The economic feasibility of stabilizing the price and supply of potatoes in Ecuador

Material Information

The economic feasibility of stabilizing the price and supply of potatoes in Ecuador
Wiegand, Kenneth Bruce, 1942-
Place of Publication:
Gainesville FL
University of Florida
Publication Date:
Physical Description:
xiii, 171 leaves : ill. ; 28 cm.


Subjects / Keywords:
Potatoes -- Prices -- Ecuador ( lcsh )
Food and Resource Economics thesis M.S
Dissertations, Academic -- Food and Resource Economics -- UF
bibliography ( marcgt )
non-fiction ( marcgt )


Thesis (M.S. in Agr.)--University of Florida.
Bibliography: leaves 165-170.
General Note:
General Note:
Electronic resources created as part of a prototype UF Institutional Repository and Faculty Papers project by the University of Florida.
Statement of Responsibility:
by Kenneth Bruce Wiegand.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Kenneth Bruce Wiegand. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Resource Identifier:
000355577 ( ALEPH )
02280806 ( OCLC )
ABZ3818 ( NOTIS )

Full Text


I am deeply appreciative of the time, patience and concern which Dr. Chris 0. Andrew has so generously invested in my training and the preparation of this thesis. It has truly been a learning process, primarily due to his unwaivering personal and professional attitude towards graduate education. Firmly, but with great tact and convincing insight, he guided my first uncertain steps in the preparation of the project proposal, and later dedicated long hours to review and discussion of each stage and chapter in the development of the final product.
I am also indebted to Dr. Kamal Dow and Dr. Kary Mathis for their many suggestions and revealing questions. The potentially difficult task of researching a project in Ecuador was substantially simplified by Dr. Dow's presence as Chief of Mission for the University of Florida program in Quito. Also, the frequent corresdpondence and timely backup support of Dr. Mathis, in addition to his patient review of several manuscripts, all helped to make the experience in Ecuador and Gainesville an enjoyable and productive venture.
Many individuals in Ecuador and Gainesville have helped to shape the thesis and make its content more germaine and easily understood. For their assistance, advice, and.moral comfort, I am happy to acknowledge here the contributions of Sheriar Irani, Luis Maldonado, Carolyn Almeter, Michael Schwartz, Jim and Diane Frito, Luis Cruz, the personnel

of INIAP, Guido Toral in Cuenca, Dr. Francisco Citelly, Bharat Jhunjhunwala, Dr. Hector Becerra, and Michael Twoomey at the International Potato Center in Lima.
I am grateful for the financial support received from the Department of Food and Resource Economics and from the National Institute of Agricultural and Livestock Research in Quito. The data collection stage of the research project was greatly assisted by the support of Lic. Humberto Mejia A. and family of Quito, who gave generously of. their hearth and home, thus providing our family with all of the necessary creative comforts for an extended overseas venture.
I would like to express my special gratitude for the hard work,
dedication, and invariably sunny disposition of the typists who have been burdened with the task of converting these handwritten words to a neat and legible presentation. Beth Zirnpfer, Wilma Galanos, Dora Benavides, and Alicia Ruales were extremely helpful at different stages of the project, and Sandy Wiedegreen was given the final task of putting it all together in its present state.
I would also like to acknowledge the inspiration and constant
faith inspired by my parents, who by their daily example, determined my attitude towards life and the goals therein worthy of achievement.
Finally, I would like to express my gratitude to my wife, Teresa,
for her numerous suggestions, privileged insight, and unfailing patience throughout this period, and to my son, Kevin, who made it all worthwhile.

The Development Process 1
The Setting 2
The Problem 4
The Hypothesis 12
The Objectives 12
Organization of the Thesis 13
Theoretical Considerations 14
Price Variability ,14 Benefits of less variable prices 14
Regional price variation 15
Variable supply and the inelasticity of demand 17
Supply stabilization 18
Storage 18
Concept of storage 18
Maximum allowable investment in storage 19
Production Adjustments 19
Empirical Considerations 20
Measuring Potato Price Variability 20
Storage 22
Length of Storage Period 23
Storage Costs 23
Actual Price Storage Analysis 24
Regional Storage and Transport 24
Maximum Allowable Investment in. Storage 25

Production Adjustment 27
Potato Growers Survey 27
Sample Size 28
Production Data 29
Marketing and Cost of Production Data 30
Agricultural Production in Ecuador 32
Potato Production 33
Preferred Month for Planting 37
Movement of Production to Market 41
Price Variation 48
Among Highland Crops--Risk 48
Within and Between Market Cities 49
Forms of Storage and Costs 51
National Storage Potential 54
Provincial and Regional Storage 60
Provincial Potentials 63
Regional Potentials 66
National Reallocation: Storage and Transport 72
Summary of Storage Potential 76
Production Management Potential 77
Preferred Month for Planting 78
Number of Months Required for Crop Maturation 80
Premature Harvest: Frequency and Reasons 80
Postmature Harvest: Frequency and Reasons 82
Role of Credit 83
Estimated Required Adjustments in Hectarage 84
Comparison of Supply Stabilization Techniques 87
The Costs and Benefits of a Storage Program 91
The Costs and Benefits of Production Management 93
The Least Cost Combination of Alternatives 95
Recommendations 96

Table Page
1 Average cyclical wholesale price variation
for potatoes in Ecuador and selected countries 6
2 Apparent cost of potatoes as a percent of food
expenditure in selected countries 8
3 Apparent losses in potato production during low
price cycles, Ecuador, 1965-71 10
4 Potato marketing losses in the Pichincha
Province, Ecuador,. 1973, per hundred weight of
potatoes 11
5 Preferred month for planting potatoes in the
Northern, Central, and Southern Highlands of
Ecuador by percentage of farmers interviewed, 1974 38
6 Average yearly potato production, area harvested,
and yield by region, Ecuador, for periods 1962-67
and 1968-73 39
7 Annual variability of value per hectare, yield,
and wholesale price of four alternative highland
crops, Ecuador, 1962-72 50
8 Correlation (r) matrix for monthly wholesale price
of potatoes, selected cities, Ecuador, 1962-72 52
9 Average monthly wholesale price differential
between subsequent months for potatoes in Ecuador,
1954-72, in constant 1970 sucres per quintal 55
10 Average price differential in sucres per quintal
between low- and high-price periods for potatoes,
Ecuador, 1962-71, and profit margin after storage 62
11 Estimated maximum allowable investment in potato
storage by province in the Central Region,
Ecuador, 1962-72, in sucres per quintal 64
12 Estimated regional costs and benefits for
potato storage, Ecuador, 1972,. in constant 1970
sucres 71

Table Page
13 Estimated margin for potatoes purchased and
stored in Tulcan and later transported to
selected major cities for sale to wholesalers,
Ecuador, 1962-72 74
14 Preferred months for planting potatoes by
region and size of planting, Ecuador, 1972 and 1974 79
15 Estimated average monthly surplus and deficit
potato production by region, Ecuador, 1974 85
16 Modifications required to stabilize potato
production in Central Region, Ecuador, 1972 86
A-i Production of potatoes by highland province and
region, Ecuador, 1962-1973, in thousands of
metric tons 102
A-2 Yield of potatoes by highland province and region,
Ecuador, 1962-1973, in quintales per hectare 103
A-3 Area in potatoes by highland province and region,
Ecuador, 1962-1973, in thousands of hectares 104
A-4 Monthly regional wholesale potato prices,
Ecuador, 1962-1969, 1971-1972, in sucres per
quintal 105
A-S Monthly regional wholesale corn prices, Ecuador,
1962-1969, 1971-1972, in sucres per quintal 108
A-6 Monthly regional wholesale wheat prices,
Ecuador, 1962-1969, in sucres per quintal i1
A-7 Monthly regional wholesale barley prices,
Ecuador, 1962-1969, in sucres per quintal 113
A-8 Constant monthly wholesale potato prices,
Ecuador, 1954-1972, in sucres per quintal 115
A-9 Average monthly wholesale price of potatoes
in selected cities, Ecuador, for period
1962-72, and in Pasto, Colombia, for 1955-64 116
A-10 Percent price differential between local and
national monthly wholesale potato prices in
selected cities, Ecuador, 1962-72 117
A-lI Average monthly potato arrivals to regional
markets, Ecuador, 1962-73 118

Table Page
A-12 Reaction of potato producers to price
fluctuations, by farm size, 1972 119
A-13 Annual variability of yield by province for
selected highland crops, Ecuador, 1961-72 120
A-14 Costs of production per hectare of potatoes
(mechanized), Ecuador, 1972 121
A-15 Average cost of production per hectare of
potatoes for medium size farm, Ecuador, 1972,
in the Central Highlands 123
A-16 Result demonstration cost of production per
hectare of potatoes, Hacienda El Carmen,
Pichincha Province, Ecuador, 1962 125
A-17 Index of prices to the consumer, potatoes,
Ecuador, 1974, in sucres per quintal 127
A-18 Prices and margins for potatoes from
Chimborazo Province, sold in Quito, Ecuador,
1962, in sucres per quintal 128
A-19 Consumer prices in Quito, 1951-74 (base: 1965
1 100 and 1970 = 100) 129
A-20 Percent monthly price variation for selected
crops and cities, Ecuador, 1962-72 130
A-21 Estimated storage costs for forced night air
potato silo in the Ecuadorian Highlands, 1972 131
A-22 Truck transportation costs between major cities,
Ecuador, 1974 132
A-23 Government credit for potato and general
agricultural production in Ecuador, 1951-70 133
A-24 Moving average of preferred month for planting
potatoes in the Northern, Central, and Southern
Highlands of Ecuador, by percentage of farmers
interviewed, 1974 134
A-25 Monthly quantity of potatoes that could have been
stored and consumed to stabilize supply in
Northern Highlands, Ecuador, 1972 135

Table Page
A-26 Monthly quantity of potatoes that could have been
stored and consumed to stabilize supply in Central
Highlands, Ecuador, 1972. 136
A-27 Monthly quantity of potatoes that could have been
stored and consumed to stabilize supply in
Southern Highlands, Ecuador, 1972 137
A-28 Estimated cost for storing potatoes in the
Northern Highlands, Ecuador, 1972 138
A-29 Estimated cost for storing potatoes in the
Central Highlands, Ecuador, 1972
A-30 Estimated cost for storing potatoes in the
Southern Highlands, Ecuador, 1972 140
A-31 Estimated additional revenue from stored
potatoes, Northern Highlands, Ecuador, 1972 141
A-32 Estimated additional -revenue from stored
potatoes, Central Highlands, Ecuador, 1972 142
A-33 Estimated additional revenue from stored
potatoes, Southern Highlands, Ecuador, 1972 143
A-34 National monthly quantity of potatoes that could
have been stored and consumed to stabilize supply,
Ecuador, 1972 144
A-35 Estimated cost for storing potatoes nationally
Ecuador, 1972 145
A-36 Modifications required to stabilize potato
production in the Northern and Southern Regions,
Ecuador, 1972 146
A-37 Reasons for selection of best month for planting
potatoes, by region, Ecuador, 1974 147
A-38 Reasons in Southern Region for selection of best
month for potato planting, by size of planting,
Ecuador, 1974 148
A-39 Reasons in Northern Reaion for selection of best
month for potato planting, by size of planting,
.Ecuador, 1974 149

Table Page
A-40 Months required for potatoes to mature, in the
Southern and Northern Sierra Regions, by
farm size, Ecuador, 1974 150
A-41 Frequency of premature harvest by potato farm
size and region, Ecuador, 1974 151
A-42 Number of'weeks premature harvest by total farm
size, Southern Region, Ecuador, 1974 152
A-43 Number of weeks premature harvest by total farm
size, Northern Region, Ecuador, 1974 153
A-44 Reason for premature potato harvest by number
of weeks harvested prematurely as a percentage
of farmers polled, Southern Region, Ecuador, 1974 154
A-45 Reason for premature potato harvest by number of
weeks harvested prematurely as a percentage of
farmers polled, Northern Region, Ecuador, 1974. 155
A-46 Reason for premature harvest by potato farm
size and region, Ecuador, 1974 156
A-47 Frequency of post mature potato harvest by
region and size of farm, Ecuador, 1974 157
A-48 Number of weeks postmature potato harvest by
region and size of farm, Southern Region,
Ecuador, 1974 158
A-49 Number of weeks postmature potato harvest by region
and size of farm, Northern Region, Ecuador, 1974 159
A-50 Reason for postmature potato harvest by number of
weeks harvested postmaturely as a percentage of
farmers polled, Southern Region, Ecuador, 1974 160
A-51 Reason for postmature potato harvest by number of
weeks harvested postmaturely as a percentage of
farmers polled, Northern Region, Ecuador, 1974 161
A-52 Distribution of credit among potato growers by
region and total farm size, Ecuador, 1974 162
A-53 Sample questionnaire for potato farmers,
Ecuador, 1974 163

