Chobani vs. Supply Management: Happy Marriage, Ugly Divorce or a New Chapter?

I love yogurt. Soon after I immigrated to Canada from Serbia in 2001, I realized that Canadian yogurt, feta and cream cheeses  are not to my taste, so I started discovering little ethnic speciality stores with dairy products from Macedonia, Greece, Turkey, Italy, Serbia, Croatia and other places with a strong dairy tradition.

This is partly why I can understand the success of Chobani, a company founded by Turkish immigrants in 2007 that has since then captured over 18% of the U.S. yogurt market. Canadian consumers might have the opportunity to try Chobani’s products soon, and many new Canadians like myself, used to fine cheese and yogurt, might discover a cheaper alternative to the very limited supply of expensive imports.

This is only if Chobani wins its legal battle against Canadian milk processors, lead by Danone and Yoplait, who are invoking Canadian supply management policy and claiming that Chobani’s presence in the Canadian market and its subsequent plan to build a 940,000 square foot milk processing facility in Ontario “will destroy the delicate balance created by the Canadian supply-management system.” In the meantime, the company has been granted a temporary, one year reduction in import tariffs on its U.S. produced yogurt from over 200% to 5% by the Canadian federal government. Currently, this yogurt can be bought in the Toronto and Hamilton area.

Canadian supply management policy focuses primarily on limiting the supply of raw milk at the farm level by using a system of producer quotas and import restrictions on final products. While Dairy Farmers of Ontario (DFO), the province’s supply management marketing board, do regulate the distribution of milk to processing plants based on the end use, the board does not directly limit the domestic output of specific final products. Thus, it is not clear how the policy applies in Chobani’s case. This aside, it is still up in the air whether Chobani will win this legal battle, and I will not analyze the legal issues here. Rather, I will show how the Canadian milk quota system works, which will make it easier to see why certain interest groups favour this system, and why some groups oppose the potential entry of Chobani into the Canadian market. This will also be useful for assessing the potential consequences of Chobani’s plan to enter the Canadian milk processing industry.

Why Many Current Farmers Favour the Quota System

The quota system, introduced in the early 1970’s, sets legal limits on the national, provincial and farm-level milk supply below the quantity that would have been supplied in an open market. In 2011, the quota system limited the total quantity of raw milk produced and sold in Canada to 77 million hectolitres per year, which is, for the sake of comparison, identical to the yearly supply in 1958.[1] To be technically correct, the dairy quota is expressed in kilograms of butterfat (milk fat). As the fat content is a fairly stable property, milk volume is often used as a shorthand quota measure.

Each province has a certain share of the national butterfat quota. This share is determined by the national supply management agency, the Canadian Milk Supply Management Committee (CMSMC). Each producer cannot exceed his or her share of the provincial quota. Milk is bought from farmers by the provincial milk supply management marketing boards, which then sell it to the processors.  The average price received by milk farmers last year was about $75 per hectolitre.

No one really knows what the quantity of milk supplied on the market would be if each farmer was allowed to produce as much milk as he or she wanted, but we could base our guess on the average per person raw milk production in the early days of supply management (i.e., 1975), which was 3.1 hectolitres of raw milk per person per year.[2][3][4]Using this number, we arrive at the estimated unrestricted production of 105 million hectolitres in 2011.

Similarly, no one really knows what the price of raw milk would be under these conditions, but a good guess would be that it would not be much different from the current U.S. price, like in the early years of supply management (i.e., in 1975), when the Canadian and the U.S. raw milk prices were both at about $15 per hectolitre. Last year’s U.S. raw milk price was around $45 per hectolitre at its peak in July and August.

Figure 1 shows this on a diagram. At the price of $75 per hectolitre, processors buy the current quantity of 77 million hectolitres. As the quantity supplied increases to 105 million hectolitres, the price processors pay falls to $45 per hectolitre.

Figure 1. The effect of the quota system on the raw milk price received by dairy farmers

From this, it is still not obvious why many dairy farmers would prefer a total supply of only 77 million hectolitres as opposed to 105 million.[5] However, if we compare the total revenues received by all producers under the two scenarios, this becomes a bit clearer. In the case when the producers supply only 77 million hectolitres, the total revenue they receive is 77 million hectolitres times $75 per hectolitre, which is $5.8 billion. In the alternative, higher supply and lower price scenario, the total revenue is $4.7 billion (105 million hectolitres times $45 per hectolitre). The total revenue received by primary milk producers is 22% higher under the restricted supply scenario. This illustrates the incentives behind the industry support for the quota system.

