voxeu | It is a well-documented fact that top-income growth has been
particularly stark in English-speaking countries, with incomes of the
top 1% and 0.1% rising sharply over recent decades (Atkinson et al.
2011, Blanchet et al. 2019). Some scholars have argued that the shared
economic and political institutions of these countries, such as lower
top marginal tax rates and light touch regulation, have incentivised
rent-seeking behaviour among their top earners (Piketty et al. 2011,
Bivens and Mishel 2013). In addition, technical changes may have created
‘winner-takes-all’ markets where a few workers earn most of the
returns. Recent technological innovations may thus have contributed to
the rise of top incomes (Rosen 1981, Kaplan and Rauh 2013, Koenig 2020).
Another characteristic feature of Anglo-Saxon countries is their
popularity as a destination for high-skilled migrants, particularly the
cities that serve as global services hubs such as London and New York
(Kerr et al. 2016, Kerr and Kerr 2018, Roarch et al. 2019). Anecdotal
evidence suggests that the rich and famous are internationally mobile
and favour these destinations, but this evidence is based on a few
highly visible cases. So far, we know little about the magnitude of
these effects and the extent to which migration-induced selection
effects could drive different trends in income inequality across
countries and periods.
In a new paper (Advani et al. 2020), we combine top-income records
from UK tax records with new information about migrant status to analyse
the link. Tax records have been instrumental in recent research on top
incomes. These data provide improved coverage of the highest incomes
(reviewed in Atkinson et al. 2011); however, the data include minimal
information on demographic characteristics. As a result, it has been
difficult for researchers to distinguish native workers from migrant
workers at the top of the income distribution, or to assess the impact
of migration between countries on income dynamics. We derive information
on migrant status from the structure of Social Security numbers
assigned to migrant workers on arrival.
Our findings suggest that migrants are highly represented at the top
of the income distribution. The public debate primarily focuses on
low-income migrants; however, migrants make up a higher proportion of
earners at the very top of the income distribution. Among low-income
groups, about one in six people are immigrants. In contrast, among the
top percentile of the income distribution, one in four people are
immigrants and at higher fractiles, every third person is an immigrant. A
lack of data has created a perception that migration is mainly a
low-income phenomenon, but these new data show that the economy relies
most heavily on immigrants for extremely highly paid positions.
The inflow of high-income migrants can also help in understanding
recent trends in top incomes. In the UK, an inflow of high-income
finance workers can account for much of the observed rise in top-income
shares over the past two decades. Immigrants make up more than a quarter
of the top percentiles’ income share, up from 18% in 1997. Over these
two decades, the importance of migrants thus increased by 50%, which
accounts for about 85% of the rise in top income over this period
(Figure 1a).
The impact of migration is even starker at higher income levels.
Among the top 0.1% and 0.01% top, migrants make up roughly a third of UK
top incomes (Figure 1a). This pattern aligns closely with the observed
expansion of the wage distribution at the top. Incomes in the very top
fractiles of the income distribution have grown the fastest in recent
decades, pulling away even from the rapidly growing incomes in the lower
end of the top 1%. The data suggest that differential migration rates
can rationalise this ‘fractal inequality’. The inflow of migrants into
the top 0.01% was nearly twice as large as the comparable inflow into
the top 1% over the past two decades. Hence, differential migration
rates may have increased the gap between the incomes of the top 1% and
the top 0.1% (Figure 1b).
0 comments:
Post a Comment