KEVIN O'TOOLE A new alliance study carried out by Gemini Consulting, in collaboration with Airline Business, suggests that global groupings may have to adopt more than one model for their relationships.

The last time that the airline industry indulged in a bout of alliance-forming at the start of the 1990s, many of the groupings dissolved in the wake of internal power struggles. This time around, the attempt to build broader global alliances appears to be doing better. But will they last?

Signing up partners has always been easy enough. It is in settling down to manage the relationships over the long haul that alliances have tended to come unstuck. To succeed, argue the sceptics, the new breed of global groupings will have to tread an impossibly fine line through a maze of competing interests. Too much integration risks internal power struggles and divorce. Too little and the relationships start to drift.

But Gemini Consulting questions whether the global groupings should indeed be making the attempt to bring all of the partners along at the same speed. In a new global alliances study carried out in collaboration with Airline Business, the conclusion would seem to be that they should not.

Gemini vice-presidents Graham Howarth and Thomas Kirsebom, who led the work, argue that it is precisely this "one-size-fits-all" approach to alliance formation which could emerge as a key cause of instability. Signing all partners to a single set of agreements, assumes that each relationship offers the same benefits and opportunities. In reality, the global groupings are simply too broad for that to be the case.

Deregulation, which is driving much of the rush towards globalisation, is itself at different stages of development around the world. That not only establishes the extent to which integration is possible - such as hard limits on foreign ownership - but also the degree to which it is necessary.

Take two examples. In a fragmented European market, deregulation has cleared the way for flag carriers to consolidate their largely overlapping operations, eventually moving toward merger and acquisition. Such logic already appears to be in operation as the SAirGroup effects a "virtual" merger between Swissair and 49%-owned Sabena, bringing the businesses under a common organisational structure. The alliance between KLM and Alitalia also aims to create a merger in all but name. Gemini doubts that any looser model of co-operation would be workable in the long term - more of which later.

Across the Atlantic the model changes. Partners have much to lose, and little to gain, from forcing the issues of ownership and control, as bitter experience has shown. Attempts by British Airways to squeeze value out of its minority stake in US Airways ended in court. KLM's boardroom power struggle with Northwest Airlines came close to a similar fate. KLM eventually agreed to rid itself of the controversial equity holding and BA opted for a marketing tie with American Airlines. In neither case have the core marketing benefits been lost.

Global alliance models need to be able to allow for these differences in intra-regional and long-haul relationships. "What clearly will not work in the long term is having a single model for both," says Kirsebom.

The airline industry is not the first to face the challenge. Sectors as diverse as telecommunications and oil and gas are now feeling their way through deregulated markets. "Alliances which are established under one set of market rules can quickly change once another set of rules are applied," says Kirsebom.

He cites recent experience in the upstream oil and gas sector where some initial testing of alliances and joint ventures has just begun to build towards outright mergers. Closer to home, the aerospace industry has been going through a similar process as corporations seek out economy of scale. Only at the end point, in a fully deregulated world market, does the issue become the creation and management of global business systems. And for aviation that remains a distant goal.

The alliance study, which was based around detailed interviews with more than 30of the executives who are leading the emerging global groupings, reveals that the airline industry is largely at the testing stage. Few believe that the current shape of the global groupings represents anything close to the final endgame. "Most are taking a relatively progressive approach, feeling their way forward rather than looking for some big bang," says Howarth, adding: "There was a recognition that in an ideal world they would not necessarily have started from here."

As strategists themselves concede, the opening moves in the latest round of global alliance activity were dictated as much by horse sense as long-range strategy. The marketing advantages of tying together existing bilaterals under a global marketing umbrella were not too hard to figure. And once the game had been put into play, the trend gathered a momentum of its own. None of the major players could afford to be left out. "We've found evidence that the decision-making process to join an alliance was often intuitive or opportunistic," says Howarth, adding that a high degree of "defensive logic" was also evident, as carriers contemplated the risks of life on the outside. Some executives even concede that fashion played its part.

