Research techniques and instruments

The study included documental analysis and the collection of field information. The latter was carried out during the last trimester of 2010 through the application of a survey to 90 ejidatarios and owners from four ejidos in the municipality of Atlangatepec, who were recognized as heads of households, as well as five interviews with key informants (three presidents of Ejido Commissaries, the Town Council Secretary, and the Municipal President). The interviews were performed to have a general perception as to how the households in the municipality are supplied with maize for human consumption.

The unit of analysis was the peasant household and the concept of household or domestic unit assumed by the National Health and Nutrition Survey was used as reference to define it, which points out that a household is the group of people who reside in a house and who benefit from a joint income, contributed by one or more members of the household, and who have a head of the household (man or woman) recognized by everyone (^{Instituto Nacional de Salud Pública INSP, 2006}). For the case of this study, the peasant household is also named indistinctly as domestic unit.

Size of the sample and selection of the study area

To determine the number of people surveyed, a sample with maximum variance was taken as reference (with reliability of 95 % and precision of 0.1), from a total number of 662 ejidatarios and 136 owners from the ejidos of San Pedro Ecatepec, Atlangatepec, Villalta and La Trasquila (the only four ejidos in the municipality of Atlangatepec). The formula to estimate the size of the sample was the following:

n=NZ² pq / d²(N-1)+Z²pq

where: n: size of the sample. N: size of the population of study =798. Z: level of confidence of 95 % (value of Z in table=1.96). p: approximate proportion of the phenomenon of study in the population=0.5. q: proportion of the population that does not present the phenomenon of study (1-p)=0.5. d: level of precision = 0.1. The size of the sample was defined as: n=86≈90.

Because it was not possible to have a detailed census of ejidatarios and owners, it was decided to take n=90 and to use the technique of snowballing to select them; therefore, the sample used was not random, it was out of convenience and the results cannot be generalized for the study population (N).

The selection of Atlangatepec as the study area is because it is eminently rural (all the localities have less than 2000 inhabitants); agriculture and livestock production are the main productive activities, where cereal production (barley, wheat and maize) and milk production stand out. Reports from the National Council for Evaluation of the Social Development Policy (Consejo Nacional de Evaluación de la Política de Desarrollo Social, ^{CONEVAL, 2010}) about the degrees of social backwardness make evident that it is a municipality whose population has several limitations to gain access to foods.

Typology of households according to their strategy for maize supply

The study contemplates the generation of a typology of strategies for maize supply that the peasant households from Atlangatepec follow, for which the typology of food security strategies proposed by ^{Appendini et al. (2008)} was used as reference, who determined the existence of three groups of households: auto-supply, semi-supply, loss-making.

It is pertinent to specify that this study assumes as strategies for maize supply the set of behaviors and activities carried out by members of the households with the goal of stocking up on basic foods for human consumption; in particular, maize.

Other variables that were included to characterize the households were: number of members of the household (size of the family) and age of the informants. Sub-groups of households were created based on the sale of maize and the productive diversification of the domestic unit.

For this last variable, three categories of agricultural and livestock productive diversification were created. In order to define each of these, the number of agricultural and livestock products generated by the households were taken into account, using as reference the six most frequent ones found through the field work (Table 1).

Table 1 Categories of agricultural and livestock productive diversification in peasant households. 

Source: authors’ elaboration.

With the software Statistical Package for the Social Sciences (SPSS Statistics), version 21 in Spanish, non-parametric correlation coefficients were estimated to define the degree of association between variables: type of household according to the strategy for maize supply (tipo_hogar), agricultural surface cultivated in the household (superficie), number of members of the families (tam_fam), age of the person surveyed (edad), and productive diversification (div_prod).

Characteristics of the study area

According to data from the National Institute of Statistics and Information (Instituto Nacional de Estadística e Informática, ^{INEGI, 2010}), the municipality of Atlangatepec is located north of the state of Tlaxcala at 2500 meters above sea level. It has a border to the north with the municipality of Tlaxco, south with the municipalities of Muñoz de Domingo Arenas and Tetla de la Solidaridad, east with the municipality of Tetla de la Solidaridad, and west with the municipality of Muñoz de Domingo Arenas (Figure 1).

Source: authors’ elaboration from information by ^{INEGI, 2010}.

Figure 1 Localization in the municipality of Atlangatepec. 

The surface of the municipality is 124.067 km², representing 3.05 % of the total state territory and where sub-humid temperate climate with summer rains prevails (^{INEGI 2010}).

