been exported. The price received for
seed for export has been well above the national price (there is a
substantial export tax on wheat seed) but since the market is very
limited we shall assume here that the price of seed paid by farmers is a
fair reflection of its opportunity cost.
Fertilizer consumption in Mexico has expanded at an annual rate of
12.6 percent since 1960-/. Although the domestic production of fertil-
izer doubled in the 1970s, Mexico has consistently been a net importer
of fertilizer during this period (Figure 4.2). Mexico began to export
Ammonia in 1978 but substantial amounts of dry nitrogenous fertilizer
For an overview of the Mexican fertilizer industry see IFDC (1981)
and Secretaria de Programaci6n y Presupuesto (1981).
Mexican Imports and Exports of Fertilizer, 1970-81.
19707I 712 7 A4 7A 7 7 7A 5 0 A
TOTAL N, P, K,TRADE
1. Includes phosphrte fertilizers oad phosphoric acid.
2. Includes lntrogen fertilizers aind emonla.
3. Mexco does not export potosh.
Source: Mexico,the Fertilizer Industry, IFDC, 1980; Plan de Desarrollo de la Industria Mexicona de los
Fertilizonte,Vol. II,Productos Terminados, Fertimex, S. A. 1982.
continued to be imported through 1981. Mexico has provided a substantial
surplus of phosphate rock for export. Increasing demand resulted in a
deficit at the end of the decade which was made up by imports. However,
current large scale plans to develop local phosphate rock deposits will
change this picture by the mid 1980s.
In 1980, Mexico imported 18 percent of its fertilizer needs, much
of it nitrogenous fertilizer. Preliminary figures for 1981 show 1
million tons of fertilizer imports. Hence, in the recentpast Mexico can
be considered as an importer of finished fertilizer products, especially
Urea, triple superphosphate and diamonium phosphate which are the most
common fertilizers in crop production.
The bulk of domestic production of these fertilizers occurs in the
State of Veracruz close to supplies of natural gas and is transported by
rail throughout the country. Production and distribution is controlled
by FERTIMEX which charge a unifor~ price for fertilizer throughout the
country. Fertilizer prices have declined sharply in real terms since
1970. Figure 4.3 shows that the ratio of the price of one unit of
nitrogen to the price of wheat declined from over four in 1970 to less
than one 'in the 1982/83 winter wheat cycle. There are large explicit and
implicit subsidies in these fertilizer prices from the following
a) FERTIMEX purchased natural gas at rates well below the world
price for use in fertilizer manufacturing.
b) FERTIMEX operated at a loss in many years and a government
Subsidy of abpiot l0Cpercent of income was required to make up
c) Transport of fertilizer is highly subsidized (see Section 4.3).
An effort was made to calculate the extent of these subsidies by
computing a world price equivalent in the producing region. Since most
fertilizer used in Sonora is transported from the south, the world price
Figure 4.3 Ratio of the Price of Nitrogen to Wheat, Sonora, 1970-83.
was calculated as the f.o.b. price plus unsubsidized transport costs- .
Results shown in Figure 4.4, indicate that fertilizer subsidies began in
the 1974/75 period of extremely high world prices. In recent years they
have increased substantially to reach about 50 percent of the costs of
urea to farmers in early 1983. In the past two years, increase in
fertilizer prices have reversed this trend.
Herbicides, particularly 2-4,D for broad leaf weed control, are the
main chemicals used in wheat production in Mexico. Additional spe-
cialized herbicides for grassy weeds and an insecticide application are
also common in Sonora. Competing crops, especially cotton and soya
beans, use substantial amounts of insecticides.
Thirty percent of agricultural chemicals approved for use in Mexico
are manufactured locally accounting for 70 percent of total -demand.
These chemicals include most of those used in wheat and cotton produc-
tion. Mexican prices were, however, substantially higher than world pri-
ces. Mexican farmers paid about 50 percent above the world price for
these chemicals except in 1983 when the devaluation brought Mexican
prices close to the US price. In the case of wheat, however, these
chemicals are a small part of total costs of production.
The cost of mechanization is primarily determined by the cost of
machinery and the cost of fuel. The decade of the 1970s was a particu-
larly rapid period for mechanization in Mexican agriculture. Demand for
tractors increased at an annual growth rate of nearly 10 percent between
1970 and 1980. The number of tractors sold annually nearly doubled to
23,000 units between 1976 and 1981. Much of this increase in tractor use
-/ This ignores costs of marketing and storage which amounted to about
Half of distribution costs in 1980.
A review of the Mexican pesticide industry may be found in Secreta-
ria de Programaci6n y Presupuesto (1981).
Mexican Price and World Price Equivalent for Urea and Triple
Super Phosphate in Sonora, 1970-83.
Triple Super Phosphate
occurred in the rainfed areas since irrigated areas were already highly
mechanized by 1970.
Mexico has followed a practice of encouraging local manufacturing
of tractors- The percentage of imported tractors in total tractor
sales has fallen from over half in 1970 to 20 percent in 1980. Of
course, tractors manufactured in Mexico, have a high import content with
the percentage of directly imported components, ranging from 40 to 60
percent. By 1980, imports were mostly tractors of over 130 HP, which
were imported without restrictions and free of duty. Since small
tractors, as well as large ones, can potentially enter international
trade, tractors may be regarded as a tradeable item.
Despite quantitative restrictions on imports of smaller tractors,
the price of domestically produced tractors has generally followed the
US price for the equivalent model tractor. Prices are set in relation to
US prices and have in fact been below US prices in periods immediately
after a devaluation. This has allowed Mexico to export tractors on a
small scale; as for example, in 1983. Ploughs and harrows are also manu-
factured locally and prices seem comparable with US prices. In 1983,
Mexico was exporting these implements to the US to take advantage of the
opportunities offered by the devaluation. Other specialized equipment
such as seed drills for wheat and combine harvesters are fully imported.
Duties on these items are usually zero with the exception of combine
harvesters which were charged a duty of 10 percent in 1983.
Diesel fuel for mechanical operations is an important production
input in both Tlaxcala and Sonora. It is also important in determining
transport costs to consuming points. Diesel fuel has been highly sub-
sidized throughout the period, as part of government efforts to keep
transport costs low. As a result diesel prices deflated by the CPI or
the price of wheat generally declined in the 1970s, during a period when
1/ For a review of the Mexican tractor industry see Secretaria de Pro-
gramaci6n y Presupuesto (1981).
world prices were rising rapidly. This position was dramatically re-
versed in 1982 and 1983 when diesel prices were increased by 1,400 per-
cent in 14 months. As a result the price of diesel in wheat equivalents
increased 460 percent.
