It was only a few weeks ago that, talking about tax reforms, prime minister  (CSV) told Delano’s sister publication Paperjam: “The two parties [CSV and DP] were elected on the argument that the tax burden on the middle class had to be lightened. The middle class represents virtually the entire population. That’s why we wanted a strong measure that would benefit everybody, rather than targeted measures affecting some 10% of the population.”

“Virtually the entire population?” Are we certain about that? Such claims are worth questioning and economist Dylan Theis, who recently completed a study on the subject for Improof--a think tank affiliated with the Chamber of Employees (CSL)--has been doing just that, seeking to understand what “middle class” actually means nowadays. The concept itself is vague, characterised by a mix of various socio-economic variables, and has been debated by philosophers, sociologists and economists alike.

“Given that almost everyone claims to be a member of the middle class, it’s a subject that interests me,” says Theis, who began work on the study in the summer of 2023.

An economic reading

The first definition drawn up by the CSL expert is the result of an “economic” reading of the concept, and encompasses all households with an annual income of between 75% and 200% of the median annual income. 

Theis has drawn on two main studies to establish this range, one by British economist Tony Atkinson (“a pioneer in the analysis of inequality”) and the other by the OECD. “I could have applied others, but these seemed the most relevant in the Luxembourg context,” he explains. The 75% median standard of living, for example, was chosen in order to stay well away from the poverty line, which is set at 60%.

In concrete amounts, the range translates to between €37,772 and €100,724 for a single person with no children. For a household with two parents and three children, it rises to €90,652 (lower threshold) and €241,739 (upper threshold).

On the basis of these criteria, the notion of “virtually the entire population” said by Frieden is in fact reduced to 61.4% of households. Compare this to 29.5% of households in a so-called “lower” class and 9.1% in an “upper” class.

While the middle class punches at its own weight, its income representing 61.4% of the country’s total income, the other classes are skewed: the lower class accounts collectively for only 15% of the total income, the upper class 21%. This latter number, Theis points out, is more than double the actual demographic weight of the upper class.

Wealth versus income

To challenge his analysis, Theis presents another interpretation, a grid based not on income but on wealth. With this methodology, he explains, “the middle class includes all households whose wealth is sufficient to ensure a decent standard of living for one year (defined as 75% of the median standard of living) in the event of zero income, but whose wealth is insufficient to enable them to maintain the median standard of living indefinitely solely on the returns on their wealth.”

Using the latest data available (from the Luxembourg Income Study in 2019 and, as above, adjusted for 2023 euros), the range is thus between €37,772 and €1,510,866 euros. This would put 70.7% of households in the middle class, versus 20.4% in the lower class and 8.9% in the upper class. However, these 70.7% represent only 53.3% of the country’s net wealth. The 8.9% of the upper class, in this model, hold 46.4% of the total net wealth.

The final calculation

Theis ultimately proposes a mixture of these two methodologies--income and wealth--using an arithmetical bias to annualise wealth, which is then reduced to a kind of notional income added to real income. And then: surprise! While the OECD has estimated that more than eight in ten residents imagine themselves to be part of the middle class, Theis’s calculation puts the actual figure at 51%, leaving 35.8% of households in the lower class and 13.2% in the upper class. The upper class, in this model, accounts for 36.3% of total income. “Polarisation is greater,” observes the economist.

This third calculation--the mixture of methodologies--is undoubtedly the most balanced. A household with a comfortable income may have only derisory assets; conversely, a family on a low income may have substantial assets (in agriculture, for example).

“The figure of 51% surprised me,” admits Theis. “I thought it would be much higher.”

In the end, this growing polarisation is the main finding of his research. On the figures regarding standard of living, Theis notes that while, in 2019, 61.4% of households could claim to be middle class, the figure was 70.9% almost 35 years earlier in 1985.

Taxation

“Changes in the labour market are among the possible explanations for this decline,” says Theis. The proportion of highly-skilled jobs, for example, shot up over the same 35-year period (from 14.1% to 31%). Another polarising factor is tax. Theis notes that the median “effective” tax rate for middle class households has risen from 13.2% in 2002 to 21.4% in 2019. That is an increase of 62% and a leap of 8 percentage points, a trend mirrored in the tax rate for the lower classes, which has risen from 2.5% to 10.9%. Meanwhile, households in the upper classes saw almost no change in their effective tax rate. Actually, it fell, from 22.1% to 21.6%.

Although the proportion of middle class households fell between 2002 and 2019 (from 68.3% to 61.4%), these households actually increased their share of funding of the tax system over this period (from 66.4% to 72.8%). And yet, as Theis points out, while the upper class grew in considerably in size over those years, its share of payment into the tax system fell from 33.7% to 25.1%.

“Downgraded”

“If the trend observed over the last 20 or 30 years continues, we will be in a situation where the middle classes will be in the minority,” says the economist. “Would that create a direct problem? I can’t answer that. But it would create a potential problem of social cohesion. On the whole, there is a general acceptance of our economic model, but look at what’s happening in France, where people from the middle class are falling into the lower strata. The middle class is being downgraded. They are no longer progressing, unlike the upper classes, who continue to do so--and this poses a risk in terms of confidence in institutions.”

This article in Paperjam. It has been translated and edited for Delano.