In order to assess the effects of rural transport infrastructure investments and policies on poverty and its alleviation, it is necessary to distinguish between poor and non-poor users of the infrastructure. It is impractical to obtain reliable information on the living standards of rural travellers directly from roadside interviews. This paper reports on a methodology for estimating the expenditure levels of households of road users indirectly from more readily available information on a selection of socio-economic characteristics. The methodology consists of (a) focus groups to identify possible socio-economic indicators of standards of living on which information could be quickly and readily obtained at roadside interviews, (b) a survey of a sample of households in the locality to collect information on selected socio-economic indicators as well as their income and expenditure, and (c) econometric analysis to identify the socio-economic indicators and equation relating household expenditure to them, which are good predictors of household expenditure as a measure of the standard of living. The paper also describes a procedure for defining per capita income thresholds for distinguishing between poor and non-poor travellers based on international poverty level indicators. The method was developed as a part of the study of value of travel time savings of rural travellers in Bangladesh. In the linear equation derived as the predictor of household per capita expenditure, the independent variables are: (a) land cultivated per head of household; (b) number of household members involved in income earning activities; (c) a household member in "permanent" off-farm employment; (d) a household member engaged in a "permanent" business, and (e) ownership of a motorised vehicle by household. The threshold for distinguishing between poor and non-poor road users in local currency was estimated from the international poverty line equivalent to the 1985 purchasing power parity (PPP) US$1 per person per day. The paper describes a rapid appraisal methodology for assessing rural household income levels which is applicable elsewhere. There is inevitably a trade-off between speed and economy on one side and precision on the other. The predictive precision of the method can be ensured by increasing the sample size and/or stratifying the sample to better represent population characteristics, improving the quality of data collection. Choice of combinations of variables and rigorous statistical tests in the econometric analysis testing are also essential for validity. For the covering abstract see ITRD E135448.
Samenvatting