We do not have an equivalent f.o.b. price for diesel to calculate
the subsidy on diesel. Instead we have used 85 percent of the US farm
price for diesel to approximate world prices. This adjustment reflects
the fact that Mexico is an oil exporter and also the fact that taxes are
included in US prices. The implied subsidy on diesel using this method
increased to 85 percent in 1981 and then dropped to 50 percent in 1983
with the change in government policy (see Figure 4.5).
4.2.6 Overall Changes in Mechanization Costs
Total mechanization costs are determined by initial capital costs,
labor costs for the operator, repairs and fuel. In order to separate out
these costs a simple model is developed in Appendix B and compared to
the cost of machinery rental a common practice for small farmers in
both Tlaxcala and Sonora. In Tlaxcala rental costs declined in real
terms from 1975 to 1980 (see Byerlee and Hesse de Polanco, 1982). This
reflects a decline in the price of diesel but also the fact that returns
on capital to the machinery owner have declined quite markedly from over
20 percent in 1975 to less than 10 percent in 1981. This probably
reflects greater competition in the rental market since the number of
tractors in the area increased sharply relative to the amount of land
prepared by tractor. It probably also reflects the greater availability
of loans at subsidized interest rates for the purchase of tractors.
In 1982 and 1983 machinery rental rates increased rapidly but not
enough to compensate for higher costs of operations. By 1983 returns on
capital to the owner had fallen to close to zero. Owners of course had
gained through high inflation rates for their machinery but had lagged
in raising prices to meet depreciation charges. The increased price of
diesel also had a strong influence on the cost of tractor operation.
Fuel accounted for only 8 percent of the cost of tractor rental in 1975
Figure 4.5 Real Price of Diesel in Mexico Compared to Estimated World
but had risen to over 30 percent of costs in 1983.
The situation in Sonora is similar except that the rate of return
on capital is somewhat lower reflecting the greater number of tractors
and a more competitive rental market.
Rental costs for machinery in Mexico are very close to rates paid
in the US. Table 4.1 shows comparable costs in Michigan, converted at
the official exchange rate and in Sonora. In 1980, rental rates were
almost identical but with the devaluation in 1982, Sonora rates had
fallen relative to those in Michigan.
Overall, machinery costs have probably been lowered in Mexico
through an explicit subsidy on diesel, an overvalued exchange rate
during much of the 1970s and more recently through subsidized credit to
purchase machinery or rent machinery services. In this study we only
explicitly allow for the diesel subsidy and the overvalued exchange rate
when adjusting to world prices. The removal of the subsidy on diesel
would have raised the cost of tractor services by 34 percent in 1981.
Irrigation services in Sonora are managed by Irrigation District
Authorities under the Secretariat of Agriculture. Farmers are charged
for water per hectare for each crop, although meters are now being
installed to measure the volume of water applied. The price charged to
farmers for water has tended to decline sharply in real terms (see
Tables 4.2 and 4.3). In 1982/83, out of a total wheat yield of close to
5 tons/ha, less than 5 percent was needed to pay the costs of water.
These low water prices have been achieved through increasing
subsidies on the operation and maintenance of the irrigation districts.
Subsidies increased steadily during the 1970s. For all irrigation
districts of Mexico they amounted to 30 percent of irrigation services
in the early 1970s, but by 1976 had risen to over 50 percent. Official
data on revenues and costs are not available since 1976 but reliable
Table 4.1 Contract Hire Rates for Machinery in Michigan, U.S.A. and
Sonora, Mexico, 1980 and 1982
Plowing/ Harrowing Combine Harvesting
(pesos/ha) (pesos/ha) (pesos/ha)
1980 Michigan 550 265 920
Sonora 550 275 900
1982 Michigan 1150 750 1634
Sonora 1000 500 1500
SMould board plough for Michigan and disc plough, Sonora.
/ Small grains harvesting.
Source: Michigan, G. Schwab, Michigan State University, personal
communication; Sonora, farmer interviews.
Table 4.2 Indices of Prices of Water and Electricity, 1965-1978.
Irrigation Prices Electricity
Real Pricea Real Price
Price Index Index Price Index Index
( 1960 = 100 ) ( 1960 = 100 )
1965 111 93 96 81
1970 159 113 96 68
1975 294 116 73 29
1976 306 99 75 24
1977 300 74 137 34
1978 316 66 140 29
a/ Deflated by Implicit Price Index of GDP.
Source: Lamartine-Yates (1981).
reports indicate that subsidies reached over 80 percent in 1982. In
Sonora, subsidy levels are probably somewhat lower than the national
average because of better developed irrigation systems and more effec-
tive management. Nonetheless, one report from the Yaqui Valley in the
mid 1970s indicated a subsidy level of close to half.
Table 4.3 Average Cost of Water in the Yaqui Valley for Wheat Produc-
Cost of Water
Total in Terms of
Wheat Cycle Water Charge Wheat
(pesos/ha) (kg of wheat)
1980/81 1,500 326
1981/82 1,700 245
1982/83 2,200 157
1984/85 5,722 155
Source: Farmer interviews.
One result of subsidies and water scarcity is the development of an
open market for water. In 1981 and 1982 farmers commonly purchased water
rights from neighbors at prices about three times the official charge
for water. This practice was prohibited in late 1982 but farmers now
rent the land with water rights with land rents reflecting scarcity
value for water for a particular crop.
High subsidy levels on water have been criticized because they lead
to inefficiency and wastage in water use (e.g. Palacios, 1982). In
particular, farmers have little incentive to economize on water use by
reducing the number of irrigations and choosing water-efficient crops.
There are additional subsidies in irrigation districts that are
more difficult to quantify. These include special land improvement pro-
grams such as salinity control, land levelling and subsoiling. Costs of
most of these operations are not included under the operating costs of
the irrigation authorities. For example an extensive program of land
levelling has been conducted in Sonora with loans from the official
credit bank. Two thirds of the costs are paid by the government and in-
terest rates are subsidized on the remaining one-third of the costs.
The cost of developing the irrigation infrastructure itself is a
significant cost of irrigation water. The Yaqui Valley can be regarded
as a mature system and these costs are essentially "sunk" costs. The
question then is how to use this valuable infrastructure most effective-
ly. At the same time new irrigation areas are being developed and if the
purpose is to produce wheat, cost of development should be included
along with costs of operation and maintenance.
