Jump to... Executive summary Introduction Objectives Food insecurity Survey methodology Role of women Household vulnerability Conclusions Agenda

The objective of this appendix is to report in a more comprehensive manner the information contained in the dataset. The data are analysed by production system and by household participation status in one of two IFAD projects.

Household demographics and characteristics

The communities under study are characterized by wide cultural and ethnic diversity. Entirely indigenous communities such as those in Buena Vista and Suchiquer (ethnically Mam and Chortí, respectively) are contrasted with entirely Ladino communities such as those in San Nicolas and Tituque. In between, we find the NTX communities (two thirds Mam and one third Spanish) and San Martín.

Partly as a reflection of this cultural diversity, the households present different physical and behavioural features. In several instances, the data suggest improvements over time in factors directly related to household sanitary conditions and standard of living, relevant to the concept of nutrition. Some of these changes could be attributed with some confidence to the activities promoted by the projects.

Table 20 presents data on the basic household characteristics. The average family size varies substantially across communities, with NTX communities having the largest families (7.3 members, on average, with beneficiaries reaching, on average, 8.0) and Suchiquer the smallest (5.9).

In Suchiquer, male and female heads both of beneficiary and of non-beneficiary households are characterized by strikingly low literacy rates and educational levels. Only 36% of men and 3% of women can read. This is in sharp contrast to San Nicolas, where the proportions are 66% and 44%, respectively. The data also show a close association between maternal education and child mortality. Communities such as Suchiquer and non-beneficiaries in NTX and Tituque exhibit very low female educational attainments and high mortality rates for children younger than 60 months.

As expected in a resource-poor community such as Suchiquer, the rate of temporary migration among male household heads is extremely high (58%). Most of these men work as agricultural wage labourers in coffee plantations in nearby Honduras. However, the average length of migration is quite short (1.1 months). This type of short but widespread migration for grossly underpaid and demanding jobs reflects the low income-earning opportunities within the community, and the need to meet some minimum yearly income requirement by the household in order to access necessities. In relatively better off communities, including those in the NTX and SL groups, a lower percentage of male heads migrate (27% and 17%, respectively), but for longer periods of time (2.5 and 3.1 months, on average). All communities are characterized by very little migration of women, with Suchiquer and Tituque showing no female migration. This pattern reflects a lack of outside employment opportunities for women.

Table 21 summarizes information about housing conditions. The overall picture in Suchiquer and Tituque is one of consistently harsh conditions and physical destitution. More varied circumstances prevail in San Nicolas, where a relatively well developed infrastructure coexists with very poor sanitary practices.

Communities also vary in the types of building material used for dwellings. Differences between communities often reflect local ethnicity and climate, while differences within communities tend to reflect variation in physical and human endowments among households. The large majority of dwellings in resource-poor and dry-climate Suchiquer are made out of thatch, while in wealthier and rain-abundant communities, such as the ones in the NTX group and San Nicolas, there is a larger proportion of adobe houses sporting tile or metal roofs. Many of the houses in Buena Vista and San Martín are made out of wood and have wooden shingle roofs. About 90% of dwellings have dirt floors; in Suchiquer and Tituque, the proportion reaches 100%.

The majority of families interviewed own the houses in which they live. Buena Vista and San Martín show the highest percentage of home ownership (97%), and San Nicolas the lowest (81%). The number of owners has increased over time in PCUCH communities and decreased in PROZACHI communities.

Indoor water connection can improve food and nutrition security by freeing up women’s time, reducing caloric requirements for water collection and improving sanitary conditions. In the last three years, the PCUCH communities have seen significant changes in the frequencies of households that have water inside or within 100 yards of the home. Virtually no change is reported for PROZACHI communities (except in San Nicolas, which showed a slight improvement).

The data also show considerable improvements over time in the share of households that boil or clarinet their drinking water. In spite of some improvement, the percentage in San Nicolas is still surprisingly low (only 28%), particularly in view of the relatively higher level of education among female heads. This latter information appears consistent with some of the other sanitation-related figures in San Nicolas. In fact, in spite of apparent relative physical and human wealth and proximity to transportation routes and information, San Nicolas shows the worst trends in terms of sanitation-related factors, even in comparison with much poorer communities: fewer households there boil or clarinet their water, the majority lack toilets and a considerable proportion of households still throw their garbage out in the open.

One of the goals of PROZACHI is the introduction of time and resource-saving technology. The project has apparently achieved substantial success in this area. In PROZACHI’s target communities a dramatic increase was observed in the number of households using formal cooking devices. At the time of the survey, 42%, 40% and 50% of households in Suchiquer, San Nicolas, and Tituque, respectively, were using improved stoves (mainly of the Z-mejorada type), compared with only 3%, 17% and 19%, respectively, five years ago. Beneficiaries score generally better than non-beneficiaries. In the PCUCH communities, by contrast, most cooking takes place on open fires on the floor, a situation that has changed little since the beginning of project activities in the area.

