Swedish lottery data on 76,859 players: wins raised marriage odds in men, divorce risk in women

A team led by Anastasia Terskaya of the University of Barcelona, with David Cesarini of New York University and colleagues, analyzed records of 76,859 players of Swedish lotteries. They tested how a sudden windfall affects marriage, divorce and childbirth, psypost.org reports. The work was published in the Journal of Human Resources. psypost.orgpsypost.orgpsypost.orgOpen source ↗

How the study was built

People with higher incomes differ from others in education and career ambitions, so income alone does not show whether money causes marriage. The authors used lotteries, where the size of the prize is random. They took three Swedish games: a monthly subscription lottery, a scratch-ticket lottery and savings accounts that paid lower interest in exchange for automatic entry into draws. Player records were matched to government registries of marriage, divorce and births.

The players resembled other Swedes of the same age and sex: 33 percent were married versus 35 percent in the comparison group, and 24 percent had attended college versus 30 percent. Players were grouped by exact participation, so those who faced the same odds but won different amounts could be compared. The study covered adults aged 18 to 44 at the time of the win and followed them at two, five and ten years. Effects are expressed per 1 million Swedish kronor (SEK) after taxes, about $140,000 in 2010 and more than five times a typical Swede's annual disposable income.

Marriage, divorce and children

For unmarried men, 1 million SEK raised the probability of marrying within five years by 4.7 percentage points. The baseline for this group was about 16 percent, so the rise is roughly 30 percent. The gain was concentrated among men whose income before the win was below the median. For unmarried women the effect was small and not statistically significant, but the gap between the sexes was not large enough to rule out chance. psypost.orgpsypost.orgpsypost.orgOpen source ↗

Among married women, the same prize nearly doubled the divorce risk within two years, raising it by almost 4 percentage points over a baseline of 4 percent. The effect faded by the ten-year horizon and was concentrated among women with below-median income. The authors suggest the money accelerated divorces that were already under consideration. For married men a lower divorce risk was observed, but it was not firmly established.

Ten years after a win, men had on average 0.056 more children per 1 million SEK, a 14 percent increase. For women the rise was not statistically significant. Across the whole sample, the number of children over ten years grew by 0.033 per 1 million SEK.