sign up log in
Want to go ad-free? Find out how, here.

Keith Woodford reveals a rather remarkable transition underway where A2 beta-casein milk from 2024-born calves will exceed 85%. He calls for a full-industry review of how we can benefit from the unique advantage that gives our dairy sector

Rural News / opinion
Keith Woodford reveals a rather remarkable transition underway where A2 beta-casein milk from 2024-born calves will exceed 85%. He calls for a full-industry review of how we can benefit from the unique advantage that gives our dairy sector
The national A2 milk herd is rapidly becoming dominant
The national A2 milk herd is rapidly becoming dominant

In recent years, the level of A2 beta-casein in New Zealand milk has been increasing rapidly and the level of A1 beta-casein has been correspondingly decreasing.

Unpublished data from New Zealand’s leading herd-improvement cooperative, LIC, which was supplied to me on request, indicate that once the 2023-born dairy calves join the milking herd in 2025, these two-year-olds will produce beta-casein that is approximately 84% A2. In contrast, only 16% of their beta-casein will be A1.  

This is a remarkable difference from the early to mid-1990s, when the proportion of A2 beta-casein in New Zealand milk was reported in various medical papers, using data from the New Zealand Dairy Research Institute, now part of Fonterra, as approximately 50%.

The 2024-born calves will produce milk that will almost certainly be even higher in the A2 variant of the relevant gene, probably about 86% and with A1 down to about 14%.

This high A2 beta-casein and low A1 beta-casein status that is now coming through into the next generation of New Zealand milking cows is unique in major dairy-exporting countries. However, the native cattle of Asia and Africa are all A2 except where the indigenous cattle have been crossed with European-type cattle.

The huge increase in New Zealand A2 beta-casein has occurred in the absence of any formal industry policy, partly by purposeful breeding by farmers and partly by breeding-company policy.  

These breeding-policy decisions were linked to Fonterra’s statement in February 2018 that “Fonterra Co-operative Group Limited (Fonterra) and The a2 Milk Company (a2MC) have today entered into a comprehensive strategic relationship that links Fonterra’s global milk pool and supply chain, manufacturing capability and in-market sales and distribution capacity with a2MC’s brand strength and capabilities”. 

Here is the link to Fonterra’s full statement  at that time. It seemed clear that a2Milk and Fonterra were going to take on the world together.

Six years later it, has not worked out that way, although Fonterra does sell A2 milk here in New Zealand under the a2Milk brand.  So, why does New Zealand dairy still need a strategic A2 policy?

An important insight comes from the announcement on 15 April this year by Seoul Dairy, which is South Korea’s market-dominant dairy company, that they plan to be 100% A2 by 2030. Their reasoning is that 60% of South Koreans are supposedly lactose-intolerant, but the real intolerance of many of these people is now known to be to A1 beta-casein.  

The other Korean companies will now have to give much thought as to whether they need to follow the market leader, which already has over 50% market share.

Of course, Korea is just one country. But Japan also has an A2 Milk Association, to whom I have given multiple presentations. Similarly, I have given A2 presentations to Indonesian audiences. Vinamilk, which is the biggest dairy company in Vietnam, also produces A2 milk. And so on.

Across much of Asia, the benefits of A2 become immediately apparent to many of the consumers who try it. This is because with native Asian cattle all being A2, and milk consumption traditionally low, there has been no natural selection in Asian human populations toward A1 tolerance.

A considerable amount of the infant formula consumed in Asian countries is now A2, with this infant formula providing a spearhead into the broader dairy market. Human breast milk is also of the A2 type.

A 2015 scientific paper that I co-authored in the journal Nutrients on the issue of A1 intolerance being misdiagnosed as lactose intolerance, together with the interaction between intolerances to A1 and lactose, has been cited 149 times in papers by other scientists. If we were rewriting the paper now, we would have further important evidence to strongly support what we then wrote.

The question now for New Zealand is how it should respond to this dairy disruption while it is still at a relatively early stage. New Zealand can still be in the right side if the disruption, but only if it completes the conversion to A2.

New Zealand could now get to 100% A2 within about eight years, but only with an explicit policy.

Readers may well say, so what!  Why would New Zealand want to do that given the recent melt down at Synlait?

In response, the evidence is clear that The a2 Milk Company (a2Milk) continues to do very well from A2 milk supplied by Synlait. It is now back within the biggest 10 companies of any type on the NZX. However, neither Synlait nor its A2 farmers have made ongoing fortunes from it.  

There is a lesson there. It is all about who makes profits along the value chain.

The Synlait partnership with a2Milk went well for close on ten years, but now it is in big trouble. Problems have been brewing for several years but it is in the last year that the ‘problem pot’ has boiled over.

It should always have been evident to anyone who did a detailed analysis, that a2Milk had managed to structure their agreement with Synlait so that a2Milk came first, Synlait came second, and the A2 farmers came third.  Right from the start it seemed obvious to me that farmers were getting the rough end of the agreement, but that reflected the reality of the power relationships.

Nevertheless, for quite some years everybody seemed happy.  It took tough times to show that  both Synlait and its farmers were getting the rough end of the deal.  

Right now, it is Synlait’s farmers that produce the milk, it is Synlait that processes the milk, and it is Synlait that holds the highly valuable factory accreditation to supply Chinese-label a2 Platinum infant formula into China.  Yet it is a2Milk as the marketer that makes all the money from this as they pass it on to Chinese distributors.

However, the current failure of Synlait is not only related to the nature of their agreement with a2Milk. Many things have gone wrong.

