Black Friday's Guesswork Gloom
The New York Times headline declared, “Thanksgiving Weekend Sales, at Stores and Online, Slide 11 Percent.” Not to be outdone, the Wall Street Journal reported, “‘Black Friday’ Fades as Weekend Retail Sales Sink.” And -- to prove I'm not playing favorites -- Bloomberg News noted that, "Black Friday Fizzles With Consumers as Sales Tumble 11%."
Please repeat the following, once again with gusto: We don't know what Black Friday sales were, not yet. And we certainly don't know what the impact of the actual number will be on final holiday retail data.
How did we get these numbers, which almost by definition are wrong? It is a combination of a retail trade group that cares little for accurate economic data, a terrible survey methodology and a naive and lazy news media, which seems to believe its role is to mindlessly repeat whatever nonsense is peddled by the aforementioned trade group.
Let’s start with the National Retail Federation (NRF), which surveyed 4,600 consumers on Nov. 28-29. Based on that, the NRF concluded that holiday weekend sales were “estimated to have dropped 11 percent, to $50.9 billion, from $57.4 billion last year.”
Note the use of decimals, which indicates both precision and that the NRF has a sense of humor.
I have been critiquing this foolishness each year since 2005. As we noted last week in "Retailing's Black Friday Hoax," the NRF forecasts have an abysmal track record: “Every year, I exhort investors to ignore this data series. It has no correlation to actual retail sales, tells us nothing about retail profits and gives us no insight into the holiday season.”
Ask a person a specific question, and you will typically get a specific answer. The problem is that the answer is either guesswork or fabricated, bearing little if any resemblance to reality. You have no idea what you spent on holiday shopping last year. You have no idea what you are going to spend on holiday shopping this year. The net result of comparing one made-up number with a second made-up number is a random outcome lacking any relationship to actual spending.
These are facts of human behavior. Credible survey methods seem to be wholly unfamiliar to the NRF. Why bother when the fiction works so well?
In truth, no one has an idea what the holiday sales were as of yet.
An equally invalid methodology is employed by ShopperTrak, which monitors foot traffic at retailers. Based on how many people it estimates were shopping, it said that sales on Thanksgiving Day and the following day declined 0.5 percent.
Neither of these methodologies has any validity. The proof is in the record of prior forecasts, every one of which for the past decade has been wrong.
Eventually, one of these random guesses will turn out to be right. It is unlikely that someone can keep making guesses and never end up being correct. So far, that lucky guess keeps eluding the masters of shady survey methodologies. I will keep repeating this until the methodology employed by trade groups is replaced with a statistically valid approach to gathering data.
So far, we have precisely no real data on the holiday weekend shopping. I have no idea how the holiday sales will be this year. Anecdotally, in the prosperous Long Island suburbs where I live, it seems that stores in mall such as the Americana Manhasset or Roosevelt Mall were very busy. Whether this translates into increased sales nationally is, at this point, a mystery.
The earliest data will come from the major credit-card companies, which can actually measure their customers’ spending. In January, the Commerce Department will release the December retail sales data. Until then, everything else should be put down to guesswork and not taken seriously.
This column does not necessarily reflect the opinion of Bloomberg View's editorial board or Bloomberg LP, its owners and investors.
To contact the author on this story:
Barry L Ritholtz at firstname.lastname@example.org
To contact the editor on this story:
James Greiff at email@example.com