In summary:

• In the communities surveyed, migration is used frequently as a hedging mechanism against household shortages of food and basic necessities. As expected, in resource-poor communities such as Suchiquer, a high percentage of households rely on temporary migration to improve their food entitlements. Temporary migration has been associated with moderate food insecurity (Watts 1983; Maxwell & Frankenberger 1992).
• Suchiquer performs the worst in terms of education, housing conditions and child mortality. These results highlight the importance of female education for improving nutritional security.
• Despite its superior human and physical endowments, San Nicolas is characterized by surprisingly poor sanitary conditions and habits. Here is clearly an area in which PROZACHI can effectively contribute to fostering nutrition security.
• PCUCH communities lag behind, compared with those of PROZACHI, in the adoption of time and resource-saving technology, such as improved stoves. Even though the improved types of stoves are often promoted based on their fuel wood and time (cooking time and time for fuelwood gathering) efficiency, in order to provide projects with a solid economic incentive to promote the measure, their introduction can be justified also on strictly health-related grounds. Pursuing food security without consideration of the broader concept of nutrition security is bound to have only a weak effect on the nutritional performance of the most vulnerable elements of the population – namely, women, children and the elderly. Factors likely to erode health conditions must be removed by projects that aspire to improve household nutritional well-being.

Characteristics of the unit of production

Table 7 analyses the characteristics of agricultural production in each of the selected farming systems. Data on land ownership are strikingly heterogeneous, both across communities and between beneficiaries and non-beneficiaries. Households in the SL communities and San Nicolas have higher land endowments than farmers in the other communities, owning on average 55.5 and 48.1 cuerdas, respectively. In the SL group, beneficiaries reported owning more than twice the amount of land of non-beneficiaries (67.0 versus 28.3 cuerdas). In all communities, except in Tituque, around 50% of owned land is cultivated, while the other half is used, with different percentages in each community, for woodland or pasture, or is left fallow. In Tituque, 74% of owned land is cultivated (77% by project beneficiaries), with little land left over for other uses.

The land rental market appears particularly active in Suchiquer, where about 40% of cultivation is on rented land. In all other communities, except in Buena Vista and San Martín, about 10% of cultivation takes place on rented land. Almost no land rental is visible in the SL communities. On average, non-beneficiaries own, cultivate and rent less land than do project beneficiaries. As expected, a significant share of land is used for pasture in SL communities (26%), with beneficiaries reserving a larger share (28%, versus 21% for non-beneficiaries). Households in the NTX communities, on average, leave fallow close to 20% of owned land, while in Tituque the amount is only 5%, suggesting a tighter land constraint there. Very little land is reserved for woodland in Suchiquer, and by non-beneficiaries in the SL communities and in Tituque. In the case of Suchiquer, this situation reflects widespread deforestation in the area, while in the other case it may be more a reflection of the smaller land endowments of non-beneficiaries, who are less able to diversify for lack of land in Tituque, and for prioritizing pastureland in the SL communities.

Table 22 presents data on the use of cultivated land. Although figures are reported for only the main crops, it is surprising how few other crops are grown by a large enough number of households in each community. In all communities, about 50% of cultivated land is dedicated to corn production, with 56% in NTX and 42% being, respectively, the highest and lowest shares. Practically every household grows corn, except in SL, where, because of the high altitude, only about 40% grow it, mainly in lower-altitude fields outside of the communities. In the NTX group, a consistent share of land is dedicated to broccoli production (34%, with non-beneficiaries allocating on average one cuerda more than beneficiaries). Crops such as broccoli, cauliflower and romanesco were introduced in the communities approximately 4-5 years ago by agro-exporters. Over 60% of households reported to have grown broccoli the year before. Beans are usually grown intercropped with corn, and take up about 20% of cultivated land in the PCUCH communities, and 27% in PROZACHI communities, except in San Nicolas, where approximately 50% of land is dedicated to this crop.¹ Potatoes are grown by 95% of households in the Buena Vista and San Martín, and take up on average 60% of cultivated land.² In these same communities, carrots and lima beans are grown by a little over 20% of households on approximately 25% of cultivated land. In Suchiquer, because of the extremely poor soil conditions and water shortages, sorghum is often grown, intercropped with corn, as a substitute for corn in the household diet. As the data show, in spite of apparently unsuitable conditions, corn is grown by virtually everyone, since it remains the favourite staple. Sorghum is grown by almost 85% of households and covers 46% of cultivated land. In Tituque, in addition to corn and beans, coffee is grown by 86% of households. Coffee area covers, on average, 23% of cultivated land, with beneficiaries and non-beneficiaries allocating 38% and 9%, respectively, of the total cultivated area. Bananas are also grown in Tituque by 25% of farmers, covering, on average, 11 % of the cultivated area.

