GOVERNANCE WATCH
By InsidEntity Editorial Desk · Aug 14, 2026 · 14 min read
Pick n Pay published a trading update on 6 August for the 20 weeks to 19 July 2026. Group turnover rose 2.7%, or 2.5% on a like-for-like basis. Boxer grew 7.2%. The Pick n Pay brand reported turnover flat year on year.
Read only those four numbers and the conclusion writes itself: the discount chain is carrying the group while the core supermarket business stands still. That has been the story since Boxer was unbundled and separately listed in 2024, and it is the story most coverage of this update told.
There is a fifth number that complicates it, and it sits in the same announcement.
Pick n Pay carries an InsidEntity Company Risk Rating of 3.25, Good. Full breakdown on Pick n Pay’s InsidEntity report.
The relationship at the centre of this company
Pick n Pay holds 53.1% of Boxer, a separately listed company with its own minority shareholders. Since 1 March 2026, the non-independent non-executive chairman of Boxer has been Sean Summers, the chief executive of Pick n Pay. James Formby, the former RMB Group chief executive who chaired Boxer until February and is now Pick n Pay’s chairman, remains on Boxer’s board as a non-independent non-executive director.
So the two most senior figures at the parent company sit on the board of the listed subsidiary, one of them chairing it, at a time when the parent has been selling that subsidiary’s shares to fund its own recovery, twice, reducing its holding from 65.6% to 53.1%.
Boxer’s 46.9% minority shareholders own shares in a company chaired by the chief executive of the shareholder that is selling it down.
The point is not that the share sales required Boxer’s approval. They did not: a controlling shareholder disposing of its own stock is its own decision, and Boxer’s board was not a party to either bookbuild. The point is about duty. As chairman of Boxer, Summers owes a fiduciary duty to all of Boxer’s shareholders. As chief executive of Pick n Pay, he owes one to Pick n Pay. Those duties align most of the time and diverge in specific, foreseeable places: the timing and size of further sell-downs and their effect on Boxer’s share price and free float, dividend policy at a subsidiary whose cash the parent needs, capital allocation between the two, and any supply, logistics or property arrangement between them.
Each of those is a decision where Boxer’s minority would want its chairman’s undivided attention, and each is a decision in which his other employer has a position. That is the structure, and it is fully disclosed. The directors are correctly classified as non-independent, cross-directorships between a parent and a controlled subsidiary are ordinary, and there is a serious argument that the parent’s leadership belongs on the board of an asset carrying most of the group’s value. All of that can be true while the divergences remain real, and the record over the next few transactions is what answers it.
The like-for-like line nobody quoted
Boxer’s 7.2% turnover growth came with like-for-like growth of 2.2%. The Pick n Pay brand’s flat turnover came with like-for-like growth of 2.6%.
Strip out new stores, and the struggling brand grew slightly faster in existing stores than the growth engine did.
That single comparison reframes both halves of the group. Boxer’s headline growth is substantially a rollout story: more than 500 stores, a target of 60 new openings in the year, and the arithmetic of a fast-expanding footprint. Its existing stores are growing at roughly the rate of food inflation. Pick n Pay’s flat turnover is the arithmetic of the opposite process: existing stores growing 2.6% while closures under the turnaround programme remove the difference.
The headline story survives this. Boxer is profitable and the Pick n Pay brand is not, Boxer contributes most of the group’s growth in absolute terms, and without Boxer the group would not have grown at all. Directionally, the coverage is right.
What the like-for-like line adds is where the difference comes from. One brand is opening stores that grow modestly. The other is closing stores while the survivors grow modestly. On the shopper-facing measure, the two are closer than the headline implies, and the separation between them is substantially a property of the store estate rather than of consumer preference. That matters for a turnaround whose stated aim is to fix the existing stores.
The year behind the update
To read the current numbers, it helps to know how bad 2024 was. Pick n Pay reported a loss after tax of roughly R3.19 billion that year. The trading margin collapsed to 0.3% from 2.9%. The group carried around R11 billion of debt across an eleven-bank lending syndicate, with covenants breached. The R4 billion rights offer and the separate listing of Boxer were not strategic refinements. They were a solvency response.
Against that baseline, the 2026 financial year, reported in May, reads differently. Group turnover reached R120.3 billion, up 3.4% on a comparable 52-week basis. The headline loss narrowed to R386 million from R408 million. Gross margin improved to 18.8%. Boxer sales grew 12.3% while Pick n Pay brand sales declined 1.6%. A company that lost R3.19 billion two years ago and now loses R386 million has closed most of a very large gap.
Underneath that, the direction is not uniformly good. The core Pick n Pay segment’s trading loss doubled to roughly R1.0 billion from R0.5 billion, and on the measure that includes lease interest the group trading loss widened to roughly R2 billion from R1.7 billion. The break-even target for the Pick n Pay segment moved from the 2028 financial year to 2029, the second time that date has been pushed out. Employee costs ran at 41.4% of trading expenses, which the company itself described as out of kilter with the sector, and a section 189A consultation with store employees and unions followed. The shares fell 3.55% on the results and are down roughly 60% over five years.
