An Argument for Fair Trade?

Thanks for your post, Nicole. I think I agree with much of what you say. Of course, the person who supports Fair Trade products also (by your definition) is providing additional funding for Fair Trade farmers: it is difficult to see why supporting Fair Trade products would then not benefit Fair Trade farmers, largely by definition.

Problems arise in a few areas. For example, our worries must not be limited to (a) whether we’re buying enough bananas or (b) whether we’re buying enough Fair Trade products, but also (c) whether the price a farmer will receive from Fair Trade provides a satisfactory incentive for the farmer to produce x rather than y. (Plus, there are background assumptions which also must hold: (a) the monies earned by the farmer are shared properly with farm hands, (b) the harvest will not be too big or too small, etc.) Thus, I can agree with everything you say and still reject using Fair Trade products because they might help create a different equality (rather than between corporations and locals dividing locals into richer farmers and poorer farmhands), the price gained fails to provide local farmers with a satisfactory incentive, and —an extra worry— the products that are Fair Trade are not products I would normally purchase (thus, I buy bananas only to help the farmers, but waste the food). (I admit it: I don’t care much for fruit…)

Perhaps all this highlights how messy such cases can become…

An additional concern to Thom’s would be this: suppose the consumer buys the $2 bag of Fair Trade bananas rather than the $1 bag of regular bananas. Doesn’t it also matter what that other $1 would have been spent on? If that would have gone to, say, buy coffee or something else produced by people who are poor by whatever standard you are using, then it not clear what to make of the net. I would think for your main claim those costs would have to be taken into account, and it’s not at all clear how to think of them.

I agree with Matt. This is not then to say that the picture presented in Nicole’s post leads us the wrong way. Instead, it seems more a call to flesh things out a bit more to shore up the case. I think Nicole is very much onto something even if more can be said to make the argument.

Interesting post. Isn’t there a concern, though, that given the realities of subsistence-level farming in third-world countries, the non-fair-trade banana farmer (or, more likely, one of his small children) will starve in the interregnum between his banana-farming and sugar-farming?

This isn’t a problem with your argument, per se, since you ‘make the standard economic assumptions’- but it does seem to raise a worry about the applicability of the points you raise to the decision I’ll make today as to whether to buy fair trade bananas.

Tyler Cowen has discussed this at his blog in a way that’s akin to Mark’s point about looking at what happens to the non-Fair Trade producers, and which adds to Thom’s call to “flesh things out a bit more.” Cowen notes that Fair Trade products offer retailers an opportunity for price discrimination (poorer or apathetic consumers will buy the conventional version of the good while richer or more socially-conscious consumers will now spend more to buy the Fair Trade version of the good), an opportunity which retailers will be happy to take. One likely result, on Cowen’s analysis, is “institutionalizing especially poor treatment for one class of workers.” See here.

Actually, it’s not hard to show that application of the Principle of Comparative Advantage is *inconsistent* with the Difference Principle.

I’ve always thought this too obvious to write up but the blog interest in this topic suggests otherwise.

I’d like to see it too, Terrance. As stated, I’m not quite sure what you have in mind. The Principle of Comparitive purports to be a descriptive law of economics. If it’s correct, it’s just a statement about the way the world works. So I’m not sure how it *could* be inconsistent with the Difference Principle, which is a normative principle that makes no assumptions about descriptive facts.

Matt Zowlinski doesn’t see “how there could be an inconsistency” between the Principle of Comparative Advantage and the Difference Principle.

Here’s how:

Suppose that we are contemplating entering into a trade agreement with another nation. The Difference Principle tells us that we ought not to do so if we suppose that the upshots will disadvantage our worst off citizens( however much they might advantage the citizenry in aggregate).

The Principle of Comparative Advantage tells us that, under *certain assumptions* about the nations involved, free trade between nations *must* be to the advantage of both nations. Note this is not a “merely descriptive” claim: given the assumptions, the PCA follows as a matter of logic.

Obviously any application of the PCA will assume some aggregative measure of national advantage just as any application of the DP must assume some measure of individual/class advantage. Equally obviously, if we are going to use the DP to normatively evaluate a policy of free trade we will have to find measures that commensurate the two principles.

The question then is, under such commensurate measures, and under the assumptions required by PCA, will a nation’s entering into trade arrangements disadvantage that nation’s worst off:
a) Always.
b) Sometimes.
c) Never.