Towards a new model

If the opening moves drove themselves, then the next strategic steps towards integration are going to be more testing. The core of the Gemini study is an attempt to unravel the complex blend of relationships and external pressures into a more coherent model of how global alliances are to be managed from here.

A starting point was to set down some benchmarks for how far integration has progressed along a continuum ranging from no more than shared experience through to common control and possibly ownership. That led to the creation of three basic models, effectively reflecting the depth of commitment being made between the partners:

Co-ordination; Shared strategy and operation; Serious equity partnership.

The co-ordination model is focused on the early and comparatively straightforward gains to be made on the revenue side through the ability to market a global network under the broad umbrella of a single brand. There may be additional gains to be made from co-ordinating capacity, but essentially the drive is to expand the reach of the network. The oneworld alliance, lacking the anti-trust immunity to go further on the transatlantic, was founded exactly around such benefits: a single ticket, a global brand and links on frequent flier programmes (FFPs).

The next model raises the commitment to take in deeper sharing of operations and strategies. This more profound level of co-operation promises to raise the level of revenue enhancements, through tighter joint co-ordination of scheduling, marketing and sales. More significantly, it paves the way towards realising some, though by no means all, of the hard cost benefits that can flow from economies of scale. Selective sharing of airport facilities and services, including information technology, are on offer as the partners move closer together. The Star Alliance, for example, has already begun to explore such benefits.

The final stage in achieving those cost benefits comes with a fully unified model, characterised by Gemini as a "serious equity partnership" - by which it implies something more committed than an ineffective minority stake. Here the ambition is to create a new organisational structure, with most of the business being run under common control. The SAirGroup has begun to pioneer such a structure for Swissair/Sabena.

At this stage, the alliance members are managing costs and investment as a single entity, with long-term tie-ins for the partners. In effect it amounts to a virtual merger, even if the delicacies of national pride and politics dictate that it should avoid looking too obviously like one.

Within each of the existing airline alliances a combination of these different models is at play. Neither are they evenly spread across the core functions within an alliance (see charts). Early progress has inevitably concentrated on the collaborative model. Where there is deeper integration it has tended to come in the less contentious process of sharing services such as maintenance or joint purchasing.

The ambitions of the alliance leaders demonstrates that they plan to press ahead towards greater integration across almost all core functions. Admittedly progress may be uneven. None, for example, show any relish for tackling the issue of cockpit crew. Some partnerships will remain limited by regulatory blocks. But the trend is clear enough.

Financial benefits

Driving this integration is the promise of a significant advantage on cost. Howarth agrees that alliances may yet have some further gains to squeeze out from revenue enhancements. Experiences so far suggest that alliance members should gain in the region of 2-5%. But logic dictates that this must eventually become a zero sum game. Once all the alliances have the offer of a global network, none can claim a decisive edge.

Gemini's analysis, based around a typical mid-sized flag carrier with a mix of services, suggests that the benefits not only rise in line with the depth of co-operation but do so exponentially. It estimates that a carrier in a looser marketing alliance could expect to make cost savings of less than 2%, largely from co-ordinated sales and services. By the time that alliance has developed to a fully unified structure, with new streams of savings starting to flow from joint flight operations and procurement, the yield could come to more than 11%.

At that stage the the exit barriers too are becoming significant ties to the alliance. Gemini estimates that an airline wishing to quit a fully integrated alliance would have to relinquish around half of the double digit cost advantage. The corollary is that an airline which wishes to stay independent within an alliance will have to forego a potential saving of around 5%.

In reality, Howarth believes that a carrier which has crossed the line from a shared strategy into a virtual merger would struggle to extricate itself. "The view is that once you give up independence, it's difficult to reverse the process," he says.