Figure Page
1 Theoretical comparison of loss and gain of
consumer's surplus caused by fluctuating prices 16
2 Map of Ecuador and the study area 36
3 Rainy period in the Highlands Region,'Ecuador 42
4 National average monthly wholesale price and
estimated monthly market arrival for potatoes by
region in highland Ecuador, 1954-72 43
5 Market flow of potatoes in Ecuador, 1973 47
6 Potential short range storage based upon
significant monthly wholesale price variance
of potatoes in Ecuador for period 1954-72 57
7Estimated average profit margins for stored
potatoes, Ecuador, 1954-72 58.
8 Wholesale potato prices showing potential
storage periods, Ecuador, 1962-72 61
9 Estimated percent of annual potato production
required to stabilize monthly supply, Northern
Highlands, Ecuador, 1972 68
10 Estimated percent of annual potato production
required to stabilize monthly supply, Central
Highlands, Ecuador, 1972 69
11 Estimated percent of annual potato production
required to stabilize monthly supply, Southern
Highlands, Ecuador, 1972 70
*A-1 Average constant wholesale potato prices showing
potential storage periods, Ecuador, 1954-72 101

Abstract of Thesis Presented to the Graduate Council
of the University of Florida in Partial Fulfillment of the Requirements
for the Degree of Master of Science in Agriculture THE ECONOMIC FEASIBILITY OF STABILIZING THE PRICE
Kenneth Bruce Wiegand
June, 1975
Chairman: Chris 0. Andrew
Major Department: Food and Resource Economics
The objective was to measure the economic feasibility of stabilizing the price and/or supply of potatoes in Ecuador. It was estimated that the risk associated with potato production could be reduced to almost onehalf its current level through a stabilization program, and that the waste associated with distributing potatoes could be reduced by 16,000 tons annually if an adequately stable supply system were established. A significant degree of price variance was measured between the low prices of May, June, and July and the rising postharvest prices in the later months of September, October, and November. Storage costs associated with a low-cost, forced-night, air-storage facility designed in Colombia were used to analyze potential storage profits, based upon a maximum storage period of-four months.
Costs and benefits of a storage program at the national level using average deflated time series data revealed no significant storage potential. The limitations of requiring a stable monthly supply of potatoes and a four month maximum allowable storage period combined to produce an unprofitable storage scheme.

The potential for supply management through changes in planting and harvest dates (production smoothing) was explored by interviews with potato producers. Their attitudes towards year-long production and incentives for premature and postmature harvest were determined. The actual costs associated with production smoothing were indeterminable at the national level although a rough measure of their size was indicated by the potentially foregone benefits of not stabilizing production and potato prices. The potential for production smoothing at the.regional level relative to storage was calculated for a typical year in the Central Region. Production costs for planting "out of season" could be allowed to exceed the normal production costs by 2.4 times before they would equal the estimated regional storage costs. This apparent cost advantage of production smoothing is tempered, however, by the intangible consumer benefits of a storage program where, For example buffer stocks could be used to alleviate temporary shortages, thus reducing the net costs associated with storage.

The Development Process
In the agricultural development process much emphasis has been given to policies and programs designed to accelerate growth in production and productivity [27, 71]. Yet, little attention has been paid to the role of agricultural product markets, the primary mechanism for coordinating production and consumption activities in the development process [6, 67]. The greenn revolution" is a good example of this strategy. The high yielding varieties together with a new package of inputs have resulted in dramatic increases in production and productivity but only minimal changes in human welfare [52]. Consumers who wish to improve their die -ts are often unable to exert an effective demand due to limited purchasing power, high prices, and inadequate access to the products. While many countries have succeeded in achieving target levels of growth in agricultural production, they have failed in the real objective of develop-, ment, i.e., improving economic well-being for all the people [57, 26].
A pilot research program designed to examine the role of food
marketing improvements during Puerto Rico's early developmental period led Harrison and other Michigan State University researchers to the conceptualization of a broad framework for examining development planning and market strategy [68]. Of primary concern was the ultimate

effect of reducing food staple prices. It was hypothesized that since food expenditures usually make up more than half of total family living expenditures in urban areas of less developed countries, a reduction in food prices can have a significant effect on real family income. In such a situation, a 10 percent reduction in food prices could increase total purchasing power (real income) by.5 percent or more. The result would be an increase in "effective demand" for both food and nonfood products. This increased demand for food would then be reflected back through rural assembly markets to farmers, who would be stimulated to expand output [57].
The Setting
The agricultural production and marketing situation in Ecuador
differs little from the conditions described by McPherson and Johnston in their characterization of agriculture in the tropics [48]. The economic and institutional conditions have been listed often in the literature and they usually include the following:
a) Rapid rates of population growth,
b) Rapid migration to the cities,
c) High rates of unemployment and underemployment,
d) Duality of economic organization characterized by
commercial and tradttinnal agriculture, modern heavily
capitalized versus traditional shops and cottage industry,
modern commercial and service firms (supermarkets, data processing, etc.) versus traditional intermediaries and
service people,
e) High degree of inequality in distribution of wealth,
income and economic opportunity,
f) Prevalence of market imperfections and monopolistic

g) Extensive overlap among political and economic elites,
h) Antiquated tax laws, evasion of taxes, rigid fiscal
and monetary policies,
i) Factor and product price distortions [26].
The potato production-distribution system in Ecuador is characterized by a large number of independent producers, a lack of government price intervention,1 and innumerable buyers and sellers [8]. Except for export crops, most agricultural produce is distributed by a multitude of small assemblers and wholesalers. Rural assemblers who buy fr6m farmers seldom have the resources to do more than ship the produce to a nearby population center, where it is resold to other assemblers. These assemblers transport the produce to the city for resale in the wholesale market. There is a serious shortage of storage and handling facilities for all domestically produced foodstuffs [42]. Spoilage and transit losses, chiefly the result of poor warehousing, have been estimated at as much as 20 to 25 percent for such staples as potatoes, barley, corn and beans [16, 23, 24, 38]. The results of these deficiencies are apparent at several levels:
1. Prices are subject to broad disparities from region to
region, and given the limited storage capacity, they reflect a sharp seasonal cycle from glut to scarcity.
2. The numerous resales and the frequent transshipments
between farmers and consumers work to the disadvantage of
3. For such a dietary staple as potatoes, the average total
markup between producer's price and consumer cost in the
Quito market has been reported to be as high as 100
percent [44].
1During 1973-74, the military government set price controls on foodstuffs, but compliance with the decrees varied widely.

Prices paid to farmers are greatly affected by price changes at the consumer level. A drop in price at the point of final demand is usually passed on to the farmer in the form of a lower price. Peasant farmers, by reason of transport costs and unfamiliarity with urban markets, usually must sell their excess production to the village storekeeper, trucker, or itinerant buyer. Small scale merchants are at a similar disadvantage with respect to the large urban wholesalers, who are frequently their primary source of working capital and who have at least a limited capacity to stockpile nonperishables [23].
The Problem
The major problem for potato producers and consumers in Ecuador is the wide seasonal and cyclical variation in prices [65, 66]. Lack of credit, traditional production practices and- price uncertainty throughout the potato production-distribution system have probably limited expansion and intensified supply fluctuations. Because of the seasonal and cyclical nature of production, prices reflect the sharp movement from glut to scarcity. The prices to farmers are subject to severe instability because: (1) the supply fluctuations, along with an inelastic demand, create price fluctuations at the consumer level which are passed on to the farmers, and (2) prices are subject to wide variations from region to region. During the 1961 1967 period, an average change of 0.53 percent in consumption of potatoes was associated with a 1 percent change in price; and after 1968, the average change in consumption dropped to 0.45 percent [8].

While not as severe as the cyclical, 3 year price variation, the
monthly price variation experienced by producer and consumer is a source of concern for both. The per quintal (100 Ibs) price of potatoes has, on several occasions, doubled within a 30 day period. In less than 90 days between February and April of 1962, the price in Tulcan dropped from S./75 to S./26 per quintal [291.2 Such a large variation in price between months can be explained partially by the three-year cycle, which would show severe price differentials in either the period just before or just after the peak price. However, even in the year cited (1962), there was a significant monthly price variation in four of the major market cities.
The cyclical price variation for potatoes in Ecuador has, for
many years, been qualified as extreme and excessive by both consumers and producers [8, 14]. During the well-defined, three-year cycle of high-to-low prices described by Arevalo, the wholesale price for a 100-pound sack of potatoes has ranged from a high of S./120 to a low of S.125 [8, 29]. According to surveys made among producers in the Central Region, it has been estimated that potato production increases by 40 percent during these high-to-low price periods. In general terms, the cyclical price variation of potatoes is not significantly different from other Latin or North American countries, as shown in Table 1.
The relative effect of large price variations does, however, affect
the Ecuadorian consumer to a greater degree than his neighbors, and more
2Unless specified otherwise, all monetary notations indicate
Ecuadorian sucres (S./). The official rate of exchange in October 1974 was US $1.00 = S./24.90, thus one sucre was roughly the equivalent of four U.S. cents.

Table 1. Average cyclical wholesale price variation for potatoes in Ecuador and selected countries
Average Average No. of years Percent that high
Countrya Unit low price high price in cycle price exceeds low price
Ecuador Sucres/qq. 43.68 63.61 3 43
Colombia Pesos/ton 882.00 1205.00 2.5 37
Peru Soles/qq. 85.00 113.00 1 33
U.S.A. Dollars/cwt. 2.48 3.51 3 42
Source: Calculated from [7, 8, 9, 35, 37, 74, 76, 77]. a Based upon price cycles and data collected between 1965 and 1972.

so than U.S. consumers because of the relatively greater importance of potatoes in the Ecuadorian diet.
By reducing the prices to a measurement of the change in food
costs for an average-sized family, a more meaningful comparison to other countries may be accomplished. Based upon recent estimates, an averagesized family of 5 persons in Quito will consume 1.5 lbs of potatoes per day [34]. At the median income level, this represents 3.1 percent of their food costs when potato prices reach their cyclical three-year low and 8.1 percent of their food costs when potato prices are high, thus representing an increase of 161 percent compared to Colombia, where the increase is only 96 percent over a similar period.3 The potato price instability problem is obviously more acute at lower income levels where
a much higher percentage of the family income is devoted to food. The differences between Ecuador and several other countries are presented in Table 2.
The price variability problem is no less serious at the producer's level. Potato production is considered to be a high-cost, high-output operation. This is primarily due to the large number ofphytopathological controls required to insure a salable crop, the cost of adequate seed, and the required fertilizer application. Until 1973, government-supported bank credit was largely unavailable to the majority of producers [47]. Most commercial producers were forced to secure credit through agricultural
3This analysis, made for comparative purposes, assumes that consumers would purchase the same amount of potatoes during a high-price period as during a low-price period.
4The poorest one half of households surveyed in Cali, Colombia, by
Michigan State University researchers typically spent 60 to 80 percent of their income on food compared to an overall allocation of 40 to 50 percent [58].

Table 2. Apparent cost of potatoes as a percent of food expenditure in selected countries
Cost of potatoes per person per day Increase in expenditure
Country Estimate as percentage of total food as percentage of low
expenditureb price
At low price At high price
--------------------------------- percent ------------------------------Ecuador A 3.10 8.10 161
Colombia A 2.40 4.70 96
B .71 1.14 60
Peru A .69 1.06 53
B 3.30 4.40 33
U.S.A. A 1.00 1.09 9
Source: [8, 9, 35, 53, 74, 76, 77].
aMethods of reporting data vary; therefore, alternative estimates (A and B) of per capita consumption, were used where available.
bApparent costs to consumer are based upon prices observed during the following years: Ecuador--1968, 1969; Colombia--1968, 1969; Peru--1964, 1968; U.S.A.--1964-1967.

supply houses or to finance the entire operation from their own sources of capital. Those few producers fortunate enough to harvest potatoes during a high-price period became quite wealthy; however, many large investments have been lost by those unlucky enough to harvest during the "trough" years of the potato price cycle [14]. The magnitude of the losses can be appreciated by examining production costs for the years when prices were at their low point (Table 3).
Additional problems may be linked directly to the inadeqi4acy of
the distribution system. Losses during transit and in market channels, chiefly the result of poor handling and warehousing, have been reported as high as 20 percent with spoilage accounting for 9 percent of these losses (Table 4). According to some estimates, however, potato losses between farms and consumers due to spoilage could be reduced 50 percent through the utilization of adequate storage facilities [77]. In 1972, such a reduction in losses would have provided an additional 16,027 metric tons of potatoes (5 percent of the national marketed production), representing 28 million sucres, based upon the average price of S./81 per
quintal for that year. Also, the limited export potential for potatoes has remained unexploited, primarily due to insufficient marketing and storage facilities [42].
The violent fluctuations in potato prices from year to year cause
great uncertainty for the producer and consumer and lead to an inefficient
The proportion marketed is 70 percent of total production, accounting for 331,343 metric tons in 1972. Spoilage at 9 percent accounts for 29,821 tons. If reduced to 4 percent, it would account for only 13,254 tons, a reduction of 16,027 tons, or 348,417 quintals. At S./81 per quintal, this represents S./28,221,780.

