An Empirical Analysis of the Counterfactual: A Merger and Divestiture in the Australian Cigarette Industry

Abstract

In this paper we empirically analyse two counterfactual situations facing an anti-trust authority following the merger of two of the largest international cigarette companies. First we estimate a nested logit model of demand for cigarettes. The implied elasticity of demand for smoking and implied marginal costs are both broadly consistent with the limited independent estimates available. We then use the model to simulate the proposed merger and the partial divestiture that was accepted by the Australian anti-trust authority. A comparison of the relative price changes predicted by the divestiture simulation with the actual post-divestiture price changes shows the model successfully anticipated the behaviour of the divested brands.This suggests structural econometric analysis using a nested logit may be usefully utilised by anti-trust authorities to assess the welfare implications of proposed mergers and partial divestitures.

A Random Coefficients Logit Analysis of the Counterfactual: A Merger and Divestiture in the Australian Cigarette Industry

“…Finally, we compare our results to those of Pham and Prentice (2010) and Friberg and Romahn (2012). The pattern of predictions at the company level in Pham and Prentice (2010) are fairly similar in that the largest changes were predicted for BATA and that Philip Morris and ITA would experience smaller changes or even price cuts. It is not straightforward to compare at the brand level because of the dierent levels of aggregation in estimation and reporting.…”

“…1. Pham and Prentice (2010) assess the performance of a nested logit model in simulating the same divestiture we consider. Jayaratne and Shapiro (2000) and Tenn and Yun (2011) also empirically analyse divestitures.…”

Will Plain Packaging Reduce Cigarette Consumption?*

“…Finally, the behaviour of market shares is examined. These exercises are implemented using data on prices, market shares and brands in Pham and Prentice (2010) supplemented with recent data. See for more details.…”

“… Sources : 2001 to 2010 Euromonitor Database. 1970–1995 documents from the Legacy Database as used in Pham and Prentice (2010). License rate is that applicable to the largest group of states. …”

“…18 18 Specifically, Pham and Prentice (2010) estimate a value for the parameter that captures this, σ, of 0.72, whereas Padilla (2010) varies σ between 0.1 and 0.9. …”