Economists favor free trade on the grounds of PCA. But unless they have to hand a proof of (c), Rawlsian Liberals ought not to be so sure free trade is always a good thing.

As I said, I think the answer is (a), and that this explains why Rawls so ardently supported Pat Buchanan’s presidential bid.

Well, not to pick nits, but…
a) The Principle of Comparative Advantage *is* a descriptive claim insofar as it purports to describe how the world works. Your grounds for believing it might be a priori or empirical, but that’s not relevant to the *content* of the claim.
b) Your argument shows, at best, that there is an incompatibility between belief in the moral permissibility of free trade and the difference principle, not between PCA and DP.
c) And technically, the right question to ask to see whether this incompatibility exists is not whether free trade disadvantages the worse off, but whether a basic structure including free trade benefits the least well-off as much as any other alternative policy.
d) The point about the basic structure matters, because it means that what Rawls asks us to do is decide about free trade agreements as a general category, and not on a case by case basis. [Or rather, we have to decide at the level of basic structure if we want to decide about them as a category or on a case by case basis, and then, if we choose the former, decide in what way we want to categorically deal with them] I think that (1) a much better case can be made for free trade’s benefitting the poor if we focus on free trade as a general category rather than on a case by case basis. Steel tariffs might benefit low income steel workers, but not as much as they will be harmed by general restrictions on trade. I also think that (2) there’s a good case to be made, on DP grounds, for deciding free trade issues on a general rather than case-by-case basis, given that exceptions to free trade are likely to be politically motivated. A policy maker who has discretion to adhere to free trade or create protectionist exceptions is likely to use that discretion to benefit politically powerful groups. Given that the poor are not politically powerful, they will probably not benefit from such exceptions, and will indeed probably be very severely harmed. Steel tariffs are a good example of that [steel union workers are certainly not the least well off group in the US], as are agricultural price supports and import restrictions.

Terrance said, “Suppose that we are contemplating entering into a trade agreement with another nation. The Difference Principle tells us that we ought not to do so if we suppose that the upshots will disadvantage our worst off citizens( however much they might advantage the citizenry in aggregate). ”

But I don’t think that’s how the difference principle works, or at least it need not work that way. If there is a gain from free trade to aggregate welfare, as PCA says there should be under certain conditions, the question then would be how this gain should be spread around. If the gain is large enough it can be spread around so as to help those who lose from trade, assuming they are the least advantaged. Now, this policy will perhaps be distorting, but then the question is just whether the distortion is large enough to out-weight the gains from free trade. If not, then free trade, taken with other policies, is compatible with the DP. It’s not as if if we accept free trade we must then not put in place any policies to make sure that the gains are not shared by all or help the least advantaged. So, I fail to see where the supposed contradition is here.

I’m curious about Rawls ardently supporting Pat Buchanan’s presidential bid. I’m ready to burn my copies of A Theory of Justice on a pile of butterfly ballots.

The flip side of the coin from both Matts’ responses here (as well as Terrance’s) is this. Suppose that the PCA is true as a descriptive claim, or more strongly that the empirical claims for free trade are roughly true in this way: bars to trade make everybody worse off (perhaps in ways that there are no special fixes to cure; the least well-off under any arrangement just do more poorly than when trade is possible). Then the DP would fully endorse free trade — perhaps with further stipulations as to how the gains from trade are distributed, as Matt L indicates. The point, again, is that there is no opposition: the PCA is an empirical claim which, if true, is something that has to be taken into consideration in implementing the normative requirements of the DP, not something that is either counter to or in agreement with the DP itself.

Matt Zwonlinski should pick better nits:

a) Every principle “purports to describe how the world works”. Indeed, some principles, purport to describe how every possible world works. The PCA is one such. It is *not* an empirical claim.

b) No, I was not suggesting that DP and PCA are logically inconsistent. Given that PCA is necessary and DP normative that would have been a very odd claim. I was, of course, suggesting, as you surmise, that practicing Free Trade (as characterized by PCA) might be impermissible according to DP. I’m glad you now understand what I was trying to say.

c) Both Matts and Mark LeBar seem to think that we should not apply DP to particular policies but only to bundles of policies or “basic structures”. The argument is that even if Free Trade screws the poor we might be able to make it up to them in some other way.