Table 23 reports on changes in land use since the onset of the projects. One of the stated objectives of both projects is to improve households’ socio-economic conditions via the promotion of more profitable crops, with support in both production and marketing. One of the results worth emphasizing is the dramatic increase in the area allocated to non-traditional export crops in those PCUCH communities in which a major effort is under way by the project and several agro-exporters to foster production of new export commodities such as broccoli and cauliflower. In the past three years, these communities have registered an overall increase of 135% in the area allocated to NTX crops, with an even larger increase among beneficiaries (179% versus 73% for non-beneficiaries). Coffee is another commercial crop being promoted by PROZACHI in Tituque. The data show an average increase of 22% in the area dedicated to this crop. Even though the percentage increase is higher for non-beneficiaries (36% versus 23%), in terms of area, beneficiaries are now allocating to coffee 2 cuerdas more than they were five years ago, while the increase for non-beneficiaries is only 1 cuerda.

Finally, Table 1 shows the average corn yield of the different classes in each production system. As expected, Suchiquer communities register by far the lowest yield, as a consequence of very poor land quality and water scarcity, and virtually no use of fertilizers. Figures for Tituque, even though higher, are lower than acceptable standards. Soil depletion and poor access to fertilizer are among the main concerns raised by farmers in these communities.

In summary:

• Non-beneficiaries in all communities own, cultivate and rent less land than do beneficiaries.
• Corn remains overwhelmingly the staple of choice in all communities, with approximately 50% of cultivated land allocated to this crop, and virtually everyone growing it. The only exception is the SL production system, where, because of high altitudes, corn is grown only by those households with access to lower-altitude land in nearby communities.
• Its draught resistance and ability to grow in poor-quality soil have made sorghum the forced substitute choice for almost all households in Suchiquer. However, because of high transaction costs in corn purchases, virtually everybody in the community still grows corn, intercropped with sorghum (even though all they harvest is, often, no more than the seeds they put in). Raising food production is a must in isolated and economically depressed areas such as Suchiquer, where the chance of market integration is much reduced by poor access to transportation.
• Beneficiaries in the SL communities show a significant increase in the area allocated to potatoes (41 %), having expanded cultivation of the crop from 5.1 to 7.2 cuerdas, on average, over the past three years. The project has been instrumental in fuelling improvements in potato cultivation through technical assistance and supply of agricultural credit.

Income level and sources

As noted above, many projects and research efforts have focused on the role of household income in promoting nutrition and food security, while ignoring equally important factors such as income source and its control within the household. This section considers the role of income source, arguing: (i) that in the presence of uncertainty and risk in a household’s market activities, diversification of income sources is an important part of a strategy to reduce the risk of entitlement failure; and (ii) that income diversification is one of the main risk-reducing mechanisms used by households to minimize their probability of entitlement failure, and as such must be accounted for in any measure of vulnerability. The importance of income level is undeniable. However, at these low levels of income and high levels of destitution and food insecurity, diversity in income sources may be even more important than income level in securing the survival of vulnerable household members in the event of an entitlement failure.

Table 4 reports on total household income, desegregated by source, the shares of total income derived from each source, and the frequencies of households with positive income from each source. Ideally, we would like to see shares of about 20-30% for each source (with figures below 10% indicating too little reliance on a source, and those over 50% suggesting excessive reliance), and frequencies of households with positive income from each source as close as possible to 100%. Tables 5 and 6 show, respectively, a further desegregation of off-farm income by household member, and of petty trading income by activity.

Households in the SL communities earn, on average, the highest incomes (Q 8 298),³ with beneficiaries earning about 18% more than non-beneficiaries. The desegregation by source of income for this community reveals heavy reliance on on-farm income (mainly from the sale of own-produced potatoes), especially for non-beneficiaries. About 90% reported to have earned some income from on-farm labour, with non-beneficiaries earning, on average, Q5 362 (representing 73% of their total income) compared with less than Q 4 900 for beneficiaries (56% of total income). Compared with non-beneficiaries, many beneficiaries also derive income from off-farm labour, petty trading and animal sale.

Beneficiaries in NTX communities derive over 60% of their income from the sale of own-produced agricultural products, compared with only 30% for non-beneficiaries, who rely much more heavily on off-farm income (61%, versus 25% for beneficiaries). Assuming that most on-farm income comes from NTX crops, and considering the higher risk involved in their production and marketing, excessive reliance on on-farm income can leave households too vulnerable to the uncertainty of the markets and the vagaries of agro-exporters. Over 90% of non-beneficiary households reported to have worked off farm (versus 75% of beneficiary households).

Among communities surveyed, households in Suchiquer earn the lowest income, on average. The only two sources of income appear to be off-farm labour, mainly by men in the form of temporary agricultural wage workers in coffee plantations in Honduras, and handicrafts, such as straw mats and baskets, made mainly by women and sold (by men) in the local market in Jocotan.