Note that this simple calculation takes into account only the price and quantity effects of the quota system. It does not, however, take into account reduced competition, both domestic and from abroad. In other words, here I assumed that all of the current Canadian raw milk producers are in fact able to produce milk at $45 per hectolitre. Recent farm financial data indicate that only a fraction of Ontario farmers would be able to maintain a profitable business under this price.

This calculation also suggests that, as a consequence of the quota system, milk processors are paying higher prices for raw milk than they would have otherwise been paying. In total, they pay 22% more for raw milk compared to the scenario with fewer restrictions on milk supply. At first, one would think that the processors would be strongly opposed to this system but a closer look proves otherwise.

Why Milk Processors Don’t Oppose the Quota System

We can use a similar approach as before to show the effects of the reduced raw milk supply in the final products market. There are many dairy products, but for illustrative purposes, let us look at 3.25% fluid milk. Last year, the total quantity of 3.25 % fluid milk sold in Canada was 364 million litres, and the average price was $1.5 per litre.

Note that these price and quantity correspond to the situation when 77 million hectolitres of raw milk are produced. In the case when 105 million hectolitres of raw milk are produced, we would expect about 499 million litres of 3.25% milk to be produced and sold.[6] As before, no one really knows at what price this milk would be sold, but, following the raw milk price analogy, a good guess would be that the Canadian price in these conditions would be close to last year’s U.S. price. The average price of 3.25% milk in the U.S. in 2011 was approximately $0.9 per litre.

The two scenarios are shown in Figure 2. The demand curve shows the consumers’ willingness to pay for milk at different market quantities supplied. As the total quantity supplied increases, the price that would clear the market falls.

Figure 2. The effect of restrictions in the raw milk market on the price of a final dairy product

Comparing the total processors’ revenues at the restricted milk supply scenario to the revenues at the milk supply unrestricted by the quota system reveals why processors would not oppose the quota system. At the relatively low supply of 364 million litres, the total processors’ revenue from 3.25% fluid milk is $546 million (364 million litres times $1.5 per litre). This is considerably higher than $449 million revenue (499 million litres times $0.9 per litre) under the unrestricted quantity of 499 million litres processed and sold.[7] This translates into an increase in the total revenue just under 22% compared to the unrestricted supply scenario.

A similar pattern of price differences between the U.S. and Canada can be observed in the markets for other final dairy products. For example, the average U.S. price of 500 g of yogurt in March of 2012 was $1.05 compared to the 2011 average of $2.35 in Canada. Similarly, in the U.S., 2 kg of ice cream cost, on average, $4.06 in March of this year compared to the 2011 average of $5.38 in Canada. In the case of hard cheese, the U.S. consumers paid, on average, $6.97 for 1 kg, while their Canadian neighbours paid about $13.80 in 2011. Thus, the increase in the processors’ outlays on raw milk tends to be offset by the increase in the total revenue received from consumers in the final products market. In other words, the processors simply transfer the high raw milk price onto the consumer.

Why Expanding and Potential Future Farmers Don’t Favour the Quota System

So far, we have seen why many farmers, especially those unable to produce at lower milk prices, would favour the quota system and why the processors would not be strongly opposed to the system. However, what about those dairy farmers that want to grow their business or those entrepreneurs who want to enter the industry?

First, due to the limitation on the total output, one producer cannot expand if another does not contract his or her output. The current quota holders are asking between $25,000 and $36,000 in order to reduce their output by one cow. In response to the rising quota prices, the governments of Ontario and Quebec have put price ceilings of $25,000 per kg of butterfat per day (approximately equivalent to the milk output per one cow) in 2009. Prior to the ceilings, quota prices in these provinces were in the range of $28,000 to $30,000.

The quota purchase expenditure for new and expanding operations is in addition to all regular capital investments like land, buildings and machinery. This means that if you want to start a 50-cow dairy farm, you need to cash out $1.5 million for a permit to produce and sell milk in Canada. Moreover, for the expanding operations, due to the low supply of quota, especially after the imposition of quota price ceilings, it is quite hard to acquire quota in larger chunks. This creates an impediment for these farmers to grow their business to a more desired size.

