Academic journal article Economic Inquiry

Do Mergers Really Reduce Costs? Evidence from Hospitals

Academic journal article Economic Inquiry

Do Mergers Really Reduce Costs? Evidence from Hospitals

Article excerpt


Cost reductions through consolidation are a frequent argument made in favor of mergers. Yet, as discussed by Caves (1989) and Dickerson, Gibson, and Tsakalotos (1997), ex ante efficiency may not materialize into ex post efficiency. In addition, other motivations for mergers unrelated to cost savings, such as increasing market power, may be the larger catalyst for consolidation. Thus, increased postmerger costs could occur because cost reductions were not possible or not the goal of the merger or because cost reductions were not realized. In this paper, we focus specifically on cost savings due to scale economies and examine whether economies of scale exist for the premerging firms and then whether they capitalize on these potential cost savings post-merger. (1) Our paper focuses on hospital mergers, but these techniques could be applied to any industry to better differentiate between potential and realized cost reductions.

Hospital consolidations provide a rich sample in which to study post-merger cost reductions. In the 1990s, part of our sample, the hospital industry experienced a flurry of merger activity. Prior to a merger, hospital executives, like in other industries (Pesendorfer, 2003, p. 502), often claim that costs will decrease after the merger due to gains in economic efficiency. The courts have generally agreed with this claim and used it as part of the reason to approve the merger (Bellandi, 1997). Given that hospital costs are a significant portion of overall healthcare spending, it is crucial to understand whether mergers can decrease hospital costs.

Assessing cost benefits from a merger falls into a larger body of literature on identifying average treatment effects where an outcome of interest for the treated group is compared to the outcome for a control group. The question of interest in this paper is whether the merger decreased costs. Ideally, we would like to know the costs for the merging firms if they had not merged. But of course, we never observe this event. Consistent with the treatment literature, previous merger studies across different industries have therefore focused on cost reductions for merging firms relative to nonmerging firms (Connor, Feldman, and Dowd, 1998; Pesendorfer, 2003; Engberg et al., 2004; Valverde and Humphrey, 2004; Dranove and Lindrooth, 2003). However, the difficulties of identifying the post-merger cost benefits are compounded by the fact that the merger decision is inherently endogenous, leading to possible sample selection bias. In hospital studies, Connor, Feldman, and Dowd (1998) and Dranove and Lindrooth (2003) both use a parametric difference-in-difference approach to control for observable and unobservable characteristics that affect hospital costs. In order to mitigate the sample selection bias, Dranove and Lindrooth (2003) additionally estimate the merger decision and use the propensity scores from this first stage regression to pair merging hospitals to similar nonmerging hospitals. Connor, Feldman, and Dowd (1998) find that merging hospitals have 5% lower costs while Dranove and Lindrooth (2003) find significant cost savings (14%) for mergers between independent hospitals.

The question still remains whether the merging hospitals realized the potential ex ante cost savings. That is, could they have done even better? This issue arises because the output levels and characteristics of the hospitals do not remain constant between the pre- and post-merger time periods as required within the treatment literature (Imbens, 2004). In fact, some of the changes to output and changes to institutional structures like ownership and teaching status are precipitated by the merger itself. Therefore, the mechanism by which mergers reduce costs is unclear.

The actual reporting entities for the firms also differ pre- and post-merger. The American Hospital Association (AHA) identifies a merger when hospital A and hospital B consolidate to form hospital C, and similarly denotes an acquisition if hospital A merges into hospital B. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.