Fix Bayonets!

During the presidential campaign, the importance of advances in military technology made a surprisingly high-profile appearance, when President Obama made his now-famous quip that, yes, the United States has fewer ships than it did in 1916, but it also has "fewer horses and bayonets." Beneath the zinger was an important point: quality can matter more than quantity, capabilities can matter more than numbers. But the reason the United States has been able to go from horses and bayonets to, as Obama put it, "these ships that go underwater, nuclear submarines" is that the Pentagon has long invested a percentage of its budget in basic science and technology. That legacy is now under threat amid budget cuts and the pressure to invest in programs that have immediate economic payoff — an approach that could have dire consequences for military innovation. As he enters budget negotiations, here are five things President Obama needs to remember if he doesn’t want to leave a legacy of bayonets.

1) Protect the Defense Advanced Research Projects Agency (and remember that DARPA is an agency, not a slogan). DARPA is the Pentagon’s far-forward-looking research and development arm, currently involved in everything from flying cars to a secretive "Plan X" to help the Pentagon wage cyberwar. Unlike the military services, where research is often focused on specific requirements, DARPA’s mandate is to look far out into the future, investing in science and technology that may not pay off for years. During the presidential campaign, the agency proved popular with both candidates: Mitt Romney’s energy plan, called "Believe in America," even made direct reference to the agency, saying the DARPA model, which provides "long-term, non-political sources of funding for a wide variety of competing, early-stage technologies," should be applied to energy. Obama has also heaped praise on DARPA and its model of innovation. But talk is cheap, and DARPA’s large discretionary budget makes it an attractive target for cuts. DARPA’s continuing success relies on protecting its funding and independence. While the administration was praised for only modest reducing its 2013 budget request for DARPA — to $2.8 billion — that number is still substantially less than four years prior, when the request was $3.3 billion.

2) Set long-term research goals (and remember that long-term doesn’t mean right now): The Pentagon’s science and technology cadre thrive on ambitious goals, particularly those set by the president, whether in the space program or computing. But those goals must be long-term, and by their nature, they will not always have an immediate political payoff. Obama appeared last year at Carnegie Mellon University’s National Robotics Engineering Center in Pittsburgh, praising a DARPA project that seeks to crowd-source manufacturing of military vehicles, a clear hat tip to his message about job creation: "As futuristic and, let’s face it, as cool as some of this stuff is, as much as we are planning for America’s future, this partnership is about new, cutting-edge ideas to create new jobs, spark new breakthroughs, reinvigorate American manufacturing today. Right now."  Of course, "right now" is precisely not what long-term research is usually about: the idea is that the country invest in basic research today in the hopes of reaping technological and economic payoffs that may be years away. The president should set clear goals in key military technology areas, such as cybersecurity, aviation, and space — and stick to them.

3) Beware white elephants (they will eat you out of house and home). Research and development funding too often falls victim to large procurement programs. As major weapons balloon in cost, the easiest way to cover the difference is to steal money from the research budgets (for example, the Army’s stealth reconnaissance helicopter, the Comanche, was blamed for tying up the service’s rotorcraft research and development budget for many years before it was finally canceled). The Joint Strike Fighter, with its trillion-dollar price tag, is fast becoming the white elephant to dwarf all other white elephants, and will make it difficult, if not impossible, for the services to invest in other aircraft programs. That means, in the case of aviation, investments in future capabilities are likely to go nowhere while the Pentagon struggles to cover the cost of its current procurement. The president, who has spoken in support of capabilities over numbers, needs to ensure the Pentagon reviews its largest weapons programs and actually cancels those that have chronically underperformed.

4) Sometimes picking "winners" works (but only sometimes). Both Obama and Romney agreed on the need to fund basic science and technology, but Romney argued that he would take a different approach. Romney focused on Solyndra, the failed solar company, as an example of the administration’s failed strategy of "picking winners." Yet in the defense and intelligence realm, no one seems to complain about precisely this strategy, which is employed by In-Q-Tel, the venture capital arm of the CIA. In-Q-Tel is essentially in the business of picking winners: it invests in early-stage companies that have technology it believes will be of use to the intelligence community. Though that model has yielded an occasional loser, like investment in a Buck Rogers-lightning weapon, it has also resulted in some much heralded successes, most notably its investment in Keyhole, the company that developed the technology that became Google Earth, which has widely benefited the intelligence community, as well as the general public. Picking winners can work in this case because, unlike the Solyndra case, the organization doing the picking, In-Q-Tel, has a direct connection to the consumer, the intelligence community. The question that the president will face is which models work for which problems: just as the "DARPA model" may not work for other agencies (the Department of Homeland Security’s "ARPA" has been a notable disappointment), investing in companies typically works best when there’s an intimate understanding — and even influence over — what the customer wants. That’s rarely the case outside of the defense and intelligence business.

5) Invest now in science and technology, or pay the price later. Back in 2001, the Pentagon pledged to keep the science and technology investment at 3 percent of its total budget, a benchmark recommended by the Quadrennial Defense Review. That number was tossed out the door in the years following 9/11, when the ballooning Pentagon budget would have created, at least in the eyes of some defense officials, an unduly large hike in spending for research and development. The Pentagon’s spending doubled over the course of that decade to about $800 billion a year. Not so for science and technology, which reached a highpoint of $14 billion in 2004 and has since dropped down to $10 billion — around the same as in 2001. There is now no clear benchmark for science and technology spending. Former Defense Secretary Gates promised 2 percent real growth each year in the Pentagon’s basic science budget (though not for applied research), but it’s unclear whether that policy is still in place. Worse, sequestration will hit research and development, just as it hits other parts of the budget, triggering across-the-board cuts that won’t allow managers to protect key efforts.