PART II CASE-STUDY: GUINEA-BISSAU
7. Village-level case-study I: Caiomete
7.6 Assets, wealth and poverty
In the previous sections, we have seen that Caiomete households exhibit a relatively low degree of differentiation with respect to the volume of their crop outputs. Among the reasons that account for this, we find the low degree of recourse to wage labour beyond certain specific tasks (as we have seen), along with the absence of a market for land and the low degree of technical differentiation. The survey questionnaire was not ideally designed for ascertaining the latter, but the participants in the focus groups and in the qualitative interviews concurred in suggesting that there is both a very high degree of homogeneity in terms of the use of agricultural techniques and a low degree of recourse to the use of more sophisticated agricultural tools.
This is partly explained by the technical properties of the labour processes involved in the production of the main crops. In particular, it is notably difficult to either mechanise or use animal traction in the harvesting of cashew nuts. Once the trees have been planted and have grown to the point of bearing fruit, the process of harvesting consists of picking up the mature fruits that fall to the ground in March-May and removing the nuts. These are then collected in containers and taken home, where they are stored until they are sold to buyers. When the nuts are separated from the fruits, the latter are crushed into a pulp in a device that is used to make juice, which is then left to ferment by itself until, a few days later, it turns into cashew ‘wine’. Except for this latter part of the process – crushing the cashew fruits (whose purpose, then again, is only to produce a by-product that is sold in relatively small amounts) –, none of the above can be readily subject to technical improvement. In fact, the same goes for the growing of garden vegetables – usually undertaken in small plots – and the exploration of forest products like palm wine and palm oil.
The main exception to the lack of susceptibility to ready technical improvement is paddy rice cultivation. Here, both tilling the land and harvesting the rice could all be easily mechanised and/or benefit from the use of animal traction. However, we find that that is not the case. Harvesting is done using a simple knife and tilling is undertaken without recourse to either mechanical or animal traction, simply using a hand-held plough. The
211 fragmentation of tenure, the remote location of the village and the inability of even the wealthiest Caiomete households to afford buying or renting a tractor also make it clear that the latter would hardly be an option. Still, while one might reasonably expect that animal traction be used, in fact it is not used at all. Regrettably, not enough systematic information was collected on this that might enable me to substantiate this argument in a stronger way.
However, all the sources from whom qualitative information was collected concurred in stating that animal traction is not used at all, for whatever purposes, in Caiomete. As one of the participants in the focus group put it, “our oxen are lazy; they do not know how to work”. Quite a few households (21, or 29%) do own cattle; however, these are usually in small numbers (Figure 7.11) and typically used only in the context of ritual sacrifices (on which occasions they are consumed by the members of the community).
Figure 7.11 Caiomete: Number of cattle heads owned by the households, boxplot
Thus it is that Caiomete households own very little by way of productive assets beyond the land itself and a few rather rudimentary tools. If we add to this the low degree of recourse to the use of wage labour and the lack of access to extension services it does not come as a surprise that the distribution of the output quantities is so homogeneous around a low mean. Moreover, if we additionally consider the low degree of diversification in terms of off-farm activities – the only notable exceptions being the owners of village shops –, it does not come as a surprise, either, that the level of wealth of Caiomete households is also relatively homogeneous and not particularly high.
212 In the context of this survey, wealth was proxied by the ownership of a series of durable household assets (see chapter 5 and Appendix IV). Table 7.8 summarises some key data, and suggests that while Caiomete may be considered relatively better-off than the average village in Guinea-Bissau (a country where there are many villages in which no-one owns any of the aforementioned assets except for the odd radio or cell phone), it can hardly be considered affluent when compared to a place like Braima Sori (see below, Chapter 8). Indeed, the only durable assets, out of those on which questions were specifically asked, which are reportedly owned by a relatively large number of Caiomete households are cell phones, radios, gas or electric stoves (usually gas ones, of the basic portable type), bicycles and zinc rooftops. Now, in Guinea-Bissau, zinc rooftops are the hallmark feature of villages that have made the transition from the hardest of hand-to- mouth existences to the ability to save a little149. This shows that that has been the case in this village. However, most households have not gone much further than that when it comes to durable consumption items: no-one reported owning a car or a fridge; only one of the respondents reported owning a motorcycle; and only 8% and 13% own TVs and diesel generators, respectively.
