We lived in a suburban, middle-class home. On the walls, however, hung the portrait of an 18th-century Anglo-French aristocrat, a maternal ancestor. In the intervening centuries, my aristocratic forebears had lost their status and wealth due to drink, gambling and poor decisions. Thus, my only link with royalty was that portrait and the fact that I have the same unusual middle name – Romaine – as the noblewoman in the picture.

A rare example of social mobility or a widespread pattern? Today, that is a very politically charged question, particularly in countries such as the US. But it is also a very hard question to answer definitively.

Social mobility is an issue about which politicians love to pontificate but about which we actually know surprisingly little. Economists have generally tracked mobility by looking at surveys on wealth, jobs and educational attainment over two or three generations. This has typically shown that mobility is highest in the Scandinavian countries and lowest in places such as Latin America, with the US and UK lying halfway in between.

Interestingly, these surveys tend to show that more mobile societies such as Sweden are also more equal, as determined by the Gini coefficient, the most commonly used measure of inequality, and vice versa. The idea that you can justify high levels of inequality in some nations because there is plenty of mobility – as US politicians are apt to do – does not ring entirely true, based on the economic numbers.

The problem with this widely cited economic data are that they are very limited: it typically only tracks families over a generation or two and cannot capture subtle social patterns. So Gregory Clark, an economic historian at the University of California, Davis, has recently attempted to use another innovative approach that blends sociology, economics and history. In a new book, The Son Also Rises, he has analysed surnames in historical databases around the world to work out how families have risen in terms of wealth and status over multiple generations. In the US he looked at doctors; in the UK, at elite universities such as Oxford and Cambridge; and in Sweden, land records.

This approach – like the econometric one – has its limits. Historical lists of surnames or elite jobs can be patchy, particularly since they are dominated by the paternal line. (My story of maternal downward mobility, for example, would have been missed.) But Clark reaches some thought-provoking conclusions. First, he argues that if you look at multiple generations, social mobility is lower than widely presumed in most nations. Second, there is less difference between nations than usually thought. Those “egalitarian” Swedes are not as mobile as presumed. But in America, Clark rejects the idea that mobility has recently declined sharply as social polarisation has grown – an idea posited, for example, in Coming Apart, an influential book published last year by the political scientist Charles Murray. Clark argues that mobility is indeed low in America but insists it has always been thus.