The Numbers Are Wrong!

For many investors in financial markets, economic data is crucial.

The latest instalment on inflation, growth, or jobs, can have huge influence on market sentiment and market moves.

Battalions of analysts and commentators pore over such numbers in the greatest of detail and then rush to publish their findings.

But it seems that the only thing we can sure about is that the numbers are wrong!  

For instance, currently many investors are focussed on how the US jobs market is performing. Investors (and policy makers) would like to see the labour market cool and take the pressure off inflation, which the US Central Bank is striving to reduce to closer to 2%. The first Friday of every month gives us an update on how many new jobs have been added to the US economy. January’s report was a blockbuster showing that 353,000 new jobs were created in the US – a staggering number way above average.

However only one month later, that number was revised down by 124,000 jobs -that’s a difference of 35%!. And December’s number was also revised by 43,000 jobs. January’s number was revised again last week.

In fact, in the past 12 months, we have seen revisions each and every month – of up to 90,000 on several occasions. 

It is thought that immigration may be part of the problem in collecting the US data. But whatever the cause – the result is a very volatile set of data where first estimates (which is what the markets focus on) are wrong.

This is not new – nor is it exclusive to the US. We have seen similar discrepancies in other economic data such as GDP which have seen subject to significant revisions over multi-year periods. Over the past ten years, global GDP has been revised in each and every year by an average of 0.5%. 

In some economies, such as the UK, the degree of revision has been as much as 1% – on average. When you’re looking at lowish growth rates of around 2%, such differences matter.

Published economic statistics in many countries are subject to much revision. Much of this has to do with changed work patters following Covid and deteriorating response rates to government surveys.

In the UK for example, the Office for National Statistics (ONS) was for a while unable to publish an estimate of unemployment because the response rate to its survey had fallen from 50% to about 15% making it unreliable. The ONS still advises users to exercise caution in interpreting the data.  

There have been similar reports of “survey fatigue” and declines in response rates in Europe and In Canada.

Clearly this is an issue not just for investors, but for policy makers too. 

Is there a solution? Well improving tha actual data collection can be problematical. Even making surveys legally compulsory, where it has been tried, doesn’t seem to work.

For investors seeking a signal on where we are in the economic cycle, it might make sense to diversify your sources. More timely, higher frequency data could be used to complete the jig-saw. For example, data on same store sales, traffic flows and credit card spending may provide a better steer on GDP than  GDP itself!