Costs of developing new irrigation areas have generally increased
simply because the easier projects have already been constructed. Table
4.4 gives some data on development costs over time. We have not been
able to obtain more recent information, but if costs have risen at the
same rate as the general price index, a conservative estimate would be
$500,000/ha in 1983, equivalent to $US4200/ha.
Table 4.4 Real Costs of Developing New Irrigated Land
$/ha (1970 Pesos)
Source: Lamartine-Yates (1981).
4.2.8 Credit and Insurance
A major component of government incentives to agriculture has been
the provision of credit through both official banks and the (formerly)
private banking sector. Credit available to the agricultural sector
increased very rapidly in the second half of the 1970s, especially in
rainfed areas. While most Sonora farmers had long worked with credit
from banks or credit unions, most farmers in Tlaxcala operated on their
own sources of funds. In 1979, 37 percent of farmers in a barley survey
worked with the official credit bank. By 1982, this figure was certainly
above 67 percent and even higher for wheat. Only in the case of maize is
private funding still important.
There are a number of explicit and implicit subsidies in the
granting of official credit. The most obvious is the provision of credit
at subsidized interest rates. Interest rates vary according to the type
of producer (i.e. small or large farmer) and also in recent years inter-
est rates have been crop specific. Table 4.5 shows interest rates in
1983. Note that official rates applied only to a portion of the cost,
varying from 50-90 percent of the estimated cost of production. If
needed, farmers must obtain funding from other sources at prevailing
rates for the remaining part of the cost.
Interest rates for basic crops for small and intermediate size
farmers are compared with the rate on 6-month savings certificates in
Figure 4.6. In earlier years, subsidies appeared to favor the small
farmer. In later years, subsidy levels have increased sharply for all
classes of farmers. In particular, the rate of interest for basic food
crops was only about a quarter of the commercial savings rate in 1982.
It should also be noted that for most years the real rate of interest
discounted for inflation is negative.
There are other implicit subsidies in bank services. For example,
the official credit bank often provide inputs especially chemicals, at
wholesale prices. Furthermore, all farmers receiving loans through the
credit institutions must purchase crop insurance, which has also been
subsidized (especially under SAM).
Active labor markets exist in both areas. In Tlaxcala, farmers
often face a shortage of unskilled labor because of alternative job
opportunities in factories in the area and the nearby Mexico City labor
market. As a result, mechanization has proceeded rapidly in the 1970s
and by 1982 farmers were beginning to mechanize their last major labor
intensive operation, maize harvesting.
In Sonora the existence of a class of agricultural laborers and
substantial inmigration and seasonal migration also leads to a
Figure 4.6 Interest Rates on Short Term Agricultural Credit Compared to
Interest on 6-Month Savings Certificates, 1971-84.
Maize, Beans, Rainfed
Wheat (SAM Program)
Source: FIRA and Banco de Mexico.
Table 4.5 Rates of Interest on Short-Term Agricultural Loans, 1983.
Rate of Interest, January, 1983
Proportion of Industrial
Loan Covered by Maize, Beans, Other Basic Crops (e.g.
Interest Rate Rainfed Wheat Commodities Cotton)
Small farmers 90 12* 27 27.5
size farms 80 12* 34 34.5
Large farmers 70 12* 37 60
* 27 percent as of April, 1983.
S Small and intermediate size farmers have incomes less than 1000
times and 1000 to 3000 times, respectively, the rural minimum wage
for the region.
competitive labor market. Irrigation and hand weeding in row-crops and
cotton harvesting are the major labor using activities. Cotton
harvesting has been mechanized rapidly in the 1970s and probably less
than half of all cotton is now harvested manually.
The price of labor as measured by the rural minimum wage which co-
rrelates closely with daily wages for farm work has increased in real
terms over the period. From 1971 to 1981 the rural wage deflated by the
consumer price index increased 16 percent. However, wage increases did
not keep pace with high rates of inflation in 1982 and 1983.
Well-developed land markets, especially land rental markets exist
in both areas. Although ejido farmers are not legally permitted to rent
their land, many do. Many private farmers also rent land. The price of
land, as measured by its rental value in Sonora is mainly a reflection
of the water rights associated with the land. Land rental values for
soyabeans are usually similar to the price of open market water
purchases. This reflects the fact that for second crops, water rather
than land is the limiting factor.
Rental rates have generally risen faster than the rate of inflation
or the price of wheat. Land rental for wheat was reported at $400/ha in
1970 (Hewitt de Alcantara, 1978) and by 1981 had reached $4000/ha, a 70
percent increase in real terms. This reflects the fact that improved
technology and declining real costs of many inputs have been capitalized
into land values. With high rates of inflation and rapid increases in
costs, land rental values fell in real terms in 1982 and 1983.
In Tlaxcala, land rental values are about two-third of values in
Sonora. This is despite the fact that Sonoran yields are well over
double those of Tlaxcala. This is again a reflection of differential
cost structures as well as "site value" in Tlaxcala which is closer to
large urban populations.
4.2.11 Research and Extension
Farmers in both areas are beneficiaries of the research service.
For example, most crop varieties grown are the product of the research
service. In Sonora, part of the cost of the research is paid by the
farmers through contributions to the budget of CIANO, the official
research institute. In Tlaxcala, development of improved barley
varieties were financed in part from contributions from breweries. These
research costs could be quantified and included in the calculations of
resource cost ratios but because they are a negligible part of total
costs we have ignored them.
Extension in both areas largely works through the official credit
bank. Farmers do not usually use the extension service as a source of
information. Because of this we have not costed extension services
except in Sonora where many farmers employ the services of private
4.3 Long Distance Transportation
Both rail and road transport are used to ship grains. However, the
Mexican railway system has generally stagnated or declined so that road
transport has become more important. By the end of the decade road
transport was apparently more important in the shipment of grains from
Sonora and from ports.
Railway transport remains important because of cheaper rates. This
results from a large subsidy which amounted to 50 percent of operating
costs in 1981. Road transport benefits from the diesel subsidy and it
also seems that prices for trucks have been kept somewhat below world
Transport charges in this study are based on actual road transport
charges with an appropriate adjustment for subsidized diesel prices. Fuel
consumption was estimated at 30km-ton/lt for Sonora to Mexico City and
20km-ton/lt for Veracruz to Mexico City where much of the route climbs
through the mountains, and where transport rates per km-ton are sub-
stantially higher. The adjusted transport rates using these assumptions
are given in Table 4.6. Tlaxcala rates are assumed to be one-quarter of
the rate from Veracruz to Mexico City. The diesel subsidy provided an
overall subsidy rate of 25 percent on transport costs in 1982 which was
reduced to 20 percent in 1983, when fuel prices were increased sharply.
