do You know our newsletter? It’ll keep you briefed on what we publish. Please register, and you will get it every month.
Thanks and best wishes,
the editorial team
Middle-class lifestyles start with $ 10 per person and day
– by Homi Kharas, Aviva Freudmann
More people have more choices: shoppers in Dhaka.
What defines the middle class? Is this mainly an economic grouping, or is the definition broader than that?
We need to differentiate between a definition and a measurement. The middle class is defined as a group of people which, while diverse, shares common features. Middle-class people share values of hard work, thrift meritocracy and individual responsibility. They are far enough away from the poverty line to be able to make choices to maximise their satisfaction and prospects. Unlike poor people who face day-to-day subsistence and have few options and unlike rich people who can generally buy whatever they want, middle-class people make economically based choices. They aim not just for material consumption, but also for enjoying life, appreciating leisure and art and beauty. While this definition is fairly loose, if we want to talk about the evolution of this group we need some kind of measurement. The most common measurement metric is an expenditure range. We use expenditure rather than income because expenditure is a more accurate measure of standard of living. Think of college students: they earn little or no money, yet they are independent, certainly not poor, and can borrow money and spend it. Expenditure is a much better measure of their material wellbeing than income. We originally chose an expenditure range of $ 10 to $ 100 per person per day, using 2005 purchasing power parity (PPP) dollars, to measure the middle class. The metric later shifted to $ 11 to $ 110 per person per day using 2011 PPP dollars, but it is the same in terms of purchasing power. You want a metric that is constant, so that you can compare expenditures across countries and across time.
Where does this specific range come from?
The basic idea of this range is to set the lower end at the level where people have money left over for discretionary spending and are not in danger of falling into poverty. The upper end is set where people no longer have to give much thought to trade-offs in their spending. This particular range has historical origins. But interestingly, the same range – or its equivalent in local currencies – appears at different points in history and at different places in the world. The earliest middle-class group consisted of clerks in Victorian England who were hired by banks to facilitate factory owners’ purchases of machinery during the industrial revolution. They earned the equivalent of about $ 10 per person per day in 2005 PPP dollars. Similarly, when the UK first introduced an income tax, the authorities wanted to tax only people with middle-class living standards and above. They set the minimum income level at which the tax was levied at the equivalent of $ 10 per person per day in 2005 PPP dollars. Again, in Latin America policymakers looked into the income or expenditure level at which a person would have a reasonable chance of not falling into poverty over a three year period. That turned out to be the equivalent of $ 10 per person per day in 2005 PPP dollars. The poverty line in the US was also set at around that range. In India, too, a national commission established a benchmark for the middle class. It turned out to be the equivalent of $ 10 per person per day. The lower end is the number that says, “you are not considered poor if you have at least this amount.” Having a range also permits us to make comparisons across geography and across time. We are able to trace the development of the middle class from the 19th century to today.
The composition of what people buy was totally different in the 19th century compared to today. And price levels differ widely among countries today. How do statisticians account for these differences when making comparisons?
The amount of money needed for material wellbeing has stayed the same, once one takes currency differences into account. Of course the basket of consumption items differs from country to country. But essentially, all people buy food, clothes, housing and transport. Once they reach middle class, they also consider buying vacations and entertainment. The basics of what people need have remained the same. As for price differences across countries, there are global studies that compare the prices of goods and services. The International Comparison Programme collects prices for a selection of essential goods and services, and these data are used to create an index that enables comparisons.
In 2017 you forecasted that by 2022, 170 million people would join the middle class every year. Is that still your expectation, despite the pandemic?
Well, what has changed temporarily is the rate of growth of the middle class. After a brief slowdown due to
Covid-19, I think the growth trend will pick up again and will hold over time. Covid-19 has probably set back the growth of the middle class by two to three years. Nonetheless, we still have more than half of the people on earth in the middle class or richer.
According to the World Bank, about 50 % of the world population have a purchasing power of $ 5,50 or more, not $ 10.
