Google and Facebook have become huge rivals in the digital ad space with the tech giants hoovering up 70 per cent of online advertising in the US between them, according to Marketwatch.

While in the UK eMarketer calculates the duopoly garnered 67.8 per cent of the digital advertising market in 2019.

And yet a recent expose in The New York Times revealed the US pair had signed an agreement to use a new system of selling online advertising. It meant Facebook enjoyed some benefits on the Google-controlled system that other companies did not and stopped Mark Zuckerberg’s company building a rival system.

But the deal between two tech behemoths is not as unusual as the report in The New York Times makes out.

In fact my research has found that companies are more likely to strike a partnership the more overlap across markets they have. If two firms are battling closely over several markets, it is more likely that they will enter into a partnership.

And these partnerships work precisely because they are such close rivals. Neither side deviates from the agreement or takes advantage of working so closely together because the fear of retaliation is too great.

Imagine, now, if Google decided to change the system and dial down the benefits to Facebook, with Zuckerburg’s firm losing ad revenue in the US. The social media giant could quickly retaliate by launching its own digital ad exchange, which The New York Times mentions in its report, in another important market, such as the UK.

This mutual understanding holds the partnership together and for the long-term benefit of both firms, who can spare themselves the extra cost of researching and creating new digital ad exchanges.

We studied technology partnerships among the world’s top 200 biopharmaceutical companies and advanced a theory that debunks the intuition that competing in the same markets deters alliances and necessarily undermines the co-operation they require.

Instead, we found it a strong predictor of partnerships being formed due to what economists term ‘mutual hostages’. This is where both parties know they would lose the value of their investments if they did not co-operate. Neither side wants to trigger retaliation in other shared markets, so it is a way to bind the companies together and align their incentives to foster co-operation.

This discovery adds to the established view that such alliances are made because complementary resources are behind firms joining together in a partnership.

How to choose the right company to partner with

‘Mutual forbearance’ as we call it, is a powerful force and the main one, when companies choose with whom to partner, especially in R&D projects. This is because there is so much uncertainty as to what will be found and who owns the intellectual property, a lot can’t be foreseen in a contract. Thus, choosing a rival where the threat of retaliation exists if either party steps out of line is a good way of keeping any temptation to take advantage of unforeseen discoveries in check.

In fact, when choosing a partner, the more shared markets the better. The more markets that the two companies compete together in the more likely they are to behave in the partnership, as there would be more opportunity for retaliation and a wider scope to cause serious financial damage with a price war, to name one tactic.

It is also important to take into account how big a share each rival has in each market. Mutual forbearance does not work if one of the parties has a large share in all the markets and its rival is relatively smaller; any retaliation, such as a price cut, will be far costlier to the one with the large market share. Thus, it is important that both parties are not only in the same markets but both have smaller shares to be able to inflict damage at low cost.

In the 200 biopharmaceutical firms we studied we found 119,408 new technology partnerships agreed between pairs of companies from 2008 to 2013. By factoring out other variables we found that as two potential partners have a greater level of multimarket contact, they are more likely to select each other as partners in the development of new technology.

In fact we found that the more markets the partners are involved in the more valuable the partnership becomes, with twice as many market overlaps than the average leading to a 168 per cent increase in economic value.

This stands to reason, with more markets for your rival to attack the risk of retaliation is higher and we found the more markets that are reciprocal - where one has a large share and the other a small one - also raised the likelihood of an alliance.

Ideally any partnership will see the two players not just involved in the same markets but also have an equal number where each one is dominant. This will leave both knowing any misbehaving is bound to cause a severe retaliation; nobody wants to trigger this and so the partnership is more likely to succeed.

In today’s fast-moving world where the pace of change means partnering with a rival on technological advances are an important way of keeping ahead of new entrants, choosing the right partner is vital and that might even be a close rival.

Further reading:

Ryu, W., Reuer, J. and Brush, T. H. (2020) "The effects of multimarket contact on partner selection for technology cooperation", Strategic Management Journal.

Lioukas, C. S. and Reuer, J. J. (2020) "Choosing between safeguards: scope and governance decisions in R&D alliances", Journal of Management.

Reuer, J., Mayer, K., Xing, A., Lie, X. and Klijn, E. (2020) "The effects of contract detail and prior ties on contract change: a learning perspective", Organization Science.

Jeff Reuer is WBS Distinguished Research Environment Professor at Warwick Business School and the Guggenheim Endowed Chair and Professor of Strategy and Entrepreneurship at the University of Colorado.

For more articles on Strategy sign up to Core Insights here.