By John Alic

Throwing money at R&D can be a good thing. Indeed, it may sometimes be the only thing to do. But not for energy and climate change.

Energy analysts sometimes urge dramatic increases in government R&D spending while at the same time pointing to "barriers" that seemingly impede applications of existing, cost-effective technologies. Supposedly, if the impediments could be somehow removed, these technologies would wash through the economy, saving energy, reducing dependence on oil imports, and cutting carbon dioxide emissions. In fact, R&D seldom leads in any simple and direct way to applications. Diffusion is almost always and everywhere slow, even for breakthrough technologies. And there is nothing very special about energy technologies except the reluctance of advocates to recognize parallels with the slow pace of diffusion elsewhere in the economy.

A quarter-century before the "invention" of the PC, business applications of computing began with automation of ordering at a London firm that supplied baked goods to some 200 tea shops. Two decades after engineers turned on the Internet's earliest ancestor, the national network linked no more than 10,000 computers. And more than three decades passed between early work by the U.S. Air Force on stealth and the invisible nighttime overflights of Baghdad by F-117s that opened the 1991 Gulf War. No one should expect energy innovations to flood into the economy no matter how R&D spending may be boosted. R&D is essential, but decarbonization requires a more comprehensive strategy. A closer look at military technology during the Cold War helps to understand why.

Since the end of World War II, the U.S. military has spent well over $5 trillion (in today's dollars) on R&D and procurement. Breakthroughs resulted: nuclear submarines in the 1950s, intelligence satellites in the 1960s, precision-guided missiles in the 1970s, stealthy aircraft in the 1980s. None of this was quick or easy. Nuclear submarines were envisioned in 1939; the Nautilus went to sea in 1955. Experiments with guided bombs and missiles (in the form of robotic planes) began in the First World War and picked up speed in the Second World War; not until the late stages of the Vietnam War, when laser-guided bombs showed they could destroy bridges that had survived hundreds of earlier sorties--871 in the much-advertised case of the Thanh Hoa bridge--did they demonstrate performance acceptable to military professionals. As early as 1943 the German navy coated submarine snorkels with radar-absorbing materials, dramatically reducing their visibility to Allied airborne radars, U.S. work on stealth began in the early 1950s with experiments on scale-model aircraft, yet not until the Have Blue demonstrators in 1977 did the first truly stealthy aircraft fly, and another five years passed before Lockheed began delivering F-117s to the Air Force.

Advanced technology does not come fast. It rarely comes cheaply, and even when it does, as in the case of the microprocessor--an innovation that depended not at all on new research, only on design--powerful incentives are necessary. For the computer and Internet revolutions, these were generated by economic competition (although the armed forces spurred much early innovation in digital electronics). For the military technological revolution, the Cold War--an existential threat perceived as such by most Americans, not least children cowering under school desks during nuclear attack drills--was the motivating force.

The Cold War created a compelling sense of mission and common sense of direction in government and, though perhaps less urgently felt, in defense firms and universities. Even opponents of militarism among scientists and engineers were likely to do their best when the Pentagon came calling. The Cold War mission disciplined the everyday work of technology and science, keeping efforts on track and pointed toward systems that the Army, Navy, Air Force, and Marine Corps (and the intelligence agencies) actually wanted and would employ, rather than set aside as ineffectual or too complicated or too risky for their own personnel.

In science, external sources of discipline are not so important. Science has its own built-in logic, deeply ingrained. Where discipline is needed is in technology, which lacks an internal logic and is always and everywhere guided and governed from outside. Military systems had to satisfy multiple customers and clients: the system program offices of the armed forces, the civilian staff of the Office of the Secretary of Defense, military commands up to and including the Joint Chiefs of Staff, the Office of Management and Budget, Congress. All had to be persuaded that hundreds of billions of dollars appropriated and spent each year for R&D and procurement would, in the end and despite the twists and turns and political/bureaucratic conflicts along the way, culminate in systems that would contribute to the overarching goal of offsetting Soviet military power. Of course not all weapons met this test, but enough did. And if the costs were high and money wasted, that is beside the point here, which is simply that those involved, inside and outside government, knew what they were trying to do, and why.

