Abstract: Poverty problem in Nigeria is multidimensional and it is further aggravated by some shocks that households get exposed to. In this study attempt were made to determine the socio-economic factors that influence experience of shocks by households and decomposed the multidimensional poverty across the welfare shocks and coping methods. The study used 2006 Core Welfare Indicator Questionnaire (CWIQ) data of the National Bureau of Statistics (NBS) were used. Data were analysed using descriptive statistics and Fuzzy Set. The results revealed that main shock experienced by the rural household heads is high prices of agricultural inputs. Those that are multidimensional poorer experiences shock of lack of adequate land. Most of the rural household heads usually cope with poverty by working on farms that belong to other households. The household heads that cope with poverty by begging on the street were multidimensional poorer than those that adopted other coping methods.

INTRODUCTION

Poverty reduction is a subject that has been reemphasized as one of the brightest indicators of human progress. Policy makers are now interested in examining the role of welfare shocks towards a deeper understanding of the causes of poverty (Dercon, 2004). Some emphases have been placed on social risk management strategies that will form an integral instrument for lifting the poor out of their different forms of deprivation (Holzmann and Jorgensen, 2001).

Nigerian governments have embarked upon several processes of economic programs, which are necessary instruments for putting the nation on the path of sustainable economic growth and development. Precisely, after the Austerity Measure of 1982 failed to yield desired results, the Structural Adjustment Program (SAP) was implemented in 1986. Though some benefits were achieved at the initial stage of implementing SAP, such benefits could not trickle down to the poor. Rather, the incidence of poverty increased (Aigbokhan, 2000).

Also, adverse macroeconomic shocks that inhibit economic growth and inability of some proposed programs to tactically address unfavorable macro-economic performances are notable among the factors that contributed to increasing poverty (Aigbokhan, 2000). Since 1999, when Nigeria returned to democratically elected government, some programs have been implemented with the fiscal policy thrusts focusing on employment generation, privatization, trade liberalization, provision of favourable environment for private sector development, reduction of public sector expenditure on salaries and overheads, encouragement of transparency and accountability in government contracts and expenditures, reduction of multiple taxes and levies faced by private companies and reduction of company and personal income taxes (CBN, 2005). Government ’s blue-print for economic growth and development have been summarized in some programs that are contained in the National Economic Empowerment and Development Strategies (NEEDS) document. Government is now attempting to ensure that the different facets of poverty are tactically addressed through policy and program design because of its multidimensional nature.

However, most of the times, policy makers do not have comprehensive information about the type of sufferings that could be brought on the populace through various policy programs, thereby subjecting them to some form of economic and social deprivations. Specifically, Nigerians have taken some of the programs with mixed feelings. This is because since the programs started some private and public sector workers have been sacked, petroleum prices have been reviewed upward so many times, professional workers have embarked on several industrial strikes for one demand or the other, some national industrial strikes in protest against hike in petroleum prices have been had communal conflicts are rampart in some regions, prices of commodities are on the increase, business uncertainties are increasing, petroleum pipelines have been increasingly vandalized with serious environmental degradation consequences on some households among others. Some of these externalities sometimes have significant negative effects that often erase any positive impacts that some of this programs may have at the households level.

Therefore, previous poverty reduction programs in Nigeria did not fully achieve their objectives and this raises two important issues. First, it is not sure whether the country lacks sufficient capacity to mitigate the social risks faced by households and communities and second whether the country has not paid sufficient attention to the issue of welfare shocks and uncertainty that are important for the understanding of the dynamics that often lead households to poverty (Alayande and Alayande, 2004). It is also important to note that poverty problem in Nigeria is largely a rural phenomenon. Clear evidences to this have been obtained from all nationally representative household data that were collected. Also, within the rural areas, women and households headed by women are frequently the most chronically poor. Similarly, the northern part of the country bordering the Niger, which is arid, marginal to agriculture, environmentally damaged and densely populated have higher poverty profile (Rural Poverty Portal, 2008).

