Over the past 20 years, grocery retailing has probably been the sector most investigated by the UK competition authorities. Given that three quarters of the grocery M&A deals that have been reviewed have been cleared unconditionally, this might be taken to indicate a fairly light touch approach to the sector. However, merger review decisions have played an important role in signaling the types of deals that are likely to be acceptable. The effectiveness of this signaling system appeared to break down quite dramatically in 2018 when the CMA prohibited a merger between the UK’s second and third largest grocery retailers. Long-term trends in consumer behavior that had changed competitive dynamics in the sector, and new approaches by the CMA to assessing local market competition, had indicated that consolidation between the UK’s larger retailers might receive regulatory clearance. However, this turned out not to be the case. Any large-scale consolidation in UK grocery retailing now looks some way off.
By Andrew Taylor & Nick Warren1
Over the past 20 years, grocery retailing has probably been the sector most investigated by the UK competition authorities. Since 2003, forty-three M&A deals involving grocery retailers have been reviewed by the Competition and Markets Authority (“CMA”).2 These have ranged from single store acquisitions to deals involving hundreds of supermarkets. During this time, grocery retailers have also been investigated for possible breaches of UK competition law and under the UK’s market investigation regime.3
Around three quarters of the grocery deals reviewed by the CMA have been cleared unconditionally. This might be taken to indicate a fairly light touch approach. However, the CMA’s decisions have played an important role in setting boundaries and signaling the types of deals that are likely to be acceptable. UK grocery retailers have, as a result, been able to shape their M&A activity so that, for the most part, competition clearance has been relatively straightforward.
The effectiveness of this signaling system appeared to break down quite dramatically in 2018 when the CMA prohibited Sainsbury’s, the UK’s second largest grocery retailer, from acquiring Asda, the UK’s third largest grocery retailer. Long-term trends in consumer behavior that had changed competitive dynamics in the sector, and novel approaches by the CMA to assessing local market competition, had indicated that consolidation between the UK’s larger retailers might receive regulatory clearance. However, this turned out not to be the case. While these underlying trends, which had informed Sainsbury’s and Asda’s expectations of a CMA clearance, look set to continue, any large-scale consolidation in UK grocery retailing now looks some way off.
II. CONSUMER BEHAVIOR AND RETAILER STRATEGIES
For readers unfamiliar with UK grocery retailing, there are around 7,500 large and mid-sized supermarkets and around 42,000 convenience stores across the UK. Ten major grocery retailers have outlets across Great Britain, namely Tesco, Sainsbury’s, Asda, Morrisons, Waitrose, Marks & Spencer Simply Food (“M&S”), the Co-operative Group (“Co-op”), Aldi, Lidl, and Iceland.4
Tesco, Sainsbury’s, Asda and Morrisons, the UK’s four largest retailers, have store portfolios that include large and mid-sized supermarkets and smaller convenience stores. Their larger stores typically have a product range that includes general merchandise as well as food and drink, and petrol (i.e. gas) stations are often on site. Waitrose and M&S target the premium end of the grocery market, while the Co-op is largely focused on convenience stores. Aldi and Lidl, which have both grown rapidly in recent years, generally offer a more limited range of products at discounted prices, while Iceland specializes in frozen food.
As well as these national retailers, the UK has many independently-owned convenience stores. Around half of these stores are members of “symbol groups,” such as Budgens, Nisa, Londis and Mace. The grocery wholesalers, which own and operate these symbol groups, supply member stores with food and drink and provide branding and other services.
In the years leading up to the COVID-19 pandemic, several long-term trends in consumer behavior were having a significant impact on the strategies of UK grocery retailers.
In particular, consumers were increasingly:
- conducting several smaller shopping trips each week rather than a single large weekly shop;
- turning to the fast-growing discount chains, Aldi, and Lidl, for part or all of their grocery needs; and
- buying groceries online for home-delivery rather than going to a store.
Grocery retailers responded to these trends in several ways.
- Tesco and Sainsbury’s, and to a lesser extent Asda and Morrisons, expanded their convenience store portfolios.
- Larger grocery retailers rationalized their product range and focused even more on prices as the discounters, Aldi, and Lidl, continued to grow. In 2018, Tesco started a new retail brand “Jacks” to compete directly with Aldi and Lidl.
- UK grocery retailers developed and expanded their online offer. The UK’s first online-only grocery retailer, Ocado, entered the market in 2000.
Much of the larger retailers’ expansion into the convenience store segment came through organic growth, but there were also some significant acquisitions. These included Co-op’s purchase of Balfour (2003), Sainsbury’s purchase of Jacksons (2004), and Tesco’s purchases of Adminstore (2004) and Mills (2011) – see Figure 1.
