www.lesswrong.com/posts/r3PKfvKCjy6jok4qm/dairy-cows-make-their-misery-expensive...

The cited 2007 USDA APHIS Dairy report does not say that only 13% of cows had no outdoor access or that half of cows had pasture access.

What it says is:

• 13.1% of operations allowed lactating cows no outside access in summer.
• 50.9% of operations used pasture as the primary outside area in summer.

But when the report switches from operations to cows, the numbers are different:

• 19.0% of cows were on operations with no outside access in summer.
• 22.3% of cows were on operations where pasture was the primary outside area in summer.

So the post converts operation-level figures into cow-level figures, which materially changes the claim. It also turns “primary outside area in summer” into “had access to a pasture at least some of the time,” which is a broader statement than the report supports.

The linked 2017 Scientific Reports paper does contain the phrase “fewer than 5%,” but that was not the most recent U.S. evidence on this topic.

USDA APHIS’s Dairy 2014 report gives newer national estimates and says:

• 19.9% of lactating cows had at least some pasture access.
• 54.6% of operations used pasture as the primary outside area in summer for lactating cows.

So the post’s statement that “the most recent data said that only 5% of lactating cows … have access to pasture to graze” is incorrect. The newer USDA data reported a substantially higher share.

Separately, the same post later misstates the 2007 USDA report by turning operation-level numbers into cow-level numbers; that likely contributed to the confusion here as well.

AGW’s dairy standards do require continuous outdoor pasture access for dairy cattle, but the post overstates the exception as if removal is allowed only during “actual fires.”

The standard says:

• “Continuous outdoor pasture access is required for all dairy cattle.”
• Animals may be removed from pasture when their welfare would otherwise be negatively affected.
• AGW gives examples of acceptable reasons for removal including extreme weather, emergencies such as wildfires, and even overnight removal from pasture for predator protection.
• A written plan is required for any planned removal for any length of time, or an emergency if removal exceeds 28 days.

So the standards are not “except during actual fires”; they allow other specific welfare-based reasons for temporary exclusion from pasture.

The parenthetical “so ~once per year” misinterprets the study’s incidence rate.

The study reports 24.4 cases per 100 cow-lactations for clinical mastitis. That means an average of about 0.244 cases per cow-lactation, not about 1 case per cow per year.

Put differently, if you had 100 cow-lactations, you’d expect about 24 clinical mastitis cases total—not 100. Some cows would have no cases, some would have one, and some could have more than one, but the average is far below “once per year.”