NZCC | Prices & margins in New Zealand's retail grocery sector

15 July 2021

What was the challenge?

In November 2020, the New Zealand Government asked the Commerce Commission to carry out a study into whether competition in the grocery sector is working well, and if not, what could be done to improve it.

As part of its market study in the grocery sector of New Zealand, the Commerce Commission commissioned us to undertake some econometric analysis to examine how the structure of local grocery markets affects prices and margins in New Zealand. Economic theory suggests that the structure of a market is likely to influence the prices and margins in a market.

How did we complete the economic modelling and analysis?

Economic theory yields a large range of propositions concerning the relation between elements of market structure and prices/margins. However, our analysis consisted of two studies, each of which focused on a particular issue.

The first study tested whether there is a relationship between concentration and prices in local grocery markets. We conducted the test by comparing the way in which variations in concentration across local grocery markets are related to variations in prices. We were provided with weekly store-level stock-keeping unit (SKU) data for 694 stores of the major grocery retailers: Foodstuffs North Island, Foodstuffs South Island, and Woolworths NZ.

The second study analysed the impact of the entry, exit or rebranded entry on the prices charged by close competitor stores in the 12 months following the event. We used a sample of 32 events of entry, exit or rebranding. These events had a good geographic spread with different demographic characteristics, such as population density.

What were the results?

For the first study testing whether there is a relationship between concentration and prices in local grocery markets, the results of the tests vary somewhat depending on the measure of market concentration.

In models with weighted revenue share as the concentration variable, we found that increased concentration in local markets tended to be associated with higher prices in the cases of Four Square, New World and SuperValue.
In models with Proximity to major banners as the concentration measure, we found that Four Square stores tended to have lower prices when a FreshChoice or SuperValue store is nearby; and Countdown, FreshChoice and SuperValue stores tended to have lower prices when a New World store was nearby.
Relationships between stores of other banners are less clear. We found that Proximity to stores of non-competitor banners in the same group tended to increase prices for certain pairs of stores but not for others. In particular, prices at Four Square stores tended to be higher if a PAK’nSAVE store was nearby; and prices at SuperValue stores tended to be higher when a Countdown or FreshChoice store was nearby.

For the second study, we found that there was no systematic relationship between the entry or exit of a store and the prices charged by close competitor stores in the 12 months after entry.

DOWNLOAD FULL REPORT