Paul Krugman once
said that to improve a country’s standard of living over time
“productivity isn’t everything, but, in the long run, it is almost
everything”. I want to use a recent Resolution Foundation study to
examine a slightly different question, which is what determines
differences in prosperity across countries. The answer is very
similar, but with an important modification.
The Resolution
Foundation report by Krishan Shah and Gregory Thwaites
compares productivity and (PPP adjusted) incomes per household in the UK with the
US, Germany and France, and with France it looks at both 2008 and
2019 so we can look at the comparison over time. But it starts with
the following chart which includes many more countries.
This plots GDP per
hour (productivity) on the horizontal axis against median income
(both logged) for a number of countries. The line passing through the
points is the 45 degree line, and the fact that the points are
clustered around this line shows that differences in productivity are
crucially important. However there are big divergences from that
line, suggesting other factors are important.
The first key point,
which can get lost in the detail of the report, is that incomes are
not the same as prosperity, if you define prosperity in a more
general sense. Three of the most important aspects of prosperity that
are not captured by incomes are leisure, public goods and investment.
Consider each in turn.
Imagine two
countries. In one, people work long hours, have few holidays and have
a long working life, and as a result their incomes are high. In
another, people work less hours, have longer holidays and retire
earlier, and their incomes are less as a result. It would clearly be
a mistake to call the country where people work more hours a more
prosperous country. We could ask the same question where incomes
differ because of different levels of tax, where tax goes to pay for
more public goods. The country where incomes are higher but less
goods are provided by the state is not necessarily more prosperous,
particularly if private sector provision of these goods is less
efficient (think US healthcare). These are key issues when comparing
the US and France, for example.
The final point is
that you could raise incomes by not investing in the future. As
future productivity depends on investment today, this might raise
people’s incomes today, but at the expense of their incomes
tomorrow. Differences in investment may occur not just in producing
more capital goods, buildings etc, but also with investment in
education, or simply in terms of income from overseas assets.
These factors are
important to consider when we look at the relationship between
comparisons of productivity and comparisons of income per household.
Here is the report’s comparison between the UK and France in 2019.
On the left we have
GDP/hour worked, a measure of productivity [1]. That shows that
France is 17% more productive than the UK. The penultimate column is
average household income, where France and the UK are almost equal.
Why is France more productive but incomes are no higher? The main
answer is the ‘worker/population’ column, which in this case
mainly reflects earlier retirement in France (but also longer life
expectancy). Does that mean that the average French person is not
more prosperous than the average person in the UK, despite being more
productive? Almost certainly [2] not, because people in France have
decided to use their greater productivity to retire earlier.
Differences in the
proportion of workers to the population doesn’t just reflect
retirement. There are fewer young people in the workforce in France.
This is partly an investment effect (more education) but also
reflects high youth unemployment. The other big factor reducing
average incomes in France is the ratio of domestic household income
to national domestic income. This partly reflects the fact that
French firms invest more so the share of profits in GDP is higher
(and the wage share lower), but it also reflects higher taxes and
(almost certainly) therefore more public goods. [3]
I hope it is now
clear why I wanted to stress the distinction between incomes and
prosperity. Although average incomes in France may be no higher than
in the UK, the French are still more prosperous because they have
used their productivity advantage to have a longer retirement, have
more public goods and to invest more in the future. So productivity
remains crucial to prosperity, but how people enjoy that prosperity
can be quite different between countries.
A final but crucial
point comes from comparing the last two columns. Median income is the
income of the person in the middle of the income distribution, where
you have as much chance of having an income above or below that
level. If the distribution of income is very unequal, and in
particular if it is skewed in favour of those at the top, median
income will be below average income. Median incomes are significantly
higher in France than in the UK, because the UK is more unequal. So
although productivity is crucial in making cross country comparisons
of prosperity, inequality is also important. (For a more detailed
comparative analysis of different income brackets, see John
Burn-Murdoch here. For a discussion of the impact of
changes in the proportion of income taken by the top 1% in the UK
over time, see here
and particularly here.)
The comparison for
2008 rather than 2019 illustrates a key point that is familiar. While
the productivity gap in 2019 was 17%, it was only 7% in 2009. The
last 10/15 years really has been a period
of UK decline. The 2019 comparison with Germany throws
up similarities and differences to France that the report goes into.
While the productivity gap is similar, the benefits are taken in
terms of working less hours rather than less years. Turning to the
US, the productivity gap with the UK is similar to the gap with
Germany and France, but US income is much higher. Some of that big
gap is because workers in the US work more hours, and taxes are lower
because public good provision is lower, but there are also
differences that must reflect problems with the data used.
This analysis by the
Resolution Foundation illustrates two general points. First,
comparisons of personal (post-tax) income levels are a partial
indicator of relative prosperity, because they ignore leisure,
investment and public goods. For that reason, a comparison of
productivity levels may be a better indicator of comparative
prosperity than relative income levels. Second, what productivity
ignores is the often significant impact different levels of
inequality can have on the prosperity of the typical household.
[1] GDP/hour worked
is a very aggregate measure of productivity, and could reflect
different compositions of output as well as how productive similar
firms are.
[2] We could drop
the almost if we could be sure that the difference in retirement ages
represented national preferences, including choices about retirement
incomes.
[3] In theory higher
profits could reflect higher dividends rather than higher investment,
of course. This links to the decoupling debate (between productivity
and real wages) I
talked about here, based on work
by Teichgräber and Van Reenen.
