Indonesia’s Central Statistics Agency (BPS) recently announced that the middle class in the country is shrinking. In 2019, the total middle class population numbered 57.33 million. According to BPS’s most recent data, the figure in March 2024 was 47.85 million. On its face, this would mean around 9.5 million people have fallen out of the middle class over the last five years, which is being attributed to various factors including the COVID-19 pandemic, job losses in the manufacturing sector, and a recent increase in the consumption tax.

But we have to be careful when using the middle class as an analytical tool because there is no universal definition of what it is, or how it should be measured. And the size of the middle class in any economy depends entirely on how it is defined and measured. Indonesia is using the World Bank as a reference point, based on a 2019 report that argued the Indonesian middle class was growing.

That World Bank report is pretty open about the conceptual ambiguity here, writing that there “has been a surge of interest in the Indonesian middle class in recent years, but little agreement on who they are or how numerous they are.” The authors of the report go on to note that there were four major reports attempting to calculate the Indonesian middle class between 2010 and 2019 and all used different definitions, arriving at wildly varied estimates of between 30 and 81 million people.

The World Bank report measures expenditures and uses multiples of the poverty line to establish categories including poor, vulnerable, aspiring middle class, and middle class. It argued that as of 2019, the middle class in Indonesia (defined as people who spent between 1.2 and 6 million rupiah per month) was the fastest growing group in the country. But the authors were also careful to note that the “size of the middle class, the characteristics of its members, and their roles depend on how we define them.”

In the latest BPS data from 2024, although the middle class (defined here as people who spend between 2 and 9 million rupiah per month) shrank by 9.5 million, the aspiring middle class category increased by around 8.65 million. That means the total combined population of middle class and middle class aspirants remained roughly unchanged from five years ago, only the distribution has shifted. We don’t know exactly why that is, but we do know the distribution is heavily dependent on how the upper and lower bounds of these categories are defined in the first place.

This is why measuring, and making definitive claims about, the middle class is a tricky business. However, the 2019 World Bank report also identifies certain characteristics of a growing middle class, such as a tendency for consumers to spend more of their income on discretionary purchases like travel and entertainment as opposed to necessities like food and shelter. Do we see evidence of these kinds of discretionary purchases declining in Indonesia, which would be consistent with a shrinking middle class? Not really. In fact, we see evidence of the opposite.

Since 2019 BPS has been tracking domestic tourist activity (which includes business, leisure and trips to see family or friends). Domestic travel in 2023 was 16 percent higher than it was in 2019, the year when the middle class apparently had 9.5 million more members than it does today.

We also see strong growth in other discretionary purchases like movie tickets, with Indonesian consumers turning out in record numbers at the box office lately and major theater chains investing big in expansion to keep pace with demand. This is not conclusive evidence, but it is generally consistent with increasing purchasing power rather than a hollowing out of the middle class.

I don’t doubt that the pandemic caused a large number of formal sector workers to shift into the informal economy, and some of them undoubtedly never came back. I also don’t doubt that there are economic headwinds causing job losses in some sectors and regions, or that there is weakness in manufacturing. But when we look at Indonesia’s recent record of economic growth, even just the few years since the pandemic, it has been pretty strong. And if the economy is growing while the middle class is shrinking, where are the gains from this growth going?

You could argue that it’s all being captured by a narrow elite at the upper crust of the social ladder, or that growth is being powered almost exclusively by investment that doesn’t generate jobs or widespread income gains. But you have to do a bit of contorting to get there. The simpler explanation would be to acknowledge that defining and measuring the middle class is hard and imprecise, and that the recent report from BPS is just one piece of a larger economic picture.