DE ECONOMIST 139, NR. 3, 1991 SUBJECTIVE POVERTY LINE DEFINITIONS BY ROBERT J. FLIK AND BERNARD M.S. VAN PRAAG* 1 INTRODUCTION Despite the current slight improvement of the worldwide economic situation, research on poverty is still of great importance. Although poverty was thought to be a vanishing phenomenon in Western society in the seventies, in the eighties it has been generally recognized that poverty is still a problem. Given this intuitive feeling, voiced by political parties and other pressure groups, it appears to be rather difficult to identify who is poor. The easiest characteriza- tion is by means of current income Yc- A threshold value yp is defined, which is called the poverty line, and an individual is identified as poor if his current income Yc<Y~. One of the drawbacks of this method is that we know of cases where people have sufficient income but are nevertheless restricted to such an extent that they do not feel well-off at all. It may be that such people have bad health, suffer from a lack of democratic freedom, or that their income can only be earned by making extremely long working weeks. We may also think of the Russian example (1990) where people frequently have enough money, but there are no commodities in the shops to be purchased. All these examples point to the fact that in many cases a state of poverty has to be described by more than one criterion, in this case income. Although a lot of attention is paid to this multi-dimensional aspect, the poverty concept is invariably operationalized by means of only one characteristic, i.e. income, and we shall conform to this tradition. The reasons for this choice are the difficulties of operationalizing multi-dimensional concepts. In the past, many methods have been developed with a view to constructing a poverty line in terms of income. Those methods may be divided into two groups: the absolute measurement methods and the relative measurement methods. In the first category, the poverty lines are determined regardless of the general welfare level and the welfare distribution in the society concerned. * Research Assistant, Econometric Institute, Erasmus University Rotterdam, and Professor of Economics, Econometric Institute, Erasmus University Rotterdam and member of the Scientific Council for Government Policy (W.R.R.), The Hague.