Table 3. Apparent losses in potato production during low price cycles,
Ecuador, 1965-71
Item 1965 1968 1971
-----------S./metric ton---------Price/Metric Ton 616 778 228
Cost/Metric Tona
Cost Estimate I 775 752 832
Cost Estimate II 672 653 722
Loss (-) or Gain (+) Metric Ton
Cost Estimate I -159 + 26 -604
Cost Estimate II 56 +125 -494
Source: [8, 21, 29, 30, 61, 62]. aBased upon two separate cost/ha. estimates. A partially mechanized cost is represented by I (Schwartz) and a sample of medium-sized farms in the Central Zone is represented by II (Arevalo).

Table 4. Potato marketing losses in the Pichincha Province, Ecuador, 1973, per hundred weight of potatoes Type of Total loss Price Value of total loss Loss as percent
middleman Spoilage Weight loss (a) (b) (a x b) of retail value
-----------------lbs. -------------------- -------- sucres per lb. -------- ----percent---Wholesaler 0.8 1.7 2.5 1.20 3.00 2.0
Jobber 1.5 3.1 4.6 1.35 6.20 4.1
Broker 2.3 2.7 5.0 1.38 6.90 4.5
Retailer 4.7 3.4 8.1 1.53 12.30 8.0
Total 9.3 10.9 20.2 28.40 18.6
Source: [44].

allocation of resources. The inelastic nature of the price-quantity relationship for potatoes results in extremely unstable. revenue expectations, thus causing alternating deficit and surplus situations in the marketplace.
The Hypothesis
The high degree of price variability for potatoes affects not only the welfare of the consumer, but also contributes to the great amount of instability found in the production sector. It is hypothesized that the cobweb cycle could be broken (or at least reduced in amplitude), benefiting both producer and consumer, by stabilizing the price and supply of potatoes through:
1. Storage from surplus to deficit periods;
2. Production adjustments allowing for year-round cultivation
of potatoes; and/or
3. Import and export of potatoes to balance national deficits
and surpluses.
The Objectives
The economic feasibility of the alternatives can be determined by examining the costs and benefits associated with each one separately and in combination. However, the scope of this study will allow an investigation of only the first two alternatives--storage and production adjustments. The main objectives were to:
1. Determine the economic feasibility of potato storage in
Ecuador with price supply stabilization as a primary
consideration, by estimating the potential storage profit
margins for the three major producing areas.

2. Determine the economic feasibility of production adjustments as a potato price stabilizer in conjunction with or in place of potato storage by measuring producer attitudes
and estimating the costs associated with smoothing pro-.
duction. Production adjustments would include reducing
production seasonality through the introduction of new
varieties, changes in cultural practices (such as planting
and harvesting dates) and credit.
Organization of the Thesis
Subsequent chapters describe the theoretical and empirical consideration for the analysis of storage feasibility, and the benefits which may be derived by price stabilization. A short description of potato production and distribution in Ecuador is followed by the actual analysis of price variation, storage potential, and production management. In the final chapter, the results of the analysis are summarized and several recommendations are made.

Theoretical Considerations
Prior to analyzing the feasibility of stabilizing the price and
supply of potatoes in Ecuador, the theories related to price variability, storage, and production smoothing were reviewed and the assumptions which underlie the treatment of the data were investigated. To avoid unintended bias in the data processing, interpretation, or analysis, an explanation of the data sources, treatment, and reliability is made below in conjunction with a description of the mathematical and statistical methods which are appropriate to the problem description and data characteristics.
Price Variability
Benefits of less variable prices. Shepherd discussed at some
length the benefits consumers may derive from price stabilization [631. Consumers receive more satisfaction from the fairly even consumption of a particular food than from a scarcity at one time and a glut at another. This would appear to be especially valid in the case of a dietary staple like potatoes. The extra worth of the stable supply may be greater or less than the extra money the consumer has to pay for it; however, the consumer would benefit to some extent, since the extra money paid would provide some benefit and might even be of lesser value in general to the consumer than the benefit received.

Waugh has shown, however, that with any negative sloping demand
curve, the loss in consumer's surplus from averaging two prices is always greater than the gain [63]. This can be illustrated by drawing a hypothetical demand curve and labeling the areas above and below the average
price, Po (Fig 1). The gain (G) for consumers when prices are below average is always greater than the loss (L) when prices are above average, due to the negative slope of the demand curve.. Thus, according to Waugh, consumers are harmed by price stabilization, not only as a grqup but also as individuals,
Howell and Lovasy have pointed out, however, that Waugh's theorem
is true only if prices are stabilized at or above the arithmetic mean of the variable prices [63]. Shepherd concurs, stating that in the case of a straight line demand curve, stabilizing the supplies at their arithmetic means will stabilize prices at their arithmetic means. However, if the demand curve is curved, stabilizing supplies at their arithmetic means will stabilize prices below their arithmetic means, thus benefiting consumers.
Johnson made an additional point, showing that stabilization of supplies at the arithmetic mean of the fluctuating supplies always benefits society as a whole (if carrying costs are neglected) [63]. In cases where consumer's surplus is reduced, there would be a substantial gain in producer's surplus, in which case the producers could afford to compensate consumers for their losses while maintaining a net profit.
Regional price variation. Although the scope of this study does not include examination of potato import and export feasibilities, the degree of interregional market integration within the country may be

Quantity Purchaseda
Figure 1. Theoretical comparison of loss and gain of consumer's surplus
caused by fluctuating prices.
Source: [63]
a Gain (G) is greater than loss (L) at any price below the
average price (P).

measured by determining how closely the prices are correlated between domestic markets. Market integration is defined as the degree of interrelationship between prices for the same commodity in separate markets. A measure of the extent to which the markets are integrated is indicated by the degree of correlation between prices measured on a regular basis in the separate markets. Where markets are integrated, the prices for a particular commodity, potatoes in this case, would not vary by more than the costs associated with transportation, risk of short-run changes in price, distance from sources, transportation bottlenecks, or differences in grade. Where the lack of correlation between regional market prices cannot be explained sufficiently well by the factors listed above, one may assume that price variation could be reduced through some kind of supply management program [43].
Variable supply and the inelasticity of demand. A primary cause of price fluctuations for potatoes is the highly variable quantity of potatoes supplied to the market. This supply instability in combination with the inelastic nature of the demand for potatoes illustrates the classical "cobweb" price-output pattern.
The nature of price fluctuations in a cobweb analysis depends upon
relative elasticities of supply and demand [64]. Where the elasticity of supply is significantly greater than the elasticity of demand, supply interactions will di verge from equilibrium and fluctuations will increase. Just the opposite is the case for potatoes, at least theoretically, indicating their price output pattern should converge towards equilibrium, i.e., adjustments in production should eventually bring about an equilibrium price. For a number of reasons, however, behavior in the real

world, does not conform to this simple case of cobweb theory. Farmers do not always react the same way to past prices; they play the odds, hedging their potential future losses by planting several different crops whenever possible. Weather makes total yield response extremely variable no matter how the producers react to prices. Technological change may also affect yields, crop quality can affect prices, and demand shifts may disrupt the classical cobweb pattern. The pattern of sharply fluctuating prices and income, however, results primarily from changes in available supplies and the relatively inelastic demand for potatoes.
Supply stabilization. Any attempt to alter this price instability must necessarily involve changing the supply or demand, either by stabilizing supplies and making supply more inelastic or by providing for a more elastic demand. Because major changes in demand elasticities are unlikely, stabilizing potato prices in the short run must involve some kind of supply management program. Presumably, such .a program would adjust supplies available to market demand at prices that are acceptable to both consumers and producers. A supply management program would reduce price variability in the cobweb price-output pattern by making supply more inelastic.
Concept of storage. Determination of the economic feasibility of
potato storage rests primarily upon the 'costs associated with the desired length of storage time. Samuelson describes the underlying theory of storage showing that whenever the expected price of a commodity exceeds the harvest price plus storage costs, storage will occur [60]. Ile made

this observation by assuming that:
1. Monthly demand remains constant;
2.. Variable storage costs per month remain constant;
3. There. is only one instantaneous harves t per year; and
4. Total monthly sales must equal the fixed quantity harvested. However, potatoes are a perishable commodity for which storage costs increase through time due to spoilage and weight loss, and there are several harvests throughout the year [23, 15, 69]. Samuelsonallowed for such a case in his description and illustrated seasonal patterns with an extended harvest.
Maximum allowable investment in storage. A similar but more consise analysis can be made by estimating how much could be invested in storage. Such a procedure would, in effect, measure the producer's average surplus by subtracting production-distribution costs from the wholesale market price over a long period.
Production Adjustments
Supply stabilization can also be accomplished by extending the production season, thus making potatoes available in quantities sufficient to meet the consumer demand throughout the year. The feasibility of extending the potato production season in Ecuador may be analyzed by means of-the following procedures:
1. Determine the amount of time by which the production cycle
in each of the three regions may be altered through changes
in the planting and harvesting periods.
2. Determine those factors that have the most influence on
producer -response to changes in cultural practices.
3. Identify additional factors (credit, farm size, risk,
market structure) which affect farmers' production decisions.

To determine the optimum combination of potato storage and supply management programs, the equilibrium quantity of potatoes required to meet the consumption (effective demand) requirements of the nation throughout the year. must be estimated. Stabilizing prices by storage requires that supplies be stabilized at the arithmetic mean of the varying supplies (allowing for spoilage). This implies that the point where prices are stabilized will be determined by the curvature of the demand curve [63]. Then, by estimating the demand and supply'functions for each period throughout the year, deficits and surpluses relative to the mean can be determined vis-a-vis the quantity required and the quantity demanded to maintain equilibrium (the stable supply).
The relative costs associated with either storage or production
smoothing may be examined by comparing the marginal cost associated with each. A potential limitation to such an analysis is the difficulty with which costs are assigned to the items in each alternative.
Empirical Considerations
Measuring Potato Price Variability
Price variation within and between market cities by month was measured by means of standard regression analysis. The degree of variation was also calculated for alternative highland crops using Kling's procedure6 to determine the amount of risk associated with growing and
6Measured by subtracting annual prices, yields, and value per acre from preceding year's prices, yields, and value per acre; then computing the standard deviation of these changes and expressing it as a percentage of the average price, yield, or value per acre for the period.

selling each crop [41]. Occasionally the results were compared with those from other countries to lend perspective to Ecuador's situation.
The primary source for nationwide, monthly price data in Ecuador is the series prepared by the Institute of Economic Research at Central (National) Quito. In 1952, the Institute began to tabulate monthly wholesale prices for 116 separate articles in the market plazas of seven major cities [45].
A new wholesale-retail price index was established in 1911 by the
National Planning Board (JNPCE) and the National Institute of Statistics. In recent years, these agencies have jointly published a price series and index based on data gathered in Quito, Guayaquil and Cuenca.7 The 1951-72 consumer price index for Quito has been used in this study, with
the food and beverage index converted to 1970 = 100 base.
Zuvekas has indicated that, although the quality of wholesale price data for Ecuador is generally considered to be rather poor, the data probably give a fairly good picture of long term price trends [80]. Since no other historic price series exists for the entire country, it is utilized routinely by the United Nations, World Bank, Inter-American Development Bank (BID), and the U.S. Agency for International Development. In the National Agricultural and Livestock Research Institute's
7All of the price data utilized for this study were compiled from
the previously mentioned sources in Ecuador and insertions were estimated for missing data. The 1960-69 series was copied directly from the final tally sheets at the Economic Research Institute at Central University and the post-1970 data were taken from the annual publications of the National Planning Board (JNPCE) and the National Marketing Service (SIMAE).
8The consumer price index was changed from a 1965 to a 1970 base year to correspond with Arevalo's analysis of potato production and distribution in the Central Region. These conversions are shown in Appendix Table A-19.

Economic Bulletin No. 9, Dow lists the Central University price data, although he supplemented the series with a list of producer level prices for agricultural goods (for 1972 only) [21].
Unfortunately,, there is no guarantee that the prices shown in the Central University series reflect prices for the same product. Varietal preferences of potatoes by region have been observed [30, 44], and it. should be recognized that until 1969, the price series shows a distinction by potato size only. The prices for toda grosa and redroja9 sized potatoes were compiled by university statisticians, but the varietal designations were ignored until 1972, when the popular "Chola" variety was distinguished as the object of SIME price sampling technique. However, the series is the longest and most complete collection of price data available by region and is routinely used by most institutions.
The feasibility of utilizing storage to attain stable supplies is dependent primarily upon the associated costs and length of storage. Either or both of these two factors could be determined by constructing an "ideal" storage facility and then establishing the physiological and economic parameters by purchasing potatoes and storing them for various time periods. Such an undertaking was deemed to be expensive and timeconsuming for the present study so alternative methods were chosen.
9These refer to Grade A and Grade B potatoes, respectively.