In 1983, the fuel subsidy combined with a uniform price for wheat and
fertilizer throughout the country benefited farmers in Sonora relative
to those in Tlaxcala, to the extent of about 10 percent of the value of
4.4 Nominal Protection Coefficients
The Nominal Protection Coefficient, NPC, compares prices received
by farmers with the equivalent world price for that commodity. This
requires an estimate of the appropriate world price, .the major con-
sumption points and appropriate transportation charges.
Table 4.6 Costs of Road Transport of Grain Adjusted for Subsidized
Diesel Price, 1982 and 1983
Veracruz to Sonora to
Mexico City Mexico City
June,1982 June,1983 June,1982 June,1983
Distance (km) 425 425 1800 1800
Diesel Consumption (It)a/ 21 21 60 60
Domestic Price of Diesel
(Pesos/It) 2.5 14.0 2.5 14.0
World Price of Diesel(Pesos/lt) 10.2 29.8 10.2 29.8
Actual Transport Costs
(Pesos/ton-km) 1.27 3.64 0.75 2.15
Cost (Pesos/ton-km) 1.67 4.43 1.00 2.68
Total Transport Cost-
Unsub. (Pesos/ton) 703 1884 1810 4821
Percent Subsidy 24 18 25 20
a/ Based on 20 km-ton/lt for Veracruz and 30 km-ton/lt for Sonora.
Source: Phone interviews with transport operators and conversations with
truck drivers at the local petrol station.
All crops analyzed in the study are import substitution crops for
most of the period under consideration, with the exception of cotton.
For cotton lint, we have assumed that the price received by farmers is
a true reflection of world prices converted at the official exchange
rate- The export tax on cotton is so small that it can be ignored.
For simplicity, we have not considered the value of cotton seed. This
value is small in relation to the value of lint and is usually
equivalent to the ginning costs. Hence, the NPC of cotton can safely be
assumed to be one -- i.e. farmers receive the equivalent of world prices
less a marketing margin.
-/ Cotton is usually exported from Sonora through the nearby port of
Guaymas. Transport and distribution charges are small and we ignore
For other crops we required a CIF import price. These prices are
not readily available in Mexico. Most commodities are shipped on a CAF
basis that is without insurance and other charges. Both land and sea
routes are used but shipment to ports in Veracruz and then overland
transportation to major population centers in the center of the country
is the most common mode of importation.
To represent the CIF price we elected to use the CIF price in Rot-
terdam. In 1983, freight rates from New Orleans, USA to Veracruz, Mexico
were similar or a little higher than freight rates to Rotterdam and
other costs of insurance and capital are not expected to vary much.
Nearly all of Mexico's wheat imports have been No. 2 Hard Red
Winter wheat although this policy has changed in recent years with
imports of lower quality (and cheaper) Canadian and Australian wheat. We
have used Mexico City as the consumption point for wheat. It is
estimated that Mexico City alone consumes about 1 million tons of wheat.
Available statistics from the mid 1970s indicate that about two-thirds
of all wheat shipped by rail from Sonora was destined to Central Mexico
(i.e. Mexico City, Puebla and the State of Mexico)/ Hence, the
farmgate price for wheat based on world prices is equal to the CIF price
of wheat plus transport charges from Veracruz to Mexico City less
transport charges to bring domestically produced wheat from Sonora or
Tlaxcala to Mexico City. Transport charges were based on unsubsidized
diesel prices as discussed in Section 4.3.
Relevant CIF prices and consuming points for other crops are given
in Table 4.7. Maize is assumed to be consumed locally while two extreme
assumptions are made for soyabeans--local consumption and shipment to
Mexico City. In the case of safflower only the oil is assumed to be
shipped to Mexico City and the cake is consumed locally in the animal
feed industry. For local consumption in Sonora, no internal
transportation charges are added and it is assumed that imports can be
landed at Sonora ports at the same CIF price as for Veracruz.
1/ Wheat produced in Tlaxcala is normally shipped to Mexico City,
because of the freight advantage.
Table 4.7 Import Prices, Consumption Points and Formulas Used to
Calculate Nominal Protection Coefficients in Sonora.
Consumption Formula for
Commodity Import Price Point Calculating NPC-/
1.Wheat a) CIF Rotterdam Mexico City
b) FOB Gulf ports Pf/(Pi+T -T )
No.2 HRW + freight
rate Gulf ports
2.Safflower Estimated as in Oil-Mexico City Pf/(P.+T -0.34T )
3.Maize CIF Rotterdam-
No. 2 Yellow Local Pf/P
4.Soya beans CIF Rotterdam a) Local a) Pf/Pi
b) Mexico City b) Pf/(Pi+T -T )
-- Pf = farm gate price
P. = import price
T = transport cost-Veracruz to Mexico City
T = transport cost-Sonora to Mexico City
No reliable international price information for safflower exists
since world trade in safflower is negligible. An equivalent border price
was constructed based on the price of competing vegetable oils (i.e.
sunflower seed oil) and the price of soya cake, adjusted to the lower
protein content of safflower cake. Calculations are shown in Appendix C.
Calculated Nominal Protection Coefficients (NPCs) are shown in
Figures 4.7 and 4.8. All crops show a similar general pattern over the
decade. At the beginning of the decade NPCs were close to one or above
for most crops. Although most domestic prices were raised in the period
of high world prices between 1973 and 1975 these increases were less
than world prices so that NPCs fell. NPCs rose in 1975 and 1976 as world
prices dropped and Mexico devalued its currency. Although world prices
were generally high in 1979 and 1980, NPCs rose sharply in this period
as a result of both the government efforts to stimulate basic food crop
production and also because the Mexican peso was increasingly
overvalued. The 1982 devaluations led to a sharp drop in NPCs of all
Among crops, wheat is the least protected. In fact, Mexican wheat
prices only exceeded world prices in the early 1970s when world prices
were very low (and US exports subsidized) and again in 1981. During much
of the period farmers in Sonora received only 80 percent of the world
price equivalent while the difference is even larger in Tlaxcala which
has a greater transportation advantage. However, there was a clear
effort on the part of government policy makers to set wheat prices
following trends in world prices.
