The conventional wisdom around online grocery is that every online shopping basket is structurally less profitable than a similar in-store basket and, in many cases, is loss-making. Hence, the migration to online is thought to cannibalise the profit pool of an already thin-margin industry. This certainly paints a gloomy picture for the major supermarkets. Indeed, with online shopping only set to increase over time, cynics might argue the future is the enemy of these companies…never a particularly attractive framework in which to invest!
Our viewpoint has been somewhat more nuanced than the reductive ‘online is bad’ stance. Nevertheless, it would be accurate to say that in recent years the industry’s fastest growing channel has also been its least profitable. Potentially dramatic change is afoot, though, that could start to challenge the consensus view on this subject. Using Tesco as a case study, we explore the economics of online grocery, its ‘irrational’ history and how Covid-19 is catalysing shifts in consumer and industry behaviour which may result in a distinctly better economic model than has historically been the case; in the process, we underscore why Tesco remains our preferred pick from the sector. Before I start, it is important to stress that the analysis that follows provides an insight into our investment thinking and is not a recommendation to buy or sell shares in any of the companies mentioned in this article.
Losing bread with each online basket
At face value, grocery is a retail category poorly suited to online. A typical basket contains a large number of low value, perishable, bulky items, which means picking and transportation costs are high relative to the ticket. Moreover, with gross margins of only 30% there isn’t a great deal of flexibility in the grocers’ financial models to start with. For a retailer such as Tesco, which primarily fulfils online orders from its stores (known as a ‘store pick model’), a fair rule of thumb is that a £100 online basket might incur additional variable picking and delivery costs of roughly £10 to £12 per order (running a refrigerated van costs in the region of £50 per hour, for example). Despite these high incremental costs, delivery fees have typically averaged only £2 per order, so recouping only a fraction of the outlay. Cost challenges have been compounded by retailers competing aggressively for new customers through money-off vouchers. The net result has been the establishment of what looks like a broken economic model – whilst Tesco’s stores generate an earnings before interest and tax (EBIT) margin in the region of 5%, its online business is estimated to have been operating at a loss of about 4% (source: Exane), with every 1% of its sales shifting to online therefore reducing Tesco’s total retail profit pool by about 2%. Meanwhile, Bain has estimated that the online channel shift could erode a grocer’s profit pool by 50-80 basis points over five years, depending on the rate of online growth – that is a major problem when industry EBIT margins are generally only 3-5% to start with.
This has been a great deal for the UK consumer, but why has the industry chosen to price delivery on a seemingly irrational basis that fails to cover its costs? We think this stems chiefly from a focus on customer lifetime value (CLTV) whereby, at least in theory, the upfront costs of capturing a loyal, valuable customer would ultimately be offset by multi-year revenue streams, alongside an economic model that could, over time and at scale, hopefully be improved. For context, the birth of a first child is the most likely catalyst to see a shopper transition to online grocery; it is also a life event where weekly spend on grocery increases significantly, so one can understand why competition for this customer would be fierce. Adoption of online grocery has also been relatively slow, with its overall share of the UK market only about 7% in 2019. Thus, it was easy for the grocers to consider this a problem ‘at the margin’ rather than of utmost urgency. The presence of Amazon, which has dipped its toes into the online grocery market, has also loomed in the background. With its offer primarily an added benefit for Amazon Prime subscribers, the grocers have been keen to remain very competitive vis-à-vis this elephant in the room.
Covid drives up online demand
Readers will no doubt be aware of the heightened demand for online grocery delivery slots during the Covid-19 pandemic. To put some numbers around this, according to Kantar the online share of UK grocery has increased from about 8% in 2019 to over 13% in November 2020, effectively turbocharging many years of future adoption into only a few months. Tesco’s online sales have grown from roughly 9% of total UK sales to 16%. In order to meet demand, Tesco has scaled up its online capacity from 600,000 slots per week to 1.5m slots.
Not only have order numbers increased significantly, but several second order effects are playing out which are having a potentially transformative effect on unit economics. Ocado reported an average basket size of £137.46 during its most recent reporting period for the six months to the end of May 2020, which was up 28% year-on-year, whilst Sainsbury’s baskets were up 17%. Delivery van productivity is improving across the board, which is important given costs of running the vehicle are both the largest incremental cost incurred fulfilling online orders and primarily fixed (leasing, insurance, wages). Tesco’s fleet has been delivering 21% more orders per van than in 2019 as ‘drop-densities’ improve (i.e. shorter distances with fuller vans and fewer un-booked slots). In addition, customers have increasingly been opting for click and collect which has increased from around 13% of Tesco’s online orders to roughly 21%. This is helpful as, at risk of stating the obvious, the delivery costs are avoided. Finally, with demand exceeding supply, the industry has been able to increase average delivery charges, whilst eliminating free delivery slots and money-off vouchers to induce new customers.
