As a retail analyst for the last 23 years, I can count on 1 finger the number of times I have written a post to comment on another company’s research.  This is it.  The exception here is a report which has been widely cited by the likes of CNBC, Washington Post, Business Insider, Forbes, NBC Nightly News, and many others based on a headline that Retail in the US MUST close 75,000 more stores by 2026.  And yet every reporter I have reached out to has not actually seen or read the research.  With a conclusion this sensational that is influencing so many, I felt it was necessary to share some comments with those of us in the industry who live in retail every single day.  Here are my thoughts after having reviewed a copy.

First, congratulations to UBS for getting outstanding coverage.  The headlines are everywhere and their report certainly serves the PR purposes for releasing such dramatic conclusions.  Second, it is really well written and documented.  Third, there are some truly outstanding historical data in the report looking at sales history, sales per square foot, sales by employee, the number of malls, and malls per region and states.  Fourth, I most definitely agree with the authors that specialty apparel has way too many stores (we have approximately 20x the sales square footage per capita of Germany in that sector).  Finally, it is obvious that they didn’t simply continue a historical linear trend – which would have been lazy, so really good stuff on those fronts.  However, where I believe the research falls short is that it relies on math almost exclusively at the exclusion of existing retail trends, with an inherent bias towards online trends as the primary and preferred future of retail.

Based upon what I can see in the report, I have 5 specific concerns on the research that led to the forecast conclusions on store closings.

1. When one relies almost exclusively on the US Retail Census data, one assumes that the government has kept up in their measurement criteria to handle all aspects of retail trade (online and stores). Truth be told (and IHL has had conversations with people in the know in the government) they are really struggling with how to deal with bricks and mortar retailers and the growth of the online business.  Their systems have not kept up with the rapid pace of change in retail in all facets.  Further, they are struggling with how to deal with things like Amazon taking over Whole Foods (where a pure-play online retailer all of a sudden has many stores) or other online companies opening stores organically (Amazon Go).  The US Retail Census data has a category for non-store retailers, as well as categories for the individual sectors.  This is in a system that has historical data back to 1992 for download.  Furthermore, keep in mind, the Census on Retail Trade is a survey that is extrapolated to represent the market as a whole.  These data are the best information we have from government reporting agencies, but are currently flawed in the handling of online growth and the rapid pace of change in retail.  This point is minor in our overall concerns.
2. Next, the report really doesn’t go into what causes retailers to close stores. It simply assumes a scarcity mentality…it’s all math.  It assumes that if it fixes the total pie growth (in this case 2% CAGR retail growth through 2026), then this must mean stores close if online grows.  Along these lines, there is no room for new companies to come in or pure-play online retailers to open stores in those sectors…the math only looks at what must close. So the assumption is that retail overall will only grow 2% annually and online growth substantially higher – so stores must suffer.  But if one really looks at retailers that are closing their doors (as IHL has done in previous posts), 85-90% of them took on way too much debt or expanded beyond a market that could sustain the growth.  It is very rare where a well-run business, that has adapted to changes in consumer trends, invested in technologies, and has a reasonable balance sheet has closed a sizable number of stores. So the study’s approach assumes the worst and excludes the positive parts of the market.  It’s all math.

3. My third point is of much bigger concern than the first two. The assumption here in the math is that all eCommerce sales and growth is for delivery of goods only.  It excludes Buy Online Pickup In Store (BOPIS) and Click and Collect, which are the two areas of eCommerce that are growing the most by far (5x the speed of overall eCommerce).  To do these, a retailer needs stores.  We’ve all seen how delivery is harder logistically and secure delivery is more difficult due to package theft.  Excluding this trend in the analysis is a huge problem for me when the conclusions are so sensational.

4. Next, there is a terrific chart on historical contribution of eCommerce to overall retail growth. In the latest year they studied (2018 through Q3) this number shows 26%.  Thus, of all the retail growth in 2018 through the 3^rd quarter, eCommerce represented 26% of that growth and stores represented the other 74%.  Yet the assumption in the forecast that determines the number of store closures is that eCommerce will represent 75% of all retail growth by 2026.  This, in turn, equates to the number of stores that must be closed.  So, the assumptions is what contributes 26% of retail growth today, contributes to an accelerated rate so much so that it becomes 75% of retail growth and stores only 25% by 2026.  The math then says stores must close as a result.

5. Finally, to get to the store closures above one has to assume some really incredible growth rates in eCommerce spending over the period. And the study does.  The assumption is that in percentage of total sales, Auto Parts will see a 100% increase in the percentage of total sales that will come from online, Home Improvement 200%, and Grocery 400%.  Grocery will go from $13b in online sales to nearly $80b.  If we assumed that stores are part of the delivery formula that is aggressive but not a big argument from my side.  Using this growth assumption means that if the average store does $8.4 million by 2026, 7,309 stores must close (because stores are not involved with the fulfillment if it is only math), This is simply not reasonable in an analysis with the growth for this sector is primarily from pickup in stores.  Simply, the stores need to be there.

My point in this review is not to disparage the research or the authors.  It’s just that the conclusions seem a bit sensationalized and are doing damage.  If one considers all of the trends and limitations in the math, one can see the challenges of some of the conclusions made by the media based on it. And as I mentioned earlier, there are some truly terrific data in the report.  If one can secure a copy it is certainly worth reading.

And to be transparent, IHL’s forecast on eCommerce growth is actually higher than those used here by UBS.  We project that online will be 25% of the total retail market by 2021, but 80% of all of retail will still have a STORE fulfillment component.

Finally, the primary purpose in UBS completing this research was to advise on stocks to buy or sell, and to compare the investment in one industry over another.  In this case it does its job well.  But when the conclusion jumps from where one should INVEST to that of STORES MUST CLOSE, it moves from its original purpose to commenting on the state of the industry in all facets.  The health of an industry and activities of an industry do not directly correlate to whether a stock goes up or down (more on this another day).  This leap from industry health for stock investment to industry health overall is the issue.  By simplifying that if online sales go up, store sales must go down, thus stores must close creates a headline that is very misleading.  And sadly, that headline affects thousands and thousands of jobs.

Retail is still very much about location, location, location to consumers.  Those stores may take on more of an ecommerce fulfillment role and radically change formats from today, but their advantage is their proximity to the consumers they serve.  This is true whether it be for traditional shopping, online orders for pickup or ship from store.

75,000 stores MUST close – in the immortal words of Lee Corso, “Not so fast, my friend.”

Retail is not dying, stores are not dead – I just thought you should know.