The Australian wine industry has transformed itself since the mid-1980s. At that time, Australia exported 2% of total production and was a net importer of wine. In the past year exports reached $1.35 billion, an increase of 36%. The industry exports over 32% of total production, compared with 17% by old world wine producers, France and Italy. Over 38% of wine producers are exporters, which is the highest level of export participation in any Australian industry. Australia produces only about 2% of world wine, but now holds about 3% of the world wine market by volume and nearly 5% by value. Wine exports equal or exceed earnings from such elaborately transformed manufactures (ETMs) as computer and telecommunications equipment and motor vehicles.

How is this performance to be explained? A decline in the value of the Australian dollar was one stimulus. Another factor was change in industry ownership in the late 1980s when four large companies emerged. It's worth noting that the large companies are themselves small in the international league. The largest, Southcorp, ranks 13th in the top 20, BRL-Hardy ranks 16th and Mildara Blass 18th.

Our report argues that three changes were absolutely critical to the transformations of the industry:

• First, wine producers have adopted a global business orientation – they started to think of their industry and its prospects and their individual strategies in an international context.
• Second, within this strategic frame, marketing is king: the industry has become market led, not producer led.
• Third, the industry is committed to innovation: there is clear evidence that Australia's competitive advantages have been driven significantly by technical developments. The industry has invested in research – and perhaps more importantly, findings have been adopted.

Industry collaboration has facilitated, or contributed to, these three changes. In the process, the wine industry has transformed itself. It has shed its cottage industry character. It has shed its domestic orientation and its production focus. Fierce competitive rivalry between individual producers persists. But this has been supplemented by industry collaboration around marketing and innovation. The industry has raised its level of integration and is developing into a knowledge-driven cluster.

The development of the competitive capabilities of the Australian wine industry illustrates perfectly the processes through which new market and business orientations are seeded – it illustrates perfectly the ways cluster linkages develop. Through the 1980s, this industry was marked by fragmentation and division. Its focus up to the middle to late 1980s was almost wholly on domestic markets. At this time, there was an increased engagement in exporting, initially at the commodity end of international markets.

From the early 1990s, a second significant shift occurred. The industry progressively changed its market positioning. It began to export more high value products. Now it has largely vacated the commodity or bulk end of wine production. This is one of the key distinctions between the results achieved by the Australian industry and the other New World producers like Chile and South Africa. This was the result of a deliberately orchestrated move.

It would be quite wrong to see the progressive development of cluster linkage and of changed market positioning as an orderly, planned or logical process. The appearance of a deliberate sequence is a fiction - only possible in hindsight. What is clear, is that a few visionary leaders Brian Croser and Len Evans for example, saw potential for the industry. They saw that through collaboration all could win much larger gains than any individual producer acting alone. This industry has many forums where producers come together - and this gave industry leaders the opportunity to air their views to a wide, if initially sceptical, audience.

As events proved that the possibilities were not overstated, scepticism gave way to a willingness to extend collaboration. This culminated in the preparation of Strategy 2025, released in 1996, which is to my mind a model of collaborative industry development.

Its important to identify precisely what this kind of strategic articulation contributed - and what it did not. Strategy 2025 created a kind of strategic envelope. It created a framework within which new opportunities could be recognised. It created recognition of possibility – and with it a willingness to entertain new requirements and hitherto unthinkable changes - for example the adoption of varietal labelling.

Above all, it positioned individual producers to better react to day to day developments. Detailed responses can't, and shouldn't, be planned. This is part of the genius of the free market system. Individual firms are best positioned to respond to immediate opportunities and imperatives. But whether something is judged an opportunity, and how managers respond to an unexpected development, depends enormously on their prior sense of strategic possibility. It depends enormously on a prior conversation – a conversation that sets expectations, suggests what causes what, and identifies significant future priorities. This is precisely what Strategy 2025 and it's less elaborate predecessors contributed. They enlarged the sense of opportunity. They prepared managers to respond more adroitly to unforeseen developments. They created an agenda for action at the level of individual firms and at the level of the whole industry. The success of this approach is evident in industry performance.

Strategy 2025 provided an overall vision for expansion in international and domestic markets. It represented a classic expression of strategy at a cluster level. A specific growth outcome was identified – but sufficiently far into the future to be seen as an aspiration, not in any way a precise prediction. The designation of a specific outcome was a critical step. It offered a precise number – but as a possibility for producers to achieve by will and effort - not as in any sense the inevitable result of impersonal forces, or the result of an obscure technical or econometric calculation.

Specification of a precise outcome was essential. This promoted understanding of the steps that needed to be taken at the level of the whole industry to translate this aspiration into reality. Hence, if you look at the document, you will see it estimates requirements in terms of plantings, water, acreage, skilled personnel and so forth. It also provided essential information to producers: for example the value categories and the markets in which growth could be expected, information about trends in consumer tastes and so forth. But this dissemination of factual information was not its only, or even its most important, contribution.

Strategy 2025 was also essential to mobilise producers. It translated aspirations from the clouds into the practical realm of self-interest – individual producers could calculate what it could mean for them. Strategy 2025 also influenced the perception of key customers (e.g. the British supermarket chains Sainsburys and Tescos), stockmarkets and the media. For these latter stakeholders, the industry vision reinforced confidence about its professionalism and capabilities, and augmented its credibility.