Yes, but its numbers do not reflect full per capita spending in each country. For example, in India, the Bank relies on household surveys that only capture one-third of total spending in the national accounts. My methodology is, I believe, better because it makes an adjustment for such gaps. The important point is that, in 1820 the middle class was only about one percent of the world’s population, and now, 200 years later, it represents the majority of the world’s population. That is an extraordinary development, which at its root has been caused by technology, and technological development continues apace. So today you still have hundreds of millions of people joining the middle class every year, mostly in Asia. You see big advances in education and life expectancy for these people. They work hard and have the ability to make a better life for themselves and their families. Covid-19 will probably actually shrink the middle class in 2020. We expect about 120 million people to fall out of the middle class in 2020 compared to 2019. Also, we expect that about 170 million fewer people will have joined the middle class in 2020 compared to what might have been the case without Corona. So in all, Covid-19 might be responsible for something close to 300 million people not being in the middle class in 2020 compared to what might have been without Covid-19. However, that effect will disappear fairly quickly. Hopefully by 2021 or 2022 the middle class will be back on the same growth trajectory as before. This group has been squeezed, but it will bounce back.
What are the political implications of the global growth of the middle class?
Generally, middle-class people try to ensure that their government delivers for them. They tend to strive for independence, so they favour private property, saving for the future and maximising personal choices. The middle class also favours government provision of health and education. It aims for economic safety and therefore pushes for social protections such as pensions and labour rights. The middle class champions women’s rights; the earliest suffragettes were from the middle class. The middle class also usually favours free trade, which broadens consumers’ choices. The first political victory for the middle class was the repeal of the Corn Laws in 19th century England. Those laws imposed tariffs on imported grain which kept prices high, thereby favouring landowners at the expense of consumers. The middle class fought hard to get those laws repealed.
What differences are emerging within the middle class?
Today we are seeing a bit of a fracturing of the middle class. The Covid-19 pandemic highlighted the division between the part of the middle class which is college educated and able to work remotely, and the part that is blue collar and cannot work from home. The interests of these two segments are starting to diverge. Certainly they are having different lived experiences. For example in the United States, the blue-collar segment has seen a higher incidence of alcohol and drug poisoning and suicides, the so-called deaths of despair. Among this small group we also see increases in illness, reported physical pain, mental ill health and reduced life expectancy.
Is this divergence within the middle class starting to cast doubt on using the term “middle class” to describe everyone falling within in a broad range of expenditure?
It could be that the interests of those with college educations and those without are quite different. So some now argue that education level should become a metric for measuring the middle class. I would prefer to keep the broader definition but come up with a different way to describe the lifestyles of different segments of the middle class. The divergence in lived experience may turn out to be temporary. It could be that right now some people have opportunities in the digital economy that others don’t have, but that in 10 or 15 years everyone will have these opportunities. This is a technology phenomenon. A new technology tends to benefit certain people first, before the benefits are spread throughout society. Consider the introduction of electricity: at first it was only for the wealthy, and now it is part of the middle-class lifestyle.
The middle classes are growing faster in poor countries than in rich ones. Is that disparity fuelling resentment among some parts of the middle class in advanced countries?
There is a level of saturation of the middle classes in developed countries, which helps to explain why the growth of the middle classes there is slower than the growth in low-income countries. The growth of the middle class in one place has been associated historically with more opportunities for the middle class in other places. The Marshall Plan in Europe helped the middle class in Europe, but also in the United States. As markets grow, everyone benefits. This is one of the great features of economics as opposed to politics. In economics, when your neighbour is better off, then you are better off. In politics, when your neighbour is better off, you might not be better off. Politics is more a zero-sum game than economics. The great hope is that this will be recognised. So for example, if you are in a developed country and your pension fund holds stock in Apple, you have to understand that when Apple profits from its sales in China and India that this also benefits you as an indirect shareholder. All the big brands are big brands because they can sell to billions of people in the global middle class. I could easily argue that a great part of the expansion of housing, higher education, health care, finance, insurance and many other services in the developed world is linked to the ability to build on the prosperity that has taken place because of trade with developing countries. Certainly with trade and with technological development there are transition costs. It is hard to say how long a transition will take and how politics will evolve. This points to the central fact that paying more attention to transition costs, including adapting to new technologies, is vital in a rapidly changing world.
Homi Kharas is a senior fellow and deputy director in the Global Economy and Development Programme at the Brookings Institution in Washington, DC.
Putdate 20 June 11:40 am Frankfurt time: We only asked the question about why the World Bank's estimate of global per-capita incomes diverges from the one of the Brookings institution yesterday, when the interview had already been posted. Homi Kharas responded fast.