foto knowledge fotos knowledge 2

Eradicating poverty is not only bold and complex but also the ultimate challenge. To be successful in looking into the future we always need a starting point so the first step will surely be to understand what we have learned up to now. Now more than ever the international community has accumulated a deep and robust understanding of poverty. Tracking the path of how we arrived here and observing this privileged knowledge deposit may serve to understand how poverty analysis has evolved in terms of concept and measurement contributing to the recent remarkable poverty reduction achievements but also to find new solutions for the future and analyse poverty in its new forms.


Understanding Poverty: The Concept

To study poverty first we have to understand the concept. Poverty is frequently seen as the defining characteristic of underdevelopment making its elimination the main purpose of economic development. However if that is so, it is less obvious and consensual what poverty means or is in reality. Defining poverty has always been a complex task, mainly because it is above all a multi-dimensional concept. For some, poverty is the state of being without, often associated with need, hardship and lack of resources across a wide range of circumstances. For others, poverty is a subjective and comparative term. For some others it is a moral issue attached to an evaluative judgement and for a few others, as the economists, it will have to be scientifically established to be analysed. There is a diversity of opinions and approaches that can be used to define poverty which is very far from delivering a unique definition.

Definitions of Poverty:

·          Poverty is the state of having little or no money and few, or no, material possessions.

·          To be impoverished is to lack or be denied adequate resources to participate meaningfully in society

·          Poverty is the state of being deprived of the essentials of well-being such as adequate housing, food, sufficient income, employment, access to required social services and social status.

·          Poverty is a situation in which a person or household lacks the resources necessary to be able to consume a certain minimum basket of goods. The basket consists either of food, clothing, housing and other essentials (moderate poverty) or of food alone (extreme poverty).

·          Poverty is the condition of possessing an income insufficient to maintain a minimal standard of living.

·          Definitions of poverty are culturally specific, and thus relative to the social norms and expectations endemic to a given nation-state. However, the condition of absolute poverty (i.e. lacking the income to maintain a minimum diet) is acknowledged worldwide.,,

Who better to answer what is poverty than poor people themselves. Following criticisms that top-down public social policies hardly reached the poor Participatory Poverty Assessments (PPAs) promoted by the World Bank have focused on the personal experience of poor people reporting their daily life and showing the human face of poverty. These PPAs reviews (43) showed that poor people report their impoverished status mainly in terms of material deprivation e.g. low incomes, lack of or unstable employment, shortage of food, inadequate housing, combined with inadequate access to health services and clean water.  However they also give weight to non-material social, psychological factors such as insecurity and loneliness; social and political conflict; lack of autonomy or exclusion from decision making.

Following this trend, poverty initially defined as material deprivation, has increasingly been systematised by hierarchal levels, that start in a more narrow and concrete level, focusing mainly on income and consumption patterns and progressively incorporating other factors such as: social spending or public expenditure on education, housing and infrastructures; assets including land livestock and housing or consumer durables such as radios (Baulch 1996). It ends in a broader and more holistic view of poverty that engulfs more psychological features and general human conditions as self-esteem and self-respect, dignity and vulnerability or as Amartya Sen stated the pure enlargement of people’s choices and freedoms.

Measuring Poverty: Literature, Axiomatic and Indicators

If finding a consensual and unique definition of poverty has always been a difficult task, it is not surprising that measuring it has been an even more delicate and complex process that involves considerable technical issues and theoretical assumptions.

