Public sector employees are markedly more educated than their private sector
counterparts. And ‘low-employability’ individuals face large potential lifetime premia
from public sector employment. These are among the conclusions of research by
Professor Fabien Postel-Vinay and Dr Hélène Turon, published in the October 2007
issue of the Economic Journal.
Their study investigates whether public sector workers are better paid than their private
sector counterparts. Simple income measures suggest the answer is a clear yes. But
these researchers argue that assessing the ‘public premium’ should take account of
worker quality and the lifetime value of employment in each sector.
There are large differences between the public sector and the private sector in the raw
data on pay and working conditions. The researchers’ analysis of working age men in
the British Household Panel Survey indicates that average incomes are 14% higher in
the public sector, and income dispersion – the gap between the highest and lowest paid
workers – is 25% smaller in the public sector (a phenomenon called ‘income
compression’).
Income mobility is lower too: the probability of remaining in the same fifth of the income
distribution from one year to the next is on average 16% higher in the public sector than
in the private sector. And so is job mobility: the average yearly job loss rate in the public
sector is a little over half the average job loss rate in the private sector.
This first look at the data does not take account of the likelihood that the individuals
observed in each sector are different. It is possible that the bulk of these raw
differences reflects differences in the composition of the workforce in terms of both
observed characteristics, such as age and education, and unobserved characteristics,
such as ‘public service motivation’.
For example, public sector employees are markedly more educated than their private
sector counterparts: 43% of them have obtained a qualification higher than A-levels
compared with 23% of private sector employees. Taking all such differences into
account in estimating the public gap, average income is only 3.4% higher, income
dispersion 10% smaller and the job loss rate 19% lower in the public sector.
These findings confirm the prominent part played by non-random sorting of workers
across employment sectors in explaining the apparent public premium. For example,
blue-collar workers tend to be willing to ‘queue’ to obtain public sector jobs whereas
highly skilled workers are notoriously hard to recruit and retain in the public sector.
Income and employment dynamics as well as income levels are quite different between
the two sectors. This matters to forward-looking individuals as they anticipate changes
in their employment status, sector of employment and income level within a given
sector. Comparisons of cross-sections are not very informative in the presence of
income mobility, particularly when there are cross-sector differences in income mobility.
It is thus desirable to use a criterion that takes account of all aspects of the differences
between sectors in order to give a more comprehensive and accurate picture of the
comparison between employment in the public sector and the private sector.
The study does this by estimating ‘lifetime values’ of employment in each sector. These
are simply the present discounted sums of income flows over an individual’s lifetime.
Taking account of the ‘selection premium’ arising from the fact that individuals who
select themselves into the public sector tend to have different characteristics from
individuals who select themselves into the private sector, the research finds that the
average value of a job for life is slightly higher in the public sector than in the private
sector, with a public premium of about 2-3%.
The research also reveals that there is considerably less compression in the public
sector relative to the private sector in terms of lifetime values than in terms of income
flows. The authors interpret this phenomenon as the result of income mobility offsetting
differences in cross-sectional incomes.
Intuitively, thinking of incomes as comprising a permanent individual component and a
transitory random component, both sector-specific, the results suggest that most of the
observed income compression in the public sector is due to a lower dispersion of the
transitory component of income, which is averaged out when taking lifetime values.
The researchers thus argue that the greater dispersion of private sector incomes
relates to the transitory component of income. In other words, income inequality is
greater yet less persistent in the private than in the public sector, and inequality in
lifetime values is much more similar across sectors than current income inequality.
ENDS
Notes for editors: ‘The Public Pay Gap in Britain: Small Differences That (Don’t?)
Matter’ by Fabien Postel-Vinay and Hélène Turon is published in the October 2007
issue of the Economic Journal.
The authors are at the Centre for Market and Public Organisation, University of Bristol.
For further information: contact Fabien Postel-Vinay on 0117-928-8431 (email:
Fabien.Postel-Vinay@bristol.ac.uk); Hélène Turon on 0117-928-8400 (email:
Helene.Turon-Lacarrieu@bristol.ac.uk); or Romesh Vaitilingam on 07768-661095
(email: romesh@compuserve.com).