The four military services, the office of the secretary of defense, Congress, industry and labor unions all have weapons programs they consider essential and untouchable. But accommodating every sacred cow is impossible in a constrained budget environment.

If policymakers want to develop a coherent long-term defense industrial base strategy, leaders in Congress and the Pentagon must decide which sectors are truly essential and be willing to accept some level of risk among the rest. There are undoubtedly some sectors in which it would make sense to rely more on close allies in Europe or elsewhere.

The Pentagon should begin by assessing future threats. What are the most likely and important threats the U.S. will face in the next 10 to 20 years? What broad capabilities are needed to meet them?

The next step would be to decide what systems and platforms best provide the critical capabilities. Then moving from broad capabilities to the specific systems and platforms to support those capabilities would highlight the defense sector’s critical elements. The department cannot afford to continue investing in sectors unlikely to be relevant in the future. Horses, for example, were once critical to the military. Are there weapon systems today that may similarly need to be put out to pasture?

The Pentagon should also consider an industry’s capacity to produce weapons in the required quantities. A protracted conflict against a sophisticated adversary, for example, may require investment in surge and mobilization capacity to resupply forces with munitions and replacement weapon systems.

The recent conflict in Libya demonstrated this. In the opening days of the conflict, the U.S. fired roughly 300 Tomahawk cruise missiles. At current production rates, it will take an estimated 18 months to restock the inventory from this limited operation. This raises the question: In a protracted conflict against a larger, more sophisticated adversary, could production of key weapon systems keep up with demand?

Once crucial sectors are identified, DoD should then consider the current state of industry in each. Are there multiple vendors and healthy competition? Are some companies likely to stop production or merge with competitors? Is there enough surge capacity to meet wartime needs?

If the health of these key sectors is in doubt, DoD should determine what intervention is needed to protect the national interest. In some cases, the solution may involve little or no government cost. Regulatory reform, export reform or changed contracting practices, for example, could help industry and actually save money. In other instances, however, DoD may need to commit resources, providing a steady level of funding to sustain production or key design teams.

After evaluating future threats, capabilities, surge capacity and costs, the Pentagon can develop a list of key sectors of the industrial base — and the actions needed to preserve them. Whatever fiscal or military challenges the country may face, the Pentagon must take steps to monitor and, when necessary, intervene to preserve the health and adaptability of its defense industrial base.

This will not be easy in an era of austerity. But if policymakers are willing to make difficult choices, the U.S. defense industry can remain the best in the world and the arsenal of democracy.