Crop rotation is not an old esoteric practice. It's widely used in European fields with 3 or 4 types of crops, including cover crops, and actually improves yield. The USA farming practices keep it to a strict minimum, I believe to reduce labor costs(?) but at the expense of soil health.
The US system is optimized to produce the most grain and soy of consistent quality at the lowest possible price. It’s clearly not a perfect system, but it makes food staples really cheap in a lot of places where the historically were quite expensive.
It’s kind of hard to compare US and Western European agriculture apples to apples, since the US is so heavily focused on commodity crops for export. But it’s pretty clear where the corn comes from.
American agriculture is dedicated toward optimizing the productivity of agricultural workers rather than agricultural land. Historically (and currently), the United States has a large amount of relatively unused land and a comparatively tiny population to work it.
So the productivity of the land isn't a concern. The goal is to have a system where one guy can produce enough food for 400 people.
There are some implications of moving in the other direction that you might not like. In the one-farmer-feeds-400-people model, food is very cheap because it represents a trivial amount of work. In a model that devoted more resources to cultivating the same land at higher yields, one of two things would happen.
The less likely option is that food would become much more expensive because of all the additional labor. This would preserve the general standard of living of Americans, except of course that they'd all be much poorer because so much of their money would be spent on food. But they'd all be poorer together.
The more likely option is that the price of food would only rise a little, and a class of extremely poor people would provide agricultural labor. This would be a blow to American egalitarianism.
> American agriculture is dedicated toward optimizing the productivity of agricultural workers rather than agricultural land
At least with respect to corn, US farms produce 4 tons per hectare more each year than the global average, and have substantially higher yields than the 4 nearest competitors[1].
> the United States has a large amount of relatively unused land and a comparatively tiny population to work it
We've actually increased forrest cover substantially in the last 50 years, which is great.
This is a really fussy question, but I notice something odd in your link:
> The typical farm in Brazil produced corn and soybeans in 2020. Corn was a second crop following soybeans and was produced on approximately 78 percent of the typical farm’s acreage during the five-year period.
For corn to be a secondary crop following soybeans, at the same time it's produced on 78% of the farm's acreage, something has to be weird. Either soy is so much more productive and valuable than corn that its yield can be worth more while less than a third as much land is devoted to producing it (why not just grow more soy?), or a given acre might sometimes produce corn but more often produce soy over the five-year period, while being counted as "an acre used in the production of corn".
Then, the productivity of the farm is given in yield per hectare. But we've already seen that "hectares" is a funny measurement. What's the yield per hectare-month?
Also, this is fully compatible with American strategy being to use as little labor as possible:
> Labor costs as a proportion of total costs were relatively higher for the typical farms in Russia and the Ukraine.
(Annoyingly, the article's chart doesn't display the detail the article discusses. But "operating costs", the category that includes labor, are anomalously low for the US, beating everywhere except Argentina.)
On that model, high productivity per acre would be more of a happy coincidence.
Not relevant to the topic, but sad:
> Economic profit for the five-year period was positive for the typical farms in Argentina and Ukraine. [But not for Russia, Brazil, or either US farm.]
Yeah the economics of farming are pretty tricky and it is easy to end up in an apples to oranges situation. Corn is tricky as well, because ThenUS is so dominant in that crop so that surely displaces some growers.
I’m not discounting the labor angle entirely, but there seems to be some consensus that for at least some crops US farms are quite productive.
This might have been true in the early 00s, but not now: Based on satellite imagery analysis I've seen, large parts of the midwest are doing crop rotation with corn and soybeans in the main season, which more differences in the winder depending on weather.
The reason a US farmer chooses what to plant each year is the same as everywhere else: Given the state of their soil, the expected weather in their plots, legislative incentives and some guesses on crop value, they plant whatever they think will keep making the most money. The increase in soybeans as a useful rotation crop come from higher price per bushel vs 2000, and the increase demand making it easier to sell it in their local elevator. The extra $4 a bushel makes it much better than doing all kinds of extra treatments to the soil to keep trying to run more corn. As prices rise and fall, practices change to match.
And this is why modifying plants to lower the need of fertilizer can make such a big difference: A crop that before might not have been economical can become more viable. If a farmer has to risk soil health to have a somewhat profitable year, because anything else looks like losing money, they'll take the risk, in the same way that they'll cheap out in their pesticide applications. When the farmers are making good money with safer practices, they'll use them, and they'll keep investing. But for every 2012, when drought raised prices through the roof, and everyone else got their insurance payouts, there's many years where the farmer doesn't cover costs.
Always assume that the farmer is doing the best they can to make money, whether they have some organic apple orchard, or they are planting Monsanto soybeans.
