Surveying livelihoods, service delivery and governance: baseline evidence from Pakistan - (W - 142)

Surveying livelihoods, service delivery and governance: baseline evidence from Pakistan - (W - 142)

Publication details

  • Monday | 09 Jun, 2014
  • Abid Qaiyum Suleri, Syed Qasim Ali Shah, Babar Shahbaz, Steve Commins, Syed Mohsin Ali Kazmi
  • Working Papers
Download File

Researching livelihoods and
services affected by conflict
 
Working Paper  14
Babar Shahbaz, Qasim Ali Shah, Abid Suleri, Mohsin Ali Kazmi and Steve Commins
June 2014 
 
Executive summary 
 
In 2012/13, the Sustainable Livelihoods Research Consortium (SLRC) implemented the first round of
an original cross-country panel survey in Pakistan – a survey designed to produce information on:  
  • People’s livelihoods (income-generating activities, asset portfolios, food security, constraining and enabling factors within the broader institutional and geographical context); 
  • Their access to basic services (education, health, water), social protection and livelihood assistance; and  
  • Their relationships with governance processes and practices (participation in public meetings, experience with grievance mechanisms, perceptions of major political actors).  
This paper reports on the baseline findings emerging from statistical analysis of the Pakistan first-round data.   
 
The survey sample 
 
The survey was conducted in Swat and Lower Dir districts of Khyber Pakhtunkhwa (KP) between September and October 2012. Both of these districts were severely affected by violent conflicts between the Pakistani armed forces and the Taliban during 2008/09, as well by floods in 2010, which together saw large-scale displacement of the population. After the conflict and floods, there was a massive inflow of aid geared towards the restoration of basic services and livelihoods.  
 
A total of 2,114 households were surveyed (1,270 from Swat and 844 from Lower Dir), with 34% of respondents being female. Our data are not representative at the district level, but are representative at
the village level. Our data are also statistically significant at both union council and village levels. 
 
In the study areas, an overwhelming majority (about 99%) of respondents from the sampled households in both districts reported that they had experienced fighting in the previous three years; around 90% of households in Swat and even more in Lower Dir had been displaced during conflicts between the Taliban and the Pakistani Army. More than one-third of households had at least one member of the
family who had migrated outside the country for employment. More than half of respondents were illiterate (no education) and very few respondents had more than intermediate education (12 years of schooling).  
 
Livelihoods and wellbeing 
 
Our survey generated data on livelihood activities (including how these changed during and after conflict), levels of wealth (proxied by asset ownership) and food insecurity (estimated using the Coping Strategies Index) among our sample population. Five key findings emerge from interpretation of descriptive statistics and regression analyses. 
 
First, farming is the most prevalent livelihood activity for individuals in our sample, followed by overseas labour and non-agriculture-based labour. However, overseas labour (remittances) is the primary income source for the majority of households. Incidence of not having paid employment is strikingly high in the study area. Very few people have their own business, do government or private sector jobs or work as skilled labourers. There was a drastic reduction in most livelihood activities during the conflict period. For instance, before the conflict, 800 and 700 persons in our sample population pursued farming and daily wage labour, respectively; during the conflict, fewer than 100 persons in our sample population worked in each activity. However, an increase in overseas migration and farming after the conflict was reported.  
 
Second, the results also show that about 50% of households depend on a single livelihood source , in spite of the fact that average household size is quite large (about nine members per household). The 8 data also indicate a positive correlation between the number of income sources per household and food security. This implies that donor interventions and public policy should facilitate diversification beyond agriculture, while at the same time keeping an emphasis on supporting agricultural activities, because farming still remains the most prevalent livelihood activity in the post-conflict areas under study.  
 
Third, as is to be expected, households with higher average education tend both to be less food insecure and to own more assets. There is a significant negative correlation between food insecurity and assets, but this is not consistent across districts: in Swat district food insecurity is more prevalent among sampled households, and asset ownership is also higher for these households.  
 
Fourth, having experienced a crime has a positive and significant relationship with asset ownership and a negative and significant relationship with food insecurity. While we cannot be sure of causality,
this suggests households that are more food secure and have more assets experience more crimes. Experience of shocks is positively correlated with asset ownership, but the number of shocks experienced by a household is significantly and positively associated with food insecurity – that is, the more shocks a household experiences, the more food insecure it is likely to be. 
 
Fifth, in terms of access to livelihoods assistance, we found a positive correlation between access to livelihood assistance and both asset ownership and greater food security. There is also a significant and positive association between improvements in farming (owing to the receipt of seeds and tools) and both asset ownership and greater food security. Though we cannot draw conclusions on causality, it may be the case that livelihood support has helped increase household assets and food security.. There is a positive correlation between receipt of a social protection transfer (Benazir Income Support Programme, or BISP) and food insecurity. It is unlikely that receipt of BISP is making households more food insecure; rather, this suggests BISP is well targeted towards the poor. 
 
