Saturday, June 14, 2014

When to Merge? Looking at External and Internal Relations

The merger between advertising giants Omnicom and Publicis created news for a long time, until the day when the cancellation of their merger created news. What happened? The story reported by Wall Street Journal involved many operational issues such as incorporation, choice of who firm to make the formal acquisition, and regulatory approval, but it was also pretty clear that the relations between the firms had become problematic. The CEOs clashed over a number of issues related to the new organization, such as its location (Omnicom is a US firm, Publicis is French) and key staffing choices. In the end, the firms called off the plans and both CEOs admitted that the relation between the potential merger partners had not been good enough.

Is that a good reason for calling off a merger? Possibly, but it is one that gets too much focus because such internal relations are relatively small-scale and temporary. For example, one of the CEOs, Publicis’s Levy, was supposed to retire soon but had not done so because of problems finding a successor. But relations are still important for merger success, except they are a different kind of relation. All firms have relations with other firms, as alliance partners, suppliers, or customers. Not all relations are important, but some are, like key client relations are for advertising firms. Managers pay surprisingly little attention to what kind of relations would be best for a merger, and even researchers have overlooked the issue.

A recent article by Michelle Rogan and Olav Sorenson in Administrative Science Quarterly addresses it by looking at mergers, and in fact mergers among advertising firms. Their focus is on whether firms are more likely to merge with each other if they share clients, and whether mergers have lower performance when the firms share clients. The reasoning is simple. Shared clients means familiarity because the firms are close competitors, and it can also mean over-confidence in the results of the merger. Shared clients also means that little new is added by the merger, because the merged firm gets a deeper relation with existing clients rather than a broader set of clients. Two problems follow. First, the client may not want to have a deeper relation because it sees the advertising firm as trying to gain power (Michelle Rogan and I have an article about this). Second, the firm will fail to build complementarities in its client portfolio, which can hold back innovations (my book with AndrewShipilov and Tim Rowley discusses this). So far theory.

What did they find? Evidence showed that the theory was correct on both accounts, meaning that the firms made exactly the wrong mergers. They merged when sharing clients, and shared clients meant that the performance was reduced after mergers, both when looking at loss of clients and in looking at billings per client. 

So the conclusion is clear. When looking at whether to merge or not, relations really matter. Except that the relations that matter are between firms, not between CEOs.