Length of Storage Period
In order to determine the period during which storage might
logically occur, the potato price series for the primary market cities were analyzed. The prices were averaged by month over an 18 year period (after being constant 1970 sucres) and over the cities to
*obtain a national set of average monthly potato prices. The degree of variance between months was then tested using the Tukey-Duncan
*procedure [49].
Storage Costs,
The costs associated with the storage process selected will
determine, in part, whether or not storage is a viable alternative to immediate sale. Thus, the technical requirements for potato storage by region and according to level within the production-distribution system. must be determined. Abundant literature exists describing the tempera-' ture and humidity requirements for optimal potato storage in the United States and Europe and designs for storage units which will facilitate optimal environmental conditions. Space and light conditions also play an important role in stored potato quality and are well represented in the literature [12, 28, 56, 69, 70, 72]. Unfortunately, only a few studies exist for potato 'storage in Ecuador, and these are limited to examining the feasibility of utilizing chemical germination inhibitors [17, 20, 31].
Storage costs associated with each alternative system must be
estimated, based upon location, size, and degree of mechanization of the system. These costs can be calculated based upon the following considerations [58]:

1. Cost of constructing and equipping the storage facility and
rate of depreciation;
2. Cost of maintaining and operating the storage facility; 3. Cost of grading, handling, and application of chemicals
to potatoes;
4. Value of potatoes rejected in sorting;
5. Value of potatoes lost due to shrinkage and damage;
6. Interest rate for holding potatoes based upon price at
Costs for specific storage requirements may vary substantially
between small, medium, and large farms and assembler or retail levels. Opportunity costs attached to labor, shrinkage amount of crop withheld for immediate or near-future consumption, and the availability and cost of credit may also vary.
Actual Price Storage Analysis
Along with estimates of storage costs, it was possible to review the historical potato price series on a national and regional level, determining those periods, where storage might occur, given the physiological constraints. In this way, a series of actual margins for each year was generated, which was subsequently deflated and averaged to obtain an estimate of the potential margin between the eventual sale price and the original price plus storage charges.
Regional. Storage and Transport
Another alternative to estimating storage feasibility on a national scale was the calculation of regional storage followed by subsequent transport to higher priced markets. The optimal levels of storage, transport, and price can be estimated to measure the net social benefit

of a government-supported price stabilization program. Several linear programming models have been designed to analyze such a problem, in particular, those based on Samuelson's treatment of intertemporal price equilibrium, with its concept of net social pay off [18, 60]. The data base for Ecuador, however, is not sufficiently reliable to generate either the supply or demand functions required for the above treatment. Thus, an analysis of the margins was made by adding the cost of transport to the storage costs. The city where the lowest
*series of prices occurs was selected as the storage center to show the maximum margin attainable by such a procedure.
Maximum Allowable Investment in Storage
The amount that could be invested in storage was calculated for the Central Region by determining the average price for potatoes by year and province and subtracting the estimated costs of production and distribution. Independent estimates of the cost of production were utilized in order to lend as much breadth of interpretation as possible to the calculation and the most appropriate estimates were chosen to represent the potato industry. Assuming that consumption equals production, i.e., all of the potatoes produced are consumed in one form or another, then expenditures for potatoes ingeneral are equal to the sum of production at given prices:
(1) Expenditures = QP
Where Q = Quantity of potatoes harvested in quintales per month in a given year;
P = Wholesaler retailer price per quintal per month in a given year.

By allowing Cpd to equal production and distribution costs and Cs to equal storage charges, the price equation can be expressed in terms of these costs:
(2) P = Cpd + Cs
However, to analXze the historical data available, the equations
must be converted to a measure of the actual price-quantity relationships. The national average annual price for potatoes is written as:
(3) Pj =EQij Pij
where Q = Quantity of potatoes harvested in quintales; P = Wholesaler retailer price per quintal of potatoes;
i = Ist, 2nd, 12th month of year;
j = 1954, 1955 . year. Based upon Shepherd and Johnson's [63] concept of consumer surplus at or below the mean of fluctuating supplies, it was assumed that whenever P < P, the consumer will benefit. It would then follow that: if
(4) Cpd + Cs then
(5) Cs

and (6) Cs + B =P Cpd
Storage costs and consumer benefits can be redefined as a new
identity, S, equal to Cs + B, where B represents a variable quantity which may accrue to the benefit of the consumer, i.e.:
(7) S =Cs +,B

Or, instead, an upper bound may be established on the amount that
could be invested in storage. Any difference between actual storage and the maximum allowable amount would represent a benefit (B) to the consumer, and it is assumed that storage would occur whenever B > 0.
S may be estimated using the following procedure:
(8) S = E(Pjk Cjk) Z(Pj Cj)
where S = Maximum allowable storage charge per quintal per month nationally;
Pj = EZQij Pij
Cj = Production distribution costs; i = lst, 2nd, 12th month;
j = 1962, 1963 Jth year An elementary FORTRAN program was written to calculate the maximum allowable investment in storage.
Production Adjustment
Potato Growers Survey
To determine potato growers attitudes towards planting month
preference, ease of postmature harvest and other production practices, a survey was designed for use in the Northern and Southern Regions. The questionnaire was divided into two sections, the first dealing with the factual data such as farm size, type of-seed, and planting space, and the second part dealing with the grower's attitudes towards the best month for planting, reasons for premature harvest, and other nonbiological controls.

The primary purpose of the survey was to measure the influence of the farmer's attitude concerning preferred months for planting and harvesting, as related to the size of farm and region within Ecuador. It also indicated the reasons for different production practices, which will allow future analysts to concentrate their study upon those specific attitudes or practices considered alterable and conducive to."production smoothing."
Sample Size
The size of the sample was determined by the need to make the survey as comparable to Arevalo's Central Region survey as possible. The grower's response to preferred month for planting is used in conjunction with the same response in Arevalo's questionnaire; thus, a target number of at least 80 growers was interviewed in the Southern Region, where approximately the same quantity of potatoes are produced as in the Central Region. A similar breakdown of the sample, i.e., small- (34 percent), medium- (38 percent), and large- (28 percent) sized farms was considered optimal; thus, more than 80 growers were surveyed. In the Northern Region, a similar number of growers were interviewed in order to approach the farm size classification percentages, since most farms are medium-sized in the North. Time and budgetary constraints ruled out using a larger statistically determined sample, requiring surveys in the South and in the North based upon the degree of price variation.

Production Data
Following the 1954 agricultural census, attempts were made by the Central Bank to gather and/or estimate production and hectarage for several basic crops and to publish the results. Thus, a highly estimative series of production statistics were published for 1955 through 1958. By 1962, a regular reporting series had been established by the Ministry of Agriculture (MAG).
The MAG receives estimates of hectarage and production from the
province level extension offices throughout the country, in addition to estimates from technicians employed in various semiautonomous agencies, such as the Canar-Azuay Development Board (CREA), the National Planning Board (JNPCE), the coffee and banana commissions, and the National Agricultural and Livestock Research Institute (INIAP). A series of adjusted production and hectarage estimates are calculated for each province.10 The MAG series is the longest continuous available series for production and hectarage at the provincial level. Neither varieties nor grades are distinguished for potatoes in the series or its appendices and footnotes.
The accuracy of the production and hectarage data may be questioned, especially when reportedly different sources show wide variation as, for example, in the case of the amount of wheat produced (harvested) in 1968. The large amount of variability in yield for highland crops reinforces a lack of confidence in the accuracy) of the data, since it is doubtful
10Similar to the U.S.D.A. process of adjusting U.S. national
production estimates of individual state production and acreage to the total nationwide estimates.

that natural factors alone would account for such large consecutive annual differences in yield.
A 1971 study of price and income elasticities of major agricultural commodities by Utah State University economists was abandoned, after publishing an interim-report, due to their lack of confidence in the' secondary price and production data [78]. Nevertheless, estimates of yields for the major crops probably indicate correctly the direction of changes in production and hectarage throughout the 1960's and can be utilized as "direction indicators" in the absence of more reliable statistics. -Comparison of the national estimates with those made by extranational agencies (USDA, BID, FAO) would be of little purpose, since their sources of production information are even more limited and biased than the national accounts.12 Although efforts have been made by Zuvekas to explain discrepancies in the banana production estimates, no similar work has been published for the nonexport crops, such as potatoes and its highland alternatives [80].
Marketing and Cost of Production Data
Production cost data have been selected from various publications to illustrate the wide variety of estimates available to the analyst. Shepherd emphatically points out the fallacy in utilizing or accepting only one set of production cost estimates, stating that:
11Morris Whittaker, Personal correspondence with author, April 18, 1974.
12C. Milton Anderson, Agricultural Attache, U.S. Embassy, Quito, Ecuador, Personal Interview, 1974.

Anyone who tried to es timate "the". cost of production
of a farm product soon finds that there is no such
thing. Each farmer has this own cost of production,
and these costs differ from farmer to farmer [63].
Production cost estimates for potato production during the early
1970's do tend to group around S.110,000 per hectare. However, current productions costs are much higher due to the doubling of fertilizer prices in late 1973.
Perhaps the most accurate estimates of marketing costs, margins, and movement of potatoes, at least in the Central Region, are those found in Maldonado's 1974 marketing study (Appendix Table A-17). Several tables from the study have been included in the Appendix, together with comparable data from two previous studies--the 1963-64 ITALCONSIJLT marketing analysis (Appendix Table A-18) and a more recent FAO report [25].

Agricultural Production in Ecuador
During the last decade, the annual rate of growth of agricultural production in Ecuador lagged behind that of the national economy [13, 80]. Yet, as a result of rapid population growth at 3.1 percent annually between 1963 and 1971, and gains in real per capita income from $174 in 1960 to $190 in 1970, the demand for farm products has increased steadily [5, 11].
The lagging state of Ecuador's agricultural sector is commonly ascribed to a highly adverse pattern of land and labor use, technical backwardness, and lack of investment in agriculture [23]. Methods of cultivation in the Sierra in many instances have not greatly advanced beyond-.those in use at the time of the Spanish.conquest in the 16th Century. The small Indian farmer cultivates all of his land, which has become -relatively unproductive through centuries of use, whereas the'estate owner leaves a considerable part of his land unused [23]. In neither instance have modern methods been adopted to any marked degree. This is not to say that agricultural production is necessarily inefficient, but that it has been unable to keep abreast with the requirements of an expanding population and economy [42].

The agricultural sector Icontributes substantially to export
earnings, domestic food requirements, and employment. Until the early 70's and before petroleum was discovered, agriculture accounted for 90 percent of Ecuador's export earnings, principally from bananas, cocoa, coffee, and sugar [42]. Domestic agricultural output currently supplies about 85 percent of the country's food needs and many of the agricultural products required for industry. Crop and livestock production, forestry and fishing provide a livelihood for over 50 percent,,of the economically active population. Additional indirect contributions are made to employment by the food and beverage processing industries which account for S5 percent of the value of manufactured products [54].
Food consumption in Ecuador is below minimum nutritional standards and is well under the average for Latin America. Present food consumption levels indicate a large potential demand for all food products, especially the more nutritious foods such as meat and dairy products; filling the large potential demand will depend upon supply and income factors in the years ahead. General estimates of income elasticities made by the FAQ indicate ranges from 0.1 for tubers and 0.4 for sugar, to 0.8 for meat and dairy products, and 0.9 for fats and oils [42].
Potato Production
Among the domestically produced and consumed farm products grown in Ecuador, potatoes are one of the most important. Since there is no significant export or import movement, nor alternative uses for the potato crop', potatoes are used to satisfy domestic consumer demands. In the highlands, estimated annual per capita consumption of tubers, of

which potatoes are the predominant item, is 660 grams, making tubers a dietary staple for a majority of the people living in the highlands [42, 34]. Of the field crops grown for domestic consumption in 1973, potato production ranked first in volume and second in value. Of the 819,000 hectares of farm land planted to major domestic food crops in 1973, 6 percent or 43,573 hectares (107,550 acres)'was utilized for producing 538,769 metric tons of potatoes [75].
Although there are a few very large farms devoted entirely to potato production, potatoes are grown, for the most part, in conjunction with other crops on farms of all sizes, from large estates to subsistence plots. The smaller subsistence farms of five hectares or less are the only size group which dedicate a major portion (up to 37 percent) of their land to potatoes. These smaller producers can be characterized as self-sufficient, marketing only those quantities of potatoes clearly in excess of their family needs. Thus, it has been assumed that their planting schedules are relatively unaffected by price variations [8].
Medium-sized farm owners with more than five hectares cultivate a variety of crops rather than allocating a large portion of their holdings to potato production where uncertainty due to low disease resistance, and large seasonal price variation is prevalent. Additionally, the per hectare production costs for potatoes are high, thus inhibiting extensive medium-size farm plantings. The largest haciendas are usually located at an altitude over 3,000 meters where the low cost of raising livestock offers an advantageous alternative to potato cropping, especially in view of the risks associated with potato production.

