The price of poverty: interpreting the updated extreme poverty line

In this blog, DI’s senior analyst Zach Christensen contextualises the World Bank’s September 2022 revision of the extreme poverty line and explains the update process. He also interprets how this change affects poverty data and – vitally – people who are living in poverty.

► Explore global, regional and national trends in extreme poverty.

► Find out more about multidimensional poverty.

► Share your thoughts with us on Twitter or LinkedIn

The data behind global poverty estimates

The World Bank is the guardian of the most high-profile statistics on global poverty. It collects and analyses data that documents the number and proportion of people living in extreme poverty in (almost) every country in the world. This data can be used to track changes and measure progress over time.

Two main data inputs are used to calculate internationally comparable poverty estimates:

1. Household survey data: this records how much money people earn and spend.
2. Price data: this shows how much things cost.

Combining this data makes it possible to estimate whether people’s income and consumption are sufficient to pay for a universal set of basic goods and services. If this combined figure falls beneath the threshold set by the World Bank, people are identified as living in extreme poverty.

Household surveys are conducted in different years in different countries, with most countries surveying their population every three or four years. Estimates for poverty are updated by the World Bank annually using the latest available surveys. In contrast, price data is collected globally in one massive exercise at a specific point in time. This results in a one-off adjustment to all countries’ poverty estimates and to the value of the poverty line itself.

The International Comparison Program enables price comparisons

The International Comparison Program (ICP) (currently hosted at the World Bank) is the biggest statistical project in the world. It collects data from millions of data points from 176 countries (for the 2017 round) to compare the purchasing power of different currencies. Purchasing power parity (PPP) adjustments underlie the World Bank’s revisions of the extreme poverty line from $1.00 per person per day in 1990 to $1.25 in 2005, and from $1.90 in 2015 to $2.15 in 2022. These adjustments do not raise the bar on what we understand to be the level of extreme poverty but instead reflect inflation-related price changes that make the same basket of basic goods more expensive. In reality, imperfections in the collection and analysis of this price data can result in material changes to our understanding of who is living in extreme poverty. This creates controversy and critics such as economist Sanjay Reddy argue that PPPs should not be used to develop an extreme poverty line.

Nevertheless, in 2022, the World Bank announced that it would use more up-to-date 2017 price data to adjust its poverty line from $1.90, based on 2011 price data, to $2.15. It made efforts to show that the results of the 2017 round of the ICP would not significantly change the understanding of global poverty. The World Bank stated that switching to the new line would not be “shifting the goalposts” for the goal of eliminating extreme poverty; it would still enable valid and meaningful comparisons over time between the old and the new price data. All World Bank data is now published using both the 2011 and 2017 price data, allowing users to continue using 2011 PPPs if they choose. In practice, the share of the global population living below the extreme poverty line decreased by 0.2% due to the 2017 PPP adjustments (from 9.3% to 9.1%).

Do the 2017 PPPs change our understanding of global inequality?

Household survey and price data not only provide insight into how people with the very lowest incomes are doing, they also contribute to our understanding of how incomes across the whole of the economic distribution are faring over time.

In 2013, World Bank economists Branko Milanovic and Christoph Lakner drew on World Bank data to estimate the amount of growth seen at various points in the global income distribution between 1988 and 2008. Their graph became known as the ‘Elephant Curve’ because it showed the strongest income growth for people near the median (the top of the elephant’s head) and for people with the highest incomes in the world (the elephant’s trunk).

At Development Initiatives (DI), we investigated whether the new 2017 price data changes our understanding of the distribution of income growth. Figure 1 presents our calculations (replicating the methods of Milanovic and Lakner) using the current data in the Poverty and Inequality Platform between 1990 and 2019. We present the data using both 2011 PPP (the pink line) and 2017 PPP (the red line).

Figure 1: Growth in recent decades looks more pro-poor with 2017 PPP than with 2011 PPP.

Percentage growth in global income: 1990–2019.

Chart showing percentage growth in global income: 1990–2019.

The 2017 data estimates that the bottom five percentiles experienced 17% more growth during that period compared with estimates using 2011 PPP. The 2017 PPP numbers indicate that global income inequality has been narrowing slightly faster than suggested by the 2011 PPP numbers.

There are challenges in accurately collecting and measuring trends at the extreme ends of data, but it is clear that the 2017 PPP adjustment paints a slightly rosier picture of economic progress among those people living below the extreme poverty line.

Much of what seems to be driving this difference is that the 2017 PPP round discovered that it was possible to buy more with local currencies in Sub-Saharan Africa than was found with the 2011 PPP round.

A complicated update to a simplistic measure

This recent price update already looks dated. Since 2017 there have been radical shifts in global and national economic conditions; in recent years, Covid-19, energy and food price inflation, and conflict have changed the consumption of people around the world. It’s fair to assume that an increase in the prices of basic commodities, which make up a bigger share of the consumption basket for those on lower incomes, would significantly affect the profile of global poverty and inequality. In 2021, the ICP began another round despite pandemic lockdowns disrupting data collection in some countries. Since then, price shocks on everything from fertiliser to aluminium to onions have catastrophically impacted people living in poverty.

Each country’s household survey data also has its own limitations with respect to timeliness and coverage. Moreover, this single economic measure fails to capture the complexity of poverty and how it is experienced around the world.

Fortunately, we now have a range of different poverty measurement tools and approaches that aim to account for multiple dimensions of poverty. Still, the World Bank’s efforts to produce accurate price comparison data has made poverty measurement – and attention to poverty in public policy – more tangible. Considered alongside complementary measures that document other dimensions of poverty, the extreme poverty line can be a valuable tool helping to inform policymakers about where people’s needs are highest and where progress is taking place.