In the irrigated areas all other crops generally had NPCs above
one, especially in the period 1979 to 1982. The higher level of
protection for maize compared to wheat reflects the fact that maize is
significantly cheaper than wheat in world markets even though the price
paid to farmers in Mexico is usually above wheat prices. There is a case
for valuing Mexican maize which is nearly all white maize, at a higher
price than imported yellow maize because of a strong consumer preference
for white maize- Hence, the levels of protection for maize may be an
The oilseeds, safflower and soyabeans, generally enjoyed the
highest level of protection. Even in 1982 when Mexican prices were well
below world prices for other crops, safflower and soya bean prices were
comparable to world prices. Over the period these crops have received an
average level of protection of 6 percent in the case of safflower and 14
percent for soya beans.
In the rainfed area, wheat prices were an average of 36 percent
below world prices over the last ten years. This reflects low
1/ Little white maize is traded in world markets and although it
usually sells for a premium above yellow maize, it is difficult to
obtain reliable price information on white maize.
Figure 4.7 Nominal Protection Coefficients, Sonora 1970-85.
Figure 4.8 Nominal Protection Coefficients, Tlaxcala, 1970-85.
transportation charges to consuming centers as well as the fact that
farmers in this region received only about 90 percent of the guaranteed
price because of a poorly developed marketing system for wheat. NPCs for
barley are generally similar or slightly higher than for wheat. Finally
NPCs for maize were substantially higher and often above one.
Corrected NPCs were also calculated to allow for the fact that the
Mexican peso was significantly overvalued in some periods. The corrected
NPC for wheat, as shown in Figure 4.9, was generally well below the
unadjusted NPC. The overall implicit tax on wheat was 30-40 percent for
much of the period.
4.5 Effective Protection Coefficient (EPCs)
Government policy often tries to compensate producers for low farm
gate prices by subsidies on inputs. The Effective Protection Coefficient
takes into account the difference in domestic and world prices for both
outputs and inputs. EPCs have been calculated for each crop and region
based on technical coefficients (e.g. units of Nitrogen per ton of
wheat) prevailing in recent years. The assumption of fixed technical
coefficients should not be a major problem in Sonora where these
technical coefficients have not changed much in the 1970s. However, in
Tlaxcala the rate of technological change has been more rapid.
Figure 4.10 compares NPCs and EPCs for crops in Sonora. Because
Mexican fertilizer prices were higher than world prices in the early
1970s, the EPC was less than the NPC. Thereafter they were similar until
1980 when subsidies on fertilizer and fuel reached such high levels
that the EPC was significantly above the NPC. Nonetheless, in the case
of wheat, the EPC was significantly above one in only two years, 1981
and 1982, of the last ten years. In the case of oil seeds and maize, the
EPCs are much higher than NPCs in most years, reflecting higher levels
of subsidy per unit of output valued at world prices. These crops
received substantial levels of protection throughout most of the period.
In 1981, farmers effectively received double the value added measured at
world prices for oil seeds.
Figure 4.9 NPCs for Wheat in Sonora Using the Corrected Exchange Rate,
NPC Bad on Actuo
1970 72 74 76 78 80 82 84
Nominal and Effective Protection Coefficient for Crops in
Figure 4.10. (Cont.)
N-OOK vo- 0
L -, -- -- -- -- -- -- -- -- ,-- -- -- -- -- --
Figure 4.11 Nominal and Effective Protection Coefficients for Crops in
In Tlaxcala, the EPC for wheat was always below the NPC except in
1981 and 1982. That is, subsidies have not generally provided compen-
sation for the low producer prices received by farmers. The major ex-
ception was in 1981 when the EPC reached 1.4 as a result of both the
high producer price relative to world prices (in part because of an
overvalued exchange rate) and also because of the special incentives
provided under SAM.
4.6 Subsidy Coefficients
Total subsidies were calculated for selected years and are shown in
Tables 4.8 and 4.9. Subsidies for seed (rainfed wheat), fertilizer and
fuel had already been calculated in the analysis of EPCs. Subsidies on
water and credit were calculated using the following rough and probably
conservative guides. Water was assumed to be subsidized by 50 percent in
1977 and 67 percent in the period 1981-85. Credit was assumed to be sub-
sidized by 30 percent in 1975 and 1977 and 50 percent in 1981-83 for
irrigated wheat and 67 percent in 1981 and 75 percent in 1982 for rain-
fed wheat where special low interest rates prevailed. Subsidies on
credit in the latest year were calculated as the difference between the
rate of inflation plus the assumed real interest rate (7.5% per annum)
and the actual rate of interest paid by farmers on short-term
The pattern of the results was similar in both Sonora and Tlaxcala.
When domestic wheat prices were well below world prices, subsidies on
inputs, water and credit did not compensate farmers for the low wheat
price. However, when domestic wheat prices were similar to world prices,
subsidies provided a substantial transfer to farmers. The effective
subsidy level reached 38 percent in both Tlaxcala and Sonora in 1981.
However, 1981 and 1982 were years of unusually high incentives for
Mexican farmers. In general, irrigated farmers have received higher
levels of subsidies than rainfed farmers for wheat production.
Table 4.8 Subsidies on Wheat Production, Tlaxcala, 1975-1984.
Producer-----------------Subsidy Due to----------------- Subsidy
Price Above Crop Coeffi-
Year World Price Seed Fertilizer Fuel Credit Insurance cient a/
( Pesos/ton wheat )
1975 -703 0 194 49 35 0 -19%
1981 496 382 235 385 239 211 38%
1982 -4760 450 886 1112 825 317 -8%
1984 -9500 0 1530 1030 4370 2800 0%
STotal subsidy divided by world price equivalent for wheat.
Table 4.9 Subsidies on Wheat Production, Sonora, 1977-1985.
Producer---------------Subsidy Due to-------------
Year World Price Fertilizer Fuel Credit Water
( Pesos/ton wheat )
Total subsidy divided by world price equivalent for wheat.
5.0 Farm Budgets and the Calculation of RCRs
5.1 Farm Budgets for Tlaxcala
Enterprise budgets were constructed for each major crop as a mea-
sure of farmer profitability and also as a basis for calculating re-
source cost ratios. Enterprise budgets are shown in Tables 5.1 to 5.2
for Tlaxcala in 1984. Technical parameters were employed as described in
Table 5.1. These parameters were unchanged from year to year. Price data
(Table 5.2) and a simple model to estimate machinery costs (Appendix B)
enabled us to then construct Table 5.3.
