Beware the “above pre-pandemic levels” rhetoric
COVID-19 Market Update – 26/03/2021
12 minutes to read
Introduction
This is the 36th of a series of weekly notes analysing the state of the UK retail market in the light of the COVID-19 pandemic. This note focusses on two key themes:
- What we learned from the ONS retail sales figures for February
- Reaction to the eight John Lewis store closures
Please do not hesitate to contact myself or any of my retail colleagues if you require any further information.
Key Messages
- Retail sales values decline -1.6% y-o-y in Feb
- “Less bad” than Jan and Lockdown V1
- Huge polarity in food vs non-food
- Food sales +7.3%, 4th best performance since 2000
- Non-food sales down y-o-y -24.3%
- Fashion sales still miserable (-54.3%)
- “Essential” categories seeing strong demand
- DIY +21.0%, garden centres +21.1%, chemists +20.6%
- Online penetration hits new peak of 36.1%
- “Non essential” retail on course to reopen from 12 April
- Retail sales data going forward will be highly distorted
- Beware “above pre-pandemic levels” narrative
- 8 John Lewis store closures confirmed
- York, Peterborough, Sheffield, Aberdeen full-line stores
- Ashford, Basingstoke, Chester and Tun Wells At Homes
- Major question marks as to JL’s strategic course
1. What we learned from the ONS retail sales figures for February
The last really dire retail sales figures we are likely to see for some time. From March onwards, the figures are actually going to be weird and frankly, all over the place. Harder to read maybe, but hopefully better that than unequivocally dire.
As ever, mixed messages in the both the media and in the economist community as to how the retail sales numbers are interpreted. The perennial issue is whether you focus on the month-on-month figures and therefore pay massive disservice to the seasonality of retailing, or are savvy enough to major on the year-on-year ones and take into account wider market context. For once, I will actually qualify and partially quantify this seasonality in this note, rather than just moan about it as I usually do.
The most meaningful headline numbers are thus: retail sales values (exc fuel) declined by -1.6% year-on-year in February, while volumes (i.e. sales net of inflation) declined by -1.1%. Factoring in fuel (for completeness rather than anything else), the figures were markedly worse – values were down -4.4% and volumes were down -3.7%. Not as bad as some months we have seen during pandemic, but still not in any way good.
Ironically, our media and economist friends cheered February’s performance, the month-on-month metrics painting a far more positive picture. M-o-m values (exc fuel) increased by +2.0%, with volumes (exc fuel) up by +2.4%. The fifth worst monthly figures since records began in 1989 is hardly cause for much celebration, neither is the fact that we are now in deflationary territory in retail, another factor that has been curiously overlooked.
Many of the trends previously observed during periods of lockdown were even more prevalent in February. Massive polarities in the performance of food and non-food, highly contrasting performances within non-food categories, household goods significantly outperforming fashion, “essential” operators growing much faster than their peers subject to ongoing enforced closure, online reaching new, albeit highly artificial and temporary, peaks.
The polarity between food and non-food actually accentuated during February. Food sales grew by +7.3%, the highest monthly growth since May last year (and the fourth best monthly performance since 2000). In stark contrast, non-food sales slumped by -24.3% last month, even against a soft comp (non-food sales had declined -0.9% in February 2020, before the full onset of COVID-19). A reminder, if any were needed, that lifting of lockdown cannot come soon enough for large portions of the high street.
Fashion sales remain grim beyond belief. Clothing sales were down -54.3% year-on-year in February, worse even than the -48.3% decline seen in January. On top of the -25% shortfall witnessed in 2020. Note to the economists naively predicting inflationary rises when stores reopen from 12 April: this is simply not going to happen with massive stock overhangs and two retailers as large as Debenhams and Arcadia liquidating massive inventory and this having a wider knock-on effect across the clothing market.
Household goods fared much better, growing +7.0% y-o-y in February. But across the component sub-sectors, a very variable picture. DIY continued its stellar run (+21%), more by virtue of being “essential” and open as us having suddenly become a nation of DIY enthusiasts, as our media friends would have us believe. Electrical goods had a good month (+7.3%), but this was not replicated in furniture (-12.0%), nor in carpets (-20.6%).
