Reading too much into single data points

Something business economists are taught to avoid is reading too much significance into individual data points about the economy.  Seasonal adjustments, measurement problems, weather and a host of other things mean it is very important to look at as many data points as possible before declaring a trend.  It's a lesson that seems to have been forgotten by quite a number of commentators as they looked at some important numbers this week.

On Tuesday the Bureau of Labor Statistics released consumer price index data for October.  The headline number was certainly encouraging: prices were unchanged in the month, which resulted in the year-on-year inflation rate easing to 3.2 percent from the 3.7 percent reported in September. Equity markets rallied hard on hopes that falling inflation would pave the way for the Fed to start cutting rates sooner rather than later.

A look behind the headline suggests that a more cautious interpretation of the data might be in order. The lower-than-expected headline print was largely the result of a 5 percent month-on-month decline in the price of gasoline.  That's not something that Fed policymakers can claim any credit for, and given the highly volatile situation in the Middle East, it's not something that can be counted on in the months ahead, even though a similar outcome now seems likely for November.

Other sub-components of the index were less positive from a policy standpoint. Both shelter and food prices rose 0.3 percent in October, with shelter prices up 6.7 percent higher than they were a year ago. CPI less food and energy stands 4.0 percent above its year-ago level.  Given that Fed Chair Jay Powell always insists that the Fed intends to get CPI all the way back to the 2 percent target, that suggests there is still some way to go before lower rates can be confidently predicted.  

Today the US Commerce Department provided its advance estimate of retail sales for October. The dollar value of retail sales (not adjusted for inflation) edged down by 0.1 percent from September, the first such  decline since March.  However, the monthly decline was actually smaller than markets had expected, given the very robust 0.7 percent increase initially reported for September. What's more, that September increase was actually revised higher to 0.9 percent, which means that the dollar value of retail sales in October was significantly higher than markets had foreseen, even if sales fell marginally in the month.

With the holiday shopping season yet to get fully underway -- Black Friday and Cyber Monday are still a week away -- it would be foolhardy to suggest that one month of not-actually-weak data on CPI and retail spending sets the stage for the Fed to start contemplating early rate cuts. We can be quite sure that that is not how the Fed sees it.