Frequency (Yes)
Percentage
Cell phone 51 70.8
Radio 32 44.4
TV 6 8.3
Gas or electric stove 39 54.2
Fridge 0 0.0
Bicycle 32 44.4
Diesel generator 9 12.5
Zinc rooftop 65 90.3
Motorcycle 1 1.4
Car 0 0.0
Table 7.8 Caiomete: Ownership of durable household assets
In order to turn the available quantitative information on asset ownership into a more manageable (unidimensional) format and make additional inferences, an asset index was computed for all the households in the survey sample, reflecting each household’s
149 Zinc rooftops are better insulators from the rain and are subject to less wear and tear, as well as much less vulnerable to fire, than thatch ones. For that reason, once the households that inhabit a given housing unit are able to put some money on the side, substituting zinc for thatch is usually one of the first priorities.
213 endowment of the ten durable items mentioned above along with the number of cattle heads owned by each household (see Appendix IV). This asset index was constructed in such a way as to have a mean value of “0” in the sample of 108 households, but its mean within the Caiomete sub-sample (-0.43) turned out to be much lower than among the 36 households in Braima Sori (0.87). A more thorough comparative discussion of the features of the two villages is presented and discussed in Chapter 9 (below). For now, it is more relevant to take a closer look at distribution of this asset index within the Caiomete sub- sample, and to examine the extent to which it is positively or negatively associated with other variables.
Figure 7.12 shows the distribution of the asset index among the 72 households in the Caiomete sub-sample. It is a relatively compact distribution: the standard deviation within the subsample is 0.5, compared to 1 (by construction) within the overall sample (i.e.
including the 36 Braima Sori households). Significantly, the wealthiest household in Caiomete has an asset index score (0.84) that is lower than the mean in Braima Sori (0.87).
Two households appear as positive outliers (indicated by the numbers 2 and 27): both consist of households that earn a part of their income from commerce (while also practicing agriculture), which reported hiring a substantial number of persons in the previous twelve months and which once participated in international migration but no longer do so – all characteristics that can be theoretically posited to be causally linked with household long- run wealth. Therefore, these features of the most prominent potential ‘proto-bourgeois’ of Caiomete do not really make it clear whether, and to what extent, the recourse to hired labour, commerce and migration constitute relevant pathways to upward wealth differentiation. Other pieces of information do, however.
214 Figure 7.12 Caiomete: Distribution of the asset index scores among the 72 Caiomete households in
the sample, boxplot
First of all, there does not seem to be any meaningful association between household long-run wealth (represented by the asset index score) and either the cashew or paddy rice outputs: Pearson’s r equals, respectively, 0.08 and 0.07. Seemingly, thus, enhanced agricultural production does not seem to constitute a significant way to become relatively richer in this context. Additionally, as shown in Table 7.9 (below), neither the hiring-in nor the hiring-out of agricultural wage labour are significantly associated with higher or lower levels of wealth. Arguably, this may be regarded as supporting the argument that participation in the labour market (both as employer and as provider of labour) follows institutional patterns that are not primarily related to the households’ level of affluence/destitution and that that participation is typically of secondary importance within the context of the households’ productive and reproductive activities. This constitutes another indication that there exists little by way of class differentiation.
