Transparency of Food Pricing (TRANSFOP)

Working papers

Competition in the Food Chain

Recent events in world commodity markets coupled with high and volatile levels of retail food inflation across many countries have given rise to the functioning of food supply chains. Although there are many factors that impact on the functioning of the supply chain (including technology, changing consumer tastes, regulation and so on), there has been a concern that competition (or the lack of it) can also have an important impact on price developments. The aim of this background paper is address competition in the food supply chain and summarise the key insights from research. (Working Paper 11: ISBN 978-0-902746-28-2)

An Analysis of Asymmetric Consumer Price Responses..

We analyze empirically a possible channel for the existence of asymmetric price-cost pass-through, that is, of prices responding differently to negative and positive upstream cost shocks. While asymmetric price-cost pass-through has been documented in many markets, possible causes for such a phenomenon have not empirically investigated. Using consumer panel data in the coffee retail sector in France, we estimate a demand model structurally allowing for asymmetric consumer responses to positive and negative retail price changes. According to the demand estimates, we indeed find significant evidence that
consumers react differentially to positive and negative price movements, in that demand is less sensitive (elastic) to price increases than to price decreases. Then using counterfactual simulations within an equilibrium model of demand and supply side behavior we empirically investigate to what extent the existence of the estimated demand asymmetries contributes to asymmetric responses of equilibrium prices of imperfectly competing firms given upstream negative and positive cost shocks. We do so by simulating positive and negative costs shocks, given the estimated demand model with asymmetric demand responses. We compare the changes in prices to changes in prices resulting from the same magnitude of cost shocks in an alternative demand structure without demand asymmetries. Our findings suggest that not allowing for asymmetries in demand imply similar magnitudes of simulated price-cost pass-through rates. However, when allowing there to be demand asymmetries, a positive cost shock is passed through to a larger degree to retail prices than a negative cost shock of the same magnitude. Our findings imply that the shape of the demand explains observed asymmetric price transmission of cost shocks in the context of imperfectly competitive markets. (Working Paper 10: ISBN 978-0-902746-27-5)

Measuring the impact of retailers marketing mix

We implement a three-way panel data model to test the effect of retailers’ marketing mix on food price inflation. Using high frequency scanner data for different dairy products in Italy we compute a weekly drift-free price index, specific for product category, chain and type of store. Exploiting the panel data structure to control for unobservable marketing mix by chain, type of store and time, we test if unobservables are statistically significant in influencing the food inflation rate on each of the products covered by our analysis. In general chain and type-of-store specific unobservables have a significant role in controlling the rise of prices. Moreover, we identify the role of some observable marketing mix variables on controlling or on facilitating price inflation rates. Results show that while higher PL shares help on limiting an upward inflation rate, reversely a higher PL line extension tends to accelerate it. Sales, as expected, alleviate the burden of a general increase in prices. However, PL sales have an effect on reducing the price inflation rate which is proportionally smaller than the overall average, indicating that proportionally sales on PL contribute less intensively to reduce a generalized upward price trend. Finally, assortment has a mixed effect depending on the competition environment of the market we refer to. (Working Paper 9: ISBN 978-0-902746-25-1)

Retailer Heterogeneity and Price Dynamics

This paper contributes to recent research on price dynamics using micro-price data sets. We emphasize a previously neglected aspect, the role of retailer heterogeneity. Our key findings are: (i) the frequency of price adjustment and the implied duration of prices varies considerably across retailers; (ii) price promotions (sales) also vary across retailers with some retailers seldom using sales, while for others sales are a common feature of pricing; (iii) the duration of reference prices is-at most-26 weeks but the duration of reference prices is around 16 weeks for some retailers; (iv) branded products have shorter durations than private label products; (v) decomposition analysis suggests price adjustment is evenly split between sales and reference prices but, for some retailers, reference prices are the main source of price changes; (vi) there is low correlation between the frequency of price and costs changes across both products and retailers. Taken together, while confirming the significance of price stickiness after accounting for sales, price dynamics vary considerably across retailers. In turn, retailer heterogeneity has important implications for interpreting aggregate price dynamics in both theoretical and empirical research. (Working Paper 8: ISBN 978-0-902746-24-4)

Multi-stage Market Power in the Italian Fresh Meat Industry

In line with the New Empirical Industrial Organisation literature, a three-stage modelling of the fresh meat industry in Italy is developed to evaluate the extent of the oligopsonistic behaviour of downstream operators on upstream ones.  Moreover, retailers are allowed to exercise oligopolistic market power over consumers purchasing three types of meats assumed substitutable in consumption.  Employing a flexible technique for estimating such a model on a uniquely compiled database, evidence that market power is mainly exercised at the retail level is unveiled.  In fact, roughly 75-85% of the price margin at the retail level can be associated with the occurrence of oligopolistic market power.  Empirical findings do not support the existence of oligopsonistic power of retailers over processors and of the latter over farmers.  (Working Paper 7: ISBN 978-0-902746-23-7)

Cost Pass Through In Differentiated Product Markets

Asymmetric cost pass through is often interpreted as an indication of market power. Since 2007 milk markets worldwide have been in turmoil. The price adjustments along the value chain in the following were closely monitored by the public and by antitrust agencies in the European Union, in particular to prove that food retailers use asymmetric cost pass through to increase average margins and profits. In this paper variations in cost pass through are analyzed between wholesale (costs) and retail prices for differentiated milk and butter products (brands) at different (individual) retail outlets in the German market from 2005 to 2008 on a weekly basis. The results indicate statistically significant asymmetric cost-price responses; however, the starting hypothesis, that asymmetric cost pass through is used more excessively by stronger brands, has to be reconsidered. Also the economic impact on average margins and profits appears to be limited. (Working Paper 6: ISBN 978-0-902746-22-0)

