The Department of Defence isn’t at a loss for advice in the run-up to its release of the defence industry development strategy, expected early this year. Following on the heels of a NIOA Group (et al.) contribution ^[1] to the debate comes a treatment from the Australian Industry Group and the Australian National University.

Defence industry in national defence: rethinking Australia’s defence industry policy ^[2] serves to frame the debate rather than provide a roadmap to any particular ends. The framework is a useful one. But it leaves open a path to pitfalls that have waylaid past Australian efforts to build sovereign defence industrial capacity.

Avoiding those pitfalls requires a different way of thinking about the problem. A useful perspective comes from the 1958 insight of Leonard E. Read: no one knows how to make a pencil ^[3].

Even something as simple as a pencil involves aggregated raw material and knowledge of such enormous complexity that no single person or organisation is capable of bringing it all together into a final product. In making a pencil, we have no choice but to rely on market mechanisms.

How much more is required for an armoured fighting vehicle, nuclear-powered submarine or guided missile?

The Ai Group–ANU report recommends that the government ‘utilise the full range of policy levers at its disposal to shape defence industry outcomes’. It proposes establishing a defence industry capability manager ‘responsible for defining the capability and capacity that government needs’. And so on.

These are good recommendations. The problem is that they feed into the false historical narrative that industrial capability requirements can be planned and organised down to the finest detail. They can’t, and imagining we can do so results in application of policy levers that are all but guaranteed to fail.

How can we turn the report’s recommendations into actionable policy? What tools should be applied by that defence industry capability manager (please, let’s not get off on the wrong foot by calling this person a ‘tsar’)? The most essential element of any such policy must be a focus on Defence’s demand signal. That signal tells industry where to invest far more effectively than any detailed set of prescriptions. Provide a clear enough signal and market mechanisms will do the rest.

This is more difficult than it sounds. The demand signal can’t just be about end users’ technical requirements. Defence markets are not defined ^[4] by free and open competition, and Defence must play a wide-ranging and intrusive role. But ultimately it must approach any problem from the perspective of ends, not means. In the words of George Patton, ‘Never tell people how to do things. Tell them what to do and they will surprise you with their ingenuity.’

There are many tools available to shape the demand signal in ways that will encourage the growth of Australia’s defence industry. One such tool is contract structure. Fixed-price contracts are useful for procurement of clearly defined products for which the key challenge is fostering competition and keeping prices low.

But competition and low prices are not always the most important objectives. Sometimes we don’t know how to solve a problem or even what the key technical requirements might be—certainly not when the capability we’re trying to build is an entire industry. This situation requires innovation and long-term partnerships between government, industry and academia. Those objectives are better served by cost-plus contracting.

Cost-plus development contracts often lead to sole-source awards and the close government–industry relationships that are anathema to those who place a strong emphasis on probity. But they are a necessary tool if Australia wants to build its defence industrial capacity. They can have an outsized impact on industrial structure, signalling not just what product Defence wants to buy, but what capability it wants to foster in its supply base.

There are many tools for setting up an acquisition enterprise that builds industrial capacity. For years, Australia has been wringing its hands over its inability to translate early-stage research into products for its military. The basic reason for the disjunction has been a demand signal that focuses on a set of technical requirements without regard to the knowledge base and organisational capacity required to translate them into operational reality.

The Defence Science and Technology Group is good at research and development. But there’s little point in spending years developing sophisticated defence capabilities if they are simply tossed over the wall for someone else to execute. Defence industrial capacity is about more than a set of technical specifications. It includes product strategy, marketing, business development, regulatory positioning, supply-chain management, manufacturing and a thousand other things that require a long process of learning by doing.

Creating a useful demand signal for industry is a protracted process in which the public and private sectors learn together. The cozy relationships aren’t always comfortable, and they aren’t always fair, but they’re the only way to establish the knowledge base necessary for a self-sustaining defence industry to take root. The US network of federally funded research and development centres is another example of a policy tool that helps define objectives in terms of capability rather than technical specifications, creating an ecosystem instead of just a widget.

A final point to address is human capital. Just as it isn’t possible to identify long in advance every asset required to successfully bring a defence product to market, it’s not possible to identify every skill set necessary to build it. Labour markets rely on specialisation far beyond the planning abilities of bureaucrats in Canberra, let alone politicians with their regional biases.

Rather, Defence must create a demand pull and accept that the final result might look different from what was imagined at the outset. There is no pushing the string. Whatever the rhetoric about new jobs, we can’t create something from nothing. Labour force reallocation operates through the medium of relative wage adjustments. Training pipelines are long. Pull too hard—without acknowledging the trade-offs—and you just get inefficiency and inflation.

The guiding principle must be humility. Defence must build its strategy around tools to shape the demand signal rather than attempting to dictate every requirement throughout a vast supply chain. It must avoid trying to do everything and focus on doing the right things well, such as contracting, program management and public–private partnerships. The first step in building a sovereign defence industry is admitting that we don’t even know how to make a pencil.