The concept of the “Poverty Line” has been rejected by the new machine learning model

Mathematicians have used machine learning to develop a new model of measuring poverty in different countries, which combines the old notions of a fixed “poverty line”.

The study by academics at Aston University, published in the journal Communications about nature, suggests that general thinking about poverty is outdated because it places too much emphasis on subjective notions of basic needs and fails to capture the full complexity of how people use their income.

They say their new model – which uses computerized algorithms to synthesize large amounts of spending and economic data – could help policymakers around the world predict future levels of poverty and plan interventions to alleviate the problem.

“No one has ever used machine learning to decode multidimensional poverty,” said lead researcher Dr. Amit Chattopadhyay of Aston University College of Engineering and Physical Sciences. “This completely changes the way people should look at poverty.”

Poverty measures seek to identify a threshold monetary level below which a person or household is defined as “poor.” These definitions have their origins back in the methods developed in the nineteenth and early twentieth centuries by reformers such as Ernst Engel and Seebohm Rowntree.

Currently, the World Bank sets the International Poverty Line at $ 1.90 a day, with about 10% of the world’s population – about 700 million people – living on less than this. It is based on a subjective assessment of the revenue needed to cover the basic needs of the poorest countries, adjusted for purchasing power parity (PPP).

In the new study, researchers looked at data from India over 30 years, dividing spending into three broad categories of “staple foods”, such as cereals, “other foods”, including meat, and “non-food”, which cover other expenses. such as housing and transportation costs. The model can be applied in any country.

Recognizing the “push-and-pull” interaction between the three categories – more spending in one area usually means a reduction in spending in another – allows for a more holistic measure of poverty that can adapt to each country’s circumstances. The researchers combined data sets on income, assets and commodity markets from the World Bank and other sources to produce a mathematical model that was able not only to accurately predict past poverty levels in both India and the United States. United, but also to predict future levels based on certain economic assumptions.

Taking into account the elasticity of demand and supply in the market, the model reviews the number of people traditionally considered “poor” in a more practical “middle class”. It can be reduced to reflect the conditions in the sub-regions of a country or even reduced to a single city or neighborhood, depending on the available data.

“The current thinking on poverty is extremely subjective, because ‘poverty’ will mean different things in different countries and regions,” added Dr. Chattopadhyay. “With this model, we finally have a multidimensional poverty index that reflects the experience of the real world of people wherever they live and largely independent of the social class from which they are considered to belong.

“Importantly, it is a model that takes into account the economic circumstances in which people find themselves and the factors that can make the biggest difference to their material well-being. As such, it can be an important tool for governments and policy makers worldwide in identifying poverty and implementing interventions that truly address it. ”