What drives food prices?
(authors: Jan Brůha, Petr Král, Jan Šolc)
Consumer prices reflect both costs and demand, while food price inflation is shaped mainly by supply factors. This is evidenced by a high correlation between growth in consumer prices, food producer prices and agricultural producer prices (see Chart 1). Going beyond this co-movement, however, cyclical demand pressures play a role alongside costs. This may explain the recent slight loosening of the correlation between consumer prices of food and agricultural producer prices. This box tries to answer the question of how strong an effect cyclical pressures have on food prices for consumers.
Chart 1 (BOX) Food prices
Agricultural producer prices, food industry prices and consumer prices of food are significantly correlated
(annual percentage changes)
The margin between consumer and producer food prices started to increase after 2016 due to growing demand. This trend can be observed for numerous food commodities based on CZSO data. These data can be used to determine the prices of selected food commodities at each stage of the product vertical. It is then possible to calculate the margins between consumer prices and producer prices in the food industry or agricultural producer prices in koruna terms. Both margins started to increase significantly for most monitored items around 2016. They reached their highest values in the period under review last year. This suggests that the level of demand has recently had a truly significant effect on consumer food price inflation. These aspects are well illustrated by the example of pork (see Chart 2). Although per-kilogram producer prices dropped markedly in 2018, the price of pork in shops increased slightly. The margin between consumer and producer prices thus rose significantly, probably reflecting strong demand. The situation changed in 2019. Prices of pork around the world rose sharply due to African swine fever in China, and the Czech market was no exception. Although this cost pressure resulted in a decline in the margin between consumer prices and producer prices, the decline was smaller than the increase in costs, and this margin thus remained elevated. This indicates that traders managed to pass on part of the growing costs to consumers.
Chart 2 (BOX) Prices of pork in the product vertical
The margins between consumer prices and industrial/agricultural producer prices remain elevated
(CZK/kg)
This analysis of the margins between consumer prices and food producer prices uses a trend-cycle VAR model.[1] The model was applied to the aggregate HICP data since 2005, i.e. after the Czech Republic joined the EU, and explains the aggregate growth in agriculture producer prices, food producer prices, consumer food prices and food import prices.[2] The advantage of the trend-cycle model is that trends and cyclical components are modelled separately. This is important, since consumer food prices have long been rising faster than producer prices (see Chart 3).[3] The model is extended to include LUCI, which serves as a measure of cyclical demand pressures. LUCI was chosen because it is a composite indicator of labour market tightness. A very tight labour market creates conditions for strong domestic demand. This indicator is available monthly and is constructed to be directly cyclical.[4]
Chart 3 (BOX) Price categories in the food industry
Consumer food prices have long been rising faster than producer prices
(index; 2015 average = 100)
The results of the analysis point to the presence of demand effects in the evolution of food prices. The above-mentioned trend-cycle VAR model allows one to construct a hypothetical consumer price index under neutral demand pressures. This hypothetical index thus adjusts food prices for domestic demand pressures (recessionary and expansionary). The deviation of the actual food price index from its hypothetical level (see Chart 4) then indicates the contribution of domestic demand to the level of consumer food prices. Slightly positive contributions of demand pressures in food prices can be identified in the run-up to the global economic crisis. Its onset meant a noticeable drop in demand, manifested in a narrowing of the gap between the food price index and its version adjusted for demand pressures in late 2008 and early 2009. The subsequent lengthy domestic recession dampened growth in food prices. With the onset of the growth phase of the economic cycle after 2015, the upward pressure on food prices began to increase gradually.
Chart 4 (BOX) Food price index
Food prices have also significantly reflected demand pressures in recent years
(index; 2015 average = 100)
The contributions of demand to growth in consumer food prices currently remain positive. The positive contribution of demand pressures to food price inflation peaked in the final quarter of 2017. Until recently it was gradually shrinking, but it has increased again in recent months (see Chart 5). Food price inflation is thus currently being significantly influenced again by inflationary demand pressures in the domestic economy and is higher than would correspond to the growth in costs arising from changes in agricultural and food producer prices and food import prices. The above estimate of the demand pressures in food prices is broadly consistent with the CNB’s conventional estimates of the output gap, despite being based on a different methodology and data. The results of this box therefore indirectly confirm that the output gap estimate is realistic.
Chart 5 (BOX) Demand pressures affecting food price inflation
Despite having eased a little, demand pressures are continuing to push up food price inflation
(contribution in percentage points to annual food price inflation)
[1] The general methodology of trend-cycle VAR models is described by Andrle and Brůha (2017): “Forecasting and Policy Analysis with Trend-Cycle Bayesian VARs”, IMF and CNB, (https://michalandrle.weebly.com/uploads/1/3/9/2/13921270/tc_vars.pdf).
[2] The above CZSO statistics illustrating the price margins at each level of product vertical expressed in CZK do not capture the entire consumer food basket and are only available since 2013. For the purposes of estimating the trend-cycle VAR model, we therefore work with aggregate indexes, for which longer time series are available. The change in these price indexes approximates only the change in margins and so – in contrast to the individual data – cannot be used to quantify the price margins in koruna terms.
[3] This fact is common to the other converging economies of the Central European region and thus probably also reflects changing demand for various types of foods and general price convergence as the inhabitants of converging economies become more prosperous. However, this long-term income effect is not a subject of the analysis in this box. This box focuses on the cyclical effects of consumer demand.
[4] The extended LUCI was presented in a box in IR IV/2019.