Current Progressive Dairy digital edition

1309 PD: What does it cost to produce the raw milk for one gallon of retail milk

Ben Yale Published on 25 August 2009

Here is a simple question: How much did the raw milk in that gallon of milk in the dairy case cost to produce?

The process is simple enough -- convert the average price of a gallon of milk and compare to the cost of production for that month. Oh, if it was so easy! Easily asked, not so easily answered. The USDA does not compute or publish this number, but it can be closely approximated.

We need three numbers: the price of a gallon of milk, the wholesale price for raw milk and the cost to produce a gallon of milk.



What do we use for the price of a gallon of milk? Now, certainly you can go to your grocery store and note the price and compare it to your cost of production, but that simply cannot explain the national phenomena. This is a national magazine so I am going to focus on a national number. Though the magnitude may change, the relationship cannot.

Since 2000, USDA Dairy Programs have sampled store prices in selected cities within the 10 federal milk marketing areas. For the selected cities, the market administrators identify one outlet each of the largest food store chain, the second-largest food store chain and the largest dairy/convenience store chain in the city or metropolitan area. Discount box stores are not included.

The market administrators survey the same outlet each month so that price changes are within the same store and not attributable to different retail strategies. The survey is conducted on one day between the first and 10th of each month, excluding Fridays and weekends. The price obtained is the posted retail price for the most common brand, based on shelf space. Temporary price discounts, specials, coupons or other price promotions are ignored.

The USDA reports these prices on the Dairy Programs website. You can reach that by selecting Milk Marketing Order Statistics, Prices, Retail Prices. The information includes retail prices by month by selected cities for whole milk, reduced fat, organic whole milk and organic reduced fat.

The data is much more limited than the one used by the Bureau of Labor Statistics. BLS data on retail milk prices is used to provide, among other things, the consumer price index, an indication of inflation. You can go to to find this information. It is a more comprehensive survey of prices than the USDA's, but still has its limitations.


The National Dairy Promotion Board receives information from A.C. Nielsen regarding actual sales prices, but the information is not publicly available. California, however, does report that information. The website has the most recent three months posted. This information is gathered by compiling the prices as scanned nationwide. It is a much more accurate number for what consumers pay than the BLS or the USDA data. According to this A.C.Nielsen scan data, the average retail price for a gallon of milk nationwide in June 2009 was $2.72 for whole milk, $2.63 for reduced fat and $2.55 for skim milk.

These per-gallon prices can easily be converted into prices per hundredweight by simple division and multiplication. A gallon of whole milk weighs 8.6 pounds, gallon of reduced fat milk weighs 8.62 and a gallon of skim milk weighs 8.63. Divide the price per gallon by the pounds per gallon and multiply the quotient by 100. The result is a per hundredweight price for the milk at that butterfat test. The result of that math is as follows: That converts the June retail prices to $31.63 per hundredweight, $30.51 per hundredweight and $29.55 per hundredweight. That's $29.55 per hundredweight for skim milk!

Now that we have a per hundredweight price for the milk at test, how does that compare with what producers receive for milk? We have another choice to make. There are a number of series of producer prices with which to compare. The USDA reports its minimum prices for Class I, and CDFA does the same for Class 1. The USDA also each month reports the cooperative-announced prices and average prices paid. A comparison of USDA's reported retail price for whole milk in 10 cities with the minimum and cooperative prices appears as Exhibit 53 in the recently ended producer-handler hearing. (The exhibit can be found on USDA's Dairy Programs website.)

Another comparison could be made with the mailbox prices reported by USDA and CDFA. Matching cities to the mailbox prices is not easy. Producers in a marketing area do not necessarily market their milk, or at least all of it, to plants in the marketing area. But, like the USDA data on retail prices, the data is limited and does not include all of the milk or all of the markets. Another comprehensive series of milk prices is the all-milk price reported by National Agricultural Statistics Service (NASS). Reported for 23 top dairy states, this report provides the average price paid to producers for Grade A milk (called "fluid") and, in a small number of states, the price paid for "manufacturing milk". This price is provided as a national average as well. While on a regional basis, matching state milk prices with city markets is not necessarily what is happening in the market, it certainly gives a good indication of what is happening.

According to these NASS numbers, the average fluid milk price paid to producers in the U.S. for fluid milk (all-milk price) was $19.30 per hundredweight in July 2008 and $11.30 per hundredweight in July 2009 at test. On a per-gallon basis without adjustment for butterfat, the prices were $2.24 and $1.31, respectively. For better accuracy, these prices should be adjusted for butterfat value to compare to finished milk. (They would be even lower.) Rather than go into that complexity, what numbers we do have show that the retail-to-farmgate spread in June 2009 for whole milk averaged at least $20.33 per hundredweight or $2.64 per gallon. Raw milk costs were 35.7 percent of 2009 whole milk retail prices.