Unsurprisingly, other strong performing non-food categories were those deemed “essential” – chemists (+20.6%) and garden centres (+21.1%). But another bad month for other “non-essential” high street sectors such as cosmetics (-19.1%), sports/games/toys (-28.2%) and jewelers (-43.8%). The ignominy of being the worst performing category remains mobile phones/computers (-62.6%), while the accolade of best performing category astoundingly went to music and video recordings (yes really, +109%).
Online penetration reached a record high of 36.1% in February, compared to 35.2% in January and just 20% in February 2020. Online has growth +77.6% over the course of the year, but what is interesting is that multi-channel has significantly outperformed pure-play (+50.1%) over this period. How much of the fabled “consumer shift to online during the pandemic” is a permanent one will only start to become apparent this time next year, but in the interim, online penetration will definitely decline sharply from April when lockdown is lifted.
Where do we go from here? As I’ve already alluded to, the figures from next month will be even more weird and wonderful than we have seen to date. March’s figures (released on 23 April, nearly two weeks after Lockdown V3 has been lifted) will be weak, reflecting another full month of lockdown. The tone of the actual numbers will be in sharp contrast to the wider mood music of post-lockdown euphoria (and inevitable media reports of big queues outside Primark stores…)
Deciphering the numbers and taking a meaningful retail health check from next month will be a major challenge. March 2020 marked, of course, the start of the Lockdown 1 and the numbers relating to this make for a highly distorting comp base. Food sales in March 2020 skyrocketed by +10.6% and the corresponding figures for this year will inevitably show a decline against this spike. And this will without doubt depress the overall headline retail sales figures.
Similarly, on the non-food side, March 2020 was when the rot really set in, before we hit a nadir in April. Non-food sales slumped by a then unprecedented -21.2% in March last year, making for a very soft comp this time around. Expect non-food sales in March to show modest growth (remember, they still reflect a full month of lockdown), with huge growth to follow in April. But when all is said and done, it would be as wrong to interpret March’s growth in non-food as a full-blown recovery as it would be to conclude that the slump in food meant desperate times for the grocers. In short, the numbers will be misleading and will only very partially reflect what is happening on the ground.
It is always dangerous to read too much into a single month’s retail sales figures. Retail really is seasonal in a way that the wider economy is not. I promised some clarity on this, so here goes by way of actual example. I am currently poring over a shopping centre tenancy schedule which has all the constituent retailers’ (of which there are over 50) monthly turnover figures for 2019/20 (a “normal” pre-COVID year). With this transparency, the seasonality is plain to see, +/-20% monthly swings completely commonplace amongst all operators, a way of life for both retailers and F&B operators.
To an operator on this 50 strong tenancy schedule, February is the by far the lowest turnover month of the calendar year, a fair reflection of large parts of the retail industry. For most, February represents less than 6% of a year’s turnover, as opposed to the 8.3% non-seasonal metrics would suggest. To put this into wider context, for many retailers sales in months such as February may be -70% lower than peak months (November and December). Some retailers turn over a full ten times more in December than they do in February. Having this transparency highlights the complete folly of focusing on month-on-month retail sales trends.
Why the focus on February? Because February 2020 is the baseline for any ONS comparative for “pre-pandemic levels”, on the simple basis that it was the last “normal” month before lockdown measures were imposed. But in the retail sphere, it is an extremely nebulous basis for comparison. In any given year, we would expect any month’s retail sales to be higher than those reported in February. By the same token, we should expect any month of non-lockdown to be “higher than pre-pandemic levels”, as a matter of course rather than economic triumph.
It is, of course, a massive positive that “non-essential” retail will re-open from 12 April. We are extremely bullish as to the prospect of pent-up consumer demand being released both rapidly and over the medium and longer term. But we also recognise that it will take considerable time – many, many months – before the retail sector fully gets back to its feet again. Which is why it is important to not read too much into the inevitable “above pre-pandemic levels” retail sales headlines that will inevitably predominate from next month onwards.