Beneficiaries in Tituque earn, on average, over 80% more than non-beneficiaries (Q 5 139 versus 2 820), thanks primarily to higher on-farm income deriving from coffee production, particularly pergamino, promoted by the project (Table 4). Because of the altitude, climate and soil characteristics, farmers in the area produce among of the best-quality pergamino coffee in Guatemala. Surprisingly, in spite of an apparent dynamism of female beneficiaries in the community, who are primarily involved in handicraft production, beneficiaries earn less, on average, than non-beneficiaries from this source (Q 274 versus Q 4O4 (see Table 5).

Households in San Nicolas often rely on multiple income sources, although on-farm income tends to comprise a considerable share of the total. The most notable difference between beneficiaries and non-beneficiaries in this community is the amount derived from off-farm labour (Q 778 versus Q 2 32l, respectively) and the share of off-farm income in the total, with non-beneficiaries deriving 27% of their earnings from these sources, compared with only 17% for beneficiaries (Table 4). The majority of income from petty trading comes from the sale of handicrafts (75%, on average).

In summary:

• NTX communities rely mainly on only two sources of income, namely, on-farm and off-farm, with beneficiaries relying more on on-farm and non-beneficiaries on off-farm sources. A word of caution must be raised towards the excessive reliance on NTX-crop income. During the fieldwork, many NTX farmers reported that, despite repeated changes over the years in the firms with whom they contract, they continue to receive highly unfair treatment from agro-exporters.Specifically, agro-exporters often reject products due to alleged low quality, or discount excessively for the presumed defects. The NTX farmers have also accused the agro-exporters of charging excessively for the inputs they advance to the farmers, paying late or, as occurred in some circumstances, not paying at all.
• Farmers in Suchiquer draw no income from on-farm labour. The households in this community do not appear well diversified in terms of income sources (only off-farm labour outside the community, and handicrafts), and the income they derive from these very few sources is extremely low.
• In Tituque, beneficiaries were able, on average, to increase their total income thanks to higher revenues from coffee production. Compared with beneficiaries, non-beneficiaries earn little from on-farm labour and petty trading. In spite of ongoing efforts to improve women’s income through handicraft manufacturing and retail, the higher income in beneficiaries’ petty trading does not come from this source but from the sale of agricultural products such as bananas (Q 292 for banana production versus Q 4 for handicrafts) and wild plants used as inputs in handicrafts (Q 267), for which the project cannot be credited.
• Households in the SL communities are characterized by a heavy reliance on on-farm income, particularly non-beneficiary households. Beneficiaries appear more diversified, with more households deriving income from off-farm and petty trading sources. The share of income deriving from animal sales is quite low, perhaps because the communities are still in an accumulation stage of animal (sheep) stocks. The few other small animals raised are used for consumption.
• San Nicolas appears sufficiently diversified, with many households deriving income from two or more sources. An extremely large share of households in this community derive some income from both on-farm labour (sale of corn and beans) and petty trading (sale of own-manufactured handicrafts).
• Women in all communities earn very little or no income from off-farm labour.

Assets ownership

Animals are a highly liquid asset form, compared with alternatives such as land and dwellings, and thus represent an important source of insurance against the risk of entitlement failure following events such as crop failure, the illness of a breadwinner or the loss of off-farm employment. However, disease (especially among fowl) can turn these assets into liabilities; for those overinvested in a single species, an outbreak of disease in livestock can precipitate a food crisis.

Table 8 contrasts animal ownership across production systems and between beneficiaries and non-beneficiaries. Not surprisingly, fowl (mainly chickens and turkeys) are the most common animals. Across communities, between 85 and 96% of households reported to own at least one fowl, with beneficiaries reporting a higher percentage than non-beneficiaries. San Nicolas shows the highest average number of fowl owned (20.0), with significant differences between beneficiaries and non-beneficiaries (21.8 versus 13.9). At the other end of the spectrum are the communities of Buena Vista and San Martín, where the average household owns only 6.4 fowl. These communities seem to favour ownership of pigs over fowl, compared with all other communities. Raising pigs instead of chickens appears a rational choice because of the scarcity of grain (the main food for chickens) and the abundance of potatoes during periods of the year in which pigs are mainly bred. As expected, in Buena Vista and San Martín, sheep are the main form of animal investment (16.8 sheep/household, on average). In spite of the current project efforts in the area – primarily in the form of credit for animal purchases – beneficiaries own, on average, a slightly lower number of ovine units compared with non-beneficiaries (16.2 versus 18.3).

The same table reports the frequencies with respect to the ownership of durable goods. Eighty-one percent of households in San Martín and Buena Vista own a radio, compared with 47.0% in Suchiquer, and 66.6% of households in the other communities. Only two out of ten non-beneficiaries in Suchiquer own a radio, compared with seven out of ten beneficiaries. Bicycles are common only in the SL communities (36%), particularly among beneficiaries (42%, versus 21% for non-beneficiaries). San Nicolas and Tituque register the highest frequencies with respect to manual mill and silo ownership (81% and 48% for mills, and 91% and 50% for silos, respectively). However, the figures support a significant role of PROZACHI only in Tituque, where significant differences are detected between beneficiaries and non-beneficiaries (63% versus 24% for mills, and 63% versus 29% for silos), reflecting the project’s effort to promote the adoption of these technologies. All other communities lag behind, particularly Suchiquer in mill ownership and the SL communities in silo ownership. Even though the percentage of households owning silos in Suchiquer is much lower than that in these other PROZACHI communities (only 14%, on average), the number of beneficiaries owning a silo is almost three times higher than that for non-beneficiaries (19% versus 7%).

