Columnist, Food


A combine drives over stalks of wheat during the harvest in Dixon, Ill. Combines can make short work of the work of harvesting. (Jim Young/Reuters)

Why are some foods cheap and other foods expensive?

Hint: It’s (mostly) not subsidies. Although they’ve certainly played a role in shaping our food supply such that we have huge quantities of just a few crops — a recipe for low prices — the discrepancy that seems to be at issue is the one between commodity crops such as corn and soy, and the fruits and vegetables that everyone’s trying to get us to eat more of. There’s a factor there that plays a much larger role than subsidies, and it doesn’t get much airtime.

It’s machines.

In general, if you can use machines instead of people, you can produce a crop for less. But let’s not talk in general. Let’s talk about tomatoes.

The beautiful, ripe, in-season tomato will set you back up to $5 a pound, but you can buy a 28-ounce can of perfectly tasty tomatoes for as little as a dollar. The latest data from the U.S. Department of Agriculture have beefsteak tomatoes at $3.16 per pound and canned tomatoes at 92 cents per pound.

A big part of that difference is machines.

The fresh tomatoes we buy are harvested by hand, which is still the only way to guarantee the blemish- and bruise-free specimens that picky American consumers demand. The canned tomatoes we buy are harvested by machine, because a few blemishes and bruises don’t much matter when the tomatoes are going to be processed.


Fresh tomatoes are harvested by hand. That’s why they can cost as much as $5 a pound. (Daniel Acker/Bloomberg News)

This wasn’t always the case. Until about 50 years ago, all tomatoes were harvested by hand because there was no such thing as a tomato harvester. (The price difference between canned and fresh wasn’t quite as stark then, either: In 1955, according to the Bureau of Labor Statistics, fresh tomatoes were 27.4 cents a pound, compared to 15.1 cents for canned.) But an agronomist and an engineer at the University of California at Davis set out to change that. The engineer, Coby Lorenzen, designed a machine to automate the harvest, while the agronomist, Jack Hanna, developed a less-delicate variety of tomato that ripened uniformly and could be easily plucked from the plant, qualities that made machine harvesting feasible.

The first few iterations of the harvester, introduced in the 1960s and refined in the decades following, reduced the labor required by 92 percent, to about 24 minutes per ton. (Tomatoes for processing also don’t require the labor and materials needed to be boxed, a step that eats up over a third of the price that farmers get for tomatoes, according to a study done at UC-Davis.)

Other things happened, too. In the first decade of machine harvest, yields increased from 20 tons to 33 tons per acre. Fewer workers had jobs, of course, but the ones who did earned more money (by 22 percent) than field workers, because operating a harvester was considered more skilled labor.

A modern tomato harvester costs nearly $500,000 and can harvest 70 tons per hour. It’s one reason your canned tomatoes cost less than a third of your fresh tomatoes.


Tomatoes destined for a can don’t need to be handled as carefully, so they can be harvested by machine. That is part of the reason that they cost far less per pound than fresh tomatoes. (James M. Thresher/For The Washington Post)

That’s just one crop, of course. What, more generally, have machines done for food prices? Sun Ling Wang, agricultural economist with the USDA, laid out the most basic principle: “On the supply side, technology can help produce more products. If farmers produce more products, other things being equal, prices will decrease.” But she — and several other economists I spoke with — emphasized the difficulty of teasing out the impact of mechanization from the large pool of factors that affect price.

Case in point is the machine that has probably had the biggest impact on American agriculture: the combine.

A combine is for row crops (grains and legumes), and it does, as its name implies, a combination of things: harvesting (removing the plant from the ground), threshing (removing the grain from the plant) and cleaning (removing the schmutz from the grain). What isn’t grain — the stalks and leaves — the combine can either chop up and leave on the field to provide cover and organic material, or bale for animal feed.

It does it all at once. At speed. Even when it’s 32 rows at a time. It’s an astonishing thing to watch. I got to drive a combine once, on an Illinois soy field, and I consider it a highlight of my ag journalism career. (To see a variety of big-farm machinery in action, try the YouTube channel of Brian Scott, a fourth-generation Indiana farmer whose family farms 2,200 acres of corn, soy, wheat and popcorn. He has a drone, and he’s not afraid to use it.)

Since tractors replaced horses, and then combines replaced tractors — in a period of about a century — the price of row crops has dropped fairly steadily. Although replacing horses was important because farmers no longer had to grow the grain to feed them, the biggest factor in the price drop, according to Prabhu Pingali, professor of applied economics and management at Cornell, was better-yielding crop varieties.

The combine had other effects, though: It led to larger farms and standardization of growing practices. “Combines give you scale economies to expand farms,” says Pingali, but he notes that these effects aren’t universal. In India, for example, mechanization hasn’t increased farm size, partly because farmers rent the equipment, which goes from farm to farm. But in the United States — and particularly in the Midwest, where the flat, uniform geography lends itself to large swaths of cropland — machines have been a major part of the dynamic that has more than doubled the size of the average farm and reduced the number of them by about two-thirds.


A combine unloads a hopper full of harvested soybeans into a tractor trailer on a farm in Ohio. (Ty Wright/Bloomberg News)

And then there’s jobs. While machines have clearly eliminated jobs in some areas, the dynamic is often the other way around: Lack of labor forces farmers to turn to machines. In the period after World War II, says Wang, workers left the fields for higher-paying manufacturing jobs. And now, in California, a labor shortage is driving wages up, and farmers are looking to machines that were prohibitively expensive when labor was cheap (like lettuce bots).

Among people who care about our food system, there’s often hostility toward the suite of changes to American agriculture that can reasonably be called “industrialization.” Machines are part of that, but so is specialization (a focus on just a few crops), dependence on pesticides and chemical fertilizer, large size, and the uniformity that enables just a few people — with the help of John Deere — to farm thousands of acres.

I understand the hostility, as the most serious problems we face (my list: fertilizer runoff, soil degradation, lack of crop variety) are wrapped up in industrialization. But industrialization brings with it some very significant benefits. The ability for fewer people to grow more food is one of them, and machines are a part of that. While people’s livelihoods still depend on farm jobs, it’s important that we safeguard those livelihoods, but field work is grueling drudgery. Post-Industrial Revolution modernization has brought its share of problems, but freeing humans from grueling drudgery is a big win. I do just enough farm labor to be reminded regularly that the world’s a better place if farm labor is done by tractors and harvesters and combines, rather than people.