According to the Population and Housing Census 2010, the municipality of Atlangatepec had 7326 inhabitants, which represented barely 0.6 % with regards to the total of the state of Tlaxcala. The population density is 59 inhabitants per km², one of the lowest of the state and significantly lower than the state average, which is 288 inhabitants per km² (^{INEGI, 2010}).

Concerning the conditions of poverty and marginalization, ^{CONEVAL (2010)} reported that 66.7 % of the population in the municipality lives under conditions of poverty, occupying number 26 of 60 municipalities at the state level. The scarcities of greatest intensity in the population are access to social security and food, as well as educational backwardness: 84.6 % of the population lacks social security, 33.5 % of the population has scarcity of access to food (2581 people), and 25 % of the population is under conditions of educational backwardness.

It should be highlighted that the municipality of Atlangatepec occupies the third place at the state level in terms of the scarcity of access to food, only after the municipalities of Nativitas and Tepetitla de Lardizabal. Similarly, it occupies the third place from scarcity over educational backwardness, only after the municipalities of El Carmen Tequexquitla and Zitlatepec de Trinidad Sánchez Santos (^{CONEVAL, 2010}).

Although the municipality of Atlangatepec does not occupy the first places in terms of the population under conditions of poverty, the problems of access to food stand out with regards to the rest of the municipalities of the state of Tlaxcala, situation that is not new for since 2005 the CONEVAL reported that 34.7 % of the population was under conditions of dietary poverty, occupying also the third place at the state level, only after the municipalities of El Carmen Tequexquitla and Zitlaltepec de Trinidad Sánchez Santos (^{CONEVAL, 2005}).

According to the Population and Household Census 2010, 95.7 % of the Economically Active Population was occupied; 36.6 % in the primary sector, 31.1 % in the secondary sector, and almost 31.2 % in the tertiary sector (10.37 % in commerce and 21.62 % in services). Within the economic activities of the primary sector, agriculture and livestock production stand out; agriculture with the production of barley, wheat and maize, and livestock production with dairy production and sheep breeding (^{SIAP, 2013}).

Regarding the incidence of meteorological phenomena that are adverse to agriculture, it is worth to highlight that the perception of the survey respondents agrees with records from the National Water Commission (Comisión Nacional del Agua, CONAGUA). The meteorological station in Tlaxco (the closest to Atlangatepec) shows an inconsistent presence of rainfall in the 2000-2012 period. For example, in the April-August period in 2008 (the one of highest water demand from crops such as maize, since the sowing is normally in April and the maturation in September), almost 800 mm of rain were recorded and barely reached 347 mm in 2012, that is, less than half (Figure 2).

Source: authors’ elaboration with data from ^{CONAGUA (2010} and ²⁰¹³⁾. Note: for the years 2006, 2007 and 2010 there were no data available.

Figure 2 Records of rainfall in the meteorological station of Tlaxco during the April-August period, from 2000 to 2012. 

There is a decreasing trend in the amount of rain during April-August from 2000 to 2012 (Figure 2). The CONAGUA records also show that, for example, in 2009, during the months of April to August, 50 % of the annual rain fell, and the remaining 50 % in the months of September and October, while in 2011 the rains were established in the month of June; 40 % took place in September and October, and that particular year frosts took place during the first fortnight of September (^{CONAGUA, 2010} and ²⁰¹³⁾. Therefore, the inconsistency in the availability of rain water generated an uncertain scenario for the rainfed agriculture that is practiced in Atlangatepec and evidently also for maize production.

Characteristics of the survey respondents

All the people surveyed are male ejidatarios (no women were identified as title holders of the ejido plot), with an average age of 60 years; 53.3 % are 60 years or older; 37.8 % between 40 and 59 years; and only 8.9 % less than 40 years old.

With regards to schooling, 90 % has primary school finished or unfinished, and the other 10 % carried out secondary studies. None studied high school or its equivalent.

The size of the average family was 4.6 members, with a minimum of one and a maximum of 12; 69 % of the households have five members or less.

In terms of the availability of agricultural surface, the 90 households studied farmed 489.2 hectares in 2010, with an average of 5.4 hectares per household. Of these, only 15.5 hectares (3.2 %) were irrigation-based and the other 96.8 % rainfed. Around 60 % of the survey respondents had maximum five hectares in production and, in addition to producing in their ejido plot, 54 % also rented or took at half or a third other production lands. It should be highlighted that all of the survey respondents have their ejido plots certified by the Certification Program of Ejido Rights and Garden Titles (Programa de Certificación de Derechos Ejidales y Titulación de Solares, PROCEDE).

