Introduction
Several authors have recently investigated the effects of changes in sugar policies. Studies have
focused on multilateral trade liberalization, either on the effect of an extension of preferential regimes
(e.g. the EBA), or on the effect of domestic reforms. Perplexingly, the different studies provide results
that are largely inconsistent, even for rather similar scenarios. Some authors find that market
liberalization will result in large welfare gains and significant changes in international trade. Others
believe that the overall gains will actually be limited, due to inelastic demand (i.e. small initial
Harberger triangles on the consumer side), persistence of supply control (production quotas) and
because large rents need to be reduced before reforms actually become binding and affect output.
The degree of incoherence in the quantitative results that have been published is troublesome. The
inconsistencies exceed what is normally observed between different modeling approaches (general vs
partial equilibrium) in the agricultural sector. Here, the effects of trade liberalization are sometimes
contradictory and the magnitude of the differences in, say, world price variations or change in welfare
is striking. While there are several explanations for the diverging results regarding the effects of
reforms on the world market, the changes in EU sugar net trade appear to be of particular importance.
There is a large degree of uncertainty as to the level of EU sugar supply under different policy
conditions. Because producers have been largely isolated from world market signals for decades,
there is little statistical variability to exploit, and the "guesstimates" of supply elasticities and
production costs rely on thin evidence. We also believe that not enough attention has been paid to
some specific characteristics of EU production, and in particular to the determinants of the supply of
"C" sugar, i.e. sugar produced outside production quotas (C sugar is bound to disappear within the
next years, due to the EU reform and the 2005 ruling of the WTO dispute settlement body).
A first objective of this paper is to model EU sugar supply, accounting for specific aspects such as a
potential cross-subsidy between production quota sugar and C sugar, and the existence of rents in both
the farm and processing sectors. Another objective is to provide estimates of the effects of i/ the
recent EU sugar reform (still to be implemented); ii/ the elimination of EU export subsidies by 2013,
a decision adopted (with some side conditions) in the framework of the WTO.
We first provide a brief survey of the various studies assessing the impact of liberalization of the sugar
sector. The specification of the EU supply response seems to be a major determinant of the world
market equilibrium. We then investigate some specific aspects of EU sugar production, in particular
those that might result in a cross-subsidy between in-quota and C sugar. We propose a way to model
the sugar market that includes these specific aspects of EU producer and processor behavior. We then
estimate econometrically the parameters that allow us to account for this cross subsidization and to
calibrate production costs. Finally, we integrate this representation of the sugar sector in a general