The Farmers' Dilemma and Agriculture’s Plague of Second Best Solutions

The prisoner’s dilemma, benevolent dictators, and why we grow so much corn

I learned a lot of good lessons from my high school economics teacher– some of which he even meant to teach. He was a sucker for classical economic theory. We talked a lot about comparative advantage– about focusing on what you're best at and trading for what you need. But one thing he harped on was the theory that the most efficient way to run an economy was actually by way of a benevolent dictator. I think he must have been referring to William Easterly’s theory about the ability of an autocrat to make quick and decisive decisions without the time-consuming and indecisive democratic process.

In this theoretical framing, democracy is actually the “second best solution” and the benevolent dictator is the “first best solution.” Of course, in the non-theoretical (AKA real) world, there is no such thing as a benevolent dictator. Dictatorship as a concept doesn’t often attract benevolent people, and unchecked power often leads even the most pure-hearted into, well, not benevolence.

This was the context in which I first learned about the idea of “first best” and “second best” solutions, and I think about these concepts a lot when it comes to agriculture. Without them, I think it can be challenging to understand why things in American ag are the way they are, and why they seemingly cannot be improved.

The preponderance of “second best solutions” explains a lot.

____

I don’t know if the guy from A Beautiful Mind invented these concepts, but that’s who our teacher credited. Maybe it was John Nash, or maybe John von Neumann, but either way, game theory has a lot to do with the ranking of “best solutions.”

The most famous game in game theory is the prisoner’s dilemma. The outlines are simple (and probably familiar): The police have captured two suspected criminals. They are being kept apart, in separate, isolated interrogation rooms. The cops extend an offer to both suspects. “If you finger the other guy, he’ll get 5 years, and you’ll go free. Or you can take your chance that he fingers you and you stay quiet, then you’ll get 5 and he’ll go free. If you both accuse one another, you both get 10. If you both keep quiet, we’ll bring you both up on a lesser charge, and you’ll both go to prison for a year.”

The cops here did a pretty decent job of summing up all the possible solutions to the game: 1 year each (2 total), 0 and 5, 5 and 0 (5 years total either way), 10 years each (20 total). On first blush I think, it’s pretty easy to rank the options from best to worst– assuming that neither me or the other guy want to go to prison. First best solution: we both stay quiet (1 year each). Second best solution: one of us talks, the other stays quiet (someone does 5 years). Third best (worst) solution: we both talk (both do 10 years).

But this is where it gets interesting, because of course, I as an individual player am not going to be thinking about what’s best “overall” and how to minimize total years in prison. I only have real control over my own decision, so I must optimize the outcomes for myself. And from my individual perspective, there are not four options, there are only two, I can either stay quiet or point the finger. Staying quiet will lead to either 1 year in prison or 5, and pointing the finger will lead to either no time, or 10 years. Of the four options– no time is the best option, so my personal first best solution is to point the finger, and hope the other guy stays quiet. So I will select the second-best solution.

Of course, since this logic holds true for both of us, it’s seemingly inevitable that our collective decision-making lands us not with the first or second best solution overall, but with the third best (worst) solution.

This is one lesson of the prisoner’s dilemma, that individual motivations often get in the way of taking actions and making decisions that would, in the end, be best for us all. In other words, we (often) choose second best solutions, for a myriad of reasons, and these second best solutions lead to second best outcomes, or worse, and we all suffer.

There’s an obvious solution though, at least to the prisoner’s dilemma as I’ve described it here. If me and the other guy are part of a criminal organization, the most likely reason we’ll keep quiet is not because we’re smart and savvy, or because we trust one another implicitly. It’s because somewhere on the outside, we both have a boss to answer to, and the autocratic leaders of crime syndicates do not suffer rats. The threat of future retribution is enough to discourage purely self-interested decision-making, and frankly, that’s good for everyone involved (except maybe, the police).

____

You don’t have to look very far in American agriculture to find a second best solution. Consider modern grain production, especially corn. Today (and for more than a century), American farmers are periodically growing vastly more corn than the market demands, and it's a problem, namely for the farmers themselves. Oversupply means low prices, and low prices mean hard times.

There are many solutions to the problem of oversupply, but I’d argue that the first best solution, the one with the fewest costs and most benefits is quite obvious. If you want less grain, grow less grain. I think it’s safe to label “reduce supply” as the first best solution to oversupply problems.

The problem is the “reducing supply” is the equivalent of “everyone stays quiet” in the prisoner’s dilemma. The main problem being, there is no body with the appetite to forcefully reduce grain-planting/harvesting. The government has tried to do it indirectly with many different Farm Bill schemes. They’ve paid farmers to take acres out of production and adopt conservation reserve programs. They’ve even tried to address the problem further up the value chain instead, by buying grain directly to hold it off the market so that bin-busting crops don’t drop prices through the floor and bankrupt the farm system. All of these schemes have had limited effectiveness, and many of them have outright failed.