During the 2010 agricultural cycle, 90 % of the households studied sowed maize (81 of 90 surveyed), whether as single crop or together with other crops like barley and wheat. Ten percent only sowed barley and wheat. In addition to agriculture, 57 % reported that they produce and sell this product.

It should be mentioned that all the people surveyed stated that at least one member of the family works part or full time in other activities outside the household, whether as day laborers, in bricklaying, street commerce, and as employees in a factory or shop, among others.

Maize production in the households surveyed in Atlangatepec

Out of the 90 survey respondents, 94.5 % stated that maize production in their productive unit decreased in 2008 and 2009, and this has been a trend since the beginning of the 21st Century. This information agrees with the trends at the municipal level, according to the data by ^{SIAP (2013)}, where in the last ten years (during the period of 2002-2011), maize grain production has been irregular, with a decreasing trend since 2007 (Figure 3).

Source: authors’ elaboration with data from ^{SIAP (2013)}.

Figure 3 Volume of maize production in grain in the municipality of Atlangatepec, Tlaxcala, during 2002-2012. 

Records from the municipality of Atlangatepec for maize production in 2011 and 2012 deserve a special mention. In 2011 the production fell to 970 tons, barely 25 % of the harvest obtained in 2010, getting an average yield of only 510 kilograms per hectare (^{SIAP, 2013}). This situation was caused primarily by the incidence of strong atypical frosts in the month of September of that year. For 2012 the SIAP reports a maize production of more than 8 thousand tons (Figure 3), which is totally outside of the parameters found, so this piece of data must be taken with caution.

According to the survey respondents, the causes for the decrease in production are diverse. Among these, the idea that maize production decreased because of “the times” and the reduction of the application of inorganic fertilizers stands out; at least 80 % of those who responded yes agree that the production decreased in the period signaled.

By “the times” the survey respondents refer to the presence of adverse meteorological phenomena for agriculture, such as drought, frosts and hail. When comparing the trends of the data in Figures 2 and 3, it can be observed that the incidence of rainfall and maize production in the municipality of Atlangatepec have a decreasing trend from 2004 to 2011, although 2008 and 2012 are exceptional in terms of production.

The reduction in the use of fertilizers for maize production is due primarily to the increase of their price, which drives them to apply a lower dose, having a negative effect on the yields.

Strategies of peasant households for maize supply

Of the 90 surveyed in 2009, 88.9 % sowed maize using Creole white seed; the rest didn’t sow this crop, but rather barley and wheat.

Of those who produced maize, 56.2 % pointed out that the production obtained in 2009 was enough to supply the demand of the following year; in several households there were even excesses to sell. 38.8 % stated that the maize produced was not enough to supply the following year, which is why they had the need to purchase maize, whether in grain or tortilla. The purchase was made during every month of 2010, increasing during the second semester, carrying out purchases from neighbors in the municipality of Atlangatepec, although they also reported that sporadically they purchase tortillas in the city of Apizaco (at approximately 20 kilometers from the location of the study area). The remaining 5 % (four ejidatarios) did not respond.

Although 94 % of the survey respondents said that there have been reductions in the maize production, only 10 households (11.1 %) pointed out that they have reduced the maize consumption and that they have substituted it by another product; that is, maize continues to be one of the essential foods of the households’ diet.

The key informants (thee presidents of the ejido commissary, the municipal president and the Town Council secretary) also agreed in stating that the people will not cease to eat maize, and even, as an alternative several households are leaning towards cultivating other cereals, such as barley and wheat that are more tolerant to droughts and frosts, to destine them to livestock feed, such as dairy cows and sheep, and the sale of milk or livestock to purchase foods (among them maize in grain or tortillas) for their upkeep.

The findings make evident that the strategies for maize supply are diverse. If we consider the maize production criteria in the peasant household and the amount of this maize that supplies the demand, there are three different strategies that the survey respondents implement: auto-supply, semi-supply and loss-making households (Table 2).

Table 2 Characteristics of the households, according to type of strategy for maize supply. 

Source: authors’ elaboration with data from field work November-December, 2010.

Auto-supply households

The first group (the largest, with 45 households, equivalent to 50 % of the survey respondents) is made up by those households whose supply strategy is based on producing the maize that they consume directly. The survey respondents did not manage to give concrete data about the amount of maize that they consume in their household and the production reported in 2009 is between 800 kg and two tons per hectare.