In economics literature the first effort in studying poverty was based on social welfare functions used to measure the living standards of the population. Its main advantage was the statistical aggregation that summed up the welfare of individuals turning the distribution into a single number that provided a judgment about overall welfare. However, by aggregating the social welfare of all population in one statistic, this approach failed to isolate poor in the overall population. But in the 70´s, research although influenced by a compelling axiomatic framework on inequality measurement was also motivated by very practical considerations regarding poverty measurement such as the best way to identify and aggregate poor. This led to a flourishing literature on new indices and the debate on the construction of poverty lines (Bourgignon, Cowell et all). The new trend was to develop poverty measures that focus on poor specifically identifying poor people through thresholds called poverty lines. The use of poverty lines allowed for the first time to narrow down the study only to the poor facilitating their identification and posterior aggregation into meaningful measures. Amartya Sen and Deaton were prominent contributors particularly in the aggregation step that led to what was called the Sen´s Measure:

Figure 1: Sen Measure

S(x;z) = H (I+(1-I)Gp


x is the income distribution, z is the poverty line

H is the Headcount ratio or frequency of the poor

I is the income gap ratio or the average normalized shortfall among the poor

Gp is the Gini Coefficient among the poor

The Sen Measure captures not only the frequency reflected in the Headcount but also the depth and distribution of the poor. Based on inequality analysis the Sen´s measure included for the first time the Gini coefficient in poverty measurements, but albeit the breakthrough from the focus on relative deprivation the measure was difficult to use in empirical applications. In fact the Sen´s measure is a normalized weighted sum of shortfalls selecting weights based in the rank order of poor income. Sen´s paper was more for theoretical discussion, having more a mind-blower effect than empirical applications. The reason for that was that it was not decomposable across subgroups and thus not useful for regional data which hampered the construction of poverty profiles.  When it comes to subgroups the measure always comes back to H which is not very interesting. Sen´s axiomatic was good but it was needed a more broadly applicable measure.

Presented to us in 1985, the FGT measures have become the standard for international evaluations of poverty and had a great impact on the work on theory but most of all in policy applications. Many factors contributed for the FGT measures becoming the most widely used measures of poverty. Its simple structure based on powers of normalized shortfalls facilitates communication with policymakers. However one fundamental advantage is that it allows evaluating poverty across subgroups of the population in a coherent way due to its sound axiomatic properties of additive decomposability and subgroup consistency. The initial use came from this unique practical advantage but additionally further research has shown that FGT indices were closely linked to stochastic dominance which not only enhanced predictability and robust results but also provided a unifying structure linking poverty, inequality and well-being.



Figure 2: Foster-Greer-Thorbecke Poverty Measures:

P(α) = 1/N ∑(( z-y)/z)α


Where yi is the income of individual i, z is the poverty line, N is the total population, k the number of poor people and α is a parameter that represents the degree of aversion to inequality among the poor.

This measure has the advantage of splitting into three aggregate measures through the simple change of the exponent:

If α=0   the measure gives the Headcount index (Incidence) which is the proportion of people below the poverty line.

If α=1  the Poverty Gap index (Intensity), which is the average shortfall of the poor’s income from the poverty line, averaged over the whole population.

If α=2  the Severity Index (Inequalities). This weights incomes below the poverty line convexly and so captures the inequality of incomes among the poor. Incomes further from the poverty line have more weight.



Absolute Poverty Analysis

The study of absolute poverty has two fundamental steps. First we choose the welfare indicator that traditionally is income or consumption per capita, although consumption is usually the favourite indicator in developing countries. Consumption is typically a better measure of current living standards because is less volatile and more easily measured than income due to the fact that most poor are engaged in the informal sector or family work in developing countries. The second step is to construct the poverty line. It is a complex exercise but to simplify we need to  ultimately answer to two philosophical questions: What is the adequate minimum level of well-being below which he/she is considered poor in a specific local context? and How can we find the minimum amount of money that corresponds to that same level of well-being? There are two main methods currently used to construct national absolute poverty lines: the Food Energy Intake method (FEI) and the Cost of Basic Needs (CBN). The FEI method is based in the assumption that as income (or expenditure) rises, food energy intake also rises, and the poverty-line is the level of expenditure, z associated to a given minimum adequate level of calorie. One of the advantages of this approach is its parsimony as it does not require any information about the prices of goods consumed, however it relies on an assumed relationship between household expenditure and food energy that may not hold throughout time so it does allow comparisons across time and regions. Thus findings of FEI may be seriously flawed and should not be used unless alternatives are unfeasible. (Ravallion and Bidani, 1994).