That might be so but it misses the *philosophical* point. Rawls offers DP is a principle by which we assess the *justice* of *particular* policies. Injustice doesn’t go away just because you put in place mechanisms to compensate its victims. Thus, suppose someone proposed laws which forbade the poor from buying liquor. That *policy* would be inconsistent with DP– and unjust– even if some of the savings in welfare costs were used bettering the lot of the poor over all.

d) I see no evidence in any of this discussion that anyone understands how the economist’s argument for free trade (i.e. PCA) is supposed to work. The argument is not that dropping steel tariffs makes society so much better off that we can afford to pay welfare to unemployed steel workers. The argument is that society will be so much more prosperous that the steel workers will be able to find better paying jobs doing something other than steel working. Sound plausible to you? How about janitors, nannies and lawn-mowers or … fill in your own “worst” job. If that sounds unrealistic to you maybe you have more in common with Pat Buchanan than you have hitherto suspected.

Terrance said, “Rawls offers DP is a principle by which we assess the *justice* of *particular* policies”

I’m not sure that’s right. In fact, I think it can’t be right- the principles of justice apply to the basic structure, after all, not individually to particular policies. And usually particular policies can’t be looked at in isolation anyway- they have to be evaluated in light of a larger system. Can you point to some text where Rawls says this? I don’t think there is one, but if you can point to one I’ll be interested to see it. As to your later question, I don’t think we can tell if there _is_ an injustice in any particular case w/o looking at larger schemes. So, if free trade by itself would make, say, the employees of US Steel or whomever worse off than they would be without it, but the gain in over-all welfare is enough for us to compensate them for their loss and re-train them and the like, then they have not suffered any injustice. They don’t have a right to work in the steel industry per se, after all. So, I think you cannot evaluate the situation without looking at a larger set of issues than you want to, and I’m pretty sure that that would be Rawls’s view, too, but even if not he’d just be wrong.

Nicole, one thought about the larger project (without reading the paper; I’m sorry I don’t have time to do that just now) is that it seems to be running into a broad form of the knowledge problem Hayek pressed against attempts to engage in just this sort of macro economic calculation, for reasons it sounds like are already bothering you. It sounds like you may be arriving at somewhat Hayekian conclusions from a somewhat different direction. If you haven’t thought of the Hayek connection before, that might be worth exploring.

I’m with both Matts in the dispute above — and I’d be curious to find any evidence that Rawls supported Pat Buchanan’s presidential bid (not least because of a commitment to PCA, of all things) which I very strongly doubt.

Instead, I’d like to briefly comment to Nicole’s reponse (hello again!) on (a). As a reminder, I didn’t disagree with her argument, but rather I felt her argument rested on a few unstated assumptions. One of these is that the extra funds raised from FairTrade would trickle down to the farmers (i.e., (a)). (I also thought that not only must monies trickle down, but that they had to be sufficient to offset potential profits from harvesting other crops, etc.)

I still think it might be worth being more explicit about this assumption being part of your picture. I remember looking into some research with Leif on humanitarian aid. In essence, the data is very difficult to come by on how much aid you purchase for your dollar as many agencies are reluctant to give full disclosure. From what I’ve seen, it isn’t close to $1 of aid for $1 donated. Part of this is because of administrative and promotional costs. Another problem is that often persons who go into assistance are driven primarily by (very honorable) idealism. I think it remains important to ask precisely what benefits are derived from any scheme. An argument in favour surely rests on the assumption that money into a scheme will trickle down. I wouldn’t take the assumption for granted as there is much evidence that much of funds — at least for different organizations and projects — goes astray.

Not being a pessimist: just trying to help!

The contrasts between humanitarian aid and Fair Trade here might be relevant. In the light of the aims of Fair Trade, humanitarian aid seems to be (and in fact may be) going in precisely the wrong direction. If we construe Fair Trade as an attempt (successful or not) to raise market prices for the goods poor folks produce (and to pass the margin down to them), then — at least when humanitarian aid takes the form of goods poor folks produce, such as food — humanitarian aid is, in effect, suppressing the market price, perhaps down to zero! This is, of course, just a ramification of the displacement effect that critics of such aid complain about. I don’t have a grasp of the market mechanics to think through how those features of the situation might work backwards to affect the points Thom is thinking of here, and this isn’t to advocate for Fair Trade. However, it seems to me there might be structural disanalogies as well as analogies at work here.