Although the required quota expenditures represent barriers for expansion or entering the industry, some people are still borrowing money to buy quota permits at these prices. This demonstrates their expectations of future profits that would be used to make this extra expenditure worthwhile in the long run. Moreover, this is another indication that the industry output would increase as a consequence of new entrants seeking to benefit from extra profits if the quota system was abolished.

The Unsuspecting Consumers

At this point, it becomes clear that the consumers of dairy products are the ones paying the price of the quota system. One defence often put forth for this arrangement is that the consumer expenditure on dairy products as a share of income is small, and thus the consumers don’t really feel the pinch of the quota system.

For the sake of rough assessment, let us assume that an average four-member family consumes dairy products that correspond to four times Canada’s 2.2 hectolitres per year per capita milk production.[8] Let us, as a common unit of accounting across all dairy products, use the price of 3.25% fluid milk. At the fluid milk prices of $1.5 per litre, an average family spends $1320 per year on dairy products. In contrast, if this average family was paying $0.9 per litre of milk, the expenditure would be $792 per year.[9] This is a $528 difference. This may be a small share of one’s total income or even total food expenditure, but it is still a sum of money for which many would have no trouble finding a useful purpose.

What Happens if Chobani Enters the Canadian Market

In a sense, Chobani has already entered the Canadian market by getting the temporary import easement. Note that the company is using U.S. raw milk to produce the yogurt sold in Canada. Thus, at the moment, the company is selling its product in a high-price market while buying its milk at the price of $45 per hectolitre, as opposed to the Canadian price of $75 per hectolitre. It is not clear whether this permit will be extended beyond one year if Chobani’s legal case takes longer than a year. In case there is no set expiration date, this arrangement provides incentives for Chobani to maintain the status quo, while squeezing out the existing processors by using the low raw milk price advantage. Presumably, the benefits of establishing itself as a processor in Canada would outweigh the incentives for prolonging the temporary import easement.

In case that Chobani is eventually allowed to build the proposed plant, the above analysis has given us a good basis for assessing the potential consequences of a new milk processor in Canada. First, this processor would need to acquire milk from the Dairy Farmers of Ontario. The agency could supply this milk either (1) by reallocating some of the current raw milk supply away from the existing processors, (2) by attempting to get approval for increasing the provincial raw milk quota from the Canadian Milk Supply Management Committee, (3) by increasing the actual milk output without increasing the provincial butterfat output or (4) by using a combination of the previous three options.

In the first case, it is relatively easy to see why the existing processors would object losing some of their share in the current milk supply.

Alternatively, as Chobani’s yogurt is a non-fat product, the DFO could potentially meet their demand without increasing the provincial butterfat quota by encouraging farmers to produce milk with a lower fat content. This would potentially result in a higher total volume of milk produced. However, as farmers are paid for their milk on the basis of its fat content, this option would involve DFO’s changing the raw milk pricing structure.

The interprovincial allocation of the national quota is decided by representatives of the provincial marketing boards. Although all provinces have gradually increased their butterfat output over the years, the provincial shares of the national quota have remained almost unchanged in the last 20 years. Thus, expecting a unilateral increase in Ontario’s quota may be unrealistic. In that case, Chobani’s entry into the market would add one more processor bidding for the same quantity of raw milk.

Regardless of how the DFO decides to manage the milk supply to meet Chobani’s demand, the fact that Chobani is attempting to enter the Canadian processing sector suggests that it is ready to pay $75 or more per hectolitre of milk. This would put an upward pressure on the raw milk prices, especially in the conditions of constrained supply. The ability of the existing processors to shift any raw milk price increase to the consumers  is limited by the consumers’ demand for milk. Figure 2 shows that if the processors wanted to increase the price of final products, they would need to reduce the quantity they sell to the consumers. However, the new industry entrant would generally offer its products below the current prices to build up a consumer base. This means that the existing processors could increase only the prices of products not supplied by the new entrant, which would result in a lower quantity of these products consumed. This would, in the long run, provide additional incentives for the new entrant to expand the range of its products.

It is thus not surprising that the dairy processors are the most energetic opponents of Chobani’s import easement and of its plan to open a milk processing plant in Ontario. It is the processors that would feel the immediate impact of this project through increased competition for raw milk and increased competition for the consumers’ money. But, does this mean that some of the identified potential changes resulting from Chobani’s entry into the Canadian processing sector “will destroy the delicate balance created by the Canadian supply-management system,” as some of the current processors claim? If this was indeed true, then there would be good reasons to be worried about this system even if Chobani never existed. A system in which one new firm would disturb its balance to the point that its overall functioning is jeopardized, is indeed a system worth questioning.

