What does the January “Philly Fed” Survey tell us about the US Economy?



By John Clarke 21/01/2007 16:59
Yesterday the Philadelphia Federal Reserve reported, in its Business Outlook Survey, that its current general activity diffusion index had slumped from -1.6 in December to -20.9 in January, the lowest since October 2001. 



Given that the US economy was extremely weak six or so years ago (contracting 1.4% at annualised rates in 2001 Q3), it is perhaps understandable that Wall Street, still reeling from the effects of the credit crunch and a softer than expected December jobs report, should have fallen sharply again overnight. After all, if the economy was in recession the last time the indicator was this weak, it makes sense to question whether the US economy might be heading for a hard landing in 2008 as well.


However, there are a number of problems with this line of reasoning. First, as it is only a survey of manufacturing, which today accounts for less than 15% of GDP, the indicator suffers from the same deficiencies as its more illustrious competitor, the ISM Purchasing Managers’ Survey. It can tell us what is happening to manufacturing, but not necessarily to the 85% of the economy that is not manufacturing related. But the situation is even worse than this because the survey actually only covers manufacturing in the Philadelphia region. In addition, it is also extremely volatile on a month-by-month basis. True, the indicator does, when smoothed by a three month moving average (as shown in the chart above) provide a reasonable guide to short term trends in GDP, but it is also clear that it has given misleading signals in the past. What’s more, the correlation has deteriorated markedly since the turn of the century.


These reservations notwithstanding, the thing about the -23.9 reading in October 2001 was that it was the eleventh in a run of 13 straight negative readings, not an isolated occurrence as has been the case in January 2008. Indeed, as the chart above suggests we would need to see the current general activity index averaging -20 over an extended period (at least three months) before it became consistent with recession in the wider economy.


But whilst this would be a necessary pre-requisite as a signal of recession, it is not a sufficient one. In addition to this we would also need to see the “future general activity” index falling below zero on a three month moving average basis, as was indeed the case in January 2001 (-2.4). In January 2008, it stood at +10.6.

Once again it looks to us as though the investment community has over-reacted to a particularly volatile economic data release. If it remains around the January level in February and March we will need to review our position, but for now we are content to stick with our long-held view that whilst the US economy will grow at a significantly beneath trend rate in 2008, recession will be avoided.



Read more articles from John Clarke


No comments have been made about this article.

Link to: Add a comment

Links


Article Search
From
To
Keyword(s)


 
interchange