Opinion. This is an argument piece and it sets out Newswire’s view. Factual claims are sourced in the text to the Commerce Commission, MBIE and on-the-record statements.
Since National promised on 16 September to split Pak’nSave from New World, much of the argument has come down to one question. Would two smaller supermarket groups each have to pay for their own warehouses, trucks, buying teams and computer systems, and would shoppers end up paying for the duplication?
Foodstuffs says yes. “We started with five co-ops, we’re down to two, and that’s given us the scale and efficiencies and the buying power to stay competitive for our customers,” Foodstuffs North Island chair Dean Waddell said on 17 September, in comments reported by interest.co.nz.
BusinessNZ director of advocacy Catherine Beard made the same point. “If structural separation reduces those efficiencies or duplicates costs, there is a risk that some of those costs ultimately find their way to consumers,” she said, in comments carried by interest.co.nz.
National does not dismiss the worry. “Whether a separation would deliver net benefits for consumers would depend on how it is implemented and what happens to supply-chain costs,” finance spokesperson Nicola Willis said when she released the policy. The Sense Partners analysis behind the policy put the net gain to the economy at $2.9 billion over 20 years, or lower if Foodstuffs’ costs rose, and its authors called their work “indicative rather than definitive”, RNZ reported. How much cost a split adds is the number everything else depends on.
Both sides are arguing inside the same assumption, which is that whoever owns the shops must also own the supply chain behind them. Newswire’s view is that the assumption is the problem. The warehouses, the trucks and the bulk buying should be taken out of the supermarkets’ hands altogether, owned by the public, and run at the lowest cost possible for every grocer in the country.
The design is simple to state. The Crown buys the distribution arms of the two Foodstuffs co-operatives at a fair price and runs them as one national network. Its mandate is cost, not profit. It charges what it costs to buy, store and handle the goods, plus a margin of no more than 2 percent to keep the network renewed. Every customer pays the same price for the same goods at the warehouse door, plus freight to their store. A New World, a Pak’nSave, an independent grocer in Auckland and a new chain that does not exist yet would all buy on the same terms, and each would be free to sell at whatever price it likes. Competition between the shops does the rest.
Freight is where the design needs care, because remote New Zealand must not come out of this worse off. The Sense Partners report says Foodstuffs pools freight so that every store pays the same for its groceries wherever it is, and puts the cost at $20 to $30 a tonne for most stores and over $100 a tonne for remote ones. Charging every store its actual freight would end that cross-subsidy and put prices up in exactly the towns with the least choice. Pooling everything, on the other hand, invites anyone to open a store at the end of a long road and send the bill to everyone else.
Newswire’s answer is a capped freight rate with a small, visible fund behind it. Every store pays its actual freight up to a national cap set by the Commission at about the rate most stores already pay. Freight above the cap is met by a remote supply fund, financed by a levy of a fraction of a percent on everything that moves through the network, with the levy, the cap and the list of stores it supports published each year. That is the same money Foodstuffs pools today, made visible and confined to the stores that need it. Telecommunications has run on this principle since 2011, when Parliament set a development levy, now $10 million a year and allocated by the Commission, to pay for rural broadband and other services that would never be built commercially.
To stop the fund ballooning, the subsidy should follow the community rather than the store. Every store served today would qualify. A new store would qualify only if it is 90 minutes or more from the nearest depot, the report’s own definition of a remote delivery, and its town has no qualifying store already. A second store in a town that already has one gets the same truck on the same day and pays actual cost for anything more. Subsidised deliveries would be limited to a set number a week by store size, the fund would be capped as a share of network turnover and reset with each price path, and Parliament would review the whole scheme every five years. Under those rules remote stores are no worse off than today, nobody can farm the subsidy, and the cost of serving the far ends of the country is a published number rather than a hidden one.
Nobody has costed a network of that kind. The Sense Partners report MBIE released with National’s policy modelled two options, and both cut the industry by brand. Its central case assumes the new Pak’nSave group builds its own ambient distribution centres in Auckland and Christchurch, and the New World group builds one in the lower North Island, at $100 million each, with fresh and frozen handling outsourced. It assumes a demerger lifts the cost of goods sold by 1 percent through lost efficiency and buying power. On those assumptions the net benefit is $2.9 billion over 20 years. At 2 percent it is $920 million, and the report says cost increases above 2 percent turn the result negative. The whole case for the brand split lives or dies on the cost of building a second supply chain. The option in which nothing is built twice was not on the table.