Figure 1: Grocery retailer M&A deals reviewed by UK competition authorities, 2003-21
- The number of stores acquired in each transaction is represented by the size of each circle. (Each acquired store is counted equally regardless of its size.)
- Of the 35 transactions that were unconditionally cleared by the CMA, 33 were at Phase 1 and two at Phase 2. (The TDR and Issa Brothers’ acquisition of Asda in 2021 is counted as an unconditional clearance as there were no conditions attached to the grocery retail element of the acquisition. TDR/Issa were, however, required to divest a number of petrol stations given their concurrent ownership of EG Group, a major UK petrol station operator.)
- The October 2021 acquisition of Morrisons by the private equity firm Clayton, Dubilier & Rice (“CDR”) is yet to be reviewed by the CMA. It seems likely that the grocery retail element of this acquisition will not encounter any merger control difficulties. CDR may, however, be required to divest some petrol stations given its concurrent ownership of MFG, the UK’s largest petrol station operator.
More recently, Tesco’s acquisition of Booker in 2017 and Co-op’s acquisition of Nisa in 2018 further increased these retailers’ exposure to the convenience store sector. Booker is the UK’s largest grocery wholesaler and, at the time of its acquisition, supplied around 5,500 independently-owned convenience stores. It also operates several symbol groups. Nisa was supplying around 4,200 independently-owned convenience stores when it was acquired by the Co-op.5
In 2011, Asda acquired Netto, which was the UK’s third largest limited range discounter after Aldi and Lidl. Asda is generally regarded as the grocery retailer that competes most strongly with Aldi and Lidl on price. However, its acquisition of Netto was not about acquiring a brand to compete more directly with the discounters (as Tesco has sought to do in establishing the “Jacks” brand), rather it repurposed the Netto estate into Asda branded stores.
In terms of online shopping, most UK grocery retailers have developed their online offer through in-house initiatives. Waitrose, however, worked in partnership with Ocado, the UK’s online grocery retailer, until 2019, when M&S acquired a 50 percent share in Ocado’s UK operations as a way of boosting its online presence. Waitrose has since had to rapidly develop its own standalone online presence.
III. THE CMA’S APPROACH TO GROCERY RETAIL MERGER CONTROL
Most M&A deals in UK grocery retailing, beyond the very small, qualify for review by the CMA. The “share of supply” test allows the CMA to review any transaction where the merged entity supplies 25 percent or more of a good or service in a significant part of the UK. A “significant part of the UK” can be quite small. For example, Slough, an urban area with a population of around 165,000, was found to be a “significant part of the UK” when the CC reviewed Tesco’s acquisition from Co-op of a single supermarket in this town.6
The likelihood that CMA clearance will be needed means that its approach to assessing competition in grocery retailing plays a significant role in shaping the sector’s deal making. The sector has had some reasonably clear guidance over the past 20 years due to the regulator’s consistent approach to segmenting the grocery market by store size, product range and catchment area combined with its long-standing view on the acceptable number of competitors in a local market.
Much of the CMA’s approach has been built on its observation of household shopping patterns, and in particular, the tendency of many UK households to conduct a single large weekly shopping trip. Based on this, the CMA has identified a separate product market that includes those stores large enough, in terms of their size and product range, to meet a household’s weekly shopping needs. The CMA has settled on a floor space of 1,400m2 as the minimum size of a store that is large enough to compete for households’ weekly shopping trips.
For stores smaller than 1,400m2, there is further differentiation between convenience stores and mid-sized supermarkets. UK retail trading law allows shops with a floor space of less than 280m2 to open for longer hours. This has provided a logical threshold for distinguishing between convenience stores, with their smaller product range, and mid-sized stores.7
Of course, a shopper looking to buy a small number of items which could be found in a convenience store could also purchase these same items at a mid-sized or large grocery store. As a result, the CMA has viewed the competitive constraint between stores across these three different size brackets as asymmetrical. Large stores act as a competitive constraint on mid-sized and convenience stores (as any shopping trip or “mission” that could be satisfied at a convenience or mid-sized store could also be satisfied at a large store), but it has concluded that the more limited product range at these small stores means that they cannot compete for shoppers carrying out a large weekly shop. Similarly, mid-sized stores are regarded as providing a competitive constraint on convenience stores, but not vice versa.
Layered on top of this segmentation by store size has been further segmentation based on the breadth of retailers’ product range, where this is independent of the size of their stores. For example, the CMA has regarded the discounters, Aldi, and Lidl, which have a more limited product range and a larger proportion of own-label products, as providing less of a competitive constraint on those retailers that sell a wider range of branded and own-label goods (e.g. the UK’s four largest grocery retailers, Tesco, Sainsbury’s, Asda, Morrisons). The CMA has similarly regarded M&S for much of the past 20 years as providing a more limited constraint on other grocery retailers due to the predominance of own-label goods in its product range.
