Many if not most (or all) of you know that Meatloaf sang the Jim Steinman-written 1970s song, “Two Out of Three Ain’t Bad.” But what about three out of four? Besides being a great winning percentage over an extended time period (i.e., decades) for a major college football program (0.750 or 75 %), three out of four represents the number of times in recent months in which seasonally adjusted U.S. retail sales were estimated to have fallen by the U.S. Census Bureau. Using this measure, retail sales in the United States fell in November 2022, December 2022, and February 2023. On a seasonally adjusted basis, U.S. retail sales (including food services) fell by 0.4 per cent in February 2023 when compared with their level from January 2023, which was a strong month in the adjusted data. Despite the decrease last month, the Advance Retail Sales report from the Census Bureau points out that U.S. retail sales from February 2023 were up more than five per cent from February 2022.
However, U.S. retail sales are measured in U.S. dollars with buying power that fluctuates with changing prices in the economy. If we would subtract an approximate annual inflation rate of, say, 5% or 6%, from the growth in U.S. retail sales from February 2022 to February 2023, then how much higher would U.S. real retail sales be, with real retail sales measured in U.S. dollars that have constant purchasing power, which is the buying power of the U.S. dollar in the base year or the base period.
Readers should realize that retail sales clearly do not represent all consumption spending in the economy. Further, U.S. real GDP includes not only consumption by households but also investment in physical capital (largely by firms), government purchases by the various government sectors, and net exports, where those in other countries buy the exports that count toward net exports. Although I’m not sure, retail sales could include at least some imports, while GDP includes no imports (but the value added within US borders should count toward U.S. GDP). Thus, although we should expect a positive correlation between retail sales and real GDP, they are definitely not the same, given that retail sales are not measured in constant buying power currency and real GDP includes more items. In turn, decreases in retail sales could mean but do not necessarily mean decreases in real GDP.
Additionally, the US Bureau of Labor Statistics announced this morning that the Producer Price Index (PPI) fell by 0.1% in February 2023. If my memory is correct, then on the screen CNBC printed that the market pre-announcement estimate was for an increase in the February PPI by 0.3% rather than the estimated decrease. I think I heard that year-over-year CPI inflation has been trending downward since around June 2022, and my usage of the inflation calculator on the BLS website confirms this. With year-over-year inflation trending downward and the PPI falling unexpectedly last month, this could indicate weakening demand in the economy and a recession looming for the United States. So too could the fact that, as I pointed out in my previous blog entry [Hartman (2023)], data not seasonally adjusted from the household survey imply average monthly employment gains over the last seven or so months that could be considered as modest.
Readers should note that when using data not seasonally adjusted, U.S. retail sales increased in November and December of last year, as was likely expected, probably largely due to holiday shopping. But in turn, given that seasonally adjusted retail sales fell in those two months, this could suggest that the holiday shopping season was not as strong as would normally be expected.
After usually strong retail sales growth in November and December in most if not all years based on data not seasonally adjusted, retail sales normally plummet in January of the following year. A quick look at data from the FRED website of the St. Louis Federal Reserve Bank website shows that starting with February 2009, estimated advance retail sales each February not seasonally adjusted posted only three month-to-month gains. The years with month-to-month February increases in retail sales starting with the year 2009 were 2011, 2012, and 2016. Each one of those three month-to-month gains in unadjusted retail sales followed a loss of retail sales of more than twenty per cent! This might help to explain a rebound in the following Februaries. Although some earlier Januaries during this period also posted losses in unadjusted retail sales of more than twenty per cent, each January starting with January 2019 had a month-to-month loss of less than nineteen per cent. In light of the fact that U.S. retail sales unadjusted are likely to fall in February, the decrease in seasonally adjusted U.S. retail sales in February of this year seems to underscore weakening demand in the economy.
In past years, March has often shown strong rebounds for retail sales based on data not seasonally adjusted. Looking at the retail sales data both adjusted and not adjusted for the current month, March 2023, could provide very useful information about how strong the U.S. economy actually is. However, that is a luxury that we don’t have now. This could make the FOMC’s interest rate decision next week all the more challenging.
NOTE: I apologize for any misstatements, inconvenience, or confusion, Additionally, data revisions announced after this blog entry could change the above analysis. Thank you very much for your understanding. Constructive comments are welcome at my Twitter account (@HarrisonCHartm1).
REFERENCES
Hartman, Harrison C. (2023) “Half Full or Half Empty: A Brief Note on the U.S. Jobs Report for February 2023.” Posted online March 10, 2023 at the web address below.
https://harrisonhartman.blogspot.com/2023/03/half-full-or-half-empty-brief-note-on.html
U.S. Census Bureau, Advance Retail Sales: Retail Trade and Food Services [MARTSMPCSM44X72USN], retrieved from FRED, Federal Reserve Bank of St. Louis; https://fred.stlouisfed.org/series/MARTSMPCSM44X72USN, March 15, 2023. Data accessed from https://fred.stlouisfed.org/data/MARTSMPCSM44X72USN.txt
https://www.bls.gov/data/inflation_calculator.htm
https://www.bls.gov/ppi/
https://www.census.gov/retail/sales.html
No comments:
Post a Comment
Note: Only a member of this blog may post a comment.