Earlier this week markets received some unsettling news on the health of the US economy.
The ISM’s non-manufacturing PMI report for August — a gauge that measures changes in activity levels across the US services sector from one month to the next — plunged from 55.5 to 51.4, the sixth largest decline seen since 1997.
It also left the index at the lowest level since February 2010, more than six years ago.
While still suggesting that activity levels continued to expand in August, it pointed to a sharp deceleration in what is the most important sector within the US economy.
Making the headline figure even more concerning, the decline was broad based in nature, with prominent declines seen in the survey’s employment and new orders subindices.
Like clockwork, financial markets reacted to the weak result, scooping up US treasuries and stocks, and selling down the US dollar. The odds of a rate hike from the Fed, not only for September but at some point this year, also tumbled.
Just days after the release of an underwhelming US non-farm payrolls report for August, it was unwelcome news.
But was it?
According to economists at ANZ, the ISM survey has lost some “signal effect” for predicting US GDP in the post great recession years, noting that the August result was “hard to rationalise such a large drop given the labour market remains solid”. In its opinion, there was likely “a strong element of noise” in the August data.
ANZ points to the chart below to underline this point. It plots the ISM’s composite PMI — a combination of both services and manufacturing sector activity levels — to US GDP going back to 1997.
There’s been a noticeable shift in the power of the ISM survey to predict GDP in the past seven years, says ANZ.
“The predictive power of the ISM seems to have deteriorated post the GFC,” it notes. “For example, the composite ISM index has suggested that the US economy has been growing at a pace of around 3.5% since 2014. However, actual GDP growth is reported at just 2.25%; over one percentage point lower.”
“Plotting the GDP data against the composite ISM index suggests that there has been a significant decline in the predictive power of ISM. Prior to 2009 the correlation between the two series was 0.66, but since 2009 is just 0.18.”
Put another way, where the survey was once a reasonable indicator on US economic growth, recently it has not. Quite the opposite, in fact.
It’s a reminder not too put too much emphasis on one particular data point, something markets have tended to do whenever the Fed appears to be moving towards delivering another rate hike.
As shown by ANZ, the ISM PMI survey has not been the best indicator of economic growth since 2009. Nor has the August US payrolls report been a reliable indicator of job growth in recent years, almost always underwhelming initially before being revised significantly higher in the months following.
Some food for thought just two weeks away from the Fed’s September FOMC meeting.
Business Insider Emails & Alerts
Site highlights each day to your inbox.