In terms of the geographic market, the CMA has consistently found that convenience, mid-sized and larger stores serve progressively larger catchment areas. These tend to be around 5, 10, and 15 minutes’ drive-time respectively, with larger catchment areas in rural areas compared to urban conurbations.
The end result has been a fairly clear set of “rules of thumb” that retailers could apply when assessing the CMA’s likely attitude to an M&A deal. For example, a large supermarket operated by one of the UK’s four largest retailers would be regarded as providing a meaningful competitive constraint on other large supermarkets, mid-sized stores, and convenience stores within a 15-minute drivetime. Convenience stores would be regarded as providing a competitive constraint on other convenience stores within a 5-minute drivetime, and not on any larger stores within that catchment area. Mid-sized supermarkets operated by one of the UK’s four largest retailers, would be regarded as constraining other mid-sized supermarkets and convenience stores within their catchment area. However, mid-sized supermarkets operated by one of discounters (i.e. Aldi or Lidl) would not be regarded as providing a competitive constraint on a mid-sized supermarket operated by one of the UK’s four largest retailers.
When the CMA has then reviewed a grocery retail M&A deal, it could be expected to conclude that there would be a realistic prospect of a substantial lessening of competition (“SLC”), the formal test for making an adverse finding in a merger control case at Phase 1, in those markets where the number of competitors would decline to fewer than four because of the transaction.
IV. HAS CHANGING CONSUMER BEHAVIOR ERODED THE CMA’S APPROACH?
The CMA’s approach has acted as a brake on consolidation between the UK’s four largest grocery retailers, which between them operate most of the UK’s large supermarkets, while allowing retailers with portfolios of mid-sized and convenience stores, or which offered a more limited product range, to be acquired. This includes, for example, Co-op’s acquisition of Somerfield in 2009 and Asda’s acquisition of Netto in 2011 as well as Morrisons acquisition of Safeway in 2003.
Questions about the CMA’s likely attitude to potential consolidation between the UK’s largest grocery retailers started to emerge in the lead up to Sainsbury’s announcement of its plan to buy Asda in 2018.
The declining importance of households’ weekly shopping trip suggested that smaller stores might be placing a more meaningful competitive constraint on larger supermarkets. Rapid growth by Aldi and Lidl suggested that their stores were also competing strongly with other grocery retailers for shoppers, notwithstanding their more limited product range. The move to online shopping suggested that retailers could compete with one another over longer distances. Adding fuel to the emerging possibility of a change in the CMA’s attitude was the CMA’s adoption in 2015 of a new methodological approach to assessing local competitive constraints that could systematically consider both a wider range of competitors and store locations.
As set out above, the CMA’s traditional approach had been to count the number of competing stores in each local market (as defined by store size, product range and catchment area). Where the number of competing retailers in a local market fell below four, this would generally be sufficient for an adverse decision in a Phase 1 review. In a more detailed Phase 2 review, the CMA would combine this count of competitors with a qualitative assessment of other relevant factors, such as the relative location of stores within a local market (and any that might be just outside of it).
This approach had two main shortcomings. First, decisions about which stores were in and out of the market could be quite binary. For example, a mid-sized Aldi store located next to a mid-sized Asda store would not be counted as a competitor (i.e. it would be regarded as placing zero competitive constraint on the Asda store), while a mid-sized Waitrose store located 10 minutes’ drive-time away would be regarded as constraining the Asda store. Similarly, a mid-sized Sainsbury’s located within the same geographic market as a large Morrisons store would be regarded as placing zero competitive constraint on the Morrisons store. This approach looked increasingly like it was lacking in nuance given changes in shopping behavior and the growth of the discounters.
Second, when it came to Phase 2 reviews, the qualitative assessment of factors such as the relative location of stores in a local market ran the risk of being inconsistent, either within a single transaction or between transactions, as the CMA had no systematic methodology for taking this into account. Our own experience of spending many hours with the CC staff team in the 2006-08 market investigation looking at large numbers of maps showing supermarket locations means that we can appreciate the potential for inconsistent judgements about relative competitive constraints when these were based on an ad hoc visual inspection.
In 2018, Sainsbury’s and Asda took the view that the combination of: (i) changed consumer behavior, which suggested that both smaller stores and the discounters were a greater constraint on larger stores; and (ii) developments in the CMA’s methodological approach, which could accommodate this development within CMA decision-making, were likely to be sufficient to secure clearance of a merger between them.
This, however, did not turn out to be the case.