Table 5.1 Technical Parameters for Farm Enterprises in Tlaxcala
Wheat Barley Maize
Mechanical Operations (No/ha)
Plough 1 1 1
Harrow 2 2 1
Cover Seed 1 1 0
Harvest 1 1 0
Animal Powered Operations (d/ha)-
Furrow 0 0 1
Plant 0 0 1
Cultivate 0 0 3
Planting .5 .5 0
Weeding 0 0 3
Fertilizing-Chemical .5 .5 1
-Organic 0 0 4
Applying herbicide .5 .5 0
Harvesting 0 0 9
Seed 120 120 30
Fertilizer-Nitrogen 70 70 50
-Phosphorous 30 30 30
Herbicide-2,4-D (It/ha) 1 1 0
Grain Yield 2.0 1.85 1.5
Straw Yield 3.6 3.6 2.7
Table 5.2 Farmer Prices and World Price Equivalent of Inputs and
Outputs, Tlaxcala, 1984.
Farmer Price Equivalent
Urea (pesos/kg) 18.6 36
Triple Superphosphate (pesos/kg) 21.9 28.2
Esteron 47 (pesos/lt) 887 887
Seed Wheat (pesos/kg) 42 42
Barley (pesos/kg) 42 42
Maize (pesos/kg) 41 41
Animal Power (pesos/day) 1120 1120
Labour (pesos/day) 560 560
Bank Interest (%/year) 36 67.5
Tractor ('000 pesos) 3750 3750
Combine ('000 pesos) 20600 18730
Diesel (pesos/lt) 26 42
Estimated Total Tractor
Cost (pesos/hour)a/ 1210 1420
Estimated Total Combine
Cost (pesos/hour) 11330 10700
Wheat (pesos/kg) 27.3 36.8
Barley (pesos/kg) 33.5 34.3
Maize (pesos/kg) 33.4 31.1
Barley Straw (pesos/kg)-/ 0.8 0.8
Maize Straw (pesos/kg) 1.0 1.0
a/ See Appendix B for method of calculating costs.
Wheat straw value is negligible when baling costs are substracted.
- Exchange rate US$1.00 = 192 pesos.
Table 5.3 Budgets for Wheat, Barley and
Prices Paid by Farmers, 1984
Maize in Tlaxcala Using Average
Wheat Barley Maize
Machinery 22080 20800 6560
Labour 2060 2060 9890
Animal Power 0 0 5590
Seed 5040 5040 1230
Fertilizer 4260 4260 3450
Herbicide 890 890 0
Insurance 4100 9700 3200
Bank Interest 8280 8280 6450
Total Variable Costs 47110 52710 36760
Gross Revenues 54600 64760 52800
Gross Margin a/ 7490 12050 16040
Land Rent 14000 14000 14000
Net Profit b/ -6510 -1950 2040
Net Return on Capital c/ .03 .11 .19
- Gross revenue less total variable costs.
SGross margin less land rental charge.
c- (Gross revenue/(total variable costs + land interest charges)) 1.
The budget given does not conform to traditional farm management
budgets in at least two respects. First, we have estimated machinery
costs as a charge to machinery services. The estimated charge is very
close to actual machinery costs. The advantage of estimating the charge
is that we are able to disaggregate these costs into depreciation, labor
and fuel, etc. for calculating resource cost ratios. Second, interest
charges are usually not subtracted out in calculating gross margins.
However, since most farmers use bank credit to purchase fertilizer, fuel
and other inputs it is reasonable to subtract interest charges so that
gross margins represent the farmers return to their own resources of
capital, land and management.
Table 5.4 Gross Margins and Returns on Capital for Wheat, Barley and
Maize in Selected Years.
Wheat Barley Maize
Real Gross Margins
1975 4643 4774 3605
1981 6841 4416 3346
1982 4571 3162 664
1984 2235 3596 4784
Real Return on Capital/ (% per crop season)
1975 na na na
1981 59 15 1
1982 6 24 72
1984 57 49 41
a/ Rate of return less inflation rate
na = not calculated because the price of land was not available.
Data on gross margins, converted to 1982 prices by the consumer
price index, and the real return on capital deflated by the inflation
rate are shown for various years in Table 5.4. In the early years, maize
was the least profitable crop. This resulted from both higher costs and
lower yields per hectare than for the small grains. Even the special
incentives of SAM failed to arrest the decline in incomes from maize
production. Maize continued to be cultivated for subsistence purposes
and even this seems to be declining as small farmers increasingly depend
on the market for food staples. This trend has been accelerated in
recent years by the high subsidy on tortillas and bread, which the rural
population close to larger towns is increasingly purchasing.
In 1975, barley prices were higher than wheat which resulted in
slightly higher incomes from barley. By 1980, wheat had become more
profitable than barley and this was emphasized in 1981 and 1982 when
wheat received special incentives under SAM.
The sharp increase in costs resulted in a real decline in incomes
in 1982 and 1984 for wheat and barley. In particular, machinery costs,
increased rapidly in this period to reach 46 percent of total variable
costs in wheat in 1984 compared to only 38 percent in 1975. Fuel costs
alone were less than 10 percent of tractor hire costs in 1981 but had
increased to over 30 percent of the cost in 1984. This sharp increase in
machinery costs had less effect on maize profitability and in fact,
maize was more profitable than wheat for the first time in 1984.
5.2 Resource Cost Ratios in Tlaxcala
Table 5.5 shows values of inputs and outputs divided into tradeable
and nontradeable items. Tradeable inputs were valued at their world
price equivalent. These included machinery depreciation, fuel and spare
parts, which were valued at their US price equivalent with a 15 percent
adjustment downward for diesel as before. Maintenance and repair costs
were divided into 75 percent for spare parts (a tradeable) and 25
percent for labor (a nontradeable). Seed prices were assumed to reflect
their real cost. Fertilizer was valued at FOB prices plus internal
transport charges. Herbicide was valued at US farm prices.
All nontradeable inputs are valued at actual farm prices, except
land and capital. A real cost of capital was assumed to be 7.5 percent
per year. This seems to be close to real returns to capital of machinery
owners in the area and would also correspond to average inflation and
risk adjusted returns on capital in barley production the major crop.
Land values were computed as a residual return in the best competing
alternative, since land is assumed to be the major limiting resource in
the area. For example, to calculate the opportunity cost of land in
wheat, a residual value to land in its best alternative, barley, was
calculated. The residual return to land in wheat production was 8,500
pesos/ha, which was then employed as the opportunity cost of land in
maize and barley production.