The funding for the next phase came from Boxer. On 18 May 2026, a week before the results, Pick n Pay sold 12.5% of Boxer’s shares in issue through an accelerated bookbuild at R82 a share, a 3.2% premium to the 30-day volume weighted average price, raising R4.7 billion and reducing its holding from 65.6% to 53.1%. Boxer’s own trading margin improved 30 basis points to 5.7% in the same year. The asset being sold is the one that earns; the proceeds fund the one that does not.
That transaction is the centre of this company’s governance picture, and it deserves more attention than it received.
What Pick n Pay discloses that Shoprite does not
This platform examined Shoprite last week. The comparison between the two is not the obvious one.
Shoprite’s growth engine is Checkers Sixty60, which turned over R25.5 billion and produced more than a third of the group’s entire sales growth. Sixty60 sits inside Supermarkets RSA with no separate segment disclosure, no segment margin, and no segment return on capital. Nobody outside Shoprite knows whether it earns a profit.
Pick n Pay’s growth engine is Boxer, and everything about it is public. Boxer has its own listing, its own board, its own results, its own annual report, and its own share price. Its like-for-like growth, store rollout, margins and valuation are all disclosed, quarterly, because a separate listing requires it.
The weaker company has the more transparent structure. Pick n Pay’s disclosure is better not because it chose better, but because unbundling Boxer to raise capital had disclosure as a consequence. Necessity produced visibility. Shoprite, under no such pressure, keeps its growth engine inside the consolidated accounts, which is entirely lawful and entirely its right.
For an investor, the practical effect is that Boxer can be valued directly and Sixty60 cannot. That is worth holding in mind when the two groups’ ratings are compared.
The governance layer
| Component | Pick n Pay | Shoprite |
|---|---|---|
| Director Independence | 5.0 | 5.0 |
| Shareholder Influence | 5.0 | 5.0 |
| Director Capacity | 3.00 | 2.81 |
| Auditor Independence | 0.0 | 2.0 |
| Overall Rating | 3.25 | 3.70 |
Pick n Pay (5.0 + 5.0 + 3.00 + 0.0) ÷ 4 = 3.25. Shoprite (5.0 + 5.0 + 2.81 + 2.0) ÷ 4 = 3.7025. Ratings update regularly; check the live reports for the current view.
One pillar accounts for the whole difference, and it is the one Pick n Pay could change.
Pick n Pay’s Auditor Independence pillar scores nil. Ernst & Young was appointed on 27 July 2015, replacing KPMG, which puts the relationship at eleven years, corroborated independently by Pick n Pay’s own 2025 annual report, which described EY as its audit partner for a decade. Shoprite scores 2.0. The framework marks auditor independence down as tenure lengthens and past ten years stops crediting it, and the two grocers sit on opposite sides of that line’s practical effect.
That tenure just acquired a second dimension. Tina Rookledge took over as group chief financial officer on 6 August 2026, the day of the annual general meeting, succeeding Lerena Olivier after fifteen years in the role. Rookledge joined Pick n Pay from Ernst & Young, where she had been Western Cape regional managing partner for six years, having made partner there in 2006. The incoming chief financial officer of a company whose external auditor has held the seat for eleven years is a former partner of that same firm. Nothing about the appointment breaches any rule; cooling-off requirements govern exactly this kind of move and the board evaluated it as a succession decision, not an independence one. It does mean the case for a re-tender, already the single pillar separating this company’s rating from its rival’s, now has a second reason behind it. That single pillar contributes half a point of separation between the two ratings, which is more than the full 0.45 gap between them, because the board pillar pulls a fraction back the other way.
Put differently: on this framework, the difference between the South African grocery sector’s leader and its challenger is not commercial performance, board composition, or shareholder structure. It is an audit appointment date. A re-tender would close it.
The other three pillars are near-identical. Both score 5.0 on Director Independence. Both score 5.0 on Shareholder Influence, and for Pick n Pay that reading is worth stating explicitly: the Ackerman family retains voting influence of roughly 36.8%, but the pillar measures economic concentration against a 20% threshold, and the family’s economic interest has fallen to 18.2%. Below the line, so the register reads unconcentrated. Shoprite scores the same for the same reason. On Director Capacity the two sit within 0.19 of each other, which is close enough that neither company can claim an advantage from it, though the more interesting question belongs to Shoprite: why the market leader, at record highs, does not have more board bandwidth than a company mid-turnaround.
The names behind Pick n Pay’s 3.00 are worth setting out, because the board refresh is not a swap of stretched veterans for lightly loaded newcomers. James Formby chairs Pick n Pay and Boxer, chairs RMB Nigeria, sits as an independent non-executive director of Vukile Property Fund, where he becomes lead independent director designate from September 2026, and sits on the board of Nimble Group. Thabo Leeuw, appointed in February 2026, arrived already carrying an independent seat on the JSE itself, a seat at Attacq, the chairmanship of Hulamin, and a prior term as lead independent director at Rhodes Food Group. Aboubakar Jakoet, still on the board after retiring as chief financial officer in 2019, chairs the Audit, Risk and Compliance Committee, has just become lead independent director at Oceana Group, sits at Sygnia, and serves on the University of Cape Town Council.