It should be noted that higher incidence of poverty profile in Nigeria ’s rural areas have been traced to some environmental problems associated with agricultural production, high vulnerability to health hazards (Alayande and Alayande, 2004) low level of education, high fertility rate, lack of access to improved seeds and inputs and poorly developed social infrastructural facilities (Okunmadewa, 2002) among others. Similarly due to lack of appropriate insurance against income shocks, rural poverty is often worsened because farmers dispose their productive assets such as land, livestock, equipment (etc.) in order to meet immediate consumption needs (Alayande and Alayande, 2004). Also, farming households face serious risks from inadequate rains/drought, degraded land, input shortages, disease outbreak and low prices for agricultural products.

Essentially, this study raises two main issues. First is the fact that poverty measurement in literature has recently graduated from the uni-dimensional income/expenditure approach to multidimensional approach that is based on nutrition, health, education, sanitation, housing quality, ownership of asset/fulfilment etc. (an offshoot of Sen ’s capability theory) (Sen, 1985, 1987). Previous approaches to analyze poverty in Nigeria have focused on the income/expenditure approach with little attention on multidimensional poverty assessment. Second is the fact that economic literature is now emphasizing the issue of welfare shocks in poverty analysis. Such efforts are channelled towards understanding the different forms of deprivation that households suffer along with the shocks/risks being exposed to. This is considered important given the multidimensional nature of poverty. Precisely, Christiaensen and Subbarao (2001) submitted that the need for addressing the issue of welfare shocks becomes paramount because they lead to a wide variability in households incomes. In the absence of sufficient assets or insurance to smoothing consumption, such shocks may lead to irreversible losses, such as distress sale of productive assets reduced nutrient intake or interruption of education that permanently reduces human capital, thereby locking their victims in perpetual poverty.

Therefore, given the programs that Nigerian Government had embarked upon in the past few years, this study intends to use multidimensional measure of well-being to answer salient policy related questions. What are the forms of welfare shocks that households faced and what are the coping mechanisms adopted to mitigate their effects. What is the multidimensional poverty profile for rural Nigeria? How can we decompose multidimensional poverty profile across the different forms of shocks and coping mechanisms? It is believed that if reliable answers are provided to these questions, policy makers will have some required pieces of information to proffer appropriate solution to addressing poverty in Nigeria.

MATERIALS AND METHODS

The data: Data was obtained from 2006 National Core Welfare Indicator Questionnaire (CWIQ) Survey. The survey was carried out in each of the Nigeria ’s 36 States and the FCT. A two-stage cluster sample design was adopted for the selection of the respondents in each of the States and the FCT. The first stage involves the selection of Enumeration Areas (EAs), while selection of the Housing Units (HUs) constitute the 2nd stage. The National Population Commission (NPopC) EAs as demarcated during the 1991, Population Census served as the sampling frame for the selection of 1st stage sample units. In each LGA, a systematic selection of 10 EAs was made. This is to ensure that the number sample corresponds with the number of LGAs in each state. Prior to the second stage selection complete listing of Housing Units (and of Households within Housing Units) was carried out in each of the selected 1st stage units. These lists provided the frames for the second stage selection.

Ten HUs were then systematically selected per EA and all households in the selected HUs were interviewed. The projected sample size was 100 HUs at the LGA level. Overall, 77,400 HUs were drawn at the national level. Also, sampling weights were constructed for each sample, thus making the data representative of the entire population in Nigeria.

Method of analysis: Multidimensional poverty indices were computed using the Fuzzy Set theory originally developed by Zadeh (1965). This approach had been widely applied to poverty analysis by researchers like Cerioli and Zani (1990), Martinetti (2000), Costa (2002, 2003), Dagum (2002) and Duetsch and Silber (2005). Zadeh (1965) characterized a Fuzzy Set as a class with a continuum of grades of membership. Therefore, in a population A of n households (A = a₁, a₂, a₃, ..... a_n), the subset of poor households B includes any household a_iεB. These households present some degree of poverty in some of the m poverty attributes (X).

