On the efficiency of fair trade

Abstract

This paper uses competitive equilibrium theory to analyze the economic efficiency of international “fair trade” between ethical consumers and low-income producers. The main analytical innovations are the reconsideration of the labor supply decision in a state of Keynesian involuntary unemployment as a choice between work and, not leisure, but inferior production activities; and the application of Pigou and Robinson's theory of employer monopsony, leading to a focus on the “local fair trade organization”, which has a similar effect to a labor union or minimum wage in eliminating monopsony rents. A price premium is found neither necessary nor sufficient for fair trade, and in a state of involuntary unemployment a premium does not lead to inefficient allocation. The conclusion is that fair trade improves welfare by strengthening competition for labor, and should be encouraged as a complementary element of an enlightened trade liberalization policy.

Keywords:

• fair trade
• efficiency
• involuntary unemployment
• monopsony

Acknowledgements

This is a revised version of the Position Paper presented to the International Workshop on the Economics of fair trade, which took place at Northumbria University, England, on 28 January 2005. Profound thanks are due to Leonardo Becchetti, Mark Leclair, Robbert Maseland, Alex Nicholls, Malcolm Sawyer, Albert de Vaal, and Phil Wells for presenting workshop papers in response, and to other participants, especially Geoff Moore, for valuable comments and discussion. The usual disclaimer applies. The financial support to enable the Workshop to take place was provided by the Newcastle Fairtrade Partnership.

Notes

1 The theory of employer monopsony has since progressed from the static partial equilibrium model presented here to stochastic general equilibrium models of dynamic monopsony and oligopsony, where oligopsony means employers are atomistic but still set wages. Manning cautions that:

If a theoretical paper claims a strong conclusion about the direction of [allocative] inefficiency in the free market equilibrium, then this is almost certainly because they have not considered a rich enough model in the sense that there are not enough “marginal” decisions to be influenced by incentives. (Manning 2003: 70)

The application of the traditional static monopsony model to fair trade appears reasonably secure for two sets of reasons. First, monopsony is believed to be an appropriate representation of the circumstances of local markets addressed by fair trade, although this paper cannot say whether these conclusions still hold under oligopsony. Second, in this model the slope of the labour supply curve reflects heterogeneous reservation wages (different rates of return to domestic labour) under involuntary unemployment, rather than frictions preventing a labour market from clearing (apart from those restricting households to their local market).

2 It should be noted that in a state of involuntary unemployment the competitive equilibrium coordinate (n ₂, w ₂) does not correspond to the Pareto optimum, which may be represented instead by the coordinate (n ₃, w ₃), with a higher marginal revenue curve based on full employment (MRPL^F). A state of full employment elsewhere in the economy would make local monopsony an exceptional case, since outside employers would have a strong incentive to compete for labour even in isolated areas. See also the section on the fair trade premium.

3 This result does not appear to require the assumption of efficient rationing in the sense of Becchetti and Adriani (2002), that the local fair trade organisation must employ the workers with the lowest reservation wages.

4 This paper does not consider the case for a premium as a “second best” measure to offset the effect of differential trade protection in processed and unprocessed commodities; nor as compensation, by attempting to set a minimum price based in effect on “long-period equilibrium cost of production”, for the absence of futures markets of a sufficiently long term to match the crop cycle. This cycle occurs when high prices attract investment in new capacity with a long gestation period (e.g. coffee bushes), and low marginal costs deter the scrapping of capacity when prices are low. The absence of the necessary futures markets leads to incorrect expectations and a dynamic misallocation of investment, with consequent disequilibrium swings in commodity prices that damage producers who would be efficient in long-period equilibrium, if it could ever be attained and they could survive the cycle. The following argument takes free trade as the benchmark and is limited to the short period, i.e. the equilibrium level of employment and output with a given level of production capacity. Economic theory cannot otherwise compare the efficiency of one position of long-period disequilibrium with another, and the use as a benchmark of the Marshallian concept of long-period equilibrium itself (involving both full employment and fully adjusted capacity) is rhetorical rather than scientific, in the evident absence of the requisite futures markets to coordinate long-term investment. In other words, this argument does not address the pragmatic or moral case (it cannot be based on pure economic theory) for or against fair trade based on countervailing protection or long-period cost of production.

5 The cost of social and environmental externalities internalized by the local fair trade organization may mean that the full employment competitive equilibrium does not represent the social optimum.

6 Where the premium is ring-fenced by the buyer and paid into a separate development fund there can be no adverse consequences for efficiency even in this case.