Table 9 reports the value (in quetzales) of total animal transactions. A positive value in net purchases (purchases minus sales) reflects an accumulation in the stock of animals. As expected, Buena Vista and San Martín households are accumulating animal stocks, on average, thanks in part to the credit programme funded by the project. Beneficiaries in these communities are deriving twice as much as non-beneficiaries from the sale of animals, while increasing the value of their animal stocks at more than five times the rate of non-beneficiaries. On average, households in Suchiquer show negative net purchases (Q 137), with beneficiaries exhibiting higher values (Q 182, versus Q 73 for non-beneficiaries). Also NTX communities show negative net purchases (Q 186), with beneficiaries exhibiting lower values (Q 174 versus Q 203). Negative values of net purchases can be an indication of divestment in animal assets, of earned income derived for animal-raising, or of both. A differentiating factor could be the share of distress sales out of the total value of sales, where higher shares would likely be associated with divestment. The data on distress sales show that, in about half of all cases, animal sales represent asset divestment, except among non-beneficiaries in Suchiquer, where the entire amount of animal sales went towards payments of basic necessities and emergency needs.

In summary:

• Chickens are the most common form of animal investment, except in the SL communities, where sheep and pigs are more important.
• In the SL communities, beneficiaries are "catching up" in sheep ownership; non-beneficiaries still own more sheep, but beneficiaries are purchasing more thanks to improved access to credit. Association with the project appears to be effective in levelling the size of the herds between beneficiaries and non-beneficiaries.
• The largest percentages of households with silos are in San Nicolas and Tituque. Project efforts must be credited for the diffusion of silos in Tituque, principally implemented through the training of a silo-maker in the community. This result reflects a typical example of efficient and cost-effective diffusion of technology by the project. Silo adoption is also positively related to project participation in Suchiquer, even though the figures are much lower compared with those in the two other PROZACHI communities.
• Only in Tituque does PROZACHI effectively promote time-saving technologies for women, such as manual mills for corn grinding. As described below (in the section on women’s time allocation), the common practice of grinding corn on stone is extremely time consuming, so manual mills can help free up women’s time.
• In San Nicolas, no differentiating role can be attributed to project participation in promoting the adoption of mills and silos, where their ownership appears mainly to reflect a relatively high average higher standard of living, and necessity to preserve grain surpluses.

Access to credit

It is apparent from Table 10 that both projects have made a significant contribution in all communities in improving access to cash credit for their beneficiaries. Virtually no non-beneficiary household either applied or received a loan in either Suchiquer or Tituque. Only one household received credit in San Nicolas, two in the NTX communities and three in either San Martín or Buena Vista, suggesting a near-total lack of access to credit in all communities in the pre-project period. In NTX communities, as well as in San Nicolas and the SL communities, almost every beneficiary applied for credit, versus only two thirds in Suchiquer; of these, most actually received it. In Tituque, about two thirds of the 70% of beneficiaries who applied for credit received a cash loan.

Of the 39 beneficiaries in the SL communities that received credit (87% of the total number of beneficiaries), an average of Q 2 638 was disbursed, with 60% going towards animal purchases and the rest towards agricultural input purchases (Table 11). A closer look at the direct beneficiaries of the loan within the household highlights major differences in the average amounts received by the male household head, when compared with those received by the female head. Even though more women than men received credit in the SL communities, women received on average only 30% as much as men (Q 1 180 versus Q 3 977). A higher percentage of the credit allocated to women goes towards animal purchases than towards agricultural inputs (58% versus 33%, respectively), while men use their credit for input and animal purchases in equal proportions. Credit to women is one of the main vehicles promoted by PCUCH to introduce women to market activities through the purchases of sheep herds.

In San Nicolas, women get even more on average than men for input purchases (Q 1 750 versus Q 1 358). However, only 4 women had access to credit (12%), compared with 25 men (76%). Conversely, in Tituque, more women got credit than did men, but they received much smaller loans (on average, men got Q 1 700 for inputs and Q^o375 for animal purchase, for a total of Q 2 500; women got Q 755 for inputs, for a total of Q 1 164). In this community, 40% of women from beneficiary households got credit, versus only 15% of men. The small size of loans to women reflects thewomen’s lack of collateral. Lack of collateral means little or no access to larger loans.