Consistent with Sainsbury’s and Asda’s expectations, the CMA found that mid-sized stores provided a competitive constraint on larger stores, and the CMA also found that Aldi and Lidl, and to a lesser extent Iceland, provided a competitive constraint on Sainsbury’s and Asda stores. However, even having found that Sainsbury’s and Asda’s stores faced a wider range of competitive constraints than previously accepted, the CMA still concluded that competition would be diminished in more than 500 local areas where Sainsbury’s and Asda both had grocery stores (which was around 40 percent of all their stores) if the merger went ahead. The extent to which mid-sized stores and the discounters’ stores constrained Sainsbury’s and Asda, and their closest rivals Tesco and Morrisons, was simply not sufficient, in the CMA’s view.
Moreover, in relation to online grocery retailing, the CMA concluded that this was a separate product market to in-store grocery shopping. It found that many customers who shopped online did so for specific reasons, such as being unable to transport bulky items or not being able to easily attend a physical store. For these customers, shopping in-store would be a poor substitute for shopping online. Having reached the conclusion that online grocery retailing constituted a separate product market, an adverse finding on the competitive effects quickly followed: around 90 percent of online groceries were sold by Tesco, Asda, Sainsbury’s and Ocado in 2017, and Ocado lacked the national reach of the other three retailers.
Given the scale of the adverse findings in local grocery markets, together with additional adverse findings in relation to the potential for coordinated effects in both online and in-store retailing, the CMA concluded that it should prohibit the planned acquisition.
As a result, the CMA’s decision on the Sainsbury’s/Asda merger proposal can be seen as a continuation of its long-term position, even if its revised methodology permits a more nuanced approach to assessing competition between grocery stores.
V. LOOKING FORWARDS
Since 2019, grocery retailing in the UK, like in many other countries, has faced profound challenges because of the COVID-19 pandemic. Demand for groceries skyrocketed as pandemic-related lockdowns were put in place and eating out was no longer possible. Growth in online grocery shopping was turbo-charged. Shoppers reverted to large weekly shops, with even larger shopping baskets. Coming out of the pandemic, UK food and drink supply chains have become stressed because of labor shortages in a range of areas, including HGV (truck) drivers, meat production facilities, and fruit and vegetable picking.
Two of the UK’s largest grocery retailers, Asda and Morrisons, have been acquired by private equity firms. While neither acquisition has raised any concerns in relation to grocery retailing, Asda’s acquisition by TDR and the Issa Brothers, who also jointly own EG Group, a major UK petrol retailer, have thrown up issues for the CMA in terms of competition between petrol stations. Around 30 stations had to be sold as a condition of achieving CMA clearance. CDR’s acquisition of Morrisons, while not yet reviewed by the CMA, seems likely to raise similar issues.
The long-term shift to online shopping also looks set to continue. There are now several entrants into the UK and other European markets specializing in ultrafast delivery – the online equivalent of convenience shopping, including Getir, Gorillas, Jiffy, Zapp, and others.
Supply chain issues, private equity acquisitions of UK retailers and growth in fast-delivery groceries do not appear to have any immediate implications for the CMA’s approach to assessing competition in the sector. As a result, the likelihood of any consolidation between the UK’s four largest retailers in the short-term looks remote. However, all these developments point to a market that is more dynamic than ever. Retailers are going to keep developing their strategies as they respond to these challenges, and long-standing certainty on the part of the UK’s competition regulator about the importance of competition between the largest UK grocery retailers seems unlikely to last forever.
1 Andrew Taylor & Nick Warren are co-founders of Aldwych Partners.
2 The start of the UK’s modern merger control regime in 2003, following passage of the Enterprise Act 2002, is a good starting point for looking at the broad sweep of competition authority intervention in UK grocery retailing. However, even before this, the Competition Commission (“CC”) looked in detail at the grocery retail market in 1999-2000, and reviewed major transactions, such as Morrisons’ acquisition of Safeway in 2003. References to the CMA in this article should also be taken to include its predecessor agencies, the Office of Fair Trading (“OFT”) and the CC.
3 One of the authors of this article led the CC staff team that conducted the 2006-08 groceries market investigation.
4 Not all these retailers have a presence in Northern Ireland. (Great Britain comprises England, Scotland, and Wales, while the United Kingdom includes Great Britain plus Northern Ireland.)
5 Direct ownership of grocery stores by both Booker and Nisa, before their acquisition, was very small for Booker and non-existent for Nisa. As a result, neither acquisition shows up significantly in Figure 1.
6 Competition Commission, Tesco plc / Co-operative Group (CWS) Ltd store at Uxbridge Road, Slough, November 2007. One of the authors of this article led the CC staff team that conducted the Phase 2 review of this acquisition.
7 The CMA has also occasionally considered the extent to which it should differentiate between convenience stores at the upper end of this size threshold versus smaller convenience stores. However, this has tended to be considered in the assessment of relative strength of competitive constraints within the market rather than defining separate product markets for smaller and larger convenience stores.