MEASURING PROTECTION IN AGRICULTURE: THE PRODUCER SUBSIDY EQUIVALENT REVISITED

In the 1980s, the Producer Subsidy Equivalent (PSE) became the dominant measure of protection in applied studies of international agricultural trade. This paper analyzes potential biases in the ratio form of the PSE introduced by using actual domestic prices rather than social opportunity costs in the denominator. It is shown that doing so introduces a consistent under-estimation of the effects of trade restrictions and other price-support policies, relative to deficiency payments and other income-support policies. It is found that under plausible conditions this bias leads the PSE to rank protection levels across countries or crops incorrectly. In a sample of 250 activities across 33 countries, such errors were found to occur in 5% of crop comparisons and 8% of country comparisons, including a number of politically sensitive cases. An improved formula would therefore provide significantly more accurate results than the conventional PSE, with no additional data or more restrictive assumptions.