The case starts with a bet that was made four years ago and lost. When the Commerce Commission finished its market study in March 2022, it considered requiring the supermarkets’ wholesale businesses to be run separately from their shops and decided against it. Its executive summary said there were “significant efficiencies associated with integrated retail and wholesale grocery operations” and that separation would bring practical challenges and transaction costs. It recommended instead that the major retailers be given the chance to negotiate commercial wholesale supply with their smaller rivals, under regulatory oversight.
The Grocery Industry Competition Act 2023 put that into law, requiring Woolworths and the two Foodstuffs co-operatives to set up systems to supply other grocers at wholesale. The Commission’s preliminary findings paper of June 2025 counted what it produced. By 30 April 2025 there were 161 wholesale supply agreements. In the year to September 2024 the three big players reported about $9.1 million in wholesale sales, which the Commission set against $26.9 billion in retail sales recorded by Stats NZ for the same period.
On those figures, about one dollar in every 3,000 moved through the channel that was supposed to let independent grocers compete. The Commission’s own summary of those findings, on its inquiry page, is that the regime “is not having a material impact at present”.
It has also said why. Suppliers pay the major supermarkets about $6 billion a year in rebates, discounts and promotional payments, as Newswire reported in July, and the Commission found competing retailers have limited or no access to that funding. “New and smaller entrants struggle to compete or grow as this practice allows the major supermarkets to use their market power to obtain prices from suppliers that other retailers can’t,” Grocery Commissioner Pierre van Heerden said on 7 July, when the Commission closed its wholesale inquiry and chose enforcement of the existing rules over new regulation.
In Newswire’s view that outcome should surprise nobody. A wholesaler owned by its customer’s biggest competitor has little reason to hand over its best price, and no amount of oversight changes what it would rather do. A public distributor has no shops to protect. Whatever it negotiates from suppliers, including the rebates and promotional money, comes off the price for every customer equally, because it has no reason to do anything else.
The law already reaches part of the way. A Commission factsheet from April this year sets out the stronger tools available under the Act. One is an order requiring the big retailers to supply wholesale customers on non-discriminatory terms, and a determination under such an order may include provisions about “separating their retail and wholesale operations”. But that power stops at operational separation. The Act’s definition, reproduced in the Commission’s paper, says it “does not include a requirement that any business unit must be operated by different owners”. The warehouse would still answer to the people who own the shops. On that wording, changing who owns it needs new legislation, and National has already said it will amend the same Act within 100 days of taking office.
New Zealand has done this before, more than once. Transpower was carved out of the old Electricity Corporation and set up as a stand-alone state-owned enterprise on 1 July 1994, according to MBIE’s chronology of electricity reform. It still describes itself as national grid owner and system operator. The Electricity Industry Reform Act 1998 then required full ownership separation of local lines companies from retail and generation, in MBIE’s words to encourage competition and to prevent cross-subsidisation. Companies had until the end of 2003 to comply. They completed the ownership split before 1 April 1999.
Telecom is the example Nicola Willis reached for herself. “One is that actually, as was the case with Telecom, ultimately that Foodstuffs voluntarily decides to make this change,” she said on 16 September, describing how a split might be carried out, in remarks reported by interest.co.nz. But Telecom was not divided into two rival phone companies. It told the stock exchange on 31 August 2011 that it would separate its network from its retail business, creating Chorus as the owner of about 1.8 million lines that a range of providers would use to reach their customers. The precedent National is citing is a split between the network and the retailers, not a split between brands.
The pattern in each case is the same. The part of the industry that is wasteful to build twice becomes shared infrastructure with open access, and the competition happens on top of it. There is a grocery version across the Tasman. The Commission’s paper describes Statewide Independent Wholesalers in Tasmania, which buys and distributes on behalf of Woolworths and 170 independent retailers and claims roughly 70 percent of that state’s grocery market. The 2022 market study doubted there was enough demand in New Zealand to support an independent wholesaler. A network that starts with every Foodstuffs store as a customer does not have that problem.