According to Arevalo, it is on the large and medium-size farms where production decisions determine the total quantity of potatoes which are marketed in any given period. When favorable prices occur, the medium-size farm owner can shift additional land into potato production more easily than the large hacendado, who must convert pastures to plantings. However, just the opposite is true when unfavorable prices occur, as the large farmer may easily convert to additional pasturage, but the medium-size farm owner is constrained by the market price situation for alternative crops.
The principal potato production area extends the length of Ecuador, at 3,000 meters above sea level in a long "canyon" between twin ranges of high Andean peaks (the shaded area in Fig. 2). The major regions are distinguished as the northern, central, and southern inter-Andean basins. The physiological requirements for high yields limit potato production to this inter-Andean area where daytime temperatures fluctuate between 150 and 180 centigrade, and where precipitation of 135 mm. to 289 mm. will guarantee proper potato top and tuber formation. Since precipitation is such an important factor in the growth and yield of the potato plant, the preferred planting time varies by region according to the rainfall schedule [2, 3, 4, 8, 55].
The vegetative cycle for potatoes requires about three months, followed by an equal period for proper tuber formation. Thus, most potatoes are harvested as near to the seventh month of their life cycle as possible. Among the larger producers who have access to improved seed varieties, fungicides, insecticides and irrigation, the production season is reportedly almost year-round [30]. Smaller producers, however,

81 80 79 78 77
Esmeraldas Carchi
Imbabura %
0 qU ITO0
-Pichincha Napo-Pastaza
Los Rios
2 44
\- .
Guayas -c )
Canar Santiago Zamora
3 Azuay
El Oro
. / ,
(Undemarcated ) d
PERU ..#
,_____ I
Scale 100 km.
Figure 2. Map of Ecuador and the study area. Source: Based upon [51].

still cling to their traditional planting periods, and each major producing region displays slight differences in the preferred month for planting (Table 5).
The regions are further distinguished by their soil types, cultivation systems, and product markets. According to Arevalo, the northern zone has demonstrated an expanding production due to increased yields and has been more adversely affected by price variations than the other regions. This condition reportedly has caused an increasing,shift to cereal crops and beef cattle, while the opposite situation in the Southern Region has led to an extension of the hectarage allocated to potato production. However, little support for these observations can be found in the regional production reports (Table 6) [8].
Preferred Months for Planting
The Northern Region apparently shares a bimodal production
pattern with Colombia's Narino province. The major producing province of Ecuador's Northern Region, Carchi, forms the southern limit to a natural geographic and ecologic region, the Narino Plateau. As would be expected, the production-price patterns for potatoes in Tulcan, Ecuador,,and Pasto, Colombia are quite similar.
However, the two northern provinces of Carchi and Imbabura account for only 17 percent of Ecuador's potato production. The greatest amount of cropland and resources allocated to potatoes is concentrated in the Central and Southern Andean Regions. Production is divided evenly between the two regions, with each region accounting for approximately 40 percent of the national production. In the Central Region production

Table 5. Preferred month for planting potatoes in the Northern, Central,
and Southern Highlands of Ecuador by percentage of farmers
interviewed, 1974
Month North Central South
------------------ (percent) ----------------January 5 7 2
February 6 5 2
March 4 8 5S
April 5 10 10
May 20 15 18
June 6 9 10
July 4 7
August 5 8 5
September 1 6 5
October 15 8 16
November 17 10 13
December 12 9 7
Total 100 100 100
Source: [8, 79].

Table 6. Average yearly potato production, area harvested, and yield by region, Ecuador, for periods
1962-67 and 1968-73.
Region Production Area Yield
1962-1967 1968-1973 1962-1967 1968-1973 1962-1967 1968-1973
Metric tons Hectares Quintals/Ha
North 44,170 93,9-30 4190 5485 229 372
Central 144,818 218,076 12658 14603 249 325
South 158,430 229,945 23028 25282 150 198
Percent of total Percent of total Percent increase
North 12.8 17.7 10.5 12.5 62.4
Central 42.2 40.2 32.7 32.2 30.5
South 45.0 42.1 56.8 55.3 24.2
Source: [50, 62].

is concentrated in Pichincha., Tungurahua, and Cotopaxi provinces while in the Southern Region it is concentrated in the provinces of Chimborazo and Canar.
In contrast to the strictly bimodal production pattern in the
Northern Region, the crop year appears to be more extended in the Central Region, allowing for almost year-round production. The season is limited only by the relatively dry period during July and August, according to Arevalo's survey of producers in the three central provinces. The Southern Region's seasonal price pattern suggests a planting season quite similar to the Northern Region. This can be partially explained by the relatively low level of technology which characterizes potato production in the Southern provinces, thus producing a more traditional approach to the planting season.
Reasons for selecting a particular month as best for planting
potatoes varied substantially from one region to another, although the traditional concern for weather and disease avoidance easily outweighs other reasons as the farmer's primary concern.13 The dual concern for weather and disease avoidance is not accidental, since a particularly virulent potato fungus attackes the plant only when certain weather 14
conditions prevail. Late Blight, called lancha in Ecuador, is almost always present in the fields, and the fungus will flourish when there is a cold spell (or night). accompanied by rain or heavy dew, followed by a rise in temperature (the following day), with continued humid, moist conditions [69]. Such weather patterns are more prevalent at
13See Appendix Tables A-37, A-38 and A-39.
14Phytophthora infestans (Mont) De Bary.

certain periods of the year according to region, and it is apparently the primary concern for many farmers, especially the medium and large sized commercial growers who cannot afford more than a few applications of fungicides. Additional precautions in selecting a suitable month for planting must also be taken to avoid frosts, hail, and drought. Figure 3 shows the average precipitation pattern for the Sierra Region throughout the year. The most noticeable difference by regions is observed in the Southern area of the Sierra (from Ambato, southwards) where a definite dry period appears from October through December.
Movement of Production to Market
Varying supplies of potatoes are delivered to the large urban
centers in an approximately continuous, year-round pattern, although a one-month delay or short fall in any given month's regional production could combine with a naturally low supply period nationwide to produce significant national shortages for a short period [30]. Or a regional crop slightly early or late might coincide with another regional harvest, causing a glut in the market place. Figure 4 illustrates the estimated arrival of regional harvests in urban centers based upon the regional schedules for planting potatoes, as indicated in Table 6. This pattern of regional potato production corresponds reasonably well with average monthly prices for an 18-year period. Averaging prices over such a long period, however, is somewhat misleading in that it reduces the severe monthly price differentials caused by coincidental shortages or surpluses.

Latitude Month City
10 JF M A M J J A S O N D
Equator 00
S Quito
30 Cuenca
4 Loja
J F M A M J J A S 0 N D Rainy period Figure 3. Rainy period in the Highlands Region, Ecuador. Source: [25]

S./qqa Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec MT/mo.
86 \ Southern I \ National Average 20000
ProductionWholesale Price \ "I
83 k
% "\/ 15,000
80 \ Production
. 10,000
. I
77 -,
IN. 5,000
. /\,-.
/Northe rn Production
Figure 4. National average monthly wholesale price and estimated monthly market arrival For potatoes by
region in highland Ecuador, 1962-72. Source: Calculated from [29, 50, 62, 79]. aSucres per quintal, from Appendix, Table A-8. bMetric tons per month, from Appendix, Table A-11.

The process utilized to get a potato crop from the field to the consumer is determined by the size of the crop. Small scale farmers customarily sell their produce in local marketplaces on a regularly, designated market day, either directly to the consumer or to the owners of small warehouses. Oftentimes, a special wholesale area or marketplace has been established where larger quantities of potatoes may be purchased from the farmers by the warehouse owners (or their agents). The warehouses and storage bins can usually be found on streets immediately adjacent to the main plazas [14].
Larger scale farmers usually sell their crop unharvested to field buyers or warehouse owners from the -nearby cities. Usually, the buyer will pay only 10 to 20 percent of the agreed upon price at the timc of harvest, and will pay the balance during the following 60 days. Mien purchasing in the field, the buyer will usually bring his own harvesting crew, although large producers in Cotopaxi, Chimborazo, and Canar Provinces have maintained the custom of harvesting their own fields (probablydue to the great surplus of available cheap labor in the area). Large warehouse owners (or their agents) provide their own transport,,/ hauling the crop immediately to the nearest large city, or to the largest metropolitan centers where they can obtain more favorable prices [14,44]. Occasionally, large scale producers will contract for part of their crop to be. harvested and transferred by an area trucker who employs a harvesting crew. The owner will then sell additional sections. (tablas) of the field to individual wholesalers (or field buyers), each with their own small crew of 5 to 8 potato diggers.

Once the potatoes reach the urban centers, they may be sold
directly to the public (in the market places), but most often they are purchased by a second group of wholesalers or brokers, who in turn sell the potatoes to retailers (or to the public). Quite often, the potatoes are received on consignment, and the consignee may try to expand his sales opportunities by offering commissions to street salesmen. Retailers who sell their potatoes in 100 pound sacks can expect to handle from 10 to 100 sacks per week, while those who sell b the pound (smaller neighborhood shop owners) can only expect to handle from 1 to 20 sacks per week [1, 14, 32].
Due primarily to the high production costs (especially when heavier than usual applications of fungicide are required) and the relatively small amount of available bank credit, producers often sell their crops prior to harvest to acquire additional working capital. And in the provinces of Chimborazo and Canar, a form of sharecropping is still prevalent in some zones, where the large landowner furnishes the land and potato seed, while a businessman will furnish the tractor, fertilizer, and laborers. Under such a contract (called aparceria), the crop is subsequently divided evenly between the two. [14].
In a 1974 study of potato marketing in Pichincha Province,
Maldonado concluded that the potato production-distribution system in that area is characterized by complete freedom of entry and exit. Ile found no evidence of widespread speculation (due perhaps to the highly perishable nature of the product and the complete absence of adequate potato storage facilities), nor any relation between potato price fluctuations and the cost of transporting potatoes [44].

The movement of potatoes from the areas of major production to
those of major consumption is apparently restricted only by the condition of the roads and the price expectations of the seller. Figure 5 indicates the direction of market flows and provides an estimate of the volume of potatoes moving from the production areas to major cities [39, 40, 46, 59].

Ibarra PACIFIC"\ ""
Sto. Domingo Quito
Portooviejo uv
IIIanta < O.... Latacunga
AmbatoI JI
'Babah iobamba /
Principal zones of production Major transactions
- > 4 Medium transactions OCEAN -- Minor transactions
1 achala
Loj C
Oa Cities of 500, 000 to 1, 000,000 pop.
0 Loja
p# /
/ Cities of 20, 000 to 100, 000 pop.
Figure 5. Market flow of potatoes in Ecuador, 1973.
Source: [14]

Price Variation
The degree of cyclical and seasonal potato price variation in
Ecuador described in Chapter I and the cyclical movement of prices for potatoes in several countries (Table 1) shows little difference on an international level and points to the probability of a common economic characteristic for potatoes, i.e., inelasticity of demand.
The impact of potato price variation in Ecuador may be examined from many different perspectives. Price variation may stimulate producers to grow alternative crops, alter their production practices or may be a reason per se for investment in storage. In this analysis, price variation was examined primarily as an incentive for the government to develop price stabilizing schemes. After examining the intensity of potato price variation, potato storage potentials were analyzed on national and regional, levels, with a final consideration given to possibilities for reallocation of supply on a national basis with transport and storage.
Among Highland Crops--Risk
Based on producer responses, 70 percent felt that they had experienced considerable losses in the last few years and 80 percent felt that price

fluctuation was the major causal factor (see Appendix Table A-12). Furthermore, the degree of variance in yield, price, and value among highland crops indicates that there is a greater amount of risk associated with growing potatoes than with other Sierra crops. By measuring the standard deviation of changes in price, yield, and value per hectare from one year to another, an indication' of the relative risk involved in growing each crop can be established [41].. Table 7 shows the annual variability in value per hectare, yield, and price for,,several alternative Sierra crops. Although barley and potatoes are affected almost equally by those factors influencing yield, their respective values per hectare are substantially dissimilar. This is apparently caused by the much greater degree of price variability in the case of potatoes, whereas the price for barley remains relatively stable. This may be due to the highly perishable nature of potatoes in the postharvest marketing stage.
Within and Between Market Cities
Potato price variation within the five major market reporting cities was the stimulus for the present study. Monthly price variation and the cyclical price pattern have been characterized as excessive and aggravating by consumers, producers, and national economic planners. The degree of correlation between the monthly potato prices for several market cities was calculated. The matrices of correlation coefficients are based on the 1962-72 series of wholesale prices. A measure of the significance of price variation between months on a national level is included below in the discussion of storage.

