Performance effects of corporate divestiture programs

Abstract

Purpose The paper aims at extending extant research on sources of divestiture gains by suggesting a novel program‐based perspective on divestitures and analyzing the performance of program divestitures in comparison to single “stand‐alone” divestitures. Design/methodology/approach Based on event study methodology, the authors analyze the abnormal returns of 160 divestiture announcements within the global insurance industry between 1998 and 2007. In contrast to prior research which relied on ex post statistical clustering to identify transaction programs, ad hoc corporate press releases issued with the divestiture announcements are used to categorize program divestitures. Findings Empirical results suggest that program divestitures generate higher abnormal returns than stand‐alone divestitures. Further analyses into the sources for these higher gains, however, do not provide support for experience effects as significant explanatory factors. Instead, results suggest that the scheduling of divestitures significantly impacts announcement returns. Research limitations/implications The scope and single industry setting of the study suggest future cross‐industry research on the influence of divestiture program characteristics on divestiture performance and the conditions under which these programs improve divestiture performance. Practical implications Managers are advised to refrain from piecemeal divestiture behavior lacking clear strategic focus. Instead, they are encouraged to bundle their divestitures as part of a divestiture program with a clear strategic intent and shared business logic. Originality/value While prior research on divestitures has treated divestitures as isolated events, the paper directs attention towards the analysis of divestiture programs. Further, experience and timing effects, which have been widely absent from prior divestiture studies, are considered.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 24 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.