regards the positive correlation between brand- name prices and the share of private-
label goods. But indeed, this is clearly predicted by the hypothesis of upgrading and
increasing vertical diffentiation so that this evidence support one of the stylized fact
predicted by our framework.
ortant to note that these empirical patterns are substantially similar what Ward et al.
(2002) refer to as the “conventional industry wisdom”, namely the idea that brand
manufacturers “defend their brands against private- label products by lowering their
prices, engaging in additional promotional activities, and increasingly differentiating
their products [..and that] the second- tier national brands [are] particularly hard hit
[by the growth of private labels]”. Ward et al. (2002), however, introduce a caveat about
this conventional wisdom arguing that these facts fail a more rigorous empirical test.
Their empirical results show that many of these stylized facts are not corrently true. In
particular, they find that larger private- label share leads to higher brand- name prices,
there is a pronounced downward trend in promotional activities and that
differentiation does not increase with vertical competition.
Clearly, this last result would not support our prediction. However, the measure of
differentiation they use - the number of items per firm - does not seem a good proxy
for true differentiation.
Their most robust empirical result regards the positive correlation between
manufacturer brand prices and the share of private- label goods. But indeed, this is
clearly predicted by the hypothesis of upgrading and increasing vertical diffentiation
as a consequence of vertical competition. In fact, there is now increasing empirical
evidence supporting the hypothesis that premium- tier national brands could build on
their strength by introducing high- end product varieties, which increases average
brand price (Pauwels and Srinivasan, 2004; Gruca et al. 2001; Hauser and Shugan,
1983). Similar findings are provided by Bontems et al. (2005) who examined the
changes in national brand product characteristics induced by the development of
private labels. They distinguish between three types of private labels (low price, “me-
too” and high quality private labels. Their results indicate that the effect of private
label expansion is different according to the type of the private label. Interestingly,
they show that the increase in national brand prices is partly explained by the strategy
of product differentiation developed by manufacturers to reposition national brands.
More precisely, they find that an increase in private label market share incentives
suppliers to changes the characteristics of their products and the increase in national
brand prices is explained by the channges in product repositioning.
Indeed, Ward et al. (2002) point out that the simplest way to explain price increases in
response to private- label entry is that brand manufacturers may raise the quality of
their goods when faced with private- label entry. But this means that it is possible to
read their empirical findings as supporting an aspect of the ‘conventional wisdom’
which is also one of the key predictions of our framework.
5. Summary and Concluding Remarks
Retailing concentration and the development of store brands are profoundly changing
vertical relationships between food manufacturers and retailers. We have examined the
consequences of vertical competition between manufacturers’ brands and retailers’
23