More importantly, the problems of Synlait have nothing at all to do with the overall importance of A2 beta-casein as a global dairy disruptor. Quite simply, Synlait lost the plot as they got carried away with debt-funded expansion.

Part of the agreements that a2Milk managed to stitch up with both Synlait and Fonterra were that these companies would supply A2 milk only to The a2 Milk Company.  I introduced CEOs from two European dairy companies to Synlait and Fonterra, with both seeking long-term agreements to purchase A2 milk powder, but in both cases the New Zealand companies were tied to their agreement with A2M.

In the case of Fonterra, we were told that we would have to get three-way agreement that included a2Milk and so we headed across to Australia to see if a three-way agreement might be possible. But there was no chance of getting that agreement, given that a2Milk saw their existing agreement with Fonterra as effectively taking their biggest potential competitor out of the market.

To understand the opportunity that New Zealand now faces, the first point is to recognise that a2Milk has trademarks but no blocking patents. Accordingly, there must be renewal and exit clauses within the current licence that lead to an exit strategy with due notice. It would be remarkable if that were not the case. 

The second key point is that some of the A2 beta-casein in New Zealand milk comes from cows that carry one copy of the A2 gene variant and one copy of the A1 gene variant. These cows produce both A1 and A2 beta casein in equal quantities.  To produce pure A2 milk, all cows have to carry double copies of the A2 variant.

About 70 percent of the 2023-born calves carry double copies of the A2 gene variant. These percentage proportions vary between breeds, with Jerseys having the highest proportion of 2023-born pure A2 calves (86%), then KiwiCross (72%), then Friesians (59%). 

Without getting into too much detail here, the big message is that farmers of all breeds could convert  to pure A2 herds within one cow generation, but only by explicit breeding policies.

These breeding policies include only using bulls that are A2A2 and genetic testing of all calves. Using sex-selected A2 semen over 15-month heifers for the first round of mating can be a big help in pushing the programme along, particularly for Friesian herds which are lagging behind.

Inevitably, there will be a transition and there will continue to be some laggards. Hence, separating A1 from A2 milk at the factory gate will be necessary for quite some years, just as companies like Synlait and Mataura Valley Milk already have to do. During the transition, the amount of pure A2 milk would increase each year.

What is needed now is for a mature debate as to how the industry should develop an A2 future. That discussion needs to span the continuum of breeding, logistics, processing, health issues and marketing.  Marketing is the big one.

There is a lot more that needs to be said and debated.  I intend to write more about these issues.

*Keith Woodford was Professor of Farm Management and Agribusiness at Lincoln University for 15 years through to 2015. He is now Principal Consultant at AgriFood Systems Ltd. You can contact him directly here.

We welcome your comments below. If you are not already registered, please register to comment.

Remember we welcome robust, respectful and insightful debate. We don't welcome abusive or defamatory comments and will de-register those repeatedly making such comments. Our current comment policy is here.


Well put Keith

It is disappointing that the  largest player hasnt driven this change over the last 10 years

It is/was a relatively easy and cheap win that has almost slipped away

For a long time Fonterra seem to have the same very casual relationship with organic milk 


That discussion needs to span the continuum of breeding, logistics, processing, health issues and marketing.  Marketing is the big one.

Yes it is. Because to realize the full value and potential of A2 products, the appropriate functional and emotional benefits need to be established properly. It's not just pretty packaging and clever campaigns. Given the obvious benefits of A2 itself for a sizable consumer market, the stars should be aligned. But the work has to be done.  


For a start they need to make a product people like and want. Tastes like curdling cottage cheese.


All starts with the product. Keith mentions VinaMilk. Their unsweetened UHT milk is wonderful - a creamy, refreshing taste. The competitor brands with any market penetration can't match them IMO. So they maintain their strong market share. 


Which product are you referring to? There are a few companies mentioned in the article, and several small operators making A2 products in NZ. 


"About 70 percent of the 2023-born calves carry double copies of the A2 gene variant."

This is the only number that is really relevant.

So 30% need to be brought in to the A2 herd.
Why bugger around at an individual farm level, separate the A1 and single copy A2 into separate herds at a national level and process and market the 70% A2 milk immediately.
Within 5 years we can be 100% A2 without any great fuss by replacing the culled A1 with A2, and have a real head start on competitors. Compensate those with predominantly A1 to cover their losses out of the national benefit.

Everyone operating at an individual level brings everyone down, the same as overproducing milk powders.

Time for some common sense in the dairy industry.


Keith, I know an increasing number of women finding that they are turning to non homogenised milk as they feel better on it.  I know my farm raised grandchildren with allergies, are less affected by their allergies if they have non-homoginised milk when they stay with us. We live in Central Otago with vineyards, olive groves, orchards etc on our boundary - so not a great area for those with allergies.  What would make the difference between homogenised v non homogenised milk?  My grandchildren did not like the taste of A2 milk.


Homogenisation is a mechanical process, with no chemicals involved. It breaks up the fat globules by forcing them through a very narrow gap and causes them to stay dispersed in the milk. I cannot see any reason as to why this should impact on allergies. It may impact on digestion but I have no evidence. If A2 milk tastes different then it is likely to be caused by unrelated issues. For some years the A2 milk in NZ was of dreadful quality due to logistical issues in the supply chain. It did great damage here to the A2 cause among A2 consumers, but was nothing to do with the A1 versus A2 beta-casein.


Surely A2mc only has trademark for stylised A2. As A2 is a descriptor of the actual gene how can they control all uses of that descriptor? 


a2Milk has a trademark for the lower case 'a2' registered in most jurisdictions. Other companies cannot use it even in non-stylised form.


What difference will Fonterra's announcement today make, Keith?


I am still mulling that over.