Headline writers were falling all over themselves yesterday to get the news out that retail sales were weaker than the Street conomist crowd consensus guess. The stock market soared as trading algos surmised that this meant that the Fed won’t raise rates anytime soon.
There’s just one problem. As so often is the case, the seasonally adjusted headline number was highly misleading. Seasonally maladjusted nominal retail sales were reported to be unchanged in April. The Wall Street conomist consensus was for a modest gain of 0.2%.
Nominal retail sales, not seasonally adjusted and not adjusted for inflation (or deflation), declined by $7.5 billion in April versus March. April is almost always a down month, so this drop is not unusual in any way. In fact the month was near the average change for the past 10 years of -$7.9 billion. April 2014 was down just $1.5 billion, but April 2013 was down $13.3 billion. So there was absolutely nothing remarkable about this year. It wasn’t too hot. It wasn’t too cold. It was middling.
On a year to year basis, nominal sales rose just 0.7%, which was a sharp deceleration from the recent growth rate. March had a 2.1% year to year gain. April had the smallest yearly gain since a +0.6% reading in February 2013. These are nominal sales and the early indications are that consumer prices slipped slightly in April. Using PPI for finished consumer goods as a proxy for April CPI, real retail sales show a higher growth rate. There’s no evidence here that the economy is sliding into a ditch. But that’s what traders are betting on, that a weak economy will keep the Fed from raising rates.
Nominal total retail sales does not consider consider the skewing effect of gas prices. Much of the slowing in the growth rate is attributable to the decline in gasoline prices, which reduced total nominal sales over the past 10 months.
Backing out gasoline sales and adjusting the nominal sales for changes in the price level of consumer goods excluding gas gives a picture of real sales over time, without the counter trend distortion of large changes in gas prices.
Real retail sales excluding gasoline sales fell by 1.8% from March to April. April is always a negative month in this data series. This April’s drop was smaller than the average April decline of -2.9% for the past 10 years and midway between the April 2013 and April 2014 drops. Again there’s nothing remarkable about this month’s performance as adjusted for in(or de)flation, and ex gas.
The biggest difference from nominal total retail sales is that the annual growth rate was +4.0%, which is extremely strong. It’s weaker than the blockbuster growth rates seen in December-February, but it’s still in a rising trend. In no way does this indicate weakness. In fact, it suggests the opposite. These growth rates are higher than those seen at the peak of the housing bubble. Clearly somebody is spending like there’s no tomorrow. It’s not the typical middle class worker, that’s for sure.
To normalize the data for population growth I also divided by total US population, resulting in a figure for average real retail sales ex gas per person. I do this to not just confirm the trend, but to see how the average Joe is doing over the long term. The current numbers confirm that the uptrend in retail sales is intact, with growth rates stronger than during the housing bubble, but average retail sales per person have not recovered to the levels of 10 years ago. In spite of the retail buying orgy at the top of the income spectrum, that has not been enough to push average sales per person to new highs. Consider that this is “average”. If we were to look at the median, the picture would be even bleaker for most Americans.
The data shows that consumers have increased their spending on items other than gas, benefitting some retail sectors while energy production and sales suffer. That’s a logical impact. What was illogical was conomists’ conclusion, including those at the Fed, that the US economy as a whole would be boosted. Increased spending on imported junk at WalMart as a result reduced spending on imported gas at the local Mini Mart does not boost the US economy. An increase in $15,000/year WalMart jobs against a decrease in $85,000/year oil industry jobs does not boost the US economy. For some reason, Wall Street and Fed conomists were unable to grasp such a simple concept, that was so obvious to us.
In spite of this surge in spending, average real retail sales per capita ex gas are still below the peak housing bubble years of 2004-2007. Cash-out refinancing of homes in those years allowed people to spend more than their income would normally permit. This time around they have no such benefit, so the numbers lag behind those peak levels. Real median household incomes have fallen.
However, the surge in the growth rate to above 5% in January could be a sign of spending exhaustion at this late stage of the growth cycle. Similar surges in 2004 and 2005 marked the “as good as it gets” stage of the housing bubble economy.
While the top 10% of the income spectrum does most of the spending, the gain that took place from the middle of last year through January was still a remarkable surge. Hidden beneath the headline numbers was a spending orgy, which may have exhausted itself in January when the annual growth rate hit +5.6%. Growth has softened since then as an aftereffect of that surge. But it is strong enough to encourage the Fed to tighten, when the manipulated data finally catches up with the facts.