Big Mac parity, income, and trade

Nontraded inputs account for the lion's share of a Big Mac price (Ong 1997, Parsley and Wei 2003). Major departures from Big Mac PPP may then be explained by the Balassa-Samuelson income differences effect, as shown e.g. by Click (1996). But it has been argued that Click''s result is not robust to changing estimation methods, sample of countries, and time period (Fujiki and Kitamura 2003). Here we address a key theoretical distinction between high and low income countries for the Balassa-Samuelson effect to be properly evaluated. Since this distinction is missing in Click''s analysis, we revisit his finding and take a sample which is distinct (in terms of both set of countries and time period) to meet Fujiki-Kitamura''s criticism. We find that distinguishing high from low income makes no harm to Click''s result. But we also find that openness to trade (viewed as a proxy for trade barriers) helps to explain departures from Big Mac PPP.