These households have 4.8 hectares available in average of arable agricultural surface, with a maximum of 10 and a minimum of 1.5 hectares. The average of the ejido plot is four to six hectares, but ejidatarios were found who are also small-scale land owners, or who in 2009 rented lands or took them at a third, which explains the variability between the minimum and maximum number of hectares available.

The average number of people who reside in these households is 4.1, with a maximum of 12 and a minimum of one. The average age of the heads of households surveyed is 61.6 years. In terms of the agricultural and livestock productive diversity, 46.7 % of the households have a high degree, since they produce maize, barley or wheat; they have livestock, mostly cattle, for milk production. This percentage is the highest compared to the semi-supply and loss-making households.

According to the destination of the maize production, there are two sub-groups of households: 1) those without excess of maize production to sell, and 2) those that destine part of the maize production to selling. The first is made up of 35 households and the second groups the other 10 households.

Semi-supply households

The second group covers 31 peasant households which, in order to supply the maize demand, produce it and in addition purchase it. The survey respondents did not report a reliable piece of data about the amount of maize that is consumed in the household, although they did mention a production between 700 kg to 1.8 t per hectare.

These households have 6.03 hectares in average available of arable agricultural surface, with a maximum of 35 and a minimum of 1.75 hectares. The variability of these data is because, in addition to cultivating the ejido plot, 60 % of the households have lands in small-scale ownership or take them at a third or half.

The average number of people who live in these households is 4.9 with a maximum of 10 and a minimum of one. The average age of the heads of households surveyed is 56.8 years. This group also showed households with the sale of part of the maize produced in the domestic unit, from this that they also identify two sub-groups: 1) those without maize production for sale; and 2) those that destine part of the maize production to selling. The first sub-group is integrated by 23 households and the second group includes the eight remaining households.

The main reasons why maize is sold are because of unexpected events or urgencies (for example, diseases of a family member, purchase of inputs), which require immediate monetary liquidity, or else, to take advantage of the season of price increase of the white maize grain ($2.80 to $3.50 per kilogram in 2009).

Of this group of households, 77 % purchased maize in gran; of these, 75 % purchased primarily Creole maize of white color and the remaining 25 %, yellow and also Cacahuatzintle. The white Creole is the preferred one to elaborate tortillas. In addition to using them for the elaboration of tortillas, the yellow one is used for animal breeding, such as sheep and cattle. The purchases were carried out with neighbors of the same municipality or with producers of the neighboring municipalities such as Tlaxco.

All (100 %) purchased tortillas and the survey respondents stated that, even when they prefer to consume hand-made tortillas, purchasing machine-made tortillas is more frequent, since it is the kind of easiest access (it is close to their households, the schedule of availability is broad and the price lower: in 2009, a kilogram of machine-made tortilla cost $8.00 and one elaborated by hand was worth $9.00).

None of the survey respondents knows if the tortilla they purchased was elaborated with Creole maize, with maize flour or with a mixture of both. It is convenient to highlight that the information about the purchase of tortillas or maize was answered recurrently by the survey respondents with the spouse’s help.

Those who purchased maize (in grain or tortillas) stated that the money for the purchase comes from other economic activities that they are devoted to, fundamentally the sale of cow milk, of the wheat and barley harvests, or from the subsidies they receive, such as PROCAMPO and the Oportunidades Program¹.

Loss-making households

The third group is made up of households that even when they have agricultural land available to produce maize did not do it, so that they had to purchase all the maize they consumed in 2009.

These households have in average 6.5 hectares available of arable agricultural surface, with a maximum of eight and a minimum of 1.5 hectares. The variability in the availability of the agricultural surface is because 60 % of the survey respondents also own small-scale terrains or rent agricultural terrains; the average surface of the ejido plot is four to six hectares. The crops that were sown were barley, wheat and oats. They argued that they did not sow maize because the production of these three crops is less risky and they take advantage of them to feed dairy cows or sheep, with which they obtain monetary liquidity, which allows them to purchase those foods that they do not produce in the domestic unit (among them, naturally, maize). It should be clarified that there are agricultural cycles during which they do sow maize; for example, 80 % of the survey respondents reported that in 2010 they did sow it because the rains began on the month of April.

This group of households is the least numerous and of least agricultural and livestock productive diversity. Only 50 % of the households reached the degree of medium productive diversity, which is based on the production of barley, fodder oats, wheat and cow’s milk. However, all of the survey respondents agreed that productive diversity changes from one year to another, due to the seasonality in the rainy period; that is, if the rainy period is delayed it is likely that maize will cease to be sown and instead fodder crops are sown to sell or to feed cows.