In summary:

• In all communities, the projects have been instrumental in improving poor households’ access to credit.
• A higher percentage of women received credit in SL communities and in Tituque (64% and 41%) compared with men (30% and 15%), but the amount received on average was much lower (Q 1 180 and Q 1 164 for women versus Q 3 977 and Q 2 500 for men). The results achieved by the projects are remarkable, considering the absence of formal sources of credit for women before the projects’ arrival, but they still reflect the differential treatment of and access by women to productive resources.
• In Suchiquer and San Nicolas the percentage of women reached is still quite low (14% and 12% of the total number of women, respectively).
• The low amount of credit allocated to women for agricultural input purchases in the NTX and SL communities confirms one feared conclusion: when cash crops are involved, agriculture remains within the male domain. The other route, i.e., to create a separate niche such as promoting animal ownership and/or handicraft production among women, seems to be the less conflictive alternative, followed by the projects in communities where cultural norms still grant men control of cash crops and relegate women to the role of helper.

Women’s participation in agricultural and non-agricultural activities

In Table 12, sizeable differences emerge across communities in the number of days women dedicate to agricultural activities. As expected, these differences reflect the varying crop portfolio found in the different communities and the well-defined division of labour by gender associated with each crop.

As reported by many studies from Guatemala, women contribute only marginally to corn production, mainly in activities such as harvesting and cleaning plants. Both beneficiaries and non-beneficiaries allocate similar amounts, with women in NTX dedicating on average 7.3 days/year. On the other side of the spectrum, women in Suchiquer spend only 1.3 days/year in corn production.

Women in NTX communities allocated the highest number of days per year to agricultural activities compared with all other communities. The difference is to the result of the considerable participation by women in the production of NTX crops such as broccoli and cauliflower. On average, women in NTX communities spent 27 days/year in agriculture (with more than one half going into broccoli), compared with 2.4 days in Suchiquer, 6.1 in San Nicolas, 11.3 in Tituque and 16.5 in the SL communities. The relatively high figures for Buena Vista and San Martín reflect women’s participation in potatoes (10.5 days/year) and vegetables such as carrots (11.0 days/year) and lima beans (5.2 days/year). In Tituque, women’s participation in coffee is highest among beneficiaries (7.0 days/year versus 0.5 for non-beneficiaries).

Table 13 analyses women’s time allocation to market activities – namely, manufacturing handicrafts, attending to animals, selling at local markets and attending to stores. Handicraft manufacturing (mainly straw mats and baskets and rope bags) emerges as the activity to which the overwhelming majority of women allocate the most time in both Suchiquer and San Nicolas. Ninety-seven percent of women in Suchiquer reported dedicating, on average, 99 days/year to the activity, with non-beneficiaries dedicating 109 days/year compared with 91 days for beneficiaries. This difference partly justifies the higher income that non-beneficiary households derive from manufacturing handicrafts. In San Nicolas, 86% of women dedicate, on average, 105 days/year, with non-beneficiaries allocating 70% more time than non-beneficiaries to this activity. In Tituque, only 25% of women reported dedicating any time to manufacturing handicrafts, even though they spent the highest number of days (121, on average). Virtually no women dedicate time to handicrafts manufacturing for sale in either the NTX or the SL communities.

As expected, attending to animals emerges as a primary activity for women in Buena Vista and San Martín. Women in these communities dedicate, on average, 201 days/year to this activity. Approximately 90% of women in both beneficiary and non-beneficiary households reported allocating some time to rearing animals. Also, very few women in all communities appeared to dedicate any time to selling in the local market. The only exception is San Nicolas, with 40% of its women dedicating some time to that activity. This is not surprising, given the proximity of San Nicolas to transportation routes and markets. As anticipated, attending to stored foods emerges as a very time-consuming activity for few women.

Finally, Table 14 analyses women’s time allocation to home activities. One of the results that should be emphasized is the sizeable amount of time women dedicate, on average, to making tortillas and cooking. San Nicolas, the community in which women dedicate, on average, less time to making tortillas than do women in the other communities, is also the community with the highest percentage of households owning a mill. The finding supports the idea behind PROZACHI’s effort to introduce time-saving technologies to reduce the time women dedicate to time-intensive home activities such as corn grinding. Except for San Nicolas, women in all other communities still dedicate more than two hours a day, on average, to making tortillas.

Washing clothes is another activity performed by virtually all women generally once or twice a week, and taking from 7 to 10 hours/week, based on factors such as family size and distance to water sources. Also, 80% of women in Tituque reported to have dedicated an average of 6.6 hours/week to carrying water for household consumption. In arid communities such as Buena Vista and Suchiquer, approximately 10 hours/week is spent by women in water gathering. Except in Tituque and the NTX communities, few women reported dedicating time to gathering fuel wood. Among the women who did so, women in Suchiquer and Tituque registered the highest number of hours per week (9.0 and 8.6, respectively).

Large numbers of women recount being responsible for purchasing food outside their communities. This is usually done once a week in the closest local market, and takes, on average, from four to five hours. The exception is Suchiquer, where, because of excessive distances, men are generally in charge of performing this task.