2. Reaction to the eight John Lewis store closures
The fact that eight John Lewis stores will not reopen when lockdown is lifted was covered in last week’s Retail Note, coupled with my somewhat tempered view as to why I think this is a worrying move and fundamental strategic mistake. The list of stores affected has since been made public.
The stores being axed include four full-line stores (York, Peterborough, Sheffield, Aberdeen) and four At Home formats (Ashford, Basingstoke, Chester, Tunbridge Wells). A number of these had already been touted, so not all were surprises. But that still doesn’t make the decisions any less the curious.
The At Home closures were perhaps less of a surprise, but some of the reaction has been somewhat glib. Some analysts point to the fact that At Home was an experiment, but from memory the format was launched as long ago 2011/2012, so I think it’s fair to say that it had gone way beyond dipping a toe in the water.
Two other observations on the At Home closures. In tandem with the latest news, the Partnership also unveiled plans to launch a “smaller, service-led” neighbourhood format. Was that not what At Home was partially conceived to be when it launched? There is certainly some contradiction in the closures vs the direction of travel. Secondly, it is taking a John Lewis physical presence away from some of the most “John Lewis” towns in the country – Tunbridge Wells couldn’t be any more “John Lewis” if it tried.
Of the full-line department stores, even more question marks. Peterborough was widely touted closure (and is ostensibly over-spaced as a store) but is still a long-established, fully living/breathing John Lewis store (in the same ilk as Watford) And, like Watford, the beneficiary of recent major investment and revamp. To see such embedded, long-standing stores disappear still seems odd.
York, on the other hand, is at the other end of the maturity spectrum. The store has been only open for around six years, a year or so longer than its doomed counterpart at Birmingham Grand Central, but a short time nonetheless.
Sheffield represents a major U-turn. John Lewis has a long standing in the city and had only just agreed on new multi-decade lease terms last year to be part of the city's regeneration project. Its withdrawal is clearly a major blow to these ambitions. That aside, it seems curious that John Lewis does not want any physical presence in city the size of Sheffield (urban population of 585k, metro population of 1,569k, according to Wikipedia) - particularly one that is in the throes of re-invention in which John Lewis is one of the key catalysts (and in strong negotiating position). To simply walk away and not find a viable alternative reflects badly on their retailing credentials as much as anything.
Similarly in Aberdeen (a store I can’t profess to know personally). But again, to not require a physical presence in a major city (urban population 201k, metro population 489k) with an absolutely captive catchment and not the slightest hint of cannibalisation seems short-sighted to say the least. Maybe the good people of Aberdeen are complete online fanatics and John Lewis will recoup all their spend that way?
Of course, the accompanying narrative to the news continues to be of changing consumer patterns, underpinned by the new mantra that the Partnership expects 60% to 70% of sales to be made online in the future. The more this is trotted out, the more it jars. On the one hand, it is a very facile justification for the 1,400+ partners that face losing their job on the back of these latest closures.
In the context of the wider Partnership, it is more worrying still, namely that this is a strategic target that they are at pains to achieve without appreciating the wider consequences. John Lewis seem to be under the illusion that lost store-based sales will simply be recaptured online. Another major multi-channel retailer, Next would beg to differ. In their assumptions, Next assume that only <25% of lost sales from a closed store will transfer to another Next store. Zero (repeat, zero) will gravitate to Next’s online platform. A clear difference between Next’s modelled assumption of 0% and John Lewis’ implicit aspiration of 100%.
The likely reality? Somewhere between the two, but probably closer to Next’s 0% than John Lewis’ 100%. The net result? John Lewis’ overall turnover is likely to significantly smaller than it is currently, so yes, 60%-70% online penetration may indeed be feasible. But 60%-70% of a smaller pie overall. And a smaller pie with reduced economies of scale and fewer cost benefits. And don’t get me started on the whys and wherefores of brand devaluation.
John Lewis store closures. Short term pain, with highly questionable long-term gain.