The average number of people who live in these households is six, with a maximum of eight and a minimum of two. The average age of the heads of households is 63.6 years. Of the households, 40 % have livestock (dairy cows, sheep), which is why the maize they purchase they also destine to feeding them.

40 % of these households did not purchase maize in grain, only machine-made tortillas. The remaining 60 % of the households acquired primarily white Creole maize and machine-made tortillas (the price per kilogram of maize in 2009 was $2.80 to $3.00 and the price of tortilla was $8.00). Just as the group of semi-supply household, all of the survey respondents ignore whether the tortilla they purchased was made with Creole maize, with maize flour or with a mixture of both. The survey respondents also resorted to their spouse’s support to respond information regarding the purchase of tortillas or maize.

Similarly to the semi-supply households, the loss-making ones stated that the money used to purchase maize (in grain or tortillas) comes from other economic activities that they are devoted to, among which the sale of cow’s milk and harvests of barley and wheat stand out, as well as the salaries they receive as employees outside the productive unit, and the subsidies they receive (from PROCAMPO and the Oportunidades Program, among others).

When recounting the three types of households according to the strategies for maize supply, the results show that there are substantial similarities and differences in terms of the availability of agricultural land, size of the family, age of the head of the household, and agricultural and livestock productive diversity.

Concerning the availability of agricultural land, the highest average surface is that of loss-making households, but in order to estimate the non-parametric coefficient of correlation with the type of household and the average surface, the data are not associated and this does not agree with the findings by ^{Appendini et al. (2008)} and ^{Díaz (2008)}. That is, the increase in maize supply in peasant households of Atlangatepec does not increase with the availability of agricultural area (Table 3).

Table 3 Non-parametric correlations between five variables of the households surveyed. 

*The correlation is significant at level 0.05 (bilateral). ** The correlation is significant at level 0.01 (bilateral).

Source: authors’ elaboration with data obtained in field work and its analysis through SPSS version 21 in Spanish.

The coefficients of correlation between the type of household and the size of the household’s family, the age of the survey respondent and the productive diversity were lower than 0.3, but, according to estimations with the SPSS software, are significant at 0.05 (Table 3).

The correlation between the type of household (auto-supply, semi-supply and loss-making) and the size of the household’s family was negative and significant at 0.05, with which it can be said that when it increases, the household is prone to decrease the capacity for auto-supply of maize (Table 3).

The value of the correlation between the type of household and the age of the survey respondent was positive and significant at 0.05, allowing to state that when the age of the survey respondent increases, so does the maize supply of the households (Table 3).

The productive diversity of the households and the type of household are correlated positively, that is, if the diversity of agricultural and livestock activities in the household increases (sowing maize, barley, wheat and oats, cows’ milk production), the maize supply for auto-consumption increases (Table 3).

The results give evidence that the strategies for maize supply tend to prioritize food security in the households from as a result of production for auto-consumption and the purchase of maize and tortilla (mostly) in the same localities where the members of the peasant households live. That is, the quality in the access to foods, one of the basic components of food security that ^{FAO (2001)}, ^{Salcedo (2005)} and ^{Torres (2006)} highlight, is favored. However, there are also households (particularly the ones that are loss-making) that prioritize quantity over quality, since they diversify agricultural and livestock production to purchase maize in grain or tortillas, particularly when the availability of rainfall is late and inconsistent.

Prioritizing quantity over quality of maize agrees with the findings by ^{Appendini et al. (2008)} and ^{Díaz (2008)}. However, generating other agricultural and livestock products to guarantee the purchase of maize and tortillas, and purchasing them in the localities where they live, from our point of view is a strategy that also bets on the stability to have access to foods, another of the basic components of food security (^{Salcedo, 2005}); that is, the households do not give up on consuming maize, but if the climatic conditions are not the most appropriate (particularly because of the delay and bad distribution of rainfall), then the decision is made of producing other more precocious crops whose production can be used to feed cattle for milk production, which is sold to generate monetary liquidity for the purchase of other foods for human consumption (including maize) and having another nutritional source available (such as cheeses or fresh milk that is frequently for auto-consumption), so as to contribute with that to stabilizing access to foods for the households.

Based on what is exposed, the peasant households implement strategies for maize supply that use endogenous resources (such as agricultural lands, their livestock and their knowledge in the production of diverse plant species) to navigate the scenarios that are presented each year, which causes for a household to possibly have a strategy for maize auto-supply within a year, but if by the following year there is a delay in rainfall or an early frost takes place, then, stemming from these conditions, the strategy may change.