Changes in food prices are often treated as exogenous shocks to macroeconomic developments, that strike according to the whims of nature. They are frequently seen as a nuisance, that is not worth modelling. Yet they are an important component of consumer baskets. Their weight on inflation outcomes is relevant in advanced and dloping countries alike (see e.g. Peersman (2022)). Oaverage, food and nonbeverage items represent 17% of consumer price indices in OECD countries, ranging from a low of 8% in the United States to 27% in the case of Poland. 3 Because of the high volatility of food item prices, their contribution to the variation of inflation can easily exceed their weight. Further, because of their salience, food prices could have a disproportionate impact on inflation expectations and more readily spread to other categories.

A key research question is whether and to which extent food inflation responds to broader domestic macroeconomic conditions such as output gaps or changes in headline inflation expectations. For instance, does economic over- or underheating spill-over into food pricing? Is the Phillips relation valid for food items? Answers to thescore questions are key also for policy makers, as they might for instance affect the calibration of monetary policy. Particularly in countries where the inflation target is based on headline inflation, the optimal policy is bound to differ depending on the extent to which the food component of the CPI is responsive to output gaps and inflation expectations − which are directly influenced by monetary actions. Further, how strongly do global food prices transmit to retail food costs? And by how much can larger crop yields, higher food imports – or conversely lower food exports – contribute to the alleviation of domestic food prices? This article provides quantitative answers to each of the above questions based on impulse responses obtained through the Jordà (2005) local projection method.