Although supply management can be and has been questioned on multiple grounds, its alleged internal fragility is not one of them. Regardless of some of its inherent rigidity, the above presented structure of the Canadian supply management system seems to be flexible enough to accommodate one more milk processor. Provided there is a will within the DFO to allow for some adjustments, there are multiple avenues for these adjustments within the system to take place: through gradual increases in the provincial quota, through adjustments in allocating raw milk to processors, and through adjustments in the composition of raw milk. Finally, adjustments in the prices of final dairy products would provide some of the guiding signals for the production adjustments. Ultimately, more competition in the processing sector would make the restrictions on the supply of raw milk more apparent through the upward pressure on the price of raw milk. This would provide incentives for relaxing these restrictions in the long run.

When it comes to the effect of Chobani’s presence on the consumers, they would feel the benefits of new product lines and a downward pressure on the prices of yogurt resulting from increased competition in the processing sector. However, at least in the short-run, there may be increases in the prices of other dairy products due to a reduced quantity of raw milk available to other processors.

Thus, the claim that Chobani’s processing plant would change the balance in the supply management system has some truth in it. However, the alleged detrimental effects of this change are much less obvious. In fact, if we are to be honest about economic change, any change, good or bad, starts with a “destruction” of balance in the pre-existing system. After all, didn’t Jude Delisle, the founder of Danone Canada, destroy the balance in the pre-existing dairy industry structure and consumers’ habits with the construction of the Delisle plant in Boucherville, Quebec, in 1968? In the company’s own words: “His first buyers were doctors who prescribed yogurt as a medication and Montrealers of European descent, who were soon enthusiastically spreading the word.”

This sounds much like Chobani’s current experience in the words of Hamdi Ulukaya, the company’s founder: “In many cases it’s not easy for a start-up company to have a chance in this major competition world. You can have a healthy, good-for-you product, but it has to taste good. So it has been key for Chobani that is it fat free, but tastes creamy, and people spread the word about this great find.”
Oddly enough, I haven’t tried Chobani’s yogurt yet, but this story has surely spurred my desire to do so soon.


[1] All data and calculations used in this article are available from the author.
[2] The per person milk production in Canada declined from 3.1 hectolitres in 1975 to 2.2 hectolitres in 2011. In contrast, the per person milk production in the United States increased from about 2.4 to 2.8 hectolitres during the same period.
[3] Using the per person production in the years immediately prior to supply management (i.e., about 3.8 hectolitres) would yield even higher unrestricted estimates. However, one could argue that some of the subsequent decline in the average per person production numbers could be attributed to the downward trend that was present even before supply management. This is why I use a more conservative estimate of average per person production.

Figure 3. Actual and projected raw milk output in Canada, 1920 – 2011

[4] A similar estimate is obtained if we take a linear projection of milk production between 1920 and the first year of supply management, 1972. Figure 3 shows this projection, along with the actual pre- and post-supply management production volumes.
[5] Note that each producer may individually have incentives to expand his or her production under the price of $75 per hectolitre, but if all producers expanded their production, the price would no longer be $75 but lower than that. This is why the system is held together by legal restrictions on output.
[6] Assuming that the proportion of raw milk processed into 3.25% fluid milk does not change.
[7] In reality, we should  subtract the retailers’ margin from the two totals to get the processors’ revenue, but since milk prices in both scenarios include the retailers’ margin, for the purpose of comparison, this subtraction is not necessary.
[8] Canada’s net imports of dairy products, in dollar terms, were 7.3% of the total raw milk revenue in 2011. Import tariffs of 200% to 300% are applied to these products and thus they are sold at comparable or higher prices than the domestic products.
[9] A more detailed analysis would involve comparing the Canadian price to the U.S. price of each dairy product, while taking into account the respective quantities of each product consumed in each country. However, since the final product price has to, on the margin, justify producing the specific product as opposed to fluid 3.25% milk, the other final product price differences between the two countries follow a similar pattern to the 3.25% milk price differences.

 [Originally published at LvMIC]

Leave a Reply

Your email address will not be published. Required fields are marked *