It is fairly well known that at the end of the marriage market couples seem to be tidily sorted over characteristics such as income, education, religion, height, beauty and even weight. And, of course, online dating sites allow singles to limit their search for the perfect mate according to these criterions. New economic research suggests that singles might appreciate being able to filter their searches even further – to potential partners whose parents have similar wealth levels as their own.
Evidence on marital sorting over parental wealth, particularly for the very wealthy and the very poor, finds that over generations the divide in wealth between the rich and the poor is being maintained through marriage.*
Let’s face it; no one is surprised that parental wealth plays a role in marriage market decisions.
Of course, this might not be the mechanism. It could just be that the children of wealth/poor families meet their future partners in those families’ social circles and that those social circles are defined over wealth levels.
Or perhaps, individuals who grow up with a particular level of wealth share common interests or values and are more likely to marry other individuals with those same interests or values.
The truth is that the data is just too poor to have a clear answer of what mechanism is driving this outcome.
What we do know is this. In the particular dataset used in this research (PSID, 1988), if we randomly matched a man whose parents’ wealth was less than $1,000 with a woman, regardless of her parents’ wealth level, there would only be a 16% chance that he would end up married to a woman whose parents also had less than $1000 in wealth. What we observe though is that 35% of men with low wealth parents have wives whose parents’ wealth is equally low. Of these same men, about half as many are married to women whose parental wealth is in the top of the range (greater than $100,000) as would have been predicted by random matching.
On the other hand, if we randomly matched a man whose parents’ wealth was greater than $100,000 with a woman, regardless of her parental wealth, there would only be a 39% chance that he would end up married to a woman whose parents also had more than $100,000 in wealth. Instead we see 60% married to women whose parents have wealth in the top bracket and only 7% married to women whose parental wealth is less than $1,000.
You would be right in thinking that assets valued at over $100,000 (including investment in the family home) is not, in general, how we would define “ very wealthy”. But sometimes you just have to work with the data you are given.
Of course another problem is that people with more wealth are more likely to have children who themselves have incomes, or are highly educated, and that is the sorting we are observing. The authors do what they can to test to see if this is the effect and find that education, at least, explains at most one quarter of the marital sorting over parental wealth.
Why do we care about this result? One reason is that there are gains from marriage and that marital sorting, of any type, suggests that these gains are not evenly distributed across socioeconomic groups.
Another reason is that it suggests that wealth inequality in society can be expected to persist over generations not just because children inherit wealth from their parents, but because they inherit from their spouses parents as well. If this marital sorting effect is sufficiently prevalent, then we might expect wealth over time to become concentrated in the hands of a very small proportion of households.
Well, my guess is that you know we are already there.
* Charles, Kerwin Kofi, Erik Hurst and Alexandra Killewald (2011). “Marital Sorting and Parental Wealth.” NBER working paper 16748.