Retail Needs Its Own Therapy

By in
Retail Needs Its Own Therapy

US retail sales have been on a choppy ride lately, as today’s disappointing -0.2% print shows. In fact, there have been a few sloppy reports over the last 12 months. A few quick points on those here:

  • We like to look at retail sales excluding food services both because the data is more representative of underlying consumer demand and the time series goes back to the early 1990s.
  • Like the headline number, retail sales ex-food services were also -0.2% lower this month and there have been 3 other negative reports over the last year. They were in August 2018 (-0.1%), December 2018 (-1.8%) and February (-0.4%).
  • Keep in mind these are all month-to-month numbers, subject to seasonal adjustment factors as well as the noise of measuring the data over a short time frame.

To correct for that last point, we look at year-over-year growth. Viewed in this manner, the real pattern becomes clearer:

  • Today’s April retail sales report shows 2.8% growth versus last year. That is very close to the 3-month average for Q1 2019 of 2.7%.
  • While that means April’s print doesn’t indicate any dramatic change to trend growth, the 2019 data looks pretty shabby when you compare retail sales growth now to 2018.
  • Retail sales ex-food grew by an average of 4.8% during the 12 months of 2018. The best months were in May (6.3% year on year growth) and July (6.2%).

To expand this discussion across multiple economic cycles and put the 2019 data into a historical context we have included a chart at the end of this section with the monthly year-on-year retail sales comps back to 1993. Here is what we see in this data:

  • US retail sales are almost always positive when measured year-on-year. The only exceptions are September 2001, October 2002 (because of a tough comp) and the 14 months in the midst of the Great Recession (September 2008 to October 2009).
  • The average year-on-year growth for US retail sales since 1993 is 4.3%. This is pretty much where you would expect it to be: 1% population growth and 3% inflation.
  • The average since 2010 is nearly identical at 4.4%. That’s about a point higher than where 2% inflation and 1% population growth would logically place it, but catch-up spending post the Great Recession likely explains the rest.
  • This does, however, put a question mark next to last year’s unusually strong 4.8% comp. We would credit a mélange of factors, from high employment levels to reduced individual tax rates.

The upshot here: US retail sales were unusually strong last year so we should expect lower growth in 2019. Today’s print was not an aberration but it does not portend any imminent demise to the current economic expansion.

Against that upbeat assessment, we’ll finish on a cautionary note: look at the chart below and you’ll see that every cyclical high for retail sales has been lower than every previous cycle top.

  • The high water mark was 11.6% in March 1994.
  • Even the heyday of the dot com bubble, the best comp was 10.7% in August 1999.
  • Fast forward to the early 2000s and the peak comp was 9.5%.
  • In this cycle, the highest comp was 8.4% in June 2011.

That’s what an aging/slower growing population will do to retail sales, so this pattern of lower highs is hardly a surprise. Still, it is something else to keep in mind as we see the data over the balance of the year. February 2019 showed just 1.8% retail sales growth to last year. The more recent data has bounced back, but we wonder what the next recession (whenever it comes) will bring in terms of retail sales. Negative comps, unusual as they are historically, may become the norm for recessions to come.