The Experience of Food Price Inflation Across the EU

The background to the TRANSFOP research project can be traced to the global commodity price surge of 2007 and 2008 with world prices of many agricultural and energy commodities rising, in nominal terms, to record levels. Although world prices fell back in 2009, by 2011, world commodity prices again rose and exceeded the peak levels recorded in 2008. The domestic impact of these events on global markets was reflected in higher levels of food price inflation though, despite the apparently ‘common’ nature of the global shocks, the effect on domestic food price inflation varied markedly across countries. This was perhaps most noticeable in the experience of many developing countries and certainly when contrasting the experience of emerging and developing economies with advanced economies. The general reasons for this broad experience can readily focus on the use of government policies via domestic intervention, the use of safety nets and changes in border measures (FAO, 2009). However, the experience of food price inflation has varied considerably across advanced economies and is also apparent among EU countries. Accounting for this poses significant challenges: given the existence of common trade and agricultural policies across the EU and the existence of the ‘single market’, why should the food inflation experience vary across EU countries? Identifying the factors that potentially determine the links between what happens on world markets and the resulting effect on domestic retail food prices is a key overall aim of the TRANSFOP project.  (Working Paper 5: ISBN 978-0-902746-21-3)

Sugar Policy Reform, Tax Policy and Price Transmission in the Soft Drink Industry

There is growing interest in evaluating the impact of price variations in agricultural commodities on food prices and consumption. Food industries typically consist of large firms that benefit from market power and price transmission along the chain is a¤ected by this imperfectly competitive environment. In this paper, we propose an empirical analysis of the impact of a reform of the EU sugar policy on the soft drink industry. The reform produces a significant decrease in the price of sugar. We consider how the manufacturers and the retailers both strategically react to this change in the production cost of soft drinks. Using a structural econometric model, we first estimate the consumer substitution patterns and the models for the vertical relationships between the soft drink industry and the retail industry. After selecting the 'best' model for vertical relationships, we simulate the impact of the sugar policy reform. We show that the retail prices decrease more than the marginal production costs. Our results thus suggest that the assumption of passive pricing by the industry leads to a poor estimate of the impact of an upstream cost shock. We also simulate the impact of a recently enacted excise tax on soft drinks. Because of strategic pricing, this tax is likely to lead to an increase of approximately 10% in prices thereby decreasing the soft drink consumption by more than 3 litres per person per year.  (Working Paper 4: ISBN 978-0-902746-20-6).

Estimating Market Power in a Dynamic Framework, The Case of the Italian PDO Cheese Market

In this paper, we evaluate the role of market power by retailers within the supply chain of Parmigiano Reggiano and Grana Padano, the two most famous Italian quality cheeses. Market power is analysed in the context of a dynamic imperfect competition model of the supply chain, in which retailers are allowed to exert market power both downstream and upstream. We jointly estimate market power parameters together with supply and demand elasticities, by means of a structural system of demand, supply and price transmission equations, estimated using the Generalised Method of Moments (GMM). We find evidences of downstream market power by retailers (toward final consumers) only for Parmigiano Reggiano, and no evidences of upstream market power (toward processors/ripeners).  (Working Paper 3: ISBN 978-0-902746-19-0).

Recent Developments in the Econometric Analysis of Price Transmission 

Working paper no. 2 in the TRANSFOP working paper series. In recent years, there have been a large number of studies that have applied time series econometric techniques to study price transmission covering both spatial and vertical aspects of this process. In this paper, we highlight some of the more recent developments and how they extend the now standard vector error correction models. Particular attention is given to threshold adjustment and smooth transition vector error correction models. We highlight the circumstances where such techniques would be applicable in analysing price transmission and we cover both parametric and non-parametric methods associated with functional forms associated with the adjustment process. We summarise the process of choosing the econometric methods when unit roots and co-integration do or do not exist.  (Working Paper 2: ISBN 978-0-902746-18-3).

TRANSFOP Working Paper on UK Food Price Inflation

Colleagues at the Universities of Exeter and Nottingham have produced a paper on food price inflation in the UK. The paper is the first of the TRANSFOP working paper series. The paper addresses the high levels of food inflation experienced in the UK in recent years. As the paper notes, even setting aside the rise in food inflation following the commodity price spike of 2008-2009, food price inflation in the UK exceeded non-food price inflation for over the last decade or so. Moreover, food price inflation has tended to be relatively volatile. The paper sets out to explain the factors that primarily determine food price inflation.

They use a seven variable cointegrated vector autoregression model to address these issues. Notwithstanding the fact that the commodity price spike may have influenced domestic food price inflation, the authors also note that food prices in the downstream domestic market will be affected by a broader range of factors including other costs that matter to the food industry. These include exchange rates, labour costs, oil prices and so on. The analysis shows that world commodity prices and the sterling exchange rate were the main drivers of UK food price inflation. The oil price matters but to a lesser degree. They also show that the effect of commodity prices on the domestic inflation rate depends on how long the world commodity spike lasts for; the longer-lasting the spike on world commodity markets, the greater the effect on inflation not just because of the height of the spike but also because of the duration effect. With lags in the way in which industry and consumers respond to price increases, the effect of developments on world markets feeds through so that the effect becomes cumulative over time hence contributing to the higher inflation rate.  (Working Paper 1: ISBN 978-0-902746-17-6).