This article was written by Nathan Marcus Lobow an Economics student at the University of Southampton.
“It is easy to lie with Statistics. It is hard to tell the truth without Statistics.” ● Andrejs Dunkels
In 2023, the World Bank shifted the poverty line from $1.90 to $2.15 as part of its periodic revision factoring in inflation and other indices. As per this line, as of 2024 only 2.9% of the world’s population fall beneath it, which is significant as this number earlier stood at 13.8% in 1990 when the poverty line was first introduced and set at $1. A dollar in 1990 would now be worth $2.32, which is not far off from the current poverty line so it would seem at least by this metric that poverty has been on a downward slide even after factoring for inflation. That sounds right, doesn't it? But what about the question ‘is $2.15 really a determinant for poverty’? Surely that should matter. Take another example of my home country, India. Since 2009 the poverty rate which was 29.6% steeply declined to 11.09% in 2018. Poverty slashed by over half in under a decade is an impressive achievement. Then again, similar questions arise – What is the poverty rate and is it really adequate? To the first question, it has been set by the government at Rs 32 ($0.38) in urban areas and Rs 26 ($0.31) in rural areas and for the second one, the answer is a big loud NO.
A common theme starts to emerge across both the examples and one that arguably is observable across multiple economic indices globally. At first sight, we see significant improvements made towards reducing poverty, but when we begin to look closer, it is not the numbers that are lying, but the indices themselves. By setting a shockingly low bar, it becomes easier to show significant improvement and even paint a picture which is far from reality. Why does this not get discussed more frequently? The answer may lie in a theory I would like to propose – ‘The low hanging fruit fallacy.’ Many of us may have heard the phrase ‘low hanging fruit’ which is used to describe something that can be accomplished easily and swiftly. In the field of Economics, policymakers often must attempt to quantify abstract concepts (poverty, inequality, biases etc.) by using definite numbers in absolute terms which are non-ambiguous. This can present a challenge, but more dangerously can be manipulated to achieve certain objectives and agendas. Poverty rates in both the cases above represent such instances, wherein the abstract nature of the problem requiring a definite metric could be easily designed to fit a narrative, i.e., poverty is not as bad as it seems. To achieve this, all one would need to do, would be to set the threshold for poverty so low, that more people appear to fall outside it. To better illustrate this fallacy with an example, let us assume we are evaluating a group of monkeys’ ability to jump. To determine whether it can jump or not, we design a test wherein we hang fruits (low hanging ones perhaps?) a certain height above the ground (say 5 meters) and make a monkey jump in an attempt to grab them. If the monkey grabs them, we determine that they can jump. If not, we conclude the inverse. Here we have an abstract concept (jumping) and have produced a definite test to determine it (the fruit test). Assuming 60% of monkeys are successful in the test, what can we conclude? Would it be that 60% of monkeys can jump? At first again we could agree with this conclusion but think about this. Would that conclusion still hold at 10 meters, 20 meters or 50 meters? Any result obtained always comes with a constraint and a frame of reference, i.e., the parameters set to determine an abstract concept. ‘60% of monkeys can jump 5 meters’ may sound similar to ‘60% of monkeys can jump’ but it comes with a caveat, which is crucial in such cases as it not only provides the conditions under which a statement holds true, but also throws light on the fact that the conclusions could very well change if the conditions were altered. Thus, our initial goal of using the test to determine a monkey's ability to jump now calls into question of what we consider an acceptable height to determine a jump.
Back to the topic of poverty, we earlier noticed how the global poverty rate stood at 2.9% when the line was set at $2.15 (2023). Hypothetically, if we set the line at $6.85 instead of $2.15 (which arguably is a much fairer line to measure poverty as it is more in line with current costs of living), the revised poverty rate would stand at 21%! Using the same for India, the poverty rate would rocket up to a shocking 38%!! See how the same abstract concept (poverty), when subjected to different definite measures (poverty lines of $2.15 and $6.85), gives us two different outcomes. Numbers tell us the truth. It is concrete and precise, which is why it is often favoured when trying to quantify and interpret the world around us. However, they can be designed and presented in ways where they distort the truth. This is especially crucial in a field such as Economics, where policies that affect the lives of billions across the globe are determined by extensive use of data and number driven models. When organizations and government institutions make policies intentionally or otherwise using models with poorly defined or low parameters, it is the billions globally who are affected. It is easy to make overachieving claims by hiding key facts and obfuscating the real depths of an issue when low parameters and boundaries are set for numbers. As the saying goes – ‘It is always the low hanging fruit that are the most rotten.’
The views and opinions expressed in this article belong solely to the writer and do not necessarily reflect the views and opinions of the Warwick Economics Summit.
References:
● https://www.worldbank.org/en/news/factsheet/2022/05/02/fact-sheet-an-adjustment-to-gl obal-poverty-lines#9
● https://data.worldbank.org/indicator/SI.POV.GAPS?end=2022&start=1990&view=chart ● https://www.youtube.com/watch?v=E91bGT9BjYk
● https://prsindia.org/theprsblog/how-is-the-poverty-line-measured?page=2&per-page=1 ● https://coresignal.com/blog/data-science-quotes/
Well that's a tickle on the traditional perspective of poverty, rather statistics as a whole. Great work and thinking