There is a second best solution, to ignore the oversupply problem and manage the symptoms instead. This is the finger pointing option– rather than reduce supply, let’s just focus on increasing demand. The ag industry (and it’s federal support system) is great at this. Corn is a versatile and useful crop, and we have found countless ways to use it. Every new use of corn is a new source of demand. You got animals? Feed them corn! You need fuel? Distill corn! You need starch to make your fabric stiff? Corn’s got you covered! The list goes on and on.

But the tricky thing about this second best solution is that, because it doesn’t address the underlying cause, it creates a cascade of unintended consequences. With every new market and new demand that’s found for corn, the price ticks up, at least for a short period of time. That price signal tells farmers, “plant more corn!” And farmers oblige, pouring in more land, seeds, water, crop chemistry, and other inputs to grow a crop to feed the temporarily hungry market. Within a relatively short period, any price benefit of the growth in demand will be eliminated by increased supply, and the market will find itself right back where it started.

In other words, when we continuously try to increase demand, it creates incentives for farmers to grow more, and they end up sabotaging the very system they rely on. It is a coordination failure, and as oversupply grows worse, prices drop, and farmers who want to maintain a certain level of income are actually compelled to plant even more, not less (to make up with volume what they lost in value). We’ve seen this doom spiral occur many times in the last few centuries.

This is agriculture’s prisoner’s dilemma. Everyone could just plant less, and everyone would benefit, but instead everyone defects, because no matter what anyone else does, each individual’s first best solution is to plant more, and so everyone suffers.

Rather than trying to coordinate all these players/farmers (a very heavy lift), another option would be to introduce a crime boss or a benevalant dictator. After all, if the players can’t be trusted to make the most collectively advantageous decision, then make it for them.

When it comes to agricultural production on private property, there is a boss at the top– the federal government. All the private land in this country is still, legally, America, and cannot in any meaningful way be removed from America (except by external force). Therefore, private land is really the property of the federal government, and its deed is simply a legal transfer of limited power from the government to an individual. It’s the government saying, “all of this is mine, but I will give over some sovereign rights to this tiny piece to one of my subjects, who will rule it as my vassal.” A vassal that must pay for the privilege (through property taxes), and who’s right to the land is finite.

Given that this is true, the government could use its monopoly on the legitimate use of force to compel private landowners to plant less grain. Reclaiming land from private holders is one option (eminent domain would be one avenue for this), another would be to police or regulate the amount of grain any individual could plant. A third might be the kind of supply management scheme that is currently used in Canada’s dairy sector, where contracts for certain quantities of production are issued, and if you don’t have a contract, you don’t have the right to grow the crop.

Of course, there are ways to avoid or circumvent any rule, and America is nothing if not a country where the self-interested win out (often by any means necessary). One of the things that the U.S. ag system has always struggled with is the government’s lack of conviction to do anything meaningful about the industry’s fundamental market failures.

“I’m not worried that all my guys are getting locked up,” Uncle Sam, mafia boss, seems to say, referring to the suffering farmers face related to oversupply. “They always turn on each other in the interrogation room, but I’ll just try and make sure their experience in prison isn’t too terrible.” This is not a very effective way to run a criminal organization, nor is it a very effective way to run an ag economy.

____

Sometimes pursuing first best solutions isn’t possible, but from my perspective, we’ve mostly given up on the first best solutions altogether. Oversupply of grain is a good example where many have actually convinced themselves that oversupplying grain is not actually a problem at all. This kind of mental gymnastics is truly wild. American farmers, in the last year, have seen the price of corn dip nearly to levels their fathers might have seen in the 1980s or ‘90s. That is not a “best” outcome. It is a problem, and one that American taxpayers shouldn’t pay farmers to get out of (which is how we’ve been solving it for decades). Subsidizing agriculture through direct payments is not even a second best solution– it’s much lower on the list than that. Not least because direct subsidization, even more than higher prices, enables farmers to grow more crops, worsening the problem and creating an ever increasing hunger for subsidies.

I say all this not to specifically advocate for one solution. The most “straight forward approaches” I’ve discussed— large-scale land reform, benevolent dictators, and production controls all come with major tradeoffs. There may well not be a true “first best” solution to the myriad of dilemmas facing American agriculture. What I am advocating for is that we broaden the scope of potential solutions on the table.

Policy discussions in agriculture are often irrationally limited to a tiny list of possibilities that are “effective.” Everyone has a different definition of that word. Some mean “theoretically sound,” some mean “work in practice,” and some just mean “politically and economically expedient to me, all others be damned.” The reality is that every so-called “solution” we’ve tried in agriculture has been at least a partial failure, that’s why so many problems in the system remain. So let’s not limit the conversation to the familiar, the least objectionable, or the most protective of the rights and privileges of any one group or special interest. After all, that’s what a benevolent dictator would do, wouldn’t they? They’d choose the option that maximizes the good, and individual whiners or vested interests can kick rocks.

We, the people, might not technically be benevolent dictators as a collective, but we do have the power to control a sovereign state, and to choose any solutions we find fitting. But the first step to finding the first best solution is to put all the possible solutions on the table.

____

If you’re interested in learning about some more “second best” solutions in agriculture with a particularly environmental bent, check out the recently published episode of The Only Thing That Lasts— The Land is Dead, Long Live the Land.