A couple of days ago, Bloomberg ran a piece citing analysts at Goldman Sachs Group Inc. and Nomura Securities International who were pointing out a limitation in the models used to smooth data to account for seasonal changes for several economic indicators.
Most models are unable to distinguish from a one-time shock, like the collapse of Lehman Brothers, and recurring issues that should be smoothed away, reports Bloomberg. The more extreme the shock, the more unreliable the smoothing will be.
We have been having similar discussions at Edmunds. For example, we noted last year that during times of economic distress, the consumer's desire for deals ratchets up. This provides an extra boost to sales during traditional event periods and drags them down when the events end. (The sales events at year-end would be a case in point.)
What I am driving at is that the monthly Seasonally Adjusted Annual Rate (SAAR) performance the industry (and media) hangs on each month should perhaps be taken with a grain of salt. SAAR may not provide the clear window into underlying consumer demand that we all are looking for.
Something to think about when the January sales numbers get reported next week...