The obvious objection is that a state monopoly with no profit motive will grow fat, and that the bloat will cost shoppers more than the supermarkets’ margins ever did. It is a fair fear and the design has to answer it. Transpower is the model here too. The Commerce Commission sets the maximum revenue Transpower can recover, in the Commission’s words “to run the National Grid efficiently”, and resets that path every four or five years. A public grocery distributor should live under the same kind of cap, set by the same regulator, with its cost per case, its delivery performance and its wastage published so that any grocer or shopper can see whether it is getting leaner or heavier. The mandate should carry efficiency targets alongside the margin ceiling, and its board should answer for both.
It would not be starting from nothing. Foodstuffs says it has 3,000 supply chain and support centre staff. The people, the buildings and the systems that move the groceries today would move them the day after. What changes is who they work for and what they are told to maximise.
There are harder objections and they should be stated plainly. Much of a supermarket’s buying advantage sits in its negotiations with suppliers, not in its sheds. A single public buyer would be the largest customer most New Zealand food producers have, and a mandate to drive cost down is also a mandate to squeeze them. The Grocery Supply Code would need to bind the public distributor as tightly as it binds the supermarkets, and suppliers would need to stay free to deliver direct to stores, which the Commission’s paper says already happens with a significant number of products.
Woolworths is the second problem. It runs its own supply chain and under this plan it would keep it, while being free to buy from the public network if that proved cheaper. That means the New Zealand-owned co-operatives give up their warehouses and the Australian-owned chain does not. The answer is that Woolworths would then face competitors, old and new, supplied at cost, and that the co-operatives’ owners would be paid for what they give up. But it is an asymmetry, and the Commission should be asked whether Woolworths’ distribution belongs in the network as well.
The third is the price. The Foodstuffs distribution centres belong to co-operatives owned by more than 500 store operators. Willis has been careful to tell them that “this isn’t a crazy policy in which we’re expropriating your business”, and that National “will ensure that they continue to have access to national distribution chains”, in comments reported on 17 September. A compulsory purchase at fair value is not expropriation, but nobody has published what fair value would be. The only public number near it is the $1.3 billion the Greens’ KiwiMart policy puts on buying 120 stores and two distribution centres, a Parliamentary Library estimate that covers far less than a national network. BusinessNZ has already said National’s milder plan sends a chilling signal to business. This one would be colder, and the case for it has to be that groceries are an essential service with a market structure that four years of lighter remedies have not shifted.
The electricity precedent is also a caution. The 1998 ownership rules were relaxed twice before the Act was replaced in 2010, and electricity prices rose far faster than food prices in the year to July. Separating the network creates the conditions for competition. It does not guarantee the result, which is why the retail end matters as much as the warehouse.
That is where National’s plan comes back in. The strongest argument against splitting Pak’nSave from New World is that each would need its own supply chain. Put the supply chain in neutral public hands and that argument falls away. The brands could be separated, or individual owner-operators could simply be left to compete with each other and with anyone else who wants to open a shop, all buying from the same door at the same price. Economist Cameron Bagrie told the NZ Herald after the announcement that he would go further than National and separate the wholesale and retail arms, The Spinoff reported. Newswire has argued before that the state should take the long-horizon risks that private capital alone will not. A national cold chain that serves its owners’ competitors is one of them.
Willis has promised the Commission $5 million and six months to test whether splitting Foodstuffs by brand leaves shoppers better off. Whoever forms the next government should put this option through the same test, on the same numbers, at the same time, with the purchase price and the running costs laid out in public. The major supermarkets have held more than 80 percent of the market through every reform so far, the Commission’s June report found. A third chain built from the same two co-operatives may help. A warehouse door that opens for every grocer in the country, at cost, would change who is able to compete at all. With every party now offering a grocery policy, you can see which party matches your views before 7 November.
Should the public own the supermarkets’ warehouses and trucks and run them at cost for every grocer? Tell us what you think in the comments below.
This article was written by AI, briefed to report the facts, hopefully without some of the bias people bring to the job 🙂
Leave a Reply