This article is published in partnership with Dialogues économiques, produced by Aix-Marseille School of Economics (AMSE).

“Man, 28, from an excellent family, pleasant appearance, good prospects, wishes to marry young lady with a dowry.” Published in Le Chasseur Français1 in 1896, this ad now raises a smile. Who today would openly display their desire for income on Tinder or Match?

And yet, at the time, speaking about money was perfectly natural. In 1903, a “keen hunter” about to return to “Tonkin” (as it was then called)2 specified the amount of his assets (8,000 francs in the colonies, 2,700 in France) to attract the “well-groomed and educated” woman of his dreams. In 1904, a “young man of high society, perfect in every respect, wealthy, fortune entirely in securities” had no qualms about seeking a young woman “with land”3.

This outspokenness is hardly surprising. When it comes to the question of who marries whom, the demographer and sociologist Alain Girard pointed out as early as 1964 that Cupid does not strike at random: people mostly marry within their own social group4.

The turn of the 1970s

It was not until the late 1970s that this social homogamy – the tendency to marry someone from the same background – began to decline. French society then became more open, although the propensity of graduates from elite schools to marry among themselves remained strong5.

The question is how this change came about. Is it because we now come across people from different social backgrounds more frequently than in the past? Or because we place greater value on emotional harmony than on financial security?

In other words, is the current evolution linked to a change in the “marriage market”, or to a shift in our preferences? It is this question that the economists Quentin Lippmann and Khushboo Surana set out to answer.

Mobile apps have replaced newspaper ads.

Mobile apps have replaced newspaper ads.

AndreyPopov / iStockphoto.com

AndreyPopov / iStockphoto.com

Share

Share

One million personal ads to go through

In order to capture people’s desires even before any actual meeting took place, the two researchers turned to matrimonial ads published in France, North America and India. These ads were extremely popular throughout the 20^th century, even if they seem to have led to relatively few unions (between 1% and 3% in France)6.

The two economists collected around one million advertisements placed by those seeking a soulmate, published from 1950 up to 1995, the year Match.com – one of the first online dating sites – was created. They subsequently “read” them, or rather subjected them to statistical analysis. This analysis was all the easier to automate given that the texts are nothing as complex as mediaeval courtly poetry – paid for by the word, the ads are short and to the point: “X (young man, widow, wealthy divorcee, etc.) would like to meet / would respond to / would marry Y (attractive musician, Paris-based executive, or someone aged 38–45).”

Screening criteria for choice

By identifying the phrases linking the two parts of the ads, the researchers were able to isolate the offer (the person writing) from the demand (what they are seeking). It is this latter dimension that attracted their attention: whom are people looking for? What criteria guide their choices?

To find out, the researchers identified four categories of vocabulary based on the most frequently used words each year: economy; personality; tastes and cultural preferences; and physical characteristics. After checking for various biases, they observed how the relative weight of each of these semantic fields evolved over time.

The vocabulary changes markedly. In the 19^th century, as evidenced by the historian Claire-Lise Gaillard7, economic criteria were predominant in France. The two economists demonstrate that this trend, which they observe across all four countries studied, begins to shift from the late 1960s onwards.

Personality takes precedence in France and North America

In France and North America, the share of economic vocabulary declines significantly, giving way to a growing importance of terms describing personality (“pleasant”, “emotionally stable”, and so on) – with particular emphasis on race in American advertisements.

Females now sought to meet a “masculine, refined, intelligent, generous, well-off” male “for a harmonious life” (Le Chasseur Français, 1970), or even “a serious young romantic – if such a thing still exists” (ibid., 1973). Might this evolution simply reflect a new form of prudishness, where it may be inappropriate to mention financial criteria openly?

Hindu wedding, in the Telougous fashion, one of the peoples in southern India.

Hindu wedding, in the Telougous fashion, one of the peoples in southern India.

Doctorpori / Wikimedia Commons 2023 CC BY-SA 4.0

Doctorpori / Wikimedia Commons 2023 CC BY-SA 4.0

Share

Share

India, a different model

To verify this, the researchers also analysed how the authors of these ads described themselves. If economic criteria had remained a priority in partner selection, individuals would likely have continued to highlight their credentials in order to enhance their attractiveness.

On the contrary, they stopped presenting themselves in terms of their economic position, indicating that this criterion genuinely lost importance, both in what people offered and in what they sought – at least in advertisements published in France and North America. In India, however, the pattern is radically different. There, economic considerations became increasingly prominent in descriptions of the desired partner from the 1970s onwards. How can such divergent trajectories be explained?

Women’s economic independence

In Western countries, the late 1970s saw fewer mentions of explicit economic criteria, at a time when ways of life were changing: marriage rates fell sharply, cohabitation became more common, divorce increased, and people had children later in life. Could these phenomena share a common cause, such as the growing involvement of women in the labour market?

In France, for instance, the employment gap between men and women narrowed dramatically between 1975 and today (reducing the gap from 36-points to 9-points)8 . In the United States, women accounted for one in seven workers in 1966, compared with one in two by 20139.

Diverging needs

This hypothesis may, conversely, help explain the distinctive pattern observed in India. There, women’s integration in the labour market remained relatively low between 1950 and 200010. A closer examination of the vocabulary used in Indian ads seems to support this idea: within the economic lexicon, words relating to employment and work are the most frequent.

Quentin Lippmann and Khushboo Surana naturally refrain from drawing hasty conclusions. Nevertheless, they suggest that their findings are consistent with the hierarchy of needs proposed by the American psychologist Abraham Maslow in 1943. This theory describes human needs according to their order of priority: eating, drinking and sleeping are fundamental necessities which, once satisfied, allow individuals to turn their attention to other imperatives, such as self-esteem and fulfilment.

In the case of India, persistent material insecurity, despite the country’s economic development, may remain a primary concern, hence the emphasis on financial criteria in choosing a partner. By contrast, the greater economic independence of women in Western societies encourages the search for relationships based not on material necessity – making a home and a living – but on the desire to build a harmonious partnership. Yet, whether this has made the search easier is doubtful, judging by the proliferation of websites promising miracle solutions for finding a soulmate.

Scientific reference

Quentin Lippmann, Khushboo Surana, “The evolution of partner preferences: evidence using matrimonial ads from Canada, France, India and the United States”, Journal of Economic Behavior & Organization, volume 233, 2025.