Wine industry experience demonstrates conclusively the synergies that can be gained, and the energies that can be liberated, through a combination of collaboration and competition. It illustrates the challenge in sustaining collaboration – the process was not, and is not, free of opposition or criticism – witness the tax debate last year. `But it shows how collaborative action can proceed without unanimity, and how dissent can be ventilated and managed. Finally, it shows collaboration as an on-going process – the greater the success today, the greater tomorrow's challenges.

I mentioned earlier the development of the cluster occurred opportunistically. This does not however mean it was wholly a matter of luck. Nor does it mean there are no lessons for others. On the contrary. The wine industry experience illustrates an approach that seems translatable to an array of industry groups, although the precise approach might vary.

For example, in the wine industry case an export target and the potential to market 'brand Australia' provided the initial impetus. How generally applicable is this starting point? There may be tougher issues in other sectors. For example, I have just participated in a study of the film and television industry in NSW. The cottage character of the production end of the industry means seeding collaboration would be a much more difficult enterprise.

But the potential too is great. Indeed taking the film and television sector as an example, without collaborative action it is hard to see how significant development might occur. Canada starting from a position much like ours fifteen years ago, has built an enormously successful sector though strategies that parallel those of our wine industry. The UK and Ireland in different ways are currently embarked on a similar exercise.

Vision and commitment are not common qualities – and both are needed to seed the development of collaboration. But, on the other side of the ledger, imitation is the most powerful teacher. Once there are a few recognised successes, momentum can be generated. Meantime the gains for individual firms can be large – and this is another very powerful incentive. The presence of clear performance tests is a positive factor. The gains from clustering are not some fanciful promise well into the future – once appropriate marketing or research steps are taken, the results should be evident. If they are not someone has miscalculated, and schemes need to be discarded or amended. In the wine case, if positive results had not been forth coming, industry leaders would have been dismissed as dreamers not visionaries.

The power of the clustering idea is evident. Clusters span large portions of the economies of most nations. For example, in the United States, Detroit, Hollywood, Wall Street, Silicon Valley are all familiar cluster metaphors. The world's two leading motor cycle companies and two leading musical instrument producers grew up in the Japanese city of Hamamatasu. All of Germany's leading cutlery firms are located in Solingen. These examples can be multiplied. At best, clustering blends the nimbleness of small firms, with the broadly based capacities that would otherwise require mega-firms.

What are the lessons for other companies? For many small and medium sized businesses and even businesses that are large by Australian standards, but small in a global context, clustering could leverage their capabilities and extend their reach.
This is because the transaction costs that arise in moving to a more outward-looking approach can prove daunting. These costs are associated with such steps as:

• gathering information about external markets,
• prioritising their attractiveness, and
• assimilating their requirements in terms of tastes, regulation etc.

Other transaction costs include those associated with establishing a brand identity; deciding on the appropriate mode of distribution; and hedging these various risks.

Technological change has similar implications. It promises to enhance the competitive capability of individual firms, particularly small and medium sized enterprises. However, such businesses typically lack the capacity to invest in research on the scale required for impact. The costs of learning and adaptation an easily prove prohibitive.

Through collaboration, shared investment in marketing and innovation can make the costs and risks manageable, even for small players.

Individual firms or individual industry leaders interested in developing linkages can take the initiative. But sooner or later an industry association would need to be formed. Industry associations are key intermediaries. For example, in the wine industry case, three industry associations – the Winemakers Federation of Australia, Australian Wine and Grape Export Corporation, Australian Wine and Grape Research and Development Corporation - played critical roles. These organisations are interlinked. They divide up the tasks of strategy, marketing and research. Together, these associations provided the platform for the development and dissemination of proposals for common action.

This experience also has important messages for government. Old-fashioned industry policy was about protection – this was essential to allow domestic firms of sufficient scale to develop. Modern industry policy is about productivity gain. This is the only path to creating wealth. Productivity gain mostly comes from market expansion, or from innovation, or from both activities. Unfettered markets are a powerful incentive to productivity gain – and policy over the past fifteen years has been designed to remove distortions and impediments, of which we had many. But as we have seen in the wine case, a cluster structure can complement market forces. It goes with the grain of markets – but enhances and leverages the responses of individual firms.

Governments can play a catalytic role in these processes. Government can seed the development of collaboration. The Mortimer report proposed steps down this path. Subsequent action has implemented aspects of these recommendations. But the wine industry approach is much more elaborated. If governments seek, in the national interest, to encourage more exporting and a greater commitment to innovation, here is a model of how it can be done. There is of course no one to one transfer to other sectors. But the principles that might shape public policy are clear.

By contrast with old-fashioned industry policy, participants in a cluster commit themselves to productivity gain. They commit themselves to particular outcomes, particularly in relation to exporting and innovation. Government could provide incentives for the development of such activity. There are numerous examples of specific approaches in other states, such as the USA, Ireland, Canada, Taiwan and Singapore. From a public policy perspective, one charm of a cluster approach is that it works with market tests. If the imperative is, in the national interest, to encourage exporting and innovation, the encouragement of the wider adoption of cluster strategies seems the logical step.

Well-orchestrated collaboration offers a supplementary way to upgrade our economic performance. Through linkage in clusters, individual firms can ground their continued growth and success on distant markets and on innovation. The approach adopted in the wine industry suggests a pattern for other Australian firms, and for other sectors and for governments. The wine industry has exhibited levels of creativity, entrepreneurship and management capacity that are in that much overworked term, world class. But individual firms could not have realised these outcomes on their own. If exporting or innovating again become national issues, the wine industry offers a home-grown, local approach that others can study and emulate. And all this has been achieved in just one decade. I hope I have convinced you their effort deserves your attention.