Whilst Tesco does not disclose its online profitability specifically, our engagement with the company has confirmed it is now enjoying a distinctly better economic model than prior to Covid-19, such that it is no longer seeing significant dilution from online sales relative to in-store. That is quite a remarkable shift compared to the margin estimates we mentioned earlier and challenges the embedded consensus view. Other datapoints from across the industry reinforce the point: whilst Ocado’s model is not directly comparable to Tesco’s, its retail earnings before interest tax depreciation and amortisation (EBITDA) margins expanded dramatically during the first half of its 2020 financial year (from 3.04% to 4.47%); meanwhile, Sainsbury’s management recently spoke of ‘better online margins and profitable growth’.
This, of course, all sounds very positive. However, what matters most to the investment case is whether these improvements prove durable. Whilst some normalisation in online grocery demand post-Covid seems inevitable, as shoppers spend less time and consume fewer calories at home when distancing measures are lifted, it is possible that a reasonable chunk of the extra online spending will stick. Anecdotally, Ocado’s management have long stated that once a customer has shopped online three or four times, then they tend to become very loyal to the channel. Bain estimate that about 45% of the online grocery spending surge will stick in the UK beyond lockdown, with it now projecting online grocery penetration rates of up to 14% in 2025, compared to pre-Covid forecasts of roughly 9%. Meanwhile, technology should continue to help the retailers lower their unit costs. Tesco, for instance, is installing 25 semi-automated warehouses over the next three years at its largest stores in the UK, at a cost of £5m apiece, which it believes can help contribute to a 20% improvement in ‘pick & pack’ productivity. Whilst it remains to be seen whether industry discipline on delivery pricing and vouchering remains, there are reasons to be hopeful. With online now a much more meaningful channel for all players, collective economic rationality has become desirable; and the opportunity to entice high-value new customers into the channel is less obvious (if you aren’t shopping online for grocery now, then you are fairly unlikely to anytime soon). Thus, whilst there may be some ups and downs along the way, we now feel considerably more confident that a path to sustainable profitability in online grocery exists than if we’d been having this discussion in early 2020.
Summing up
So, what does this all mean for how the stock market might appraise Tesco in future? To start with, it is perhaps helpful to qualitatively assess where the consensus view sat coming into the pandemic. We’d summarise it broadly as: Tesco’s current online grocery penetration of roughly 9% might increase to 15-20% over the next 5-10 years, representing a 10-20% profit headwind to its core retail business…and don’t forget about discounters Aldi and Lidl, who will keep chipping away at market share.
Next, let’s think quantitatively about what’s priced into the shares. At Majedie, we’re not dogmatic about one valuation methodology, preferring to apply a broad toolkit. One shorthand metric we find useful is the concept of free cash flow total return (FCFTR). This concept, attributed to fabled investor Bill Miller, states that the expected return on an equity investment is the sum of: i) its free cash flow (FCF) yield over a forecast period; and ii) the anticipated sustainable growth rate for FCF beyond this. Applying this logic to Tesco, sell-side analysts we follow are projecting £1.7bn to £1.8bn of FCF for calendar year 2021-22, which equates to a prospective 10.5% FCF yield at current share prices. We estimate the expected rate of return on UK equities is 8.5% per annum (in line with the 50-year average total shareholder return from the FTSE All Share Index). Rearranging the FCFTR formula we can back solve from the share price, which shows that embedded expectations are for Tesco to deliver a sustained FCF decline rate of about 2.0% per annum (8.5% required return minus 10.5% FCF yield). So, valuation appears to match the gloomy consensus narrative, with Tesco effectively being treated as a company whose profit pool faces terminal decline.
Our job is to assess fundamentals and determine whether those embedded expectations are reasonable. And we are actually relatively sanguine about Tesco’s ability to defend or modestly grow its cashflows. In addition to the improvements in online profitability discussed earlier, shoppers have responded favourably to meaningful improvements in Tesco’s price points, product quality and store experience over recent years. As a result, Tesco is now winning grocery market share from the discounters for the first time in many years and has reported a quantum improvement in customer loyalty, with 1.1m shoppers now shopping more regularly with Tesco (as at December 2020). As well as a more robust operating framework, Tesco’s balance sheet has been transformed by the well-priced disposal of its Thai business unit for approximately £8.0bn, taking its total indebtedness/EBITDA ratio down to about 2.5x from a peak of 6.0x in 2014-15. We expect now this target leverage has been achieved Tesco will be in a position to return substantially all of its FCF to shareholders, through a combination of dividends and share buybacks.
We don’t wish to overstate the potential for a company such as Tesco, which is operating in mature markets and generating low double digit returns on capital, to meaningfully compound shareholder value over time. However, we see it as an attractive source of yield within our portfolios, with more robust earnings streams than the market is currently pricing. We are quite happy to clip a 10.5% FCF yield on our investments, as through the cycle this should contribute nicely to the Fund’s overall return. But if Tesco, through strong, steady execution, can continue to demonstrate its durability to the market, then we think any change in the share price multiple (from the undemanding 10.5% starting FCF yield) is likely to be favourable.
Key Risks
