The author is professor of Strategy at IMD and holds a PhD in Strategy from University of Western Ontario, Canada.
Many leading companies have dramatically expanded their alliance activities with competitors in the past few years. This trend is particularly prominent in consolidating global industries such as airlines, telecoms, automobiles and chemicals, but also in rapidly expanding Internet-related industries featuring players like Cisco, Microsoft and AOL.
But there are two very different kinds of alliances in use. One might be called a flirtation -- a low commitment, shallow alliance that has not had a lot of resources devoted to it, and can be broken on short notice.
As an example, think of the current airline alliances, which seem to feature new partners every month. At the other end of the spectrum are much deeper alliances involving higher levels of financial and managerial commitment by the partners. General Motors built such an alliance with Fiat in 2000.
From a strategic point of view, entering an alliance with a competitor is a risky and difficult proposition. The risk, of course, is that your ally of today may again be your outright competitor tomorrow -- now strengthened with knowledge of your technology, your markets, and your way of operating.
But refusing to enter into an alliance with a competitor, remaining alone in a consolidating industry -- like Honda and BMW are doing in the auto industry -- carries its own risks. Will they be big enough to survive?
If you are considering an alliance with a competitor you need to decide how deep your alliance should be. Deep alliances are generally slower moving than shallow alliances, more difficult to manage, more difficult to end and carry more risks.
But in some circumstances they also offer more potential rewards. In this article the differences between deep and shallow alliances are explained, and advice is offered on when you might want to use each type.
Deep alliances
Think of a deep alliance as a broadband linkage between two firms. There are many cooperative activities being undertaken (see figure 01), many points of contact between the firms, and many people in the two firms are working together.
Even the CEO's of the partner companies probably meet on a regular basis to discuss unresolved partnership issues and to discuss new areas where the firms might work together. A lot is at stake in the success of a deep alliance, and the commitment level on both sides is high.
Deep alliances may be developed over a period of years, or they may be 'deep' from birth. The alliance between Ford and Mazda, for example, developed from a shallow alliance to a deep one over a period of more than twenty years, and ended with Ford taking (friendly) control of Mazda.
The alliance between General Motors and Fiat, on the other hand, sprang into life in the year 2000 as a deep alliance.
The advantage of Ford's approach is that the partners can build trust gradually, before adding new activities and committing more resources to the partnership. The disadvantage is that it can be a long time before significant benefits are realised from the alliance.
The advantage of General Motors' approach is that the strategic and operating benefits of the alliance are available almost immediately, if the relationship between the partners is managed well.
It's a higher risk, potentially higher reward, approach to alliance building. The risk is that managers are being asked to share information and work together on high stake projects before they know each other, trust each other, or in some cases are even convinced that the alliance is a good idea.
The usual elements of a deep alliance are shown in figure 01.
Cross ownership: Cross ownership between partners, or even a one-way ownership position, can bring significant strategic benefits. If the financial investments necessary to create the cross ownership positions are significant, they send a signal to the outside world, and to the employees of both companies, that top management is serious about the alliance.
The investments keep other potential partners, or potential acquirers, at bay. GM's 20% equity position in Fiat Auto, combined with Fiat's approximately 5% ownership position in GM, creates a strong link between the companies. The disadvantage is that such links can be hard to break if things do not workout.
GM is discovering this in 2005 as it tries to reduce its commitments to the financially troubled Fiat. Ownership positions are often accompanied by one or more seats on the partner firm's board of directors.
When asked about the value of having a seat on the board of a partner company, one senior manager replied, "Firstly, the symbolism is of major importance, both inside and outside the company. In practical terms it also gives us an early look at our partner's plans for the future. If they are looking for a partner for a new venture, it gives us the right of first refusal to cooperate with them. We might exercise it, and we might not. Just because we are their partner does not mean that we will fall in love with all of their proposals."
Equally balanced joint ventures
While cross shareholding positions signal to the industry that the alliance is serious, the accompanying joint ventures are where a lot of the cooperative work is done, and the immediate financial value of the alliance will be realised. These might be 50-50 joint ventures, or perhaps offsetting 51-49 ventures, with one partner taking the lead in each.
GM and Fiat, for example, have created two 50-50 joint ventures, one to create economies in purchasing, and the other to supply powertrains to both parents. Between the two ventures, the companies estimate a first year saving of 500mn euros, rising to 2bn euros by 2005.
Of course, setting up joint ventures is easier than making them work. Employees in either partner company may feel threatened and resist any move to share technology or market knowledge or standardise product specifications with a competitor.
Consider Concert, the huge 50-50 joint venture that combines the international operations of AT&T and British Telecom. In the early months of this newly created JV, major product and market decisions have to be made, and relationships sorted out between companies that do not yet know each other well and have not had time to build trust.
