The January retail sales report demonstrates perfectly the nature of this whole recovery, but especially the last year or so when everything holding to the primary narrative boils down to the unemployment rate – a statistic that is more and more determined by peculiar assumptions and calculations. The advance release from the Census Bureau had enough positive vigor to provide palpable enthusiasm in place of the very reluctantly increasing glumness.

Retail sales increased for a third straight month in January as Americans kicked off 2016 by spending freely on cars, clothing and online merchandise…

Greater job security, improving wage growth and falling gasoline prices may be persuading more consumers to loosen their purse strings after a fourth-quarter slowdown. A pickup in household purchases, which account for the lion's share of the economy, would help the U.S. stave off the negative effects of a strengthening dollar, sluggish foreign demand and tumultuous financial markets.

When initially searching for how or where consumers were “spending freely”, I failed to see anything resembling strength or “greater job security, improving wage growth”, etc. Instead, the retail sales report was the usual flavor of awful, with sales growth overall including autos just 1.44% in January. Strong consumers would be represented by at least 5% if not steadily above 6%. That rate was the same as the snowy cold of February 2015 and worse than the awful November that started the hugely concerning Christmas season. Auto sales themselves were up just 3.5%; a broad-based disappointment no different than the recessionary levels that have become alarmingly common since the end of 2014.

It wasn't hard to find what the mainstream is relieved about, however, as there was a huge discrepancy with the “seasonally” adjusted figures. Seasonal adjustments are not just altered for seasonal swings month-to-month but they are also reconfigured by formula for differences in numbers of weekends or even the arrangements of holidays within the month, or even as compared to the same month in the prior year.