215
Yes No p-value of
significance of differences (T-test,
2-tailed, equal variances not
assumed)150 Households currently participating in
migration
-0.498 -0.213 0.101
Remittance-recipient households -0.277 -0.496 0.106
Households that include former migrants -0.430 -0.445 0.910 Households that include former international
migrants
-0.436 -0.432 0.978
Households that include former migrants to Europe
-0.505 -0.430 0.662
Households that hired in agricultural labour in the previous year
-0.423 -0.460 0.749
HH that hired out labour in the previous year -0.433 -0.424 0.955
Female-headed households -0.602 -0.132 0.001**
Households that report earning part of their income from commerce
0.229 -0.484 0.052*
Table 7.9 Caiomete: Mean asset index score for selected household sub-samples in Caiomete
In this context, migration does not seem to be associated with upward differentiation, either. This is discussed further in section 7.7 (below), but it is immediately clear from the Table above that those households that are currently participating in migration are poorer on average than those that do not, and that those which participated in migration in the past and have subsequently returned to the village did not become wealthier than average as a result (regardless of the national or international character of their migration project). There are some issues to do with the direction of causality that can be raised here, of course: it might be the case that poorer households are more likely to engage in migration to begin with, and that the successful completion of one or more migration cycles actually enables them to partially catch up with the rest. However, given
150 A p-value lower than 0.10 (0.05) implies a 90% (95%) probability of the difference not being caused by random variation and reflecting a ‘true’ association. * and ** indicate significance at the 0.1 and 0.05 threshold, respectively.
216 that remittance-recipient households (a subset of the group of migrant households) do appear to be somewhat wealthier than the average, it is much more likely the case that migration fails to show any significant beneficial effects for the households of origin because few of those households are actually sent any money remittances by the migrants, whereas all of them suffer an at least temporary reduction in their labour pool as a consequence of migration.
Now, while (proto-capitalist) agriculture and migration do not seem to play a significant role when it comes to the accumulation of wealth, commerce is a different matter. As shown in the last row of Table 7.9, households that report earning at least a part of their income from commerce are on average much wealthier than the rest; in fact, they are the only sub-group represented in this Table for whom the mean asset index score is positive (i.e. greater than the average in the overall sample of 108 households). Taken alongside the previous pieces of evidence, this suggests very clearly that this is a context in which, as often occurs, merchant capital has preceded productive agricultural capital as a significant feature at the community level, and where this is in turn reflected in the households’ differential levels of long-run wealth.
While it is apparent that the incipiently wealthy of Caiomete are basically drawn from among those who engage in (usually relatively petty) commercial activities, it is equally clear that the most destitute of all consist of landless female-headed households.
The mean asset index score among female-headed households is 0.47 lower than that of male-headed households (the difference amounting to 1/2 of the overall standard deviation). This feminisation of poverty is therefore extremely pronounced, and is related to other power imbalances and forms of inequality between men and women, both within the household and more broadly within the community. The facts that women are customarily prevented from inheriting most assets and that no women were identified in the survey who autonomously engage in commercial activity (in the sense of buying and selling commodities in order to make a profit) constitute examples of such imbalances and inequalities. The fact that many women are required by their husbands to provide a type of quasi-corvée in return for the right to harvest cashews in the latter’s parcels constitutes another. And, as might be expected, landlessness and the payment of rent in the form of labour time, which as we have seen have a clear gender dimension, are also associated with a lower asset index score (Table 7.10).
217
Rent-paying status n Mean asset
index score Used other people’s
land
Paid rent (in the form of labour time)
24 -0,504
Did not pay rent 20 -0,409
Not applicable (did not use other people's land) 28 -0,393
Total 72 -0,435
Table 7.10 Caiomete: Mean asset index score according to the rent-paying status of the household
The analysis of the asset index and its association with other variables therefore illustrates very clearly which attributes are associated with wealth and poverty in Caiomete.
The typical wealthy household in this village is a male-headed household that engages in commerce and cultivates or harvests its own parcels (possibly in addition to earning rent in the form of labour time). Conversely, Caiomete’s poorest households typically consist of female-headed households that are also partially or fully landless, which are required to pay rent in exchange for using other people’s parcels and which, despite possibly participating in migration, do not earn any money from remittances.