Table 7. Annual variability of value per hectare, yield, and wholesale price of four alternative
highland crops, Ecuador, 1962-72
Annual Variabilitya
Crop Value per hectare Yield Price
----------------------------(percent)-------------------------Potatoesb 62.5 42.5 44.5
Corn 41.3 41.1 19.3
Wheat 33.0 28.7 9.8
Barley 43.1 35.6 18.1
Source: Calculations based upon data from [29, 41, 50, 62] and Appendix Tables A-1, A-2, A-4, A-S, A-6 A-7, A-13 and A-20.
aMeasured by subtracting annual prices, yields, and values per hectare of 1962-72 from preceding year's prices, yields, and value per hectare; then computing the standard deviation of these changes and expressing it as a percentage of the average price, yield, or value per hectare for 1962-72. bCompare with the estimate of risk associated with U.S. potato production during the 1918-40 period; Value per acre--46%, Yield--9%, Price--53%. [41]

Potato price variation between market cities was examined by
generating the coefficients of correlation between the cities for each year in the 1962-72 series, and for the entire series broken into two subgroups (Table 8). The 1962-69 prices were analyzed to examine variation between cities prior to the completion of a new major highway linking the Northern and Central regions and prior to instituting a new price reporting system. Subgrouping the routine increased the degree of correlation for recent years, although prices in Cuenca remain comparatively less correlated to those in other cities. It would be difficult to define with accuracy the reasons for increased correlation in prices between the major cities for the two periods without additional investigation; however, some major reasons could be:
1. Improved communication between cities, both terrestrial
and verbal;
2. Gradual dispersion of the Santa Catalina potato variety
with subsequent standardization of the growing season
between regions; and
3. Gradual implementation of the year-long planting
seasons due to the Santa Catalina variety and
abandonment of traditional seasons and varieties.
Ramifications of the variation in price between major market cities are discussed below in reference to storage with interregional transport.
Forms of Storage and Costs
Potato storage research in Ecuador has concentrated upon methods utilizing chemical germination inhibitors [20]. Major emphasis was placed upon the determination of potential length of storage using different inhibitors and the costs associated with each. Although alternative storage methods have been reported in conversations with

Table 8. Correlation (r) matrix for monthly wholesale price of
potatoes, selected cities, Ecuador, 1962-72.
Period 1962-1969 (n=600)
Market Tulcan Quito Guayaquil Ambato Cuenca
Tulcan 1.00 0.75 0.75 0.38 0.50
Quito 1.00 0.93 0.37 0.75
Guayaquil 1.00 0.43 0.70
Ambato 1.00 0.12
Cuenca 1.00
Period 1971-1972 (n=120)
Market Tulcan Quito Guayaquil Ambato Cuenca
Tulcan 1.00 0.98 0.97 0.97 0.89
Quito 1.00 0.99 0.99 0.86
Guayaquil 1.00 0.99 0,85
Ambato 1.00 0.88
Cuenca 1.00

potato specialists, only one of these has been observed by the author, and none have been described in the literature. According to some technicians, there is widespread use of open pit storage in fields where the potatoes are harvested. This may occasionally be in a small depression only, where the potatoes are covered by a mulch composed of dry brush and dead potato plants. Potatoes stored in such a pit, however, are not for sale since they will be used for seed as soon as growing conditions permit. Thus, pit storage seems to be primarily a method to avoid backhauling seed potatoes to the planting site.
Farmers in the Cayambe region north of Quito reportedly store potatoes in caves excavated under carefully selected hillocks, where surface water can be directed to the ground surface over the caves and allowed to filter down into the cave roof and walls, thus cooling and moisturizing the storage chamber. Although no such storage chambers were observed by the author, reliable sources testify to their existence, but the amount of potatoes stored in such caves was indeterminable.15
A third storage method reported to be popular among Carchi farmers is to store potatoes in a loft over the house or kitchen (i.e., soberado) for seed and household consumption, although the potatoes may also have a limited impact on commercial supplies.
Francisco Citelly, agricultural engineer at the University of Nariho in Pasto, Colombia, recently designed and tested a low-cost, forced night air storage facility for potatoes in the Colombian highlands [19].
15Portilla Rocha, Fabian, Director. National Institute of
Agricultural and Livestock Research (INIAP), Quito, Ecuador, personal conversation, 1974.

Where the required number of nights with temperatures below 80 C are present and where relative humidity remains near 80 percent, potatoes may be stored in the silos up to four months with only minimal weight loss (1.5 percent of product value at time of sale from storage) and with no appreciable deterioration in appearance. The costs associated with a four-month storage period, where holding charges total S./9.71 per quintal, are described in the Appendix. Utilizing the rounded figure of S.110.00 per quintal, the four-month cost can be broken into a monthly storage charge, to estimate the monthly amounts as follows:
1st month storage + in/out costs
40 percent of S./10 S./4.00
Each additional month
20 percent of S.110 x 3 mo S./6.00
Total for four-month period S./lO.00
There seem to be no basic differences in climate at the recommended altitude for storage in Ecuador. Wherever these conditions and the building specifications are met, potato storage could be instituted for periods up to four months in duration at an approximate cost of S./4.00 per quintal for the first month and S./2.00 per quintal for successive months. For analytical purposes, these data are used as bench marks.
National Storage Potential
Analysis of historical price behavior allows some inferences about the future price situation for potatoes and the feasibility of storage at the national level. The monthly wholesale price average for potatoes in constant 1970 sucres for a period, of 18 years was analyzed (Table 9). By comparing any base month in the left hand column with any subsequent month indicated in the top row of Table 9, one may determine the average

Table 9. Average monthly wholesale price differential between subsequent months for potatoes in Ecuador,
1954-72, in constant 1970 sucres per quintal
Base month Subsequent months
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
January Jan 2.17 4.72b 2.67 -2.56 -0.22 -0.28 2.83 6.45a 7.95a 6.00a 3.27
February Feb 2.22 -0.28 -5.06b -2.66 -2.77 0.33 3.95 5.45a 3.50 -0.77
March Mar -2.06 -7.28a -4.78b 5.00b -1.89 1.72 3.22 1.28 -1.17
April Apr -5.22b -2.72 -2.94 0.72 3.78 5.28b 3.33 0.61
May May 2.50 2.50 5.39a 9.00a I0.50a a 5.83a
June Jun -0.22 2.89 6.50a 8.00a 6.06a 3.33
67a a a2
July Jul 3.11 6.72 8.22 6.28 3.56
August Aug 3.61 5.11b 3.17 0.44
September Sep 1.50 -0.44 -3.17
October Oct -1.94 -4.67
November Nov -.2.72
December Dec
aSignificantly different from base month mean price at .05 level (i.e. 5.36+). bSignificantly different from base month mean price at .10 level (i.e. 4.22+).

constant price differential between the two months. For example, the average constant price differential between January and M1arch is shown as S./4.72 per quintal. The price differentials which demonstrate a measurable degree of variance at two separate levels of significance have been outlined in Table 9.
From this analysis of monthly price differentials for potatoes, additional confidence is generated to support the observation that a significant difference exists between the low prices in May, June, and July and the rising postharvest prices in the later month of September, October, and November. If storage is feasible at all on a regional basis, this analysis then suggests that the positive and significant values in Table 9 depict periods when the storage potential is greatest. Negative and significant values further document price variability but depict the downward movement in prices when storage would only aggrevate the problem.
However, potential national short-range storage based on the
indicated price differentials is not encour aging (Fig. 6). In spite of the "statistically significant" variance in average national monthly potato prices, their range from S./75.00 in May to S./85.00 in October does not appear to offer sufficient margin to stimulate investment in storage, which is estimated to cost at least S./1O.00 at 1972 prices per four-month period.
Aggregate potential profit margins for potato storage at the
national level, again based on 1954-72 data, further suggest a pattern for when storage is needed and potentially feasible (Fig. 7). The difference between the wholesale price and the purchase price with

S./qq Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
p I I eI I *
Natinal ontly woleale priceofptoe 80O
tI t !I p
Solid line indicates potential uninterrupted _ _-short range storage. a
Broken line indicates potential storage for
long range storage conditions only.
Figure 6. Potential short range storage based upon significant monthly wholesale price variance of
potatoes in Ecuador for period 1954-72.
aShort range storage limited to maximum four month period, in forced, night-air sheds.

July through October I
2 I
January through March ~~I ,''
II/ ,
// j,
/ # August through October
May through August
I June through September
-2 \/
-3 I
J F M A M J J A S 0 N D
Figure 7. Estimated average profit margins for stored potatoes, Ecuador,
Source: Calculated from Appendix Table A-8, A-il, A-21 and Fig. A-I. Note: The profit margins were calculated as the average constant
wholesale potato prices less value of potatoes at harvest
price and their accumulated storage costs. They are shown in
sucres (s./) per quintal.

storage costs included produces a series of positive and negative margins dependent upon the length of storage. The greatest storage profit potential apparently lies between the months of July and October, although. potatoes stored through October to November .remain a profitable venture.
For determining storage feasibilities, however, it is important to question the validity of measuring national price variation from deflated price averages based on time series data. Although the averages indicate long-run conditions, and help to reduce the effect of short run fluctuations, they obscure the actual price situation faced by producers and consumers on a week to week basis. Nor do the national average prices. accurately reflect local, provincial and regional conditions, where temporary differences in supply and alternate demand schedules based upon local varietal preferences may be incorporated into the price structure. The effect of these differences and others, such as regional differences in yields, supply of labor and transportation at hardest, and factor input costs are implicitly examined in subsequent sections of the analysis. However, to complete an examination of potato storage potential on a national level, and thus determine the degree of incentive for application of the analysis to a less aggregate local level, the non-deflated price series for potatoes was analyzed.
By determining historical periods based on real prices when storage would have been economically feasible, a storage pattern was identified which more accurately depicts the impact of time differentials on storage needs at the national level. Using only those periods in which storage could have profitably occurred, the national storage potential was calculated for each year of a representative 10 year span which

included three complete price cycles (Fig. 8). Using Citelly's low cost storage estimate of S./10.00 per quintal as a bench mark, each period of storage was allowed to "creep" forward as long as the subsequent prices exceeded the storage costs.
Although the analysis is based upon perfect previous knowledge of price movements, the results even under these ideal conditions are not supportive of the popular belief that potato storage is a business venture that would guarantee consistently high profits. With perfect previous knowledge of price movements, the data indicated a difference of only S./20.00 between the average low and high prices, for an average storage period of approximately three months (Table 10). When the estimated costs for storage during each time period are subtracted from the average price differential, very little remains as an incentive, much less as a margin or hedge, against unforseen changes in price.
Thus, even though month to month price variability appears to be
significant on a national basis, the margins for profitable storage appear to be insufficient for a national storage program. This does not preclude the possibility that regional storage could be feasible due to the unique characteristics of regional demand and supply which may negate potential benefits of interregional transport and national storage of potatoes.
Provincial and Regional Storage
Although calculation of potential storage profits on a national basis indicated only limited prospects for.successful commercial involvement in such a venture, the positive margins did show that perhaps greater potential exists at either regional or local levels.

]-Potential storage periods 90
1962 1963 1964 1965 1966 1967 1968 1969 1971a 1972
Figure 8. Wholesale potato prices showing potential storage periods, Ecuador, 1962-72.
aPrice data for 1970 are not available.

Table 10. Average price'differential in sucres per quintal between low- and high-price periods for potatoes, Ecuador,
1962-71, and profit margin after storage
Low pricea High priceb Price Period of Constant price Storage cost
Year Range Average Range Average Differencec storaged difference for length Profit margin
(1970 = 100) of storagee
---------------- sucres per quintal ------------------- months ------------- sucres per quintal--------------1962 37-44 40 51-60 56 16 3 25.20 6.71 18.49
1963 51-58 55 66-69 68 13 3 18.95 6.71 12.24
1963 69-69 69 80-80 80 11 4 16.03 8.38 7.65
1964 55-63 59 66-73 70 11 3 15.30 6.71 8.59
1965 39-39 39 49-50 49 10 4 13.48 8.38 5.10
1965 44-65 50 65-80 72 22 3 29.65 6.71 22.94
1966 60-60 60 74-79 76 16 3 19.88 6.71 13.17
1968 46-51 49 58-64 61 12 4 13.62 8.38 5.24
1968 46-64 54 58-87 73 19 3 21.57 6.71 14.86
1969 74-81 77 85-100 94 17 3 17.56 6.71 10.85
1971 29-40 35 46-53 49 14 3 13.15 6.71 6.44
1971 41-85 56 61-108 90 34 3 31.92 6.71 25.21
aAverage low price for potential "into storage" period.
bAverage high price for potential "out storage" period.
CSimple difference, between high and low limited to four month maximum spread.
dFigured from first month "in storage" to middle month of "out storage," with minimum S./10 difference.
eFirst month of storage 40% x S./8.38; each additional month add 20 percent of S./8.38, which is cost of storage in constant (1970) sucres.

