Fairtrade means among other things that farmers from poor countries receive a guaranteed minimum price for their produce, or that workers receive a minimum wage, that they have good working conditions, and that production is environmental-friendly. In exchange, consumers in the developed countries agree to pay a higher price in the shops. Fairtrade is now one of the fastest growing markets in the world, and the Fairtrade market-share has in some countries become significant for some products. But in general, “fair” products make up a very small share of total trade between the North and the South, and for many products there exists no fair alternative. The reason that so little of trade is Fairtrade could be that most consumers do not have the willingness to pay for fair production. However, to conclude in this way, we have to suppose that the markets are well-functioning in the sense that consumers’ preferences for Fairtrade are accommodated. Alternative explanations of the low share of Fairtrade are market failures in the provision of fair goods, of which the most important seems to be information about the ethical characteristics of goods. The argument is as follows. When a consumer considers buying a product from the South, he or she does not know whether the product has not been produced using child labour, with bad working conditions and subsistence wages, or large environmental problems. This can lead the consumer to feel ethical risk. To reduce ethical risk, the consumer can either search ethical information about the good, or buy a good that he or she considers ethically “safe”. The consumer may for instance trust that products in the North have been produced in a fair manner because of effective legislation. As for ethical information, there exist many sources, such as specialized ethical consumer magazines and a large number of internet sites. However, there may be high costs associated with finding and making use of such information for ethical consumption. For all except the most concerned consumers, these costs might be near prohibitive. Information may also simply be unavailable. On the other hand, if the relevant ethical information could be summarized in a simple “ethical” product label, it would be costlessly available to all consumers. I therefore argue that ethical information has the main characteristic of a public good. There are large costs in searching and summarizing information, but when this is first done, the costs of allowing an additional consumer access to this information are negligible, and the use of one consumer of the good does not impair the use of another consumer of the good. Free-rider problems mean that we could have a too low supply of ethical information in a private market, which means that there might be a role for public policy. An important objection against the need of public involvement is exactly that there have been established private labelling schemes like Fairtrade that guarantee consumers a fair alternative. These altruistic agents are willing to incur the costs associated with fair labelling, and thus help correct the market failure in ethical information. Does this mean that public policy after all is unnecessary? Here the fact that Fairtrade labelling currently exists only for certain products, mostly primary products like coffee, could be of importance. One possible explanation for this fact might be that certain characteristics of primary products make producers especially vulnerable. For example, supply and demand conditions may make the income from agricultural production highly volatile. One could therefore argue that there is less need for Fairtrade in other products. However, also in other industries like the textile industry and even the production of mobile phones there have been many reports of bad working conditions. To me, a more plausible explanation for the fact that there is Fairtrade in some products, but not others, is that there are differences between products in how easy it is to verify that production is fair. It is probably easier to ensure fair production of a “simple” product like coffee, than of products that have a longer and more complex production chain. To the extent that labelling other products requires more resources, it is possible that it will be too costly for private altruistic agents like Fairtrade. As for public policy, several options are available. One strategy could be to label all products from the South, so that consumers can choose between fair and unfair products. However, there may not always be a fair alternative available. A second possibility is to ensure that consumers have the option of a fair good. Public policy then aims to facilitate ethical consumption. But many are sceptical of what ethical consumption can achieve in addressing the problems of globalisation, and call for international legislation, e.g. through standards for labour and the environment in the World Trade Organization (WTO). In this case, public policy seeks to go further and regulate all production. However, developing countries oppose such standards, fearing they will work as hidden protectionism from the North, and reduce their competitiveness. If standards lead to reduced trade, this could have negative effects for the South. However, if fair North consumers realize that the higher costs are due to the fulfilment of standards, trade need not be reduced much. I set up a partial equilibrium model to study the welfare implications of four stylized public policies: 1. Labelling all products from the South with respect to their fairness; 2. Guaranteeing a fair alternative to consumers; 3. Ensuring all South products are fair through international standards; and 4. Using public ethical investment to ensure fairness in South firms. The model draws on the consumer framework of Adriani and Becchetti (2004), but introduces