The refresh runs on a published schedule. David Robins retired after twenty-two years at the 2024 annual general meeting, David Friedland after ten years in 2025, and Audrey Mothupi-Palmstierna, after thirteen years, is retiring at the 2026 meeting alongside Jakoet’s own scheduled rotation off the audit committee chair. Grant Pattison and Pooven Viranna joined the board alongside Leeuw and Rookledge. That is a multi-year, minuted programme rather than a response to this year’s headlines, and it is consistent with what the capacity score shows: incoming directors who are independently well qualified and, in Formby’s and Leeuw’s cases, carrying substantial board loads of their own.
What the voting percentage does not tell you. The family’s decline from 49% to 36.8% reads as control loosening, and in one sense it is. The 2025 annual general meeting suggests a narrower reading.
At that meeting, held before the November sell-down and therefore at the 49% level, support for the remuneration resolutions ran at roughly 62% to 66% when counted on ordinary shares alone, and director Aboubakar Jakoet was re-elected with 65.24%. Those figures pass. What they also show is that on ordinary shares alone, dissent ran above 34%, comfortably past the 25% threshold at which the JSE listings requirements and King IV oblige a company to engage its dissenting shareholders and report back.
In other words, the B share vote did not merely add to a comfortable majority. It was the difference between resolutions carried with routine support and resolutions carried with the kind of dissent that triggers a formal engagement obligation. Ordinary shareholders were expressing a level of disagreement that the aggregate result did not surface.
That reframes what the November cancellation of 105,186,279 B shares actually changed. It reduced the family’s voting influence by roughly twelve percentage points. Whether it reduced it below the level needed to carry contested resolutions comfortably is a different question, and the 2026 annual general meeting will be the first occasion on which it is tested. Shareholder Influence at 5.0 is correct on the framework’s economic test. The voting record is where the rest of the picture lives.
Pay tied to the turnaround
Sean Summers was paid R25.2 million for the 2026 financial year, and he forfeited one million performance shares, worth roughly R21 million, because the group had not reached break-even. Senior executives took a salary freeze. His contract was extended to 2029, the new break-even target.
That is remuneration policy operating exactly as it is written: a chief executive who did not meet a performance condition lost the pay attached to it.
The group’s own disclosure also records the pay distribution plainly: the top five employees averaged R1.22 million while the bottom 5% of staff earned around R72,000, a disparity the company attributes to the workforce profile of retail. Publishing that ratio at all, in the middle of a section 189A process, is a disclosure choice a company is entitled to avoid.
Reading the turnaround and the governance together
None of this says Pick n Pay is failing. A group turning over R120 billion, with a narrowing headline loss, an improving gross margin, R4.7 billion of fresh funding and a chief executive whose incentives are visibly tied to a break-even date, is executing a turnaround rather than sliding.
Read together, the two layers say something more specific. Pick n Pay is more transparent than its larger competitor because circumstance forced it to be, scores better on board capacity than the company beating it in the market, carries a nil auditor pillar that accounts for its entire rating gap to that competitor, and now sits in a structure where the leadership of the parent chairs the listed subsidiary the parent is selling down.
The turnaround’s next test is the 2029 break-even date, which has already moved twice. The governance questions are not waiting for it.
What to watch
Five things. Whether the Pick n Pay brand’s like-for-like growth continues to run at or above Boxer’s, which would suggest the core is recovering faster than the headline suggests. Whether Pick n Pay sells down further in Boxer, and how Boxer’s independent directors handle the next transaction. Whether the section 189A process concludes without further cost pressure or industrial action. Whether the audit relationship that produces the nil pillar is re-tendered, since it is the single item separating Pick n Pay’s rating from its larger competitor’s. And whether that re-tender question gets any easier or harder to avoid now that a former Ernst & Young partner sits in the chief financial officer’s chair.
Pick n Pay’s full Company Risk Rating, pillar breakdown, board record and auditor appointment history are on InsidEntity. A trading update tells you what the tills took. The governance layer tells you who sits on both sides of the table while it happens.
About the InsidEntity Company Risk Rating
The InsidEntity Company Risk Rating (CRR) scores companies on a 1 to 5 scale across four weighted governance pillars: Director Independence, Director Capacity, Auditor Independence, and Shareholder Influence, with Director Independence carrying 40% of the rating and the other three 20% each, so a weakness in any single pillar shows up in the headline number. Ratings update regularly as corporate changes occur, so the platform is always the current view. Explore ratings for companies across 145 stock exchanges at InsidEntity.
Related reading: SPAR’s chairman and deputy chair resigned citing threats from its own retailers. The retailers say there were none. Both things are now on the record..
Know your entity. This is independent research, not financial advice.