It isn't. Some of their data has been suspect for a very long time, unfortunately (and ironically) there is also a partisan reason why the Economist doesn't mention the series that is most suspect: population estimates. These numbers are known bad for over two decades now...they have done nothing.

The US has had a far more serious approach to economic statistics, the ONS is a complete joke. It has never been easier or cheaper to collect stats but the ONS, for some series, is using collection methods that haven't changed since WW2. This is typical of government in the UK, completely isolated from the rest of society.