Provincial Potentials
The maximum allowable investment in storage was calculated for
provinces in the Central Region as an indicator of storage profitability at the regional rather than national level (Table 11). This approach involved inclusion of. production costs to specify a production-storage feasibility. The only region with sufficiently accurate production cost data for this analysis was the Central Region. Separate estimates of production costs were used in order to include a wide range of levels of technology and farm size.16 Alternative I represents a technified, highly-mechanized system, with heavy chemical inputs [61]. Alternative II is based upon actual field surveys made by Arevalo in 1971 in the Central Region for medium size farms [8]. Alternative III is based upon a 1962 experimental plot set up in Pichincha province by Torres, and it offers a basis for comparison with the other models [73].
To determine the average storage potential by province, the market
price for potatoes in each province was calculated over the 1962-72 period by averaging the available market city data. The prices for Pichincha and Cotopaxi provinces were calculated by averaging the price series for Quito ana Ambato. The Tungurahua prices were established from the Ambato price series. Thus, a series of wholesale prices was generated for a 10 year period (1970 was omitted due to lack of data) for each province.
To determine the amount of potatoes marketed each month, Arevalo's survey of potato farmers indicating their preferred month for planting, was utilized [8]. Seven months were allowed for the planting-to-market
A detailed description of each cost estimate may be found in the Appendix, Tables A- 14, A-15, and A-16.

Table 11. Estimated maximum allowable investment in potato storage
by province in the Central Region, Ecuador, 1962-72, in
sucres per quintal
Cost of production Pichincha Cotopaxi Tungurahua
I Schwartz 4.27 10.88 4.58
II Arevalo 9.34 15.07 .9.56
III Torres 5.42 11.83 5.71

period, and the estimated amount of potatoes marketed in each month was calculated by multiplying the percentage of preference for a given month by each year's estimate of potatoes harvested. Thus, a series showing the estimated amount of potatoes marketed every month in each province was generated for the 1962-69, 1971-72 periods.
The market price series was multiplied by the 'monthly production
(converted to quintales) to determine a total monthly expenditure, which was then converted to a measure of estimated annual expenditure for potatoes by province to calculate the average annual prices.
The alternative costs of production per hectare were deflated to
produce a 10 year series and were then converted to cost per quintal by dividing through with each year's yield, by province. Thus, a real-costof-production series was established for each alternative and for each province in the Central Region over the 10 year period. Maldonado's estimated cost of distribution within the region (from the farm to wholesale level) was deflated and added to each year's estimate of production costs [44]. Subsequently, the combined production and distribution costs for each year, province, and alternative cost of production were subtracted from the average annual price. The constant amounts calculated for each province and year were averaged, and the regional potential for storage was thus determined for each alternative 17
cost of production.
17The procedure outlined above to estimate the maximum amount which could be invested in storage is described mathematically in Chapter II.

The estimates of the maximum amount which could be invested in
storage by each province are shown in constant 1970 sucres. The range of allowable investment in storage from S./4.27 to S./10.88 per quintal, utilizing Schwartz'.cost of production estimates, tends to corroborate the analysis at a national level. After subtracting storage costs, very little profit would remain as an incentive for private commercial interest in potato storage, at least in the Central Region. Accurate production and distribution cost estimates for the Northern and Southern'regions were not available at the time this study was made, however, their introduction into the model might produce greater storage investment incentives, especially in the Southern Region where production costs are assumed to be lower due to surplus labor conditions. Regional Potentials
Although regional storage profit margins appear too low to stimulate private investment in storage, the costs and benefits associated with a program which would stabilize the supply were calculated. The volume of potatoes and length of storage in each region required to stabilize the monthly supply were estimated using the survey in which producers indicated their preferred month for planting. Assuming a constant monthly 18
requirement of 8.33 percent of annual production, a storage schedule for each region was drawn, showing the percentage of potatoes which
18The monthly requirements are not actually constant, due to the increased seed requirements during-the heavy planting months; however, increasingly the seed is being supplied from specially inspected farms which produce only certified improved seeds.

must be stored each month in order to maintain a constant supply.19 To achieve a stable supply at the least cost, the use of Citelly's forced night air storage system was required, thus limiting the storage period to a maximum of four months. Figures 9, 10, and 11 show the percentage of annual production and length of storage required to stabilize the supply for each region. In constant 1970 sucres, such a program would have cost an estimated S./5.00 to S./9.00 per quintalin 1972, depending upon the region, with a total storage requirement of 70,797 metric tons (Table 12). However, at no time would monthly storage requirements have exceeded 16,790 metric tons (during a maximum
4 month storage period) in any region, which was only 3.6 percent of the national annual production. The monthly quantities of potatoes moved into and out of storage in each region and the estimated storage costs for each quantity are shown in the Appendix, Tables A-25 through A-30.
When the benefits were calculated for a "typical year" by
subtracting the total revenue of the stored potatoes at the harvest price from the revenue produced by holding them for later sale (in order to stabilize the potato supply to consumers), the benefits were small, and in the Central Region, a negative value appeared (Table 12). These values would have been higher if the requirements for a completely stable monthly supply and the maximum four month storage period had been relaxed. Given these requirements, in a typical year, the expected profit from
19Mannarelli's stud), indicates that an average of 70 percent of the annual potato production was marketed in 1973[46].

Month/% annual Storage Month/% annual
production (percent annual regional production) c)nsumptionb
May May
(11) -(8.3)
Jun Jun
(15) (8.3)
J ulJu
(11) (- (8.1)
Aug Aug
(8) (8.3)
Sep6 66. Sep
(5) (8.3)
Oct Oct
(5) .(8.3)
Nov Nov
10) (8.3)
Dec 3 Dec
(10) (8.3)
Jan Jan
(10) (8.3)
Feb Feb
(5) 6 (8.3)
Mar 6 Mar
(3) (8.3)
Apr Apr
(7) (8.3)
Figure 9. Estimated percent of annual potato production required to
stabilize monthly supply, Northern Highlands, Ecuador, 1972. aThese estimates of monthly production are based upon moving three month average of preferred month for planting (Appendix Table A-24) with six months added for the production season from planting to harvest. bThis estimate assumes constant demand throughout the year for potatoes for consumption in all forms, i.e. food, seed, silage, industry.

Month/% annual Storage Month/% annual
productiona (percent annual regional production) consumptionb
Nov Nov
Dec Dec
(ii). (8.3)
Jan Jan
(10) (8"3)
Feb Feb
(7) (8.3)
Mar Mar
(6) (8.3)
Apr Apr
(7) (8.3)
(7) -- 2.o),
May May
(8) ( (8.3)
(9) (8.3
Jul Jul
(9) ( (8.3)
Aug Aug
(7) (8.3)
Sep Sep
(7) .3(8.3)
Oct Oct
(8) (8.3)
Figure 10. Estimated percent of annual potato production required to
stabilize monthly supply, Central Highlands, Ecuador, 1972. aThese estimates of monthly production are based upon moving three month averages of the preferred month for planting (Appendix Table A-24) with six months added for the production season from planting to harvest. bThis estimate assumes constant demand throughout the year for potatoes for consumption in all forms, i.e, food, seed, silage, industry.

Month/% annual Storage Month/% annual
productiona, (percent annual regional production) consumptionb
Nov Nov
Dec Dec
(13) 0.o (8.3)
Jan .Jan
Feb (7) (8.3)
Mar Mar
(6) (8.3)
Apr Apr
(9) (8.3)
May May
(il) (8.3)
Jun Jun
(12) (8.3
Jul Jul
Aug Aug
(3) (8.3)
Sep Sep
(3) 0.3)
Oct Oct
(6) (8.3)
Figure 11. Estimated percent of annual potato production required to
stabilize monthly supply, Southern Highlands, Ecuador, 1972. aThese estimates of monthly production are based upon moving three month averages of the preferred month for planting (Appendix Table A-24) with six months added for the production season from planting to harvest. bThis estimate assumes constant demand throughout the year for potatoes for consumption in all forms, i.e. food, seed, silage, industry.

Table 12. Estimated regional costs and benefits for potato storage, Ecuador, 1972
in constant 1970 sucres
Region Quantitya Total Unit Cost Total Unit Net
stored cost benefit benefit benefit
MT S./ S./MT S./QQ S./MT S./QQ S./QQ
Northern 14,954 2,734,640 183 8.41 56,820 3.79 .17 -8.24
Central 26,173 3,251,290 124 5.71 -927,080 -35.42 -1.63 -7.34
Southern 29,670 5,898,090 199 9.14 2,853,920 96.19 4.43 -4.71
Total 70,797 11,884,020 168 7.72 1,983,660 28.02 1.29 -6.11
aCalculated from monthly totals, Appendix Tables A-28 through A-33. bCalculated from constant average wholesale prices (Appendix, Table A-9) using Tulcan prices for the Northern Region, Quito and Ambato prices for the Central Region, and Cuenca prices for the Southern Region.

holding the potatoes for later sale will not offset the costs associated with such a program. Although profits might be generated in some years, they would be discounted in subsequent years of losses.
Also, the probable size of the subsidies which would be required of a government-finanqed plan to stabilize supply through the use of a low-cost, forced night-air storage program can be inferred from Table 12. For example, such a program in the Southern Region, where benefits are highest, would cost at least S./9.14 per quint al, less S./4.43 per quintal in benefits, leaving S./4.71 per quintal as a loss to be covered by a subsidy.
National Reallocation: Storage and Transport
The effects of allocating potato supplies equally among months on a national rather than regional basis was determined using the same approach as that used for the separate regions. Quantities were totaled across regions for each month to determine the periods of shortage and surplus, and a storage scheme was designed to standardize the supply for consumers (Appendix,. Tables A-34, A-35). A total of 51,760 metric tons was required for storage, at a cost of S./10,369,510 or S./9.21 per quintal.
Although the total cost calculated at a national level is lower than the total of the regional costs at S./11,884,020, several important ,differences between national and regional storage must be emphasized. This national approach ignores the transport costs necessary for a plan which would stabilize the potato supply on a national basis, and does not allow for the-limiting effect of specific regional varietal

preferences. Thus it may be assumed that storage at a national level would not be considered an economically viable alternative to regional storage. However, it does furnish a more accurate estimate of national costs then heretofore specified.
Another possibility would be to combine storage and transportation in order to more adequately assess a national program to stabilize potato supply throughout the country.20 The apparent regional variations among local and national monthly wholesale prices, at first observation, suggest a potential for storing potatoes and shipping them among the larger cities (Appendix Table A-10).
However, closer analysis reveals a set of economic conditions which might limit commercial interest in such a venture. Citelly's estimated storage cost of S./10.00 per quintal, together with the estimated transportation costs between major cities and prices for 10 years were used in the analysis. Tulcan was selected as the origin of production and transport due to its relatively high average price differential of 17 percent below the national average price. The periods when storage could have economically occurred were calculated together with an indication of the price differentials between Tulcan and other major cities during the storage periods, with both storage and estimated transport costs taken into consideration (Table 13).
The results indicate a series of nmargins in each city which may not be sufficiently high to cover the risks involved in such a venture. The highest average margin is 20 sucres per quintal in Guayaquil, where
20 Table A-22.
For transportation costs see AppendixTalA-2

Table 13. Estimated margin for potatoes purchased and stored in Tulcan and later transported to selected
major cities for sale to wholesalers, Ecuador, 1962-72
Year Quito Guayaquil Ambato Cuenca
Storage Price spread Storage Price spread Storage Price spread Storage Price spread
period high low period high low period high low period high low
(mos.) (constant S./QQ) (mos.) (constant S./QQ) (mos.) (constant S./QQ) (mos.) (constant S./QQ)
1962 13 38 8 58 is 41 9 54
1962 --........ 7' 6
1964 10 29 -- 13 46 27 42
1965 9 32 10 47 ........
1966 ............ 6 .53
1967 8 23 8 33 9 28 29 57
1968 10 34 7 44 9 33 -1969 7 22 ..........
1971 8 19 13 44 13 41 13 23
1972 12 26 ............
Average 9 28 10 45 11 28 11 48
& storage -17 -25 -21 -31
Margin 11 20 17 17
Source: [79].

