The detection of highly pathogenic avian influenza (HPAI) in dairy cows has dominated most of my conversations in late March and early April. We now know the outbreak started in February in the Texas Panhandle, and on April 15, USDA had confirmed that 26 farms in eight states (Texas, New Mexico, Oklahoma, Idaho, South Dakota, Michigan, and North Carolina) have had milking cows test positive. We continue to hear that 10% to 15% of the cows show significant symptoms, with the animals primarily being mid- and late lactation. Milk production for the infected farms is down 15% to 25%, which suggests production is off for the cows not showing symptoms as well. Most of the sick cows recover in 10 to 21 days, although about 10% of the sick cows dry off early. We also hear that the true number of farms infected is higher than the official count, either due to the lag between infection and the USDA confirming it or due to farms deciding to handle the outbreak quietly.
The hard part of analyzing the potential impact of the disease on milk production is trying to make a reasonable assumption about how many farms will eventually be infected. We know some of the spread between farms has been due to moving cows from farm to farm, but it isn’t clear yet how much spread between the southern states and northern states is due to bird migration. In 2015, there was a significant outbreak of HPAI in commercial poultry flocks in the Upper Midwest, and most of the spread happened in April and May as birds migrated north. So, if we are going to see a rapid rise in cases, it should be happening right now.
In situations like this, it is useful to look at different scenarios since it is very hard to say how many farms will become infected. I am assuming that 15% of the milking cows on an infected farm will show significant symptoms and none of their milk makes it to the bulk tank for 14 days. I assume 10% of the sick cows will dry off and we lose four months of production on the tail end of their lactation. I also assume that production for the remainder of the herd will be down by 5% for 14 days as well. So, the short-term impact is a 20% drop in milk production for an infected farm, with production remaining down about 1.5% for the following four months due to the lost production from the cows that dried off early. That is a significant short-term hit to production, but on an annual basis, the farm only loses about 1% of milk production.
The national picture
We have roughly 26,500 licensed dairy farms in the U.S. If 5% of the farms were infected, that would be 1,325 farms. With their production down about 1% for the year, it would reduce total U.S. milk production for the year by 0.04%. If 30% of the farms become infected, it would reduce U.S. milk production by 0.23% for the full year. The short-term impact on U.S. milk production would be bigger, but it would depend on how quickly the virus spreads and how many farms were infected in any particular month.
On an annualized basis, the impact of HPAI on U.S. milk production looks minimal. I can envision the potential for regional disruptions or significant but short-lived impacts. There are only 300 dairy farms in Texas, but there are 5,900 in Wisconsin. If the virus is being spread north by birds, we have the potential for a rapid rise in infected farms that could dent Upper Midwest production for a month or two, which would be bullish for the dairy markets (especially cheese). So far, USDA hasn’t confirmed any cases in Wisconsin, and we aren’t hearing of any. Even if half the dairy farms in Wisconsin were infected, that would only reduce annual U.S. milk production by about a tenth of a percent, which is a pretty small impact on an annual basis.
There is also some bearish risk around this. We could see domestic consumers reduce dairy purchases due to the negative headlines around bird flu in the dairy herd, or importing countries could block U.S. dairy products. The virus is killed by pasteurization, so there is very little actual risk for consumers and importers. As of mid-April, we haven’t heard of a significant impact at retail, and none of our trading partners have blocked U.S. product. I think the price risks lean bullish, and we could get a pop in the market if a lot of farms become infected quickly, but on an annual basis, the impact might be minimal.