Outputs were also divided into tradeable and nontradeable compo-
nents. The grain produced was valued at its import price adjusted for
transport costs. Transport costs were divided into fuel, a tradeable,
and nonfuel items which were regarded as nontradeable. Domestic
production of grain produces a benefit in savings of transportation
Table 5.5 Calculation of Resource Cost Ratios in Tlaxcala, 1984.
Wheat Barley Maize
(Pesos/ha) (Pesos/ha) (Pesos/ha)
Machinery depreciation 7020 7020 1970
Fuel 4940 4940 2410
Spare parts 5570 5570 1660
Other inputs 12950 12950 6980
Machinery maintenance (labor) 1860 1860 550
Labor 2060 2060 9890
Animal Power 0 0 5590
Capital 20760 20760 14420
Insurance 9700 9700 9700
Land-Opportunity Cost a1070 8330 8330
Grain (CIF) 69040 60230 43230
Fuel for transport 1080 1000 810
Straw 0 2880 2700
Non-fuel transport costs b/ 3490 3230 2610
National Profitability 7270 -7270 -12550
Opportunity Cost Domestic Resources 32360 37010 43570
Value Added (Tradeables) 39624 29740 31020
Resource Cost Ratio 0.82 1.24 1.40
-- Residual return to land in best competing alternative valued at
world price equivalent, i.e. barley in the case of wheat and wheat
b in the case of barley and maize.
- Net transport costs to bring imported grain to Mexico City less cost
of transport from Tlaxcala. Fuel for transport is valued at world
Sum of nontradeable inputs less nontradeable outputs.
costs of imported grain to the consuming point. Straw was treated as a
nontradeable and valued at its actual farm price.
The'result of all these calculations for Tlaxcala produce no sur-
prises given our earlier discussion of policy incentives and farmer
profitability. We have seen that despite favorable maize prices, farm
level profitability of maize is very low or negative. Relative to wheat,
yields are lower, world prices lower and costs higher; therefore the
resource cost ratio has to be above one. Likewise for barley, relative
to wheat, world prices are usually slightly lower, yields slightly lower
and costs the same. Hence, wheat has a resource cost ratio slightly
below one and barley slightly above one. The only exception was in 1975
(Table 5.6) when world prices for malting quality barley were higher
than for wheat.
Table 5.6 Resource Cost Ratios for Wheat, Barley and Maize, Tlaxcala
Year Wheat Barley Maize
1975 1.12 .85 2.20
1981 .88 1.07 1.90
1982 .83 1.06 2.32
1984 .82 1.24 1.40
Based on the above analysis of comparative advantage, there is a
strong case for promotion of research and extension in wheat production
in Tlaxcala/Hidalgo. There is a comparative advantage in production of
wheat relative to barley and its development would be consistent with
the government's food security objective. The major obstacles that need
to be overcome are a) a price incentive for wheat that reflects its cost
of importation relative to barley, b) a marketing system that effec-
tively transmits these prices to farmers and c) a variety that reduces
risks relative to barley, especially one with early maturity and disease
Considerable productivity gains need to be achieved for maize in
the Tlaxcala region to become competitive, from the national pers-
pective. With no change in fertilizer application, maize yields would
have to increase by around 0.5 ton/ha to become competitive with wheat,
an increase of over 30 percent. If we allow for a higher fertilizer
dosage of 100-50-0 of NPK, an average yield of maize of 2.4 ton/ha is
needed to compete with wheat. There are prospects, however of reducing
costs of maize production through reduced tillage and chemical weed
5.3 Farm Budgets and Returns for Sonora
A similar methodology was used to calculate returns in the case of
Sonora. The situation is of course more complicated because of a larger
number of crops and crop combinations. The number of operations
performed on each crop is also larger. Some simplifying assumptions had
to be made. For example, we did not attempt to calculate detailed costs
for a mechanical cotton harvester. Rather we assumed that these costs
bear the same relationship to each other as in the case of mechanical
wheat harvesting. Likewise many different insecticides are used. We
noted that one insecticide application for cotton and soyabeans usually
costs double that for wheat and safflower and used this relationship for
The technical parameters and prices used in Sonora are shown in
Table 5.7 and 5.8 for the 1984-85 cycle. This leads to budgets for
1984-85 shown in Table 5.9 and estimated gross margins and real returns
on capital for several years (Table 5.10).
Wheat consistently provided good returns to farmers. Gross margins
for wheat increased each year, even when adjusted for high levels of
inflation. (In 1982/83, land preparation, planting, and many inputs were
purchased before the sharp price rises of late 1982). Real returns on
capital including land rents in wheat production have been 20-35 percent
from 1981 to 1983. Gross margins and returns on capital were higher for
cotton in most years. However, profitability of cotton was subject to
considerable price risk. (Yield risk is also higher but that is not
considered here). In particular, cotton was unprofitable relative to a
Table 5.7 Mechanical Operations and Inputs Used in Constructing Farm
Budgets, Sonora, 1985.
Wheat Safflower Cotton Maize Soya
Mechanical Operations (number/ha)
Insecticide and Herbicide
Total Tractor Hours/ha
12.15 14.85 25.05 14.85 13.55
Total Combine Hours or Equiv./haa/ 1.25
Total Labour Days/hab/
Herbicide (No. of Applic.)
Insecticide (No. of Applic.)
21.6 17.3 15.6
3 / 375 2.0
3. da 3.74/ 2.0
a/ Conversion made on the basis of rental charges for harvesting of
b/ Includes tractor driver's labor, assuming 8 hours per day.
c/ Yield of seed cotton, assumed to be 33 percent lint.
-- Yield for August planting.
Note: Year 1985 refers to the 1984-85 crop cycle.
Table 5.8 Farmer Prices and World Price Equivalent of Inputs and
Outputs, Sonora, 1985.
Farmer Price Equivalent
Urea (pesos/kg) 27.5 67.7
Triple Superphosphate (pesos/kg) 32 56.1
2, 4-D Herbicide (pesos/lt) 2115 2115
Insecticide Wheat (pesos/application) 2380 2380
Insecticide Cotton (pesos/application) 4280 4280
Seed Wheat (pesos/kg) 62 62
Safflower (pesos/kg) 103 103
Cotton (pesos/kg) 156 156
Maize (pesos/g) 85 85
Soyabean (pesos/kg) 120 120
Tractor ('00 pesos) 3800 .3800
Combine ('000 pesos) 25575 23250
Diesel (pesos/lt) 32 46.8
Bank Interest (%/year) 36 67.5
Labour (pesos/day) 975 975
Water (pesos/ha) (for wheat) 5720 17160
Wheat (pesos/kg) 37 37.6
Safflower (pesos/kg) 63 36.3
Cotton (pesos/kg)a/ 343.5 343.5
Maize (pesos/kg) 37 36.3
Soyabean (pesos/kg) 82.7 64.6
- Net price of lint, with cotton seed price offsetting cost of
Note: Exchange rate US$1.00 = 275 pesos.
