In the last few weeks, it has almost seemed like there are two (or three) contradictory economies when it comes to hiring and spending. Did the economy suddenly tank in January (Daily Treasury Statement suddenly pulling in much less than expected, Wal-Mart fearful of a sudden decline in sales)? Or is the consumer on a tear (Gallup Daily Spending making, and continuing, new post-recession highs)? Or is it somewhere in between (ICSC and Johnson Redbook same-store sales at the lower end of their 2012 YoY range)?
The payroll tax increase is real. It ought to be showing up in either decreased spending, decreased savings, increased credit, or some mix of the three. This is a real quandary, and being a total nerd, I did what total nerds do: I made a spreadsheet of the data to see if there were any tell-tales about what is the actual state of the consumer and employment. And lo and behold, there are some very telling clues. It looks like there really are three results. As we'll see below, low-end consumer spending is getting hammered, mid-level consumers are impacted but holding their own, and high-end consumers, gifted with income moved forward from 2013 to 2012, have been spending with wild abandon - so far.
So first of all, let's go to the data. Below is a chart of the year-over-year percentage change in the weekly ICSC and JR same-store sales. Next is the YoY $ change for the 14-day rolling average of Gallup Daily Consumer Spending. The final column is the YoY percentage change for the 20 days of withholding taxes paid. These are all series that I report on each Saturday. The chart starts with the week of November 3, 2012, and continues up through last week. The last two lines are the December average vs. the average of the last four weeks:
|Week||ICSC||JR||Gallup||Daily Treas. Stmt|
|last 4 weeks avg||2.5%||1.8%||$17||11.9%|
All four of these series show the marked impact from Hurricane Sandy in November. By the beginning of December, however, all of them have bounced back. The differences start appearing at the end of December. Both same-store sales series start to show decelertion YoY. They are still positive, but at the low end of their recent ranges. Both Gallup and withholding taxes, however, show a very strong increase. In the case of withholding taxes, the increase totally reverses to outright declines by the end of January, while Gallup consumer spending continues to be very strong.
What caught my eye was the extremely sudden spike in positive YoY withholding taxes paid in the 20-day average once the end of December and beginning of January were added in. This coincides with the huge gain in personal income reported for December. Twenty reporting days later, the spike completely reversed. I wondered if there was an anomaly, so I checked each individual day, and compared it with the prior year. And sure enough, it turns out that in just two days - December 31, 2012 and January 2, 2013, $70 Billion in withholding taxes were reported! A comparison with the 2011-2012 shows that while $30 Billion was also reported on January 2, 2012, only $11 Billion was reported on December 31, 2011 vs. $45 Billion on December 31, 2012!
In other words, there was a huge shift in payroll tax reporting from 2013 to 2012. Undoubtedly a large share if not an outright majority represented things like enhanced bonuses paid to beat the presumed tax increase in 2013 (remember that the "fiscal cliff" wasn't resolved until just hours before the end of the year, so employers had no way of knowing that tax rates would be kept constant, e.g., on incomes up to $400,000 or higher).
This is of a piece with the huge spike in personal income reported for December -- about a 2.5% increase in just one month, where 0.1% or 0.2% increases are more the norm. Further, much or most of those huge bonuses, etc., being "found money," were likely to be splurged on goods - and the money wasn't all necessarily going to be spent in January. That looks like the explanation for the continuing surge in Gallup's daily consumer spending report.
But how do we square that with the ICSC and JR same-store sales reports, and the WalMart emails of sales falling off a cliff at the beginning of February?
We know that WalMart accounts for a huge share of lower-end spending. Further, it no longer participates in the same-store sales reports. This means that the ICSC and JR reports are almost certainly skewed away from low-end sales. Comparing December and January, the ICSC and JR YoY percentage increases have decelerated by about 30%. That suggests that the consumers they are measuring are absorbing about 70% of the payroll tax increase and replacing it with spending out of savings and/or increased credit. WalMart's shoppers, who are more likely to be living paycheck to paycheck, can't do that and are bearing the full brunt of the payroll tax increase. (This morning it is reported that WalMart is laying most of this off on delayed tax refunds. Since its results aren't transparent on a weekly or monthly basis, there's no way to know how much this impacts my analysis).
But what about Gallup? Gallup's sample certainly includes WalMart shoppers. But Gallup is reporting the average, rather than the median, daily spending of consumers, and it is also catching the very high end. Suppose the ICSC and JR reports do not sufficiently capture the high end? Then the surge in high-end spending is only showing up in the Gallup report, which is, for now, being overwhelmed by that surge.
If that is true, then the data from all four series, and the WalMart reports, all make sense. WalMart shoppers, with little savings, are buying markedly less than before the payroll tax kicked in, perhaps spending a full 2% less. Middle-brow shoppers are taking some money out of savings or increasing their use of credit, and are pulling back less than 1%. High-end shoppers are still spending the enormous sums of discretionary income that they received in December to beat the tax increase.
There will be two keys to the unfolding of this scenario. One is to watch for when the bonus money runs out. This will be reflected in Gallup's Daily Consumer Spending report. There is some evidence that this may already be happening. In the last 6 reporting days, the average of consumer spending has fallen back below $80.
The second key will be to watch for when tax withholding returns to a more normal YoY pattern. That will tell us when the decrease in 2013 income due to it being pulled forward into 2012 is over. That could be happening now, or it might not happen for several more months.
Until that happens, I fully expect the bite from the payroll tax increase to intensify.