Fathers and Sons: new cross-country evidence on the intergenerational links in earnings
There is a strong link between the earnings of fathers and sons, according to new research by Professor Robin Naylor and colleagues. What’s more, the likelihood of a son having earnings similar to his father’s is greater for those born into particularly rich or poor backgrounds. And especially in the UK and the United States, the sons of earners in the top 20% are very unlikely to end up in the bottom 20% of earners.
The study, presented at the Royal Economic Society’s 2006 Annual Conference at the University of Nottingham, examines how intergenerational mobility compares between the UK, the United States and the Nordic countries of Norway, Denmark, Sweden and Finland. The main results are that:
Despite the commonly-perceived view of the US as an open society with ready opportunities for individuals to rise from poverty to affluence (from rags to riches), the evidence shows that the opposite is true. On average, a son’s earnings are more closely related to his father’s earnings in the United States than in any of the other countries.
In the UK, the connection between sons earnings and fathers earnings is weaker than in the United States, but stronger than in the Nordic countries. There is substantial earnings persistence across the generations in all countries.
In all of the countries, intergenerational earnings persistence is most pronounced in the tails of the distributions. In other words, the likelihood of a son having earnings similar to his father’s is greater for those born into particularly poor or particularly affluent backgrounds, relative to those from backgrounds with average earning fathers.
In most countries, the likelihood that a son with a very high earning father (in the top 20% or highest quintile of earners) will also subsequently be a top earner himself is especially high. For example, in the UK, 30% of sons with fathers in the top earnings quintile become top quintile earners themselves. In a world of perfect intergenerational mobility this figure would be 20%. Persistence tends to be even greater at the top than at the bottom of the earnings distribution.
The UK differs from the Nordic countries in the very low likelihood that the son of a high (top quintile) earner will become a very low (bottom quintile) earner. The probability is just 10%, half what it would be in a world of perfect mobility. In this, the United States is very similar to the UK. Downward mobility is very low in the UK and the United States.
The major difference between the United States and all the other countries is in the very poor prospects of sons with fathers earning in the lowest 20% or bottom quintile. These sons have a 40% likelihood of themselves becoming bottom quintile earners, twice that which would arise with perfect mobility and much higher than in the other countries. In the UK, the figure is 30% while in the Nordic countries it is 25-28%. The rags-to-riches depiction of intergenerational mobility in the United States is a myth: upward mobility is especially low there.
The Nordic countries tend to differ from the UK and the United States in the apparent protection of the life chances of sons of low earning fathers. The sons with fathers in the lowest quintile tend to have rather similar earnings to those with fathers in the next lowest quintile. Thus, there is a non-linear relationship between sons and fathers earnings in the Nordic countries. This contrasts with the UK and the United States, where there seems to be a relatively linear relationship between sons and fathersearnings: the lower the earnings of the father, the lower the earnings of the sons, on average, across the distribution.
This is not the first research to look at how intergenerational mobility compares across countries. But it is one of the first to do so using a common statistical methodology applied to highly standardised datasets for the different countries. For all countries, the data refer to sons born around the late 1950s and use information on the earnings of sons both in their 30s and in their 40s. Information on fathers earnings at the age of around 40 is also used.
The dataset for each country is chosen so as to enable close data equivalence to the National Child Development Study data for the UK. The NCDS data refer to all children born in the UK in a particular week in March in 1958. Fathers earnings are for 1974 when the child was aged 16. Sons earnings data are collected when the sons were aged 33 and again when they were 41 in 1999/2000. The NCDS is recognised as a very rich dataset that forms the basis for much of the leading work on the empirical analysis of intergenerational mobility in the UK.
Non-linearities in Intergenerational Earnings Mobility: Consequences for Cross-country Comparisons by Bernt Bratsberg, Knut Røed, Oddbjørn Raaum, Robin Naylor, Markus Jäntti, Tor Eriksson and Eva Österbacka was presented at the Royal Economic Society’s 2006 Annual Conference at the University of Nottingham, 18-20 April.
For further information: Contact Romesh Vaitilingam on 07768-661095 (email: romesh@compuserve.com).
Related research, items by these authors and papers references by them are available from IDEAS and EconPapers.