According to one observer, this JV is not going well because the competitive instincts of the parent units are getting in the way. The parents reportedly will only cooperate in areas that are not crucial to the JV's success. Concert was said to be 'suffering from parental abuse.'
These competitors, like many others, see the strategic need for a deep alliance, but when it comes down to day-today decision-making, ambivalence sets in.
Smaller projects, quick wins
Once two companies have decided to work together, a number of apparently attractive small joint projects will probably emerge. Some of these will have a significant strategic or financial impact, some not. I recently visited a manager in a six-month-old deep alliance, who had more than fifty such projects on his desk, and was trying to decide how to prioritize them.
He decided that one of the criteria he needed to consider was whether or not a project would offer 'quick wins'. These were projects that 01 had a high probability of success, and 02 would produce an immediate payoff.
His reasoning was that such quick wins, made visible in both companies, would encourage everyone to work harder at solving the more difficult situations that he was sure were going to emerge in the larger and more important joint ventures that were being established. Thus wins in smaller pieces of the alliance might increase commitment in some of the larger, higher stake components.
Shallow alliances, 'traveling light'
A manager in a new economy company recently said, "Alliances are important to us, but we need to move fast and travel light. We do not want big cumbersome alliances with committees at three levels that get things done eventually. We use flexible alliances to give us a look at smaller competitors - if we like what we see, we will probably buy them. If we do not, we may sell our share or write it off. Exceptionally, we will do some joint development work with them."
As this manager suggests, shallow alliances are often used by companies that want to create options in fast changing industries where the way ahead is not clear. For example, a competitor's fledgling technology may or may not prove to be important. The shallow alliance solution is to buy 10% of the competitor's stock in a friendly transaction, get a seat on the board, and maybe at the same time create an option to buy the remainder of the equity.
The assigned board member, from a privileged position, can now assess the management of the company, its market prospects, its emerging technology and so on. The board seat provides information not available to others. The alliance is not intended to be permanent.
On the other hand, you may be interested in forming a deep alliance with a competitor, but are not sure if the two companies would be able to work together effectively. Perhaps the cultural differences between you are huge. A shallow alliance -- perhaps a single joint project -- may be a sensible first step. If it works, you add something a little more complex, and a little more important.
Ford and Mazda started with a single joint project in the early 1970's, but by the 1990's there were more than fifty separate cooperative deals between the companies.
Of course the labels 'shallow' and 'deep' are just the end points of a continuum, and there is a lot of middle ground that you can move through on your way from a shallow alliance to the richness of a deep alliance.
The major airline alliances such as One World or the Star alliance, for example, are currently rather shallow, encompassing code sharing, coordinated schedules, combined frequent flyer plans and joint airport lounges.
To become deeper they would need to move to the joint ownership of aircraft, and perhaps equity ownership between the participating airlines. If that were to happen, these strategic groups would really start to mean something, and leaving one alliance to join another would become much more difficult.
The common elements of shallow alliances are shown in figure 02.
Alliances as stepping stones
Alliances between competitors are usually not a final destination. They are often stepping stones to something else. Ford took control of Mazda, its long time partner. GM will probably buy Fiat Auto in a few years. BT bought out its joint venture partners in the Netherlands and Germany.
The list continues. Many companies that have an alliance strategy, or are building a network of alliances, often really have an acquisition strategy in waiting. That is to say, they would prefer to acquire their partner, or all of their joint venture, but for one reason or another, cannot.
If you enter into an alliance with a competitor, be aware of the competitor's strategic objective. Do they want to learn from you - and then exit the alliance? Do they want to exit the business area in question and hope that you will gradually take over 100% of the joint venture? Do they want to buy you? Assume that your alliance is temporary, and act accordingly. It probably will be.
Recommendations
If you are planning an alliance with a competitor, consider the following:
1. Recognize the differences between shallow and deep alliances and decide which makes most sense for you. If you are in a consolidating industry, and you can see the future with some clarity, a deep alliance may make sense. If you cannot see the future well, avoid deep alliances. You may make a wrong choice of partner and be trapped in an inappropriate relationship. Shallow alliances are best when you are trying to create options for the future, not achieve economies of scale or extend your global reach.
2. Remember that your ally of today may become your outright competitor again tomorrow. Decide in advance what you are, and are not, willing to share. Misplaced trust can be very dangerous.
3. Have clear strategic objectives when you are entering an alliance, and do the best you can to determine your partner's real strategic objectives. The operational payoffs of working together, such as cost savings, will usually be obvious. The longer-term game may be more difficult to figure out, but is usually more important.
Powered by 
Published with the kind permission of The Smart Manager, India's first world class management magazine, available bi-monthly.
More from rediff