Year 1985 refers to the 1984-85 crop cycle.
Farm Budgets for Wheat, Safflower, Cotton, Maize and Soya,
Wheat Safflower Cotton Maize Soya
Cost of capital
Spare parts and maintenance
Herbicide and Insecticide
5460 18040 15020 13550
Subsidy on Inputs
Transport to Market
7500 19000 7870 9690
Total Variable Cost
Return on Capital (%)
28000 28000 50400 28000 28000
88370 78440 197380 82270 101380
175750 126000 343500 138750 165400
Note: Year 1985 refers to the 1984-85 crop cycle.
Table 5.10 Farmers Returns by Crop, Sonora, 1977 to 1985.
Year Wheat Safflower Cotton Maize Soya
Real Gross Margins (1981 Pesos/ha)
1977 9445 5121 36474 3698 10363
1981 9725 4440 12354 4326 8047
1982 11667 6759 34712 4976 8316
1983 12301 6421 27959 2545 7511
1985 14224 7743 23790 9190 10420
Real Returns on Capital
(percent/cycle adjusted for inflation)
1977 na na na na na
1981 22 -12 3 -13 13
1982 34 1 78 -21 10
1983 32 -17 35 -58 0
1985 15 -15 3 -9 -7
na = not calculated because no rental value for land was available.
Note: Year 1985 refers to the 1984-85 crop cycle.
wheat-soya bean rotation in 1981 and 1985 when cotton prices were low
because of the overvalued peso. In other years, wheat-soya beans
returned less than cotton.
Soyabeans were also a profitable crop although less so than wheat
and cotton. Of course soyabeans are a short season crop so that gross
margins or returns on capital on a monthly basis are more favorable-
Safflower and maize were generally unprofitable crops. Gross
margins for safflower were only about half of those for wheat and
returns on capital were negative. The area sown to safflower usually
reflects problems in wheat such as missing the wheat planting date,
-/ Our results for soya beans for 1982 and 1983 are overestimates
since we have used the same input prices as for wheat even though
six months of rapid inflation resulted in substantially higher
costs for soya beans.
severe weed problems or water problems. Safflower yields were also
higher on alluvial soils providing a stronger position to compete with
wheat. Maize is generally a minor crop. The incentives of SAM in the
early 1980s led to some improvement in returns but this has been eroded
more recently. Maize is usually harvested from January to March and sold
before the new guaranteed price is fixed.
5.4 Resource Cost Ratios in Sonora
The methodology used to calculate resource cost ratios was adjusted
to Sonora to take account of the fact that water, not land, is often the
limiting factor. In the absence of reliable data, we assumed that water
costs were subsidized by 67 percent which is probably a conservative
assumption. We then calculated returns to land as a residual to
represent the case where land is limiting. Here we assumed that wheat,
safflower, cotton and winter maize competed for land. Likewise, summer
maize and soyabeans compete for the same land. Hence, in evaluating the
best alternative for land we considered two different seasons of the
year. In the case in which water is limiting we allowed all crops to
compete. Cotton was assumed to require 50 percent more water than wheat,
while summer maize and soyabeans requirements were 25 percent above
wheat. Safflower was assumed to require 20 percent less water and maize
(winter) was assumed to use the same amount of water as wheat.
The calculation of resource cost ratios is given in Table 5.11 and
results for five years are provided in Table 5.12. An important result
is that safflower, soyabeans and maize provided negative national
returns to land and water in 1981, 1982 and 1985. That is, if the land
and water is assumed to have no value, (and no cost in the case of
water) the value of output of these crops (at world price equivalent)
was not sufficient to cover the cost of tradeable inputs (at world price
equivalent) and labor and capital used in their production. This is in
spite of the fact that we have used the favorable assumption that
products of these crops (except safflower oil) are consumed locally and
therefore only incur local transport costs. Returns were positive for
safflower and soyabeans in 1977, which was a year of unusually high oil
Table 5.11 Calculation of Resource Cost Ratios, Sonora, 1985.
Wheat Safflower Cotton Maize Soya
Tradeables (grain, etc.)
343500 136130 129200
81060 -8860 -12040
Resource Cost Ratio
Resource Cost Ratio
' Charge for water with land the limiting resource
nc = not calculated because returns on land were negative
Table 5.12 Resource Cost Ratios for Various Crops, Sonora, 1977-1985.
Wheat Safflower Cotton Maize Soya
1977 2.5 2.7 .5 4.4 1.3
1981 1.0 -ve 1.0 2.1 -ve
1982 .7 -ve 1.2 -ve -ve
1983 1.3 4.4 .8 2.3 1.0
1985 1.3 2.7 .8 -ve -ve
1977 1.9 .8 .6 1.8 1.6
1981 .7 -ve 1.2 -ve -ve
1982 .7 -ve 1.4 -ve -ve
1983 1.0 2.1 1.0 1.6 2.8
1985 1.1 1.7 .9 1.8 2.5
seed prices and unusually low wheat prices. Returns were also positive
in 1983 when the value of traded items increased much more rapidly than
that of non-traded items because of the sharp devaluation. However, if
water is limiting, none of these crops compete with wheat or cotton.
Wheat and cotton clearly have the comparative advantage in Sonora,
whether land or water is assumed to be limiting. If land is limiting,
cotton has the advantage in three years and wheat the advantage in one.
If water is limiting, wheat has the advantage in two years and
wheat and cotton are equal in 1983. These years represent world prices
for cotton lint ranging from 6.8 to over 10 times the CIF price of
wheat. World cotton/wheat prices generally have stayed in this range
except in the unusual period of 1974-76. Hence, we can conclude that
wheat competes strongly with cotton in most years. These results
contrast with the high resource cost ratios calculated for wheat
relative to cotton in Egypt and Pakistan. The high yield levels for
wheat in Sonora distinguished the Mexican case.
The question arises as to what extent our negative results on oil
seeds are sensitive to assumptions about yields and prices. Sensitivity