Basic services, social protection and livelihoods assistance 
 
Our survey asked respondents about access to a range of services and support – including health, education, water, social protection and livelihood assistance – as well as their experiences of using
them. Again, five key findings emerge. 
 
First, in general, there are relatively high levels of access to and satisfaction with some basic services within our sample population, particularly for health and education. Average travel time is 34 minutes to health centres and about 10 minutes to primary schools. Health and education services seem to be in good shape after the conflict, possibly because of high government, non-governmental organisation (NGO) and international agency investments in these areas. But access to piped and safe drinking water is much lower compared with pre-conflict and pre-floods levels. Just over 10% of households have piped water, and 20% never or rarely have drinking water available. The overwhelming majority of households maintain drinking water themselves.  
 
Second, there seems to be a link between journey times to the health centre or school (for boys and girls) and greater satisfaction with the service as well as between greater assets and greater satisfaction with the service. For example, the data suggest that households with a higher Morris Score Index value send their male children to more distant schools. Respondents from wealthier households are also more likely to report being satisfied with the service. This suggests wealthier households tend to use more distant but better-quality services. 
 
Third, a total of 25% of households receive a social protection transfer (of which 80% receive the BISP cash transfer); 24% receive some form of livelihoods assistance (the majority of these receive seeds and tools).  
 
Fourth, there is fairly high satisfaction with the usefulness and timeliness of livelihood assistance. It is important to note, however, that respondents from households receiving such support from the government are likely to be less satisfied. Satisfaction with the BISP social protection transfer is moderate – 60% of respondents said it helped them a bit in buying extra food, but this is arguably to be expected given the low transfer level. Around 34% of households thought the transfer was too small to make any difference.  
 
Finally, although there is no consistent set of variables explaining why some respondents are more satisfied with services than others, there is some indication that people’s specific personal experiences with the service heavily influences their overall level of satisfaction. Regression analysis of respondents’ experience with both education and health suggests factors such as ‘satisfaction with the availability of medicine’, ‘satisfaction with the waiting time in the clinic’, ‘satisfaction with the number of teachers’ and ‘satisfaction with the quality of the teaching staff’ are strongly and positively associated with higher levels of overall satisfaction with those services. We also observe this for social protection: descriptive statistics show the majority (80%) of respondents from households that have never received the BISP transfer on time and/or in the right amount said the transfer was too small to make any
difference in their lives. 
 
Perceptions of governance 
 
In order to examine people’s relationships with governance actors, our survey generated information on respondents’ interactions with and perceptions of local and central government.  
 
Data show that the vast majority of respondents (more than 90%) believe that the decisions of those in power in government (either local or central) never reflect their priorities. Comparatively more espondents said central government never reflected their priorities than did so for local government. Similarly, the overwhelming majority of respondents did not agree with the statement, ‘The local/central government cares about my opinions’.  
 
There are some tentative findings that start to explain people’s perceptions of government. First, a household’s livelihood situation seems to have a consistent if weak impact on perceptions of government. Respondents from households with more assets have more positive perceptions of local and central government, while those from households with greater food insecurity have worse perceptions of local and central government. 
 
Second, security seems to matter but only for perceptions of local government. Perceptions of safety (in moving to other places like a workplace/market/town) are significantly and positively correlated with a
‘yes’ response to, ‘The local government cares about my opinions’, and those feeling safer are more likely to say the local government’s decisions largely/always reflect their priority. In other words, those
respondents who feel safe have greater trust in local government. However, causality is unclear, and it could also be the case that those who trust local government feel safe. This is a notable finding, given
that 99% of households have experienced fighting in the past three years. 
 
Third, there are some counter-intuitive and confusing results around the shocks experienced by households. The more shocks a household has experienced, the more likely it is the respondent feels local and central government decisions largely or always reflect their priorities. In other words, those who have experienced shocks generally have more positive perceptions of government. It is not clear why this is the case, and this will have to be further explored in the qualitative fieldwork. 
 
Fourth, there are some consistent, if weak, patterns linking better access to some services to more positive perceptions of government. Someone in the household receiving a social protection transfer means the respondent is more likely to have trust in central government. Respondents from households whose daughters travel further to school (i.e. have worse access) have lower trust in local and central government. Those travelling further to the health centre are less likely to agree the government’s priorities reflect their own in some areas (as opposed to in no areas). Experience of services, on the other hand, has no consistent impact on perceptions of government. 
 
Finally, the way services are being run – and having grievance processes and consultations in place – seems to matter, especially for perceptions of central government. Respondents who reported the
existence of an official way to make a complaint were likely to have trust in central government. Similarly, respondents reporting that someone had consulted their household about basic services tended to be more optimistic about the local and central government.