study in particular several types of non linear pricing relationships with two part tariffs contracts
allowing retailers to enjoy some endogenous buyer power, and where RPM may be used or not.
The method is implemented on the market for bottles of water in France in 2006 and estimates of
demand parameters using micro-data allow us to recover price-cost margins at the manufacturer
and retailer levels for different models. We then test between the different models. Our empirical
evidence allows to conclude that manufacturers and retailers use two part tariffs contracts without
RPM and that the buyer power of retailers is endogenously determined by the upstream horizontal
competition between manufacturers. The buyer power of retailers is thus affected endogenously by
their offers from other manufacturers. Remark that the endogeneity of the buyer power of retailers
that we take into account does not come from the retailers production of private labels but only
from their strategic role in retailing and the competing manufacturers making also offers. With a
different modelling, Meza and Sudhir (2009) study how private labels affect the bargaining power
of retailers. In the present paper, the set of brands and products is taken as given including private
labels. Endogenizing entry of private labels is more complicated and out of the scope of this paper.
Finally, remark that the contracts considered here between manufacturers and retailers are
"bundling" contracts where manufacturers make take-it-or-leave-it offers to retailers for their mul-
tiple products. Considering unbundled contracts is also possible in our methodology even if more
demanding in terms of estimation. However, this alternative model is likely to reinforce the buyer
power of retailers, allowed to accept part of the brands of a manufacturer instead of the whole
bundle, a situation which would thus go even more in the direction found and the evidence that re-
tailers enjoy additional buyer power in front of their upstream competing providers. Endogenizing
the bundles of goods offered to retailers as well a the possible foreclosure effects in this industry is
thus an interesting research direction (Rey and Stigliz, 1995, Rey and Tirole, 2007). The markets
for bottles of water in France does not seem to be importantly affected by such strategies but other
markets are (Asker, 2005) and further work needs to be done in this direction.
34