There is general agreement among economists regarding the superiority of free trade over protection, to the extent that even when the strictly economic case for free trade fails, they are generally quick to embrace it for the practical reason that it is the least bad alternative. Nevertheless, trade liberalization is one of the most politically contentious actions that a government can undertake, and has almost always been associated with changes in political regime or undertaken at a point of economic crisis.
In Discussion Paper No. 391, Raquel Fernandez and Research Fellow Dani Rodrik propose and formalize a new explanation of this unpopularity, which is based on uncertainty and is complementary to the traditional explanations. They argue that many individuals simply do not know how they will fare following a trade reform, and that this may reduce support for a reform that would otherwise have been popular, even in the absence of risk aversion.
Outward-oriented policies favour entrepreneurs and workers already employed in the exportable goods sector, who can generally be expected to identify themselves, but reforms also favour some sectors and individuals that previously produced primarily for the domestic market, who will turn to export under the new price structure. Given the difficulty of predicting the post-reform structure of trade and production, it is unreasonable to expect all these individuals to be able to identify themselves as gainers or losers ex ante. Fernandez and Rodrik show that reforms that would receive adequate popular support once enacted to be sustainable may nevertheless fail to attract sufficient support in advance to ensure their implementation. Moreover, the role of uncertainty in determining the outcome is not symmetric: reforms that are initially rejected may remain on the drawing-board indefinitely, while reforms that are initially accepted may come to be reversed over time.
Fernandez and Rodrik use a simple model to demonstrate that a bias towards the status quo in this case protection exists even when individuals are not risk-averse and when they are rational and fully aware of the aggregate efficiency gains generated by reform.
They illustrate their argument with evidence drawn from South Korea, Chile and Turkey, where uncertainty of the kind mentioned above is likely to have been an important element in the early opposition to trade reform. For all three countries, they observe both a substantial change in the composition of foreign trade and the appearance of new products that were not exported prior to reform. This helps explain why the trade reforms which have been so beneficial to the private sectors of all three countries received so little support from those private sectors early on, but came to be supported by an important section of private industry once their consequences became evident.

Why is Trade Reform so Unpopular?
Raquel Fernandez and Dani Rodrik

Discussion Paper No. 391, March 1990 (IT)