In all fairness, a lot goes on between your bulk tank and the consumer's grocery bag. There are processing, transporting, warehousing, distribution, stocking, advertising, accounting and retailing costs. It is not all profit to retailers. Also, the spread reported above does not consider the premium paid by processors for Grade A milk. The reported all-milk price for fluid milk includes the average prices paid for all Grade A milk, even that used to make cheese. There is in the FMMO an approximate average difference of about $1.50 per hundredweight.


But as everyone knows well, what is paid for the milk is not necessarily what it costs to produce the milk. Just like no agency matches raw milk prices with retail prices, no agency matches cost of production with either minimum milk prices, retail prices or prices received. The Farm Bill of 2008 on the service mandated that the USDA Dairy Programs determine what those costs for feed and fuel were in each of the marketing orders, but refused to do so when it lowered the milk prices last fall.

There are three series of data available to assist in determining the cost of production besides your own cost. The first are the private ones put out by the major dairy farmer accounting firms. These provided accurate information on the farms included in those studies. They, unfortunately, are not comprehensive enough. The second series includes audited studies of farmers' costs done by CDFA as part of its legislated mandate. These can be found at Although by far the best information publicly available, it is limited to just one state.

The most comprehensive study on costs is one done by the USDA's Economic Research Service (ERS). Going back decades, ERS has estimated production costs monthly for the major dairy states. Each year ERS estimates costs of production. These annual estimates are adjusted based on price and production differences between the month and the baseline. The data can be found, generally retrievable in Excel workbooks, at

Since for this simplified analysis we are looking at a national average retail price, we should compare it to a national average cost of production. According to the USDA's Economic Research Service, a simple average of 23 states' average cost of production for June 2, 2009 was $27.13. (For those thinking gallons, it was $3.15 without adjustment for butterfat). This is for "total costs." The other number is "operating costs." The difference is that the first includes depreciation, taxes, overhead and hired labor, which the second does not. From a business standpoint, or even analysis, ignoring those real costs makes no sense.

The accuracy of these numbers is often challenged, though no one comes forth with better numbers. The problem is that they do not represent the actual costs on a weighted basis in the state but the cost of an "average" dairy farmer. That makes them higher. For example, ERS estimates California's total operating costs to be about $20.00 for the first quarter compared to CDFA's $18.50. The range in reported costs is wide. June 2009 ERS estimate for COP is $13.75 for New Mexico with 670 million pounds and $41.82 for Tennessee, whose pounds are even reported, though they are so small. Again the accuracy is a question that needs to be answered.

A simple average of the states (New Mexico and Tennessee get the same weight) will not work. To get an approximate weighted average cost of production, additional data is needed. NASS also reports monthly on the production of 23 states. While there are 23 states with reported costs of production and 23 states with monthly production reports, they are not the same 23 states. Using the common 19 states in 2009, the weighted averages cost of production are reported at $24.20 per cwt or $2.81 per gallon. Now I am not confident that number really shows what milk costs are. A study by ERS several years ago said that larger farms’ (1,000+ cows) cost of production was 15 percent less than the average of all farms. That brings the number down to $21 on total costs with current prices. That is still unrealistically high.

I will be the first to admit I find the numbers questionable. When you work with numbers you have to use the numbers that are given. It is dangerous to play with them and make adjustments without real facts behind it. In the midst of this great dairy policy debate, getting accurate information would be of the first priority. If the data is wrong, then give us better data. The simple answer required a complex solution and got only an approximate, qualified answer. What this shows us is that the data to answer that question is not what it should be. This explains in part why there appears such a disconnection between policy maker actions and what we feel on the farm -- they really do not know. But they should. This information should be forthright and published monthly by the USDA so everyone, producers, consumers, politicians and agency employees, fully understand the disconnection between food prices and production costs.

So the answer is (subject to all kinds of qualifications): The average retail price reported by A.C. Nielsen for June 2009 was $2.72 for a gallon of whole milk. The weighted-average ERS cost of production estimates it cost $2.81 per gallon to produce. According to NASS, nationwide for June, producers received an average of $1.31 per gallon for milk. Those certainly are not exact numbers but at least tell us that the problem measured in real stress at the farm can be seen in approximate estimates. PD

Ben Yale
Attorney at Yale Law Office