The welfare attributes considered in this study based on the CWIQ. The findings of Ayoola et al. (2000) for some Nigerian rural and urban areas will assist in selecting relevant welfare attributes because their study was based on focused group discussions and therefore considered to be the voice of the poor. Following Costa (2002), the degree of being poor by the ith household (i = 1, …., n) with respect to a particular attribute (j) given that (j = 1, ……, m) is defined as: μB[X_j (a_i)] = x_ij, 0 ≤x_ij ≤1. Specifically, x_ij = 1 when the household does not possess welfare enhancing attribute and x_ij = 0 when the household possesses it. Betti et al. (2005) noted that putting together categorical indicators of deprivation for individual items to construct composite indices requires decisions about assigning numerical values to the ordered categories and the weighting and scaling of the measures. Individual items indicating non-monetary deprivation often take the form of simple yes/no dichotomies. In this case x_ij is 0 or 1.

However, some items may involve more than two ordered categories, reflecting different degree of deprivation. Consider the general case of c = 1 to C ordered categories of some deprivation indicator with c = 1 representing the most deprived and c = C the least deprived situation. Let c_i be the category to which individual i belongs. Cerioli and Zani (1990) assuming that the rank of the categories represents an equally-spaced metric variable assigned to the individual a deprivation score as:

(1)

where, 1 ≤c_i ≥C. Therefore, x_ij needs not to be compulsorily 0 or 1 but 0 ≤x_ij ≤1 when there are many categories of the jth indicator and the household possesses the attribute with an intensity. The multidimensional poverty ratio of an household, μ_B(a_i), which show the level of welfare deprivation and membership to set B is defined as the weighted average of x_ij:

(2)

w_i is the weight attached to the jth attribute. The intensity of deprivation with respect to X_j is measured by the weight w_j. It is an inverse function of the degree of deprivation and the smaller the number of households and the amount of their deprivation, the greater the weight. In practice, a weight that fulfils the above property had been proposed by Cerioli and Zani (1990). This can be expressed as:

(3)

Ideally,

is the relative frequency represented by the sample observation a_i in the total population. Therefore when x_ij = 0, the welfare attribute should be removed.

The poverty ratio of the population μ_B is simply obtained as a weighted average of the poverty ratio of the ith household μ_B(a_i):

(4)

Similarly,

(5)

In this way, it is possible to decompose the multidimensional poverty ratio of the population μB as the weighted average of μB (X_j) with weight w_j:

(6)

From Eq. 2, the dimension that tends to increase the level of poverty of each household can be determined by decomposing the household poverty index:

(7)

where, y_ij is the contribution of the j-th attribute to the overall amount of the household poverty index μ_B (a_i):

(8)

Following Mussard and Alperin (2005), it is possible to decompose multidimensional poverty indices by sub-population. Suppose the total economic surface is divided into K groups, S_k, of size n_k (k = 1, …, K). The intensity of poverty of the ith household of S_k is given as:

(9)

where, x^k_ij is the degree of membership related to the fuzzy sub-set B of the i-th household (i = 1, …,n) of S_k with respect to the jth attribute (j = 1, …, m). Hence, the fuzzy poverty index associated with group S_k is:

(10)

Following Eq. 8, the overall poverty index can be computed as a weighted average of the poverty within each group:

(11)

Thus, the contribution of the k-th group to the global index of poverty is:

(12)

RESULTS AND DISCUSSION

Distribution of house heads experience of welfare shocks across the Geo-Political Zones (GPZs): Table 1 shows the frequency distribution of household heads experience of welfare shocks across the GPZs. High price of agricultural inputs was the main shock experienced by the household heads. It accounted for 22.16% of the total welfare shocks that were reported in all the GPZs. North East and South South have the highest percentages (5.22 and 5.20, respectively) of household heads that experienced shock in the form of too high agricultural inputs price. Those who indicated that agricultural inputs were available accounted for 8.73% of the total welfare shock. North East, North West and North Central have the high percentage of the shock agricultural inputs not available. The corresponding percentages are 2.58, 2.47 and 2.04. It can therefore be generalized that none availability of agricultural inputs was the main shock in the Northern zones of the country. The third important welfare shock is hard economic times. It accounted for 6.40% of the total welfare shock. Across the zones the percentage welfare shock hard economic times is high in North West, South South and South West. The corresponding percentages are 1.25, 1.21 and 1.14. The forth welfare shock is lack of capital for agricultural expansion. It accounted for 4.51% of the total welfare shock of which North East and South East have the highest percentage of 0.90 and 0.91, respectively. The fifth welfare shock is low agricultural production in which Northern zones have the highest percentages. 34.41% respondents did not state the welfare shock that they experienced.