uncertainty about the fairness of goods labelled “Made in South” (ethical risk). The main mechanism in the model works through prices, with fair goods normally receiving a higher price than unfair goods. The “Made in South” good normally receives a price somewhere in between these two. Consumer fairness is here defined relative to the costs of ensuring fair production, which are assumed to be higher than the costs of unfair production, through a higher “fair” wage. This is explained in the following. For Public policy 1, labelling all products from the South, the main results are as follows. If there is no consumer fairness, ethical labelling is irrelevant, since consumers in any case buy the cheapest products. With intermediate fairness, fair products receive a higher price than the price for goods “Made in South”, while unfair products receive a lower price. Because they receive a higher price for their products, fair firms can expand their production and employ more workers, who have increased welfare. However, unfair firms receive a lower price for their products, and some workers may lose their job and end up in the informal sector with a very low income. Some of the costs for unfair firms from being boycotted by ethical consumers are thus passed on to the workers. Fair consumers in the North who are now able to buy guaranteed fair products have a higher welfare, as have the least fair consumers who can buy a cheaper South good. If there are ex ante no fair firms in the South, ethical labelling may have only negative effects, since the lower price for unfair products means reduced demand for labour in the South. Also the welfare of North consumers is reduced. However, the higher fair price may also allow fair firms to enter the market. Effects are then as in the case with ex ante fair firms in the South. With high fairness, the price for fair products is sufficiently higher than the price for unfair products to make fair production most profitable. Profit-maximizing firms switch to fair production, and all regular production in the South is fair. Workers in these firms experience a welfare improvement (except possibly some workers who lose their job in a profit-maximizing firm if the net effects on its demand from changing to fair production are negative). Also most North consumers are better off, except the least fair consumers who no longer have the option of a cheap unfair good. Public policy 2 is ensuring a fair alternative. This could be done by certifying existing fair firms as fair, or by ensuring that there are fair firms e.g. through ethical investment (see below). The main results are as follows. With no fairness, a fair alternative is irrelevant if its price is higher. With intermediate fairness, certified fair firms receive a higher price and can employ more workers. However, there might be negative effects for the workers in the unfair firm, if the fair alternative leads to consumers to reduce their subjective probability that non-certified products from the South are fair. This is because the non-certified South good then receives a lower price. With high fairness, profit-maximizing firms change to fair production and have their products certified, as fair production is most profitable. Effects are then the same as with ethical labelling and high fairness. Public policy 3, international standards, differs from the other two strategies in that profit-maximizing firms no longer have the option of unfair production, and consumers have only the fair alternative. With no fairness, the effects of standards may be predominantly negative. Though some workers receive the higher fair wage, many others may lose their job. North consumers are also worse off since the price of the South good increases. With intermediate fairness, effects depend on whether or not consumers are aware of the standards. If they are, all South products receive the fair price. This means that fair firms can employ more workers. It also means that the reduction in employment in profit-maximizing firms due to the higher costs of respecting standards is smaller, because they receive a higher price for their products. However, if consumers are not informed, effects are similar to those when there is no fairness. With high fairness and informed consumers, standards are not binding since fair production is most profitable. Effects are then the same as for the two policies described above. However, if consumers are not informed, standards may have predominantly negative effects, even though consumers would be willing to finance the costs of respecting standards. A policy implication seems to be that an introduction of international standards should be combined with informing consumers, to exploit consumer fairness. Public policy 4, ethical investment here means that the authorities require fair production in the South firms they own. Effects are similar to those of international standards, but on a smaller scale. Also, as mentioned ethical investment may be a way of ensuring there are fair firms in the South. Providing fair consumers with information may have predominantly negative welfare effects if there is no fair alternative. On the other hand, requiring fair production may be harmful if consumer fairness is low, or consumers are not informed. Ethical consumption is therefore likely to be harmful without fair production, and vice versa. Of the policies considered, ensuring a fair alternative seems to offer most of the positive effects while minimizing potential negative effects. However, the analysis leaves many important questions open. Among these are what would in fact be the costs of ensuring all South production is fair, and the practical possibilities of implementing the different public policies. Many possible effects are not taken into account. The results should therefore be taken only as suggestive.