The CBN method is the favourite although it is slightly more complex in terms of calculation, requiring extra data on prices and detailed consumption data, but in turn it makes it more rigorous. This approach stipulates a bundle of goods that are associated to a minimum status of welfare and reflect local consumption patterns. This is the tricky part: how do we define an adequate bundle that represents basic needs? Usually the method used is to take the average nutritional requirement for an individual to be in good health, (often approximated to be 2,100 calories per person per day), as the benchmark to compile a basket of goods that reflect local diet near the poverty line.[1] Then the cost of this basket is estimated based on the prices of local foods, which makes the food poverty line. Additionally it is also included in the overall poverty line a nonfood component corresponding to mostly the costs for housing, clothing or electricity. In the absence of an objective caloric requirement, it is difficult to set what is the adequate or basic need of non-food components and indeed there is no consensus or best practice to estimate the non-food poverty line. One option is to arbitrarily stipulate a bundle that reflects basic non-food consumption and then price it accordingly[2]. There are no normative criteria and different methodologies are found in practice, but typically it is set according to food demand behavior in each sub-group of the population and is found by looking at the non-food spending of people in a neighborhood of the food poverty line. Thus in the Cost of Basic Needs approach the poverty line is estimated as the cost of a basic bundle of goods that corresponds to a low cost adequate food diet but also to  non-food basic consumption requirements.

The use of poverty lines allowed narrowing down the study only to the poor and most of all identifying the poor and aggregate them in meaningful measures, such as the most often quoted poverty indicator the Headcount ratio or the FGT measures. Although these measurements are the most popular and mostly used to make poverty analysis these techniques are limited money metrics that focus only on income/consumption and do not tackle the multi-dimensional side of poverty. Nevertheless measurements of absolute poverty have spread use because of its simplicity as it is very difficult to use subjective variables of poverty like self-esteem, dignity or psychological features as defining indicators. Indeed measuring the welfare of an individual/household is not an easy task, but it could be done if one restricted the concept to material or economic welfare, making it feasible. However, by doing that, a panoply of non-material factors that influence happiness and satisfaction is subtracted from the analysis for the sake of practical reasons. However albeit its limitations the absolute poverty approach allowed overcoming constraints in poverty analysis addressing the issue of identification and aggregation of poor which were fundamental to produce national poverty profiles and poverty household statistics crucial for targeting the poor and implementing  poverty reduction policies in the last decades.


Relative Poverty Analysis: Comparing Poverty across Countries

Although absolute poverty analysis was a huge revolution that provided for the first time robust national poverty profiles, it is not possible to use these techniques to compare two or more countries. If absolute poverty is more likely to be a priority in the agenda of national governments, relative poverty is more palatable to the international community forums eager to compare and rank countries. Monitoring global poverty estimates will not be possible without a common poverty line across all countries. Since the World Development Report of 1990 that the international 1$ a day PPP poverty line (more precisely, the line is $32.74 per month, at 1993 PPP) has been used to measure global poverty, providing a comparable standard of welfare between countries. Currently the World Bank updated the initial threshold and uses an international poverty line of $1.25 a day, in 2005 prices. Deliberately conservative this line corresponds to an average of the national poverty lines of the 15 poorest developing countries; it intentionally represents a very low threshold standard of living to assure that it is anchored to low-income countries (Chen and Ravallion 2010). Constructed to reflect the standards of the poorest it also accommodates differences in the Cost of Living through the PPP adjustment designed to enable comparison of purchasing power across countries and over time. There are several ways to measure PPP exchange rates. One is the Geary-Khamis (GK) method used by the Penn World Tables (PWT) that uses quantity weights to compute the international price indices. This technique tends to be more suitable to richer countries as it gives a too high weight to consumption patterns when measuring poverty globally. Instead the EKS method tries to correct this bias using the extended version of the bilateral Fisher index that is widely used to compare particularly developing countries.