losses due to spoilage on the shelf are higher th an other markets. These losses are due to the coastal equatorial climate and the perishable nature of the product which produces a high price differential between the Sierra
and the Coast. The actual margin would be even smaller than the average figures indicate since the calculation of the storage/transport periods was made for only those periods when subsequent prices exceeded storage and transport costs, thus assuring a positive margin. In many cases, a storage period longer than four months (Citelly's low-cost tirhie limit) was required to secure the greatest price advantage possible. For potatoes sold in Quito, the length of storage period needed averaged nine months, while for Guayaquil the period averaged 10 months. Storage periods of this length are far beyond the four month capability of the forced night-air storage facilities upon which the storage cost estimates were based. Beyond a four month storage period, losses would become increasingly frequent. Finally, "perfect" storage/transport periods were calculated for Tulcan, the city with the greatest negative average monthly price differential from the national average. The profit margins in other cities would be even smaller, and in some cases (Cuenca) nonexistent.
As mentioned above, the regional preferences for particular potato varieties would affect the feasibility of shipping potatoes from a low price to a higher price region. Monthly price data for Grade A potatoes in several cities were used, irrespective of the varieties they represented, because no historical data were available regarding varietal price discrimination. However, a visit to any regional marketplace reveals that strongly established "tastes" do prevail and are manifested

through substantial price differences among varieties [30, 44]. Thus, the margins are unrealistically high for the Central and Southern provinces where the Northern varieties .(Violeta and Curipamba) are not really accepted.
Summary of Storage Potential
An examination of national monthly wholesale potato prices over an 18 year period revealed a statistically significant difference between the seasonally low prices of May, June and July, and the subsequently higher prices in September, October and November. The maximum average constant differential of S./1O.00 per quintal does not appear to provide sufficient margin to stimulate commercial investment in storage on a national level.
Aggregating potential profit margins from potato storage based upon the difference between the average constant wholesale prices when potatoes would go into storage and prices when they would come out of storage gives the maximum potential returns to be secured nationally from a storage program. However, such a method of calculation may obscure the actual price situation faced by producers and consumers in local markets. Using historical prices, a national difference of only S./20.00 per quintal between average low and high price periods was encountered. With storage costs removed, little would remain as an incentive or hedge against unforseen changes in price.
Calculations at a regional level provided only-marginal additional incentives. The maximum allowable investment in storage was calculated for the Central Region, producing a potential storage investment

ranging from S./4.27 to S./l5.07 per quintal depending upon the province and cost of production estimate.
When calculations were made to determine the costs and benefits associated with a storage program designed to stabilize annual supply of potatoes, the estimated costs exceeded the measurable benefits in every region. Thus, subsidies would be required if such a program were instituted.
Finally, the alternative of reallocating potatoes between regions to take advantage of regional price differences, was examined in combination with the required amounts of short range storage. However, the length of storage required to fully exploit the regional price differentials exceeded the physiological limitation of the low cost storage system. Where the supply was stabilized on a national basis within the limitations of the storage system, the total cost was less than the combined total cost for the regional estimates. Yet neither transportation costs nor regional varietal preferences were included both of which reduce the potential effectiveness of a national program.
Production Management Potential
Storage is only one alternative to a series of activities which could be employed to stabilize the price and supply of potatoes. Another alternative would employ production management techniques to "smooth" production throughout the year. The success of such a venture would. be heavily dependent upon an accurate analysis of the economic, social, and physiologic conditions which regulate. current production. Some basic knowledge of these conditions was needed prior to the formulation of a plan to change current practices..

To accomplish this, a survey of potato farmers was designed, in
cooperation with the Ministry of Agriculture's Tuber Division, to measure current levels of production and farmer attitudes in the Northern and Southern regions of the country. The Central Region was omitted due to lack of personnel andfunding, and the availability of Arevalo's potato survey, conducted just one year before in the region. Small, medium, and large size farm owners were interviewed to determine their attitudes towards seasonality and to identify the degree and underlying cause of production seasonality in each region. Preferred Month for Planting
The survey indicated only minimal differences between regions in the preferred month for planting potatoes. Farmers in the Northern and Southern Regions appear to prefer a similar planting schedule and both 21
contrast with the schedule preferred by Central Region farmers. The similarity between the Northern and Southern regions is apparent when the preferred month for potato planting is charted by farm size and region (Table 14). The preferred month for planting in the Central Region appears to be more stable throughout the year than in the other two regions and increasingly more stable as size of area planted to potatoes increases. Apparently weather is a primary factor in creating this stable condition (Fig. 3 in Chapter III). A short,.dry period in the Central Region during the summer months disrupts continuous plantings yet two such dry periods affect the Southern Region. The strong
21This may in part be due to sampling bias inherent in the two separate, independently conducted surveys (Arevalo and Wiegand).

Table 14. Preferred months for planting potatoes by region and size of planting, Ecuador, 1972 and 1974 Southb Central Northc.
Montha Small Medium Large Small Medium Large Small Medium Large
-- % ---------------------------- ---- ---Jan 3 0 0 4 7 9 12 11 50
Feb 3 6 5 3 7 5 17 19 0
Mar 14 6 0 9 8 8 14 16 0
Apr 16 28 32 12 9 10 15 9 25
May 31 16 16 20 14 11 22 16 0
Jun 4 0 5 9 9 9 8 16 0
Jul 1 3 0 3 4 7 6 0 0
Aug 1 0 0 9 9 8 1 0 0
Sep 0 3 5 4 5 7 0 3 0
Oct 13 22 11 9 10 7 1 0 0
Nov 4 6 0 12 8 9 3 3 0
Dec 7 0 0 6 10 10 1 I5 25
No preference 3 10 26 ...... 0 3 0
Total 100 100 100 100 100 100 100 100 100
Observations 134 32 19 27 30 23 73 37 4
aCrop size is classified as per Arevalo's survey for purposes of comparison, i.e., Small Farm (0.1-2.9), Medium Farm (3.0-9.9), Large Farm (10 or more) and refers to number of hectares planted in potatoes. bx2 = 41.9 at .01 with 24 d.f.
Cx2 = 24.5 at .43 with 24 d.f.

preference for two separate planting seasons in the Southern region is clearly revealed with the first during April and May, and the second in October.
Number of Months Required
for Crop Maturation
The amount of time potatoes are allowed to remain in the ground prior to harvest (the number of months required for maturation) must be considered in planning a supply management scheme to stabi-lize production, especially if some variation occurs among regions or by size of farm. Farmers in the Central Region reported a seven month maturation period [8]. Southern Region farmers showed some variation in their concept of the amount of time required for crop maturation, with 52 percent of those interviewed preferring a six month period and only 31 percent preferring seven months. 22 Northern Region potato growers indicated a much higher preference for the seven month maturation period.
Premature Harvest: Frequency and Reasons
When questioned whether they occasionally harvested their potatoes prematurely, 61 percent of the Southern growers and 77 percent of the Northern growers indicated that they do harvest their crops prior to the normal period of maturation (See Appendix Tables A-41, A-42 and A-43). In the Southern Region, small farm owners showed a tendency to harvest four weeks early when harvesting prior to maturation, whereas the medium and large size farm owners polled showed a preference for only two weeks
22 For Northern and Southern Region preferences see Appendix Table A-40.

premature harvest. In the Northern Region there was a strong preference for a three week premature harvest, although no significant difference appeared between farm sizes.
Reasons for harvesting the crop prior to its full period of
maturation were varied (See Appendix Tables A-44, A-45 and A-46). A major reason in the Southern Region for the farmers who favored two-week premature harvest was price. Disease, pest, and weather considerations also played an important part in determining the Southern farmers' preference for two-, three-, or four-week premature harvest; however, price remained overall the most frequently mentioned factor, representing 38 percent of the farmers polled. Ten percent of those favoring a two-week premature harvest gave miscellaneous reasons for this practice. This figure often represents the high incidence of premature harvest for purposes of consumption among small farmers. In the Northern Region, price was again mentioned by more farmers than any other reason for premature harvest, although disease, pests, and weather were also important contributors. The main difference between the Northernand Southern Regions was the concern for disease and pests in the Northern area, and with miscellaneous reasons (custom, consumption, crop maturity) in the Southern.Region.
An examination of premature harvesting relative to farm size in the SouthernRegion reveals that medium and large size farmers harvest early primarily for price and price-related reasons, whereas the small farmers list their primary reason as weather, disease, and consumption (miscellaneous). Although the tendency in the North appears to show price as the primary reason for premature harvest among different size

farms, the measure was not significantly different from one farm size to another.
Postmature Harvest: Frequency and Reasons
A large number of farmers will allow their crop to remain unharvested beyond the normal date of maturation if necessary (See Appendix Tables A-47, A-48 and A.-49). In thie Southern Region 68 percent of the farmers polled indicated they had occasionally harvested potatoes postmaturely,. as had 91 percent in the Northern Region. A very significant difference between small, medium, and large size farms was registered by the survey, showing that medium and large size farms are much more likely to consider late harvesting their potatoes than the smaller farmers. In the Southern Region, farmers will apparently harvest up to four weeks late, especially on the large sized farms. The second most common delay in harvest was
*two weeks. The Northern farmers surveyed demonstrated most often a threeweek delay in harvest, especially among the growers on medium sized farms, when conditions werenot conducive to immediate harvest.
Significant differences in the reason for postponing harvest were noted. In Northern and Southern Regions, the primary reason for postmature harvest was for the purpose of conditioning the tubers for use or sale as seed potatoes (See Appendix Tables A-50 and A-51). This appears to be especially applicable to those potatoes held three and four
*weeks. Farmers indicated that weather conditions were another important reason for harvesting their potatoes late. However, price considerations were an important joint determinant in the Southern Region, although they were not as important a factor as in the case of premature harvest.

Weather was the primary concern for farmers contemplating postmature harvest, i.e., bad weather causing inaccessible potato fields, especially in the rich dark soils of the Northern Region.
Thus, the factors which most influence the farmer's decision to
postpone the harvest of his crop are identified for at least two large potato producing regions of the country. Weather is seen to be a primary factor, although the role price plays in its influence is unknown. It appears, however, that the major reason for postinature harvest is for the purpose of "hardening" the potatoes for seed. Role of Credit
Another potential influence upon a farmer's production decisions is the presence or absence of credit and the degree to which it has become integrated into the production-distribution process. The share of total agricultural production credit provided by the government for potatoes relative to other crops, declined from 12.2 percent in 1963 to 4.2 percent in 1970 (See Appendix Table*A-23). The survey of potato farmers in the Sierra indicated that credit plays a larger role in Northern Region potato production (75 percent) than in the Southern Region (50 percent), where its use is limited primarily to large farms (See Appendix Table A-52). In the North credit is widely used by small and medium size farms, contrary to custom in the South.
However, government farm credit officials in the Northern province of Carchi indicated that total potato production is only marginally affected by their lending policies. Given the large number of Northern producers using credit (as measured by the survey), it may be assumed

that "credit" as such may not be from the formal, supervised, farm credit sources, but rather in the form of personal loans, advance purchase arrangements, or supply house credit. Such private arrangements are not easily. incorporated into a government production management policy. Estimated Required Adjustments
in Hectarage
The monthly surplus and deficit production by region may be estimated from data collected in the survey of farmers (Table 15). Thdse estimates indicate that 17 percent of the normal annual potato production in the Northern Region, 8.3 percent in the Central Region, and 18 percent in the Southern Region would have to be shifted between months to obtain a stabilized monthly production of 8.3 percent. Such a program would involve shifting portions of production out of some months and into others. However, no additional land would have to be placed into production that had not already been used, i.e., at no time would the sevenmonth total hectarage planted to potatoes under a stabilized production schedule exceed the maximum total seven-month hectarage planted under the actual, unregulated system.23 Table 16 indicates the changes in production which might have occurred in the Central Region under a stabilized system in which 8.33 percent of the annual hectarage is planted each month.
23This is calculated on a regional aggregate level; individual producers might plant more or less. Note that a similar analysis was.performed for the Northern and Southern Regions as reported in Table 16 for the Central Region (see Appendix Table A-36).

Table 15. Estimated average monthly surplus and deficit potato
production by region, Ecuador, 1974
Month Northern Central Southern
--------------------- percent ---------------Januarya -0.33 -1.33 -5.33
February -3.33 -1.33 -5.33
March -3.33 -0.33 -2.33
April 1.66 2.66 2.66
May 1.66 2.66 4.66
June. 1.66 1.66 3.66
July -3.33 -1.33 -1.33
August -5.33 -2.33 -2.33
September -1.33 -1.33 0.66
October 2.66 -0.33 *2.66
November 6.66 0.66 3.66
December 2.66 0.66 -1.33
aBased upon constant monthly demand of 8.33 percent of the annual production, and moving average of preferred month for planting, Appendix Table A-24. Deficits are expressed as negative percentages.

Tablec 16. Modifications required to :tabilize potato production in Central Region, Ecuador, 1972
Month Actual planting Chango in production
In Out
Has. Has.
January 920 175
February 920 175
March 1,092 43
April 1,447 350
May 1,447 350
June 1,315 218
July 920 175
August 789 306
September 920 175
October 1,052 43
November 1,184 87
December 1,184 87
Totals 13,150 1,092 1,092
aCalculated from Tables 15, A-3, and A-24.