In summary:

• The introduction of time-saving technologies, such as corn mills and improved stoves, can be instrumental in reducing the time women dedicate to home production. However, the data reveal significant savings only in San Nicolas, where widespread mill ownership can be partially credited for reducing the time women allocate to preparing tortillas.
• Project participation significantly improved women’s participation in market activities. However, in spite of their greater involvement in income-generating activities, women continue to allocate many hours per day to home production.
• Adoption of time-intensive NTX crops significantly increases women’s participation in agriculture.

Intrahousehold resource control

Substantial cross-community differences exist in the sample with respect to intrahousehold control (Table 15). SL communities and those such as San Nicolas showed the highest percentages of petty trading income upon which women claimed either partial or total control. With the exception of San Nicolas, women in beneficiary households in all communities appear to retain control over a higher proportion of the household petty trading, compared with their counterparts in non-beneficiary households.

With respect to animal ownership, in the majority of cases, fowl are owned by women in all communities. The shares of pig ownership by women vary substantially across communities, going from 41% in the NTX communities to 62% in Buena Vista and San Martín, to 75% in Tituque. The share of larger animals (cows and horses) under the direct control of women is much lower than that of smaller animals, except for cows in Tituque. Sheep in Buena Vista and San Martín are often considered jointly owned with the husband, even though the women claim a substantial share (38%) as their own.

A considerable share of women in Buena Vista and San Martín (in 31% of the households) reported to be in charge of handling the household savings and the share of household animal stock for which they claimed ownership (70% for fowl and 62% for pigs).

Women in Suchiquer reported controlling only 35% of income from petty trading, in spite of the fact that they are the ones primarily responsible for generating this income through the manufacturing of handicrafts. Women in beneficiary household’s fare substantially better (46% versus 27%). The majority of women in Suchiquer are not able to market their products directly, having to rely on their husbands’ trips to town for the sale of their products. This phenomenon often translates into a partial or total loss of control over the generated income. Only a limited number of project beneficiaries have been able to benefit from a retailing facility set up by the project in the capital to market their products.

In summary:

• On average, the projects appear quite successful in fostering an improved control of resources by women, even in Suchiquer, where women’s control over household resources is still very low compared with that of women in all other communities.
• Underdeveloped marketing channels and poor demand represent bottlenecks, depriving women of earning a fair price for their products.

Food security

As repeatedly mentioned, a more comprehensive characterization of the concept of HFS must go beyond the mere quantification of household food supplies and access. It should also incorporate information reflecting the risk involved in different HFS coping strategies and the likelihood of a household’s being exposed to entitlement failures.

Table 2 analyses the flows of the main staple crop, corn, and a series of other related variables in an attempt to identify those households with the heaviest dependence on food markets. Again, heavier dependence from the food market does not necessarily translate into increased food insecurity. However, a heavy food dependence, if combined with low income levels, poor diversification and limited access to credit, is most likely to translate into increased food insecurity.

It is not surprising that San Nicolas is characterized by few net buyers and many net sellers of corn. The low ratio of food dependency⁴ indicates the limited reliance on the market for corn consumption by households in this community. At the opposite end of the spectrum, we find Buena Vista and San Martín with a corn dependency ratio of 81% and an extremely low number of self-sufficient households (5%). While the low level of self-sufficiency was to be expected, the reliance on purchased corn is surprisingly high in view of the fact that potatoes were expected to take a more important role as corn substitute in the households’ diets. However, as seen in the previous results, the communities also present the highest average income among all those surveyed, and sufficiently diversified crop and income portfolios.

As expected, Suchiquer presents extremely distressing features. In addition to the low income levels, poor diversification and inadequate asset levels, the figures for this community also show high food dependency and low food self-sufficiency, with a dependency ratio of 44% (indicating that, on average, almost half of cereal consumption comes from purchases) and only 14% of the households’ being self-sufficient in cereal.⁵

The NTX communities and Tituque present intermediate features, with both exhibiting an overall dependency ratio of 29% and with beneficiaries in the NTX communities relying less on purchases and those in Tituque relying on them more, compared with non-beneficiaries. About one quarter of households in the NTX communities reported being self-sufficient in corn, compared with almost one half in Tituque.

Finally, Table 2 reports information on the households’ eating habits, by presenting the frequencies of households consuming meat at least once a week. Animal-rich communities, such as Buena Vista and San Martín, exhibit the highest frequencies (74%), with beneficiaries performing even better (77%, versus 68% for non-beneficiaries). Many households interviewed reported raising small animals such as chickens and pigs for own consumption. Few also reported consuming sheep meat from their own herds. Once again, even though these communities showed the highest degree of dependency from the corn market, they also exhibited some variety in their diet (corn, potatoes and meat). Had a temporary shock in the corn market occurred, with other sources of nutrients to fall back on, most of these households would have been able to cope sufficiently well.