In North West zone the first five welfare shocks experienced in order of there importance were high price of agricultural inputs (3.65%), none availability of agricultural inputs (2.47%), hard economic times (1.25%), low agricultural production (0.93%) and lack of capital for agricultural expansion (0.77%). In North East zone the first five welfare shocks experienced in order of there importance were high price of agricultural inputs (5.22%), none availability of agricultural inputs (2.58%), hard economic times (0.98%), low agricultural production (0.95%) and lack of capital for agricultural expansion (0.90%). In North Central zone the first five welfare shocks experienced in order of there importance were high price of agricultural inputs (3.16%), none availability of agricultural inputs (2.04%), low agricultural production (0.90%), hard economic times (0.82%) and lack of capital for agricultural expansion (0.79%).

In South East zone, the first five welfare shocks experienced in order of there importance were high price of agricultural inputs (2.74%), hard economic times (0.99%), lack of employment/jobs (0.97%), lack of capital for agricultural expansion (0.91%) and none availability of agricultural inputs (0.59%). In South West zone the first five welfare shocks experienced in order of there importance were high price of agricultural inputs (2.20%), hard economic times (1.14%), none availability of agricultural inputs (0.66%), low price for agricultural produce (0.54%) and high price of commodity (0.48%). In South South zone the first five welfare shocks experienced in order of there importance were high price of agricultural inputs (5.20%), lack of employment/jobs (1.22%), hard economic times (1.21%), lack of capital for agricultural expansion (0.70%) and lack of capital for business (0.49%).

Table 1:	Frequency distribution of house heads experience of welfare shocks across Nigeria ’s geopolitical zones

From Table 2, it could be inferred that irregular payment of pension and delay payment of gratuity affects those that are in there late and early 60's, respectively. These shocks if not attended to could send these groups of people into their early grave. Lack of market/buyers and retrenchment/redundancy affects those that were in there early fifties. It implies that if these shocks are not addressed they might not be able to recover from it. Those that their age falls within 48 and 49 were affected idiosyncratic shocks such as lack of adequate land, low prices for agricultural produce, animal deaths from diseases, lack of capital for agricultural expansion, lack of capital for business, lack of credit to expand business, high commodity prices, agricultural inputs prices too high, hard economic times and low profit. Low salary and cultural/religious reason affects those that are in their early forties.

The retirees that experienced delay in the payment of their gratuity have the highest household size. The lowest household size is observed with those that experience shock in form of low profit. It is also observed that those that experience shocks relating to agriculture have relatively high household size. Table 3 shows the distribution of welfare shocks across the education and sex of the house heads.

Table 2:	Average ages and household sizes of rural households across welfare shocks

Table 3:	Frequency distribution of welfare shocks, education and sex in rural Nigeria

The welfare shocks affect those that do not have any form of education except shocks relating to payment of pension and gratuity. Irregular payment of pension and delayed payment of gratuity affects those that have post secondary education most. Among those with some primary, completed primary, some secondary, completed secondary and post secondary education agricultural inputs prices too high was the main welfare shock that affect them most. Among those with some primary education other shocks that affect them most are hard economic times, agricultural inputs not available, lack of employment/jobs, lack of capital for agricultural expansion and low agricultural production.

Table 4:	Frequencies distribution of welfare shocks and rural households employment groups in rural Nigeria

Those with completed primary education experiences other shocks such as hard economic times, agricultural inputs not available, lack of capital for agricultural expansion, lack of employment/jobs, low agricultural production and lack of capital for business. Those with some secondary education experiences other shocks such as agricultural inputs not available, hard economic times, lack of employment/jobs and lack of capital for agricultural expansion. Those with completed secondary education experiences other shocks such as hard economic times, agricultural inputs not available, lack of employment/jobs and lack of capital for agricultural expansion. Among those with post secondary education other shocks that they experiences most are low salary, hard economic times, lack of employment/jobs and agricultural inputs not available.

