concentration as exogenous while the other allows the market concentration to be
endogenous.
Identification Strategy: We seek to identify market concentration with an
instrument that is constructed by the interaction of total number of mergers in the US
economy with segment specific capital asset intensity in processed food industry. The
ratio of net property, plant and equipments to annual sales provides the measure of capital
asset intensity that varies across the different segments of the processed food industry as
well as over time. The US mergers represent a pure time series variable i.e. has only time
variations and hence is expected to help in identifying the temporal variation. The capital
asset intensity measure that varies across the processed food industry segments as well as
time is expected to help in identifying temporal as well as cross-sectional variations in
market concentration. Hence our instrumental variable, an interaction of a time variant
variable with a cross-section cum time variant variable, is a potentially good instrument
to identify market concentration in our panel data model.
Since, merger waves in an economy affects mergers across all the sectors, hence
we expect overall trend of mergers in the US economy to affect the mergers in the US
processed food industry and hence the market concentration in the various segments as
well. At a first glance one might expect this variable to get affected by NPIs as the
anticipatory mergers theory argues that NPIs occur in anticipation of future mergers in
the specific industry. However, the fact that the processed food industry is a very small
proportion of the US economy, rules out the possibility of NPIs in food industry
influencing overall mergers in the economy through its influence on mergers in the
processed food industry. Hence we can argue that aggregate US mergers can affect NPIs
10