Table 3 analyses some measures of diversification, as proxies for a household’s exposure to risk of entitlement failure. Entitlement failure can occur because of shocks in agricultural and food production, work and assets. Diversification is the cornerstone of risk management for poor rural households in conditions of high uncertainty and lack of formal insurance markets. The likelihood of a simultaneous shock in more than one source of food entitlement is low, particularly if the risks are not correlated.

To construct the income diversification indicator, the sources contributing to total income are classified into on-farm, off-farm, petty trading, animal sale and transfers. In terms of this indicator, households in San Nicolas and in the NTX communities fare the best, with 59% and 44%, respectively, drawing their income from three or more sources. Once again, Suchiquer and Tituque score the worst, with only 25% and 27%, respectively, obtaining their income from three or more sources, and an alarming 33% and 34%, respectively, deriving their income from one source only.

The food dependency ratio was computed as the ratio between cereal purchases and total household consumption.

Substantial differences are observed in the average number of sources of income of beneficiaries and non-beneficiaries households of PCUCH, with beneficiaries scoring better. The data for the PROZACHI communities are more conflictive, particularly in the number of highly diversified households (three or more sources), although, on average, beneficiaries appear slightly more diversified.

A substantial share of households in the NTX communities has two or more members working off farm. This feature is often considered a good indicator of a household’s resilience to entitlement failure associated with, for example, crop failure. The highest percentages of households with no member earning off-farm income are in San Nicolas and Tituque (56% and 59%, respectively).

About two thirds of households in the PCUCH communities do not earn a significant amount of income from petty trading. Conversely, over 80% of households in Suchiquer and San Nicolas obtain income from this source, with 16% in San Nicolas deriving it from two or more activities. A substantial number of households (42%) in Buena Vista and San Martín also earn no off-farm income. (These households are mainly concentrated in Buena Vista.)

Another likely indicator of vulnerability to food insecurity is the number of crops grown by a household. The average numbers are generally low, and range from 2.2 crops (for non-beneficiaries in Suchiquer) to 2.8 (for beneficiaries in Buena Vista and San Martín). The only exception is beneficiary households in the NTX communities and Tituque, with 3.8 and 3.5 crops per household, respectively. Even though these two groups exhibit substantially higher figures compared with the other groups, the degree of agricultural diversification can be rated as extremely low in all the surveyed communities.

Finally, the total number of crops grown in each community is reported as proxy for community-level vulnerability. Once again, Suchiquer scores the worst, with only six crops being grown⁶ in the community. The PCUCH communities and Tituque rate the best, with a clear impact of project participation. What is important to note is the extremely low number of crops grown in each community by a sufficiently high number of households, even in these relatively better-off communities.

In summary:

• Even though sufficiently diversified with respect to income sources, households in San Nicolas exhibit extremely low levels of agricultural diversification (basically, corn and beans).
• In addition to low income, poor yields and high market dependency, Suchiquer is also characterized by very low agricultural diversification (both at the household and the community level) and low income diversification.
• A very limited number of crops is generally grown both at the household and the community level.
• A large number of households in both Suchiquer and Tituque derive their income from only one source. The result highlights the extremely high exposure to risk of food entitlement failure for a large share of households in these communities, being also characterized by very high food dependency ratios.

Project activities and degree of participation

Table 24 reports information on the main project activities and their direct beneficiaries within the households. Considerable intercommunity differences emerge from the figures. In Buena Vista, San Martín and Tituque, women from beneficiary households had, in percentage terms, access to credit more than twice the time as did men. The pattern is reversed in all other communities, with more men receiving credit than women. Among these latter communities, Suchiquer exhibits the largest gender gap, with almost 50% of male beneficiaries receiving a loan from the project, compared with only 9% of women.

Sizeable gaps also exist within communities in the intrahousehold distribution of technical assistance by the projects. Particularly in Suchiquer and San Nicolas, the percentages of women actually reached by the projects’ technical assistance component are quite low (24% and 33%, respectively).

Through their infrastructure components, both projects promote a food-for-work programme. A considerable number of men and a few women in Suchiquer appear to have benefitted from the programme. In all other communities, the percentage of households affected is quite low. Because of delays in the distribution of the food rations, and the drop in supplies from donor agencies, recently both projects have been forced to reduce the scope of the programme, resulting from an inability to cover its full costs.

Finally, PROZACHI appears to have accomplished considerable results in Suchiquer, helping women to obtain their identification cards, with 33% of women from beneficiary households having taken advantage of the programme.

---

1/ When intercropped, the crop area was computed as the ratio between the total plot area and the number of crops grown in the plot.

2/ Virtually all carrot and lima bean producers are in San Martín.

3/ USD 1 = 6 QuetzaIes (Q) in July 1996.

4/ The food dependency ratio was computed as the ratio between cereal purchases and total household consumption.

^5/ Includes both corn and sorghum.

6/ Elderly farmers in the community recall that only 30 or 40 years ago a number of vegetables such as tomatoes and cabbage were grown in the area thanks to better soil conditions.