In the last decade nothing has resonated more in the poverty agenda than the “One Dollar a Day” rhetoric. Some say it is too simplistic and for others it may seem more a bad publicity jargon than a real life benchmark, so what is the magic about it? It is a step forward in monitoring global poverty because it goes beyond national poverty lines allowing poverty comparison among countries. In assessing the extent of poverty in a given country one naturally focuses on a poverty line that is considered appropriate for that country, but poverty lines vary across countries in terms of their Cost of Living. To cope with this economic gradient, in such way that richer countries tend to adopt higher standards of living in defining poverty, the PPP correction is used to tackle this bias.  With PPP adjustment, two people with the same purchasing power over commodities should be treated the same way even if they live in different countries and only under this assumption one can infer if one individual is either poor or not poor. But why not use a common exchange rate to convert different poverty lines?

International comparisons of economic aggregates have long recognized that market exchange rates—which tend to equate purchasing power in terms of internationally traded goods—are deceptive, given that some commodities are not traded; this includes services but also many goods, including some food staples. Furthermore, known in the literature as the “Balassa-Samuelson effect”, there is likely to be a systematic effect: low real wages in developing countries entail that labor intensive non-traded goods tend to be relatively cheap. In addition it is the now widely-accepted explanation for an empirical finding known as the “Penn effect”—that GDP comparisons based on market exchange rates tend to understate the real incomes of developing countries. Similarly, market exchange rates overstate the extent of poverty in the world. So for all these reasons global economic measurement, including poverty measurement, has used Purchasing Power Parity (PPP) rates rather than market exchange rates.

The beyond 2015 agenda has burst discussions on the best way to measure poverty globally in which unavoidably the “One Dollar a Day” debate came up to stage. There are several views on whether this frugal line adequately embraces current standards for defining poverty. Martin Ravallion, the mentor of the 1 $ a day measures, argues that the $1,25 poverty line is necessary and useful, but suggests that to gauge sensitivity it can be used a higher line set at the double of the 1,25 Dollar a day. It has also proposed a new measure of poverty called “weakly relative poverty” that combines absolute and relative poverty to adjust over time or across countries for differences in the costs of avoiding social exclusion and relative deprivation.

It is widely agreed that eliminating extreme poverty in the world should take priority in thinking about our development goals going forward. The ‘$1 a day’ poverty line is a simple metric for monitoring progress toward that goal. It was chosen in 1990 as a typical line for low-income countries (as explained in Dollar a day revisited). By this measure, poverty in the world as a whole is judged by a common standard anchored to the national lines found in the poorest countries. On updated data, the current value of this international line is $1.25 a day at 2005 purchasing-power parity. Today about 1.2 billion people in the world live in households with consumption per person below this frugal line. Thankfully, the world has made progress in bringing this count down; 1.9 billion people lived below $1.25 a day in 1990.”

Martin Ravallion is the Edmond D. Villani Professor of Economics at Georgetown University, Washington DC. Ex- director of the World Bank’s research department.

Other contributions argue for higher international poverty lines as richer countries tend to use higher lines because food bundles are more expensive and the allowances made for non-food needs are more generous (Lant Pritchett, 2013). Other advocate internationally coordinated national poverty lines (Stephan Klasen, 2013) that include both the headcount and depth of multidimensional deprivation (Sabina Alkire, 2013). Others focus on relative poverty that allows the distinction across different types of poor people (Amanda Lenhardt and Andrew Shepherd, 2013). Finally, there is also the potential contribution of Big Data to poverty measurement in light of the “Data Revolution”(Emmanuel Letouzé, 2013).

Whatever potential revisions, setting the global poverty line at $1.25 per person per day in real terms means focusing on the standards of the world’s poorest.

[1] This is a delicate task because different calorie requirement can be met with multiple food baskets leading to different costs that should accommodate local price levels, resulting in poverty lines that can vary widely.  (Pradhan and others 2000; Haughton and Khandker 2009).

[2] Alternatively another approach is to divide the food component by the average share of food in total household expenditure and try to find out what is the share of non-food (Orshansky 1963).


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: