The page you were trying to view is not available for your role.
This week’s blog is written by portfolio manager Sacha Chorley
Economic data plays a central role in shaping market views and investment decisions, but how reliable are the numbers we rely on? Sacha Chorley explores the quirks, gaps and pressures that can undermine confidence in headline statistics.
For obvious reasons, both in financial markets and the ‘real world’, there’s a lot of reliance on the vast array of economic data points published by governments. One aspect that doesn’t get quite as much attention (for equally understandable reasons) is the technical nature behind the calculations that underpin these statistics.
Statistical quirks and seasonal distortions
A recent example of the more obvious statistical issues that can arise was shown in the latest UK inflation data, published this week. Headline Consumer Prices Index (CPI) rose ahead of market expectations by 3.8% year-on-year, up from 3.6% in June. This move higher included a big increase in airfares - unfortunately, this is probably not due to higher consumer confidence driving demand for flights. Rather, the Office for National Statistics (ONS) noted that the impact is likely to be the result of a quirk in the timing of the school holidays.
Falling response rates and political influence
Outside of more understandable factors like this, we have also seen statistical agencies such as the ONS and US Bureau of Labor Statistics (BLS), reporting issues regarding the inherent quality of their data. Just this week, the ONS announced that the July retail sales data will be delayed from publication until early September.
Although there was no indication of the reason, we’ve certainly seen that many economic data series have experienced material declines in the number of survey respondents where the survey data is used to infer the equivalent statistics at the national level. In the US, the commonly followed JOLTS data (Job Openings and Labor Turnover Statistics) is particularly poor, with response rates moving from around 60% in 2015 to roughly 30% today. Clearly, as we approach these levels of response rates, it is prudent to reduce our confidence in the numbers.
If the statistical quirks weren’t enough, there’s also been speculation about political interference with economic data in the US. This culminated in President Trump firing the BLS commissioner after claiming the data was ‘RIGGED’. It’s unlikely that President Trump would be calling for BLS staffers to literally change the numbers to support his desired outcomes, but political views can certainly drive changes in calculation methodology, which could lead to more subtle shifts in data.
Ultimately, while the minutiae of how these economic data points are calculated can seem irrelevant, each of the factors highlighted has the potential to increase volatility around data releases. This emphasises why we consider multiple factors when coming to an investment decision rather than blindly following the numbers, and it does seem this approach will help navigate markets in the months ahead.
Portfolio manager blog - this week written by
Sacha Chorley
Portfolio ManagerSacha is a portfolio manager of the Quilter Investors Cirilium and Creation Portfolios. Prior to joining Quilter Investors in 2011, Sacha worked at Broadstone with their team of economists before moving into asset allocation and fund manager research.
Sacha is a CFA charterholder and has also completed the Chartered Alternative Investment Analyst qualification. Sacha has a degree in Maths from the University of Bath.
Last week's portfolio manager blog
Unskilled or unlucky?
What happens when just a few stocks drive market returns? CJ Cowan explores the fine line between skill and luck in active management - and why missing a handful of outperformers can make all the difference.