In the Table 4, the shock with the highest effect across the various employment groups is agricultural input too high. Also, those that are self employed in agriculture experiences the highest effect of the various shocks except for low salary and business not doing well. Other shocks common to those in the public sector are low salary, hard economic times and agricultural inputs not available. Those that are employed in the private formal sector experiences other shocks such as lack of employment/jobs, low salary and hard economic times. Agricultural inputs not available, hard economic times, lack of employment/jobs and low agricultural production are main idiosyncratic shocks affecting those in the private informal sector.

Other idiosyncratic shocks common to those that are self employed in agriculture are agricultural inputs not available, hard economic times, lack of capital for agricultural expansion and low agricultural production. Those that are self employed in other economic activity experiences other shocks such as hard economic times, agricultural inputs not available, lack of employment/jobs and lack of capital for business. The unemployed experiences other shocks such as lack of employment/jobs, agricultural inputs not available and hard economic times.

Across the various employment groups the most common idiosyncratic shocks are agricultural input too high, hard economic times, agricultural inputs not available and lack of employment/jobs.

Multidimensional decomposition across welfare shocks: Table 5 shows the multidimensional decomposition across the house head experience of welfare shocks. The welfare shock lack of adequate land has the highest average multidimensional poverty of 0.4244. The welfare shock of irregular payment of pension has the lowest average multidimensional poverty of 0.3282. The welfare shocks that also have high average multidimensional poverty are: too high prices of agricultural inputs, death of animals due to diseases, drought, low agricultural production, low prices for agricultural produce, retrenchment/redundancy and lack of capital for agricultural expansion.

Table 5:	Multidimensional deprivation decomposition across welfare shocks
F ratio = 112.202***

Table 6:	Multidimensional poverty decomposition across the house heads coping methods
Extracted by the Author from the 2006 CWIQ data

The corresponding value of their average multidimensional poverty is 0.4131, 0.4111, 0.4104, 0.4065, 0.4026, 0.4024 and 0.4017. The highest variability index is observed in those that do not specify the welfare shock they experienced, while the lowest variability index is observed in those that specified that cultural/religious shock was the welfare shock they experienced. Those that do not specify any welfare shock have the highest absolute and relative contribution to multidimensional poverty of 0.12 and 30.93%, respectively. The welfare shock cultural or religious reason has the lowest absolute and relative contribution to multidimensional poverty of 0.0001 and 0.03%, respectively. It is also observed that too high prices of agricultural inputs account for high absolute and relative contribution to multidimensional poverty (0.09 and 24.11%). Table 6 shows the multidimensional poverty decomposition across the house heads coping methods against poverty. Those that resorts into begging from the street to enable them cope with the scorch of poverty have the highest average multidimensional poverty of 0.4270. The lowest average multidimensional poverty of 0.3458 is recorded in those that adopt other form of coping methods against poverty.

The variability index of the coping methods with highest and lowest average multidimensional poverty is 23.91 and 31.60%. The highest and lowest variability index of 36.81 and 23.75% is observed in those that do not have any coping methods and those that withdraw their children from school. Those that work on the farm belonging to other households have the highest absolute and relative contribution to multidimensional poverty of 0.0701 and 18.46%. Those that beg from the streets have the lowest absolute and relative contribution to multidimensional poverty of 0.0011 and 0.28%. Other copping methods that have high absolute and relative contribution to multidimensional poverty are those that reduce the number of their meals and those that take up other pieces of work. Their absolute and relative contributions are 0.0546, 0.498, 14.38 and 13.11%.

CONCLUSION

The main shock experienced by the rural household heads is high prices of agricultural inputs. This shock is most felt in the North East and South South Nigeria. The shocks that affected the aged were irregular and delayed payment of pension and gratuity.

Those that are multidimensional poorer experiences shock of lack of adequate land. Most of the rural household heads usually cope with poverty by working on farms that belong to other households. The household heads that begged on the street in order to cope with poverty were multidimensional poorer than those that adopted other coping methods.

In order to reduce the shocks experienced by the rural households, the government should subsidize the prices of agricultural inputs so that the prices of these inputs will be relatively stable. The government should ensure that land is available at lease for farming. The government should embark on programs that would alleviate the poverty level of street beggars.