Network variations

At what point should carriers consider giving up their independence to an alliance? The answer hinges on which part of the network you look at. It became clear from the interviews with airline strategists, that it was possible to differentiate between three different components of the network:

Inter-regional: essentially intercontinental routes, such as the transatlantic. Major intra-regional: short-to-medium haul trunk routes within a region, typically between hubs or major city pairs. Thin intra-regional: regional services typically flown by aircraft with under 100 seats and with fewer than 100,000 passengers a year.

Gemini argues that an alliance should respond differently on each of these parts of the network. Development on the inter-regional segment, for example, is tightly constrained by ownership limits and the extent of air service agreements. That effectively rules out mergers, leaving alliances to concentrate on code-sharing and (anti-trust permitting), a degree of joint planning and revenue management. The core issue is about gaining market access, says Kirsebom, and until markets such as the transatlantic are liberalised that is likely to remain the primary goal.

Within a deregulated regional air market, such as those of North America or Europe, the opportunities and goals begin to shift. However, Gemini makes a further distinction between thin regional routes and those major hub services flown by mainline carriers.

Success on the thin routes crucially hinge on the ability of the operator to retain the flexibility and lean cost base necessary to exploit new potential markets. Merging such operations within a major airline structure would appear to endanger those advantages for little additional gain. The Gemini study concludes that alliances can take the main benefits - revenue increases and access to new destinations - through a model of co-ordination.

European merger pressure

It is rather among mainline operations that the pressure for mergers is starting to build. That is nowhere more so than in the fragmented and high cost markets of Europe. The shake-out that followed US deregulation provides a practical reminder of how far consolidation has to run on the other side of the Atlantic.

Despite a rash of alliance-making, Europe has, so far, shown little appetite for outright mergers and acquisitions. That is hardly surprising, perhaps, given the legacy of national politics and pride. But loose alliances are unlikely to last, says Kirsebom.

It is a first principle of successful alliances that power and benefits must be shared mutually. Without equality they risk power battles and internal strife. In the inter-regional model, such conflict is easy to avoid: Networks are complementary and the prize is entry into previously accessible markets. Low cost regional carriers may also offer a complementary style of operations which their major partners cannot hope to match.

Europe's flag carriers, almost by definition, are alike, They operate similar services in the same markets. The goal of the alliances is precisely to combine overlapping positions in order to rationalise costs and build market strength. In this process, it is the dominant hub carrier which must always be favourite to take the biggest share of the spoils.

"Small flag carriers are stuck in the middle. They can't become major hub players and they can't offer lower costs," says Kirsebom. "I don't see how they could take the same value from an association as the larger partner. They can't take out sufficient benefit to overcome their competitive disadvantage."

He concludes that the choice is either to stay independent as a niche player, or give in to gravity and be absorbed. What the second-tier carrier cannot hope to achieve, or at least not for long, is to retain independence within an alliance. "Either there will be a power shift towards the main hub airline or a power collision," says Kirsebom.

Despite some obvious caution, the first signs of this logic may already have begun to appear in the shape of KLM/Alitalia and more decisively with Swissair/Sabena. In both cases the aim is to create a virtual merger. That may not be in name or brand, to satisfy the conceits of national politics, but it nonetheless shifts real control towards the joint venture and away from the national capital.

The lesson from other industries shows that once a few deals are in place, momentum can build surprisingly fast. Recent signs of movement among Europe's airlines has persuaded Kirsebom that the process may be closer than many have believed. "As soon as one deal is in place then it becomes a vehicle which has to be utilised by others," he adds. What already seems certain is that alliance strategy still has a long way to run yet before the endgame is played out.

Cost Saving Estimates

Cost split by business process

Savings estimate per alliance model

Business Process

Proportion of Costs (%)

co-ordination (%)

Shared Strategy/ops (%)

Equity partnership (%)

Deliver service: in-flight





Deliver ground handling





Sell and distribute





Acquire assets





Maintain assets





Market/manage CRM





Develop products/service





Management /support










Source: Airline Business