Sunday, July 02, 2006

Farm Subsidies Transfers

Greg Mankiw points out this Washington Post exposé on farm subsidies. But the particular program that the piece covers is not a subsidy at all. A subsidy is a negative tax on a good. If I pay you an extra $10 for every acre of corn that you grow, I've subsidized your corn-growing activities. This distorts your decision to grow corn because you'll grow more than you otherwise would. You may not even grow corn at all if I weren't subsidizing you. Since growing corn provides no positive externality and markets for agricultural goods are reasonably close to perfectly competitive, too much corn will be grown. What does that mean, exactly? "Too much corn" means that we could potentially improve everyone's well-being by moving some resources from corn-growing to other productive activities. You, the subsidized corn farmer, won't respond to price signals that tell you to abandon corn-growing for a profession that creates greater value. Moreover, the subsidy will squeeze out unsubsidized farmers, like those in third-world countries.

That's a farm subsidy. What the Post's story is really about is a program that was included as farm subsidy methadone in the Republicans' ultimately unsuccessful 1996 attempt to rid the U.S. of farm subsidies once and for all. Cutting farm subsidies is not a Pareto-improving action. Farmers are hurt (and so is Archer Daniels Midland.) They most likely have to find new work and a new home. They have to be trained to do a new job. They may never reach the same level of income that they once had because they've invested so much in farming-related human capital. But removing farm subsidies is a potential Pareto-improvement. It creates so much new wealth that the losers can (theoretically) be compensated for their lost income by the winners (everyone who benefits from lower produce prices, for example.) This program is an attempt to do that. (Though Republicans were not really interested in compensating anyone, they included the program simply to win Democratic votes.) It is not, as the Post makes it out to be, one of the truly pernicious farm subsidies that we should be most worried about.

However, this program is indeed a very bad way of compensating farmers. The program in a nutshell: if you own land that used to be farmed you get paid a per-acre transfer as long as that land is not developed and regardless of whether that land is still being farmed.

To see why this is a bad plan, consider what a good plan would look like. A non-distortionary compensation program would give everyone who was farming at the time subsidies were cut (or even better, a few months before subsidies were cut) a single payment or series of payments equal to the approximate amount lost through the removal of subsidies. This doesn't alter anyone's decision to farm (from the non-transfer, non-subsidy equilibrium) because it doesn't affect the cost or benefit of farming an additional acre of corn. If farming 500 acres of corn after subsidies were cut is profitable, farming 500 acres of corn would be no more or less as profitable after the transfer as it was before the transfer.

The plan as it was implemented has a number of problems. The biggest problem is that the transfer doesn't follow the people that were hurt, it follows the land. Why do we have an interest in whoever happens to own land that was farmed once upon a time? We don't; we have an interest in the people who were harmed by cutting farm subsidies. The transfer should be issued directly to the farmers, not to the land. This problem goes even deeper, however: it distorts the allocation of land. Suddenly, land that had been farmed had a fixed payment attached to it. Each acre of this land basically comes with an annuity and the price of this land is accordingly higher. Since farmers' property is now artificially worth more, this is in fact an incentive to sell land and stop farming. The plan actually creates an inefficiently low amount of farming. A related problem is the provision that land cannot be developed if you collect payments on it. This is simply silly, why prevent land from being used for more valuable purposes?

In this plan farmers are compensated in an extremely inefficient way: attaching an annuity to each acre of land inflates property values by the amount of the annuity. So, when farmers eventually sell their land, they are essentially paid the amount of the annuity by whoever purchases the land. And whoever purchases the land has a strong incentive to leave it undeveloped, even if it would be best to develop the land. But I wish that the Washington Post had pointed out that the basic idea is good. The losers from farm subsidy cuts are no different from the losers from free trade. Both should be compensated for their losses, not only because it is equitable to do so, but also because they will be more willing to support policies that increase the size of the economic pie.

3 Comments:

Blogger Bill Harshaw said...

One reason many economists criticized pre-Freedom to Farm farm subsidies was that, they believed, the subsidies had had the effect of increasing the value of farm land. So the FTF buyout didn't suddenly increase the value of farmland; it was supposed to prevent a sudden decrease in the value. That's probably the reason the original FTF payments declined over the 6 year period of the FTF. By 2002 the rationale for the FTF payments was long forgotten.

And farm subsidies had always been tied to the land. In the 1930's there was a big fight over cotton plantation owners dropping sharecroppers when AAA program cut back cotton acreage.

9:49 AM  
Blogger henry said...

Bill, thanks for your comment. I didn't realize that pre-FTF farm subsidies were also tied to the land. Is this done in the same way that FTF transfers are?

I think I may have been wrong when I said that attaching subsidies to the land decreases the incentive to farm. If the owner of the land gets the subsidy no matter whether it is farmed or not, then this will be reflected in property prices, but if the farmer stays they still get the subsidy. So it's a wash as far as opportunity cost. But I'm not sure why people think that this would increase agricultural output.

Of course if the land must be farmed, then it does help.

3:44 PM  
Blogger Bill Harshaw said...

From the 30's to 1995 most "production adjustment" programs tied eligibility to a historical acreage. For example, if you had a history of growing 100 acres of corn, you could enroll in the program for a year if you agreed to grow no more than 90 acres and put the other 10 acres in conserving uses. The theory was that a national reduction in acreage would reduce supply and thereby increase prices.

An interesting experiment I haven't kept up with is the buyout of the tobacco and peanut programs (this century). A quick Google shows they aren't that different from the FTF--payments based on quota ownership (i.e., basically land) stretched over a period of years. The last peanut payment is 2006, so we'll see if that provision sticks.

When I Googled "tobacco buyout program" I picked up an ad for a company that will buy the payment contract.

In theory, under WTO rules, existing subsidies are delinked from current production so the incentive is reduced. For example, if I've been growing rice and getting payments through the government program, I can convert to some other use. (I can't convert to fruits and vegetables, which is the logical switch, because it's against the program rules. So, as the WPost showed yesterday, people converted to 9-acre backyards.)

4:42 PM  

Post a Comment

<< Home