Moldova Poverty Line

Data

Poverty analysis presented in this paper is based on data from the Moldova Household Budget Survey, 1997 -2005H1. The survey is conducted separately for Right bank and Transnistria monthly by the National Statistics Department, and is nationally representative. As in the case of other official statistics, the results presented refer only to the Right Bank zone of Moldova.

For the purposes of this study, we constructed an absolute poverty line that determines a fixed minimum standard of living and permits consistent comparisons across groups and over time. This poverty line is calculated at 196 lei per capita per month at 2001 Q4 prices. It is based on costing out a 2,100 Kcal/day food poverty line to which an estimated non-food portion is added.

Measuring Poverty

In countries such as Moldova, where income is highly seasonal (due to a large agricultural sector) or informal, or otherwise hard to measure, consumption is the preferred measure of welfare. In this study we therefore use per capita consumption as the welfare measure.

The Hentschel and Lanjouw (1996) methodology is used to calculate the poverty lines and the consumption aggregates stepwise, adding non-food items categories cumulatively. The non-food items categories are: (1) tobacco and drugs, (2) clothing and footwear, (3) health care, (4) education, (5) transportation and communication, (6) entertainment services, (7) utilities, (8) housing, and (9) consumer durables. In the case of both food and non food expenditures, cash and in-kind expenditures are included along with the self-reported value of consumption of any home produced item.

The primary aim of including durables in the consumption aggregate, is to assess the value of their consumption flow. To analyze expenditure on durables instead we would need a detailed study on all household durables which we do not have. Currently the HBS provides extremely lumpy and hence unusable expenditure data. The same is also true for housing where we want to impute a value of housing for those who own their home. Given that housing markets in Moldova are quite small, it would be difficult to estimate the market rental rates for the housing consumption services during 1997 to present. Also, utilities were highly subsidized until 2000, after which the subsidy regime changed. Including expenditures in this category may therefore be misleading, since some household might be obtaining the services but not reporting any associated expense until 2000. Furthermore, this consumption category is presumably subject to a lot of seasonality.

Given our data limitations for dealing with these issues, the consumption aggregate and poverty line estimated in this study excludes utilities, housing and consumer durables. Signoret (2002), however computes new poverty lines and rates including utilities, housing and durables in the total consumption aggregate. He proves that overall poverty rates in Moldova do not change much.

Equivalence Scale

For the purposes of this study we use consumption per capita (i.e., per household member) as an acceptable