period about 17 % of the 10 food items under study are put on sale, which roughly means 2 out
of ten products. This share does not vary substantially over time. Further, sales are focused
temporarily in single or particular weeks, but appear to be spread almost equally across time.
Even though the aggregate series across all store types shows some slight temporal
autocorrelation, it is likely insignificant from an economic point of view. The uniform
distribution of sales across time also suggests that at this level of aggregation we do not find
synchronisation of promotional measures between shops.16 This feature also occurs for store
type aggregates (see the graphs for SSM and BSM in Figure 2). In addition, these store type
aggregate series are not found to be significantly correlated across store type. Only SSM and
BSM, and BSM and CSM store types are found to some correlation (0.15). When the indicator
of sales is compared across different product categories, no systematic differences are found. An
exception is that for fruits and vegetables some correlation of sales across store type is found.
On average sales occur a little more frequent in BSM and CSM store types. In conclusion, our
indicator of sales suggests that sales are not coordinated across stores or correlated across time.
Instead, we find that a significant number of sales occur each week. The number of sales does
not differ across products and differs only slightly across store types. <Insert Figure 2 about
here> Next, we examine the impact of sales on the expenditures of consumers. Of interest is
whether sales reduce consumer expenditures. To proceed, we compute average per capita
expenditures for the basket of food products examined in the sample. Because price changes for
high value goods are larger in magnitude, we distinguish sales for meats from those for fruits
and vegetables. Further, we consider the wholesale price paid for the product to affect the retail
price and thereby the expenditures.17 To construct the average per capita expenditures or the
retail price index, we use the average per capita consumption data for Germany for the product
categories under study.18 For each store, the weighted retail price index is a calculated as:
9
ptI= ∑ ptiqi.19 The same operation is employed to calculate a store level wholesale price index
i=1
that is interpreted as a measure of costs faced by the store for the product. Because of seasonal
variations in the prices of fruits we also incorporate monthly seasonal dummies in the model
specification. To summarize, we hypothesize the following model:
pRττ = α0 + α1 pWP + α2 SM + α3 Sf & V + ∑ βjD1t + εt. The left-hand side variable is the retail price
index. The right-hand side variables by order of appearance are the wholesale price index, the
number of sales for meats, the number of sales for fruits and vegetables, and a set of seasonal
monthly dummies. We assume these variables to be exogenous determinants of the retail price.
As price changes and very likely thereby sales have a much higher impact in the case of meats
compared to fruits and vegetables, we did not aggregate the number of sales over all products.
We estimate this model for independently for each store type (SSM, BSM, CSM, DC).20 To
16
17
18
19
20
The standard deviation of the share of sales can be employed to measure the extent of synchronisation. An
increase of the standard deviation indicates a higher synchronization. However, if we compare the measure
for different aggregates (groups of stores) we have to consider the number of observations between
aggregates. As the measure has a tendency to decline with the number of observations.
As we assume the quantities to be fixed we expect a direct impact on the expenditures. In this view we
could also define our measure of expenditures as a particular type of price index.
While for items such as lettuce this measure might be very close to the actual quantities bought of that
item, other products such as beef are consumed through many other items but fresh beef of steak quality.
For instance, fresh beef of other qualities, packed beef, etc.
We use here only 9 of the ten food items as for sausages neither an average per capita consumption
measure nor a wholesale price could be found.
We expect at least for the level of the price index (expenditures) differences between the store types, as
DC are generally much cheaper than e.g. BSM. To test for these or other potential deviations in the