When your map is unreliable, get a better one.
The case for doing statistics differently.
In September the UK Office for National Statistics announced that we were £50bn better off than they thought the day before. In other words, they found about £1000 for every adult down the back of the sofa. That is a very large error.
And then last week they stated that the official statistics for worklessness were not fit for purpose. They are survey based – and with 5 in 6 people refusing to participate the results just aren’t reliable.
As Chris Giles wrote in the FT, this is “deeply problematic for economic policy.” The Bank of England must now decide interest rates without a reliable measure of GDP, and without knowing whether unemployment is rising or falling. We should hold the Bank accountable for their decisions, but we can’t expect them to make good decisions if they have no clue as to what is going on. The same is true for the Treasury, the Office for Budget Responsibility, and for that matter the opposition and companies deciding whether to invest.
We can and should do things better.
We cannot rely on employment surveys which do not work. We should instead use