Nothing like that common purpose and sense of direction, of shared assumptions and goals, exists, or has ever existed, for energy technology. That is why the complaints heard so often since the 1980s, when the Reagan administration slashed energy R&D, get at only half of the problem. The other half is, what to spend it on? Without a concrete, agreed answer, little will be accomplished.

For decades, energy R&D has been mismanaged, beginning literally at the beginning, with the Atomic Energy Commission's fumbling move into commercial nuclear power in the 1950s, fumbled because there was never any agreement on what the government should be trying to accomplish. While Congress fussed and opponents of public power lobbied against possible encroachment by government, the Commission itself, almost entirely consumed with nuclear weapons, had little time for the issue. There were no strong counters to the political pressures impinging on the power reactor program, which began in disarray and ended with large financial subsidies fueling what now seems like a speculative bubble. Of course politics always in some sense dominates, including in defense during the Cold War. Yet the Soviet threat was always there, an ever-present if not always wholly effective check on excess and stupidity. The United States did not go on to build a nuclear-powered bomber for the Air Force in the 1950s; the program was halted for lack of military utility.

For energy and climate change, the danger is that new policies and new institutions put in place without a relatively high level of agreement on their goals and purpose, or expansion of old institutions such as the Department of Energy (DOE), descendent of the Atomic Energy Commission and inheritor of many of its dysfunctional ways, will disappoint after a few years, and a reaction set in.

Efforts to clone DARPA (the Defense Advanced Research Projects Agency), early sponsor of what became the Internet, illustrate the dangers. DARPA's place and function within the Pentagon has often been misunderstood. It was not established to sponsor research. Although DARPA has always done some of that, there are many other agencies and subagencies that fund military research. DARPA's job is to get advanced technology into weapons systems that can be fielded within a reasonable time and that promise to help one or more of the services do whatever it is the generals and admirals want to do (encrypt radio transmissions, hide planes from radar, find Soviet submarines in the deep ocean). Lacking that mission, which is unique in the Defense Department, DARPA would be superfluous and, more to the point, would not have achieved the successes that made it a model for civilian versions such as an energy-ARPA, or ARPA-E.

Plans and proposals for replicating DARPA mistake managerial and organizational features, such a famously lean staff of highly qualified personnel and an absence of internal laboratories to soak up funds better spent on the outside, for its essence. In fact, these and other aspects of DARPA were devised by managers as means to an end, that end being a reasonable chance, first, of surviving in the face of powerful military opposition (incited by the prospect of civilian responsibility for even a small share of defense R&D, today less than 5 percent), and, once that battle had been won, having a reasonable chance of accomplishing the agency's mission without a renewal of the attacks.

Put differently, DARPA has a set of clients: the upper command levels of the Army, Air Force, Navy, and Marine Corps. To survive, it must keep them happy, or at least keep them from grumbling too much. DARPA managers have always understood that. The energy side of DOE does not have such a set of clients--an identifiable group with real power and authority. Only the defense side, responsible for nuclear weapons (and naval reactors), does. Who will be the clients of ARPA-E? Of other organizations that might be put in place to develop innovative technologies for confronting climate change?

Innovation is a Darwinian process of idea generation and selection. In the private sector, the selection environment is rigorous, thanks largely to economic competition. In defense, selection was reasonably stringent because of widespread agreement on the need, should war come, for conserving the lives of U.S. and allied forces while, to put it bluntly, killing the enemy in large numbers. Nothing like that sort of selection environment exists for energy technologies and decarbonization. Without it, government risks spending billions of dollars on energy R&D while accomplishing little. More money for R&D must be accompanied by a complementary set of policies, perceived as predictably stable for the foreseeable future, to pull new technologies into applications and the marketplace on the vast scale needed to make a difference for greenhouse gas emissions. The selection environment must be rigorous and, in the absence of a mission as compelling as defense, must exist outside of government. There is only one candidate, energy pricing. Energy R&D without a carbon tax will fritter money away.

--
John Alic worked at the U.S. Congress's Office of Technology Assessment from 1979 until the agency closed in 1995. His book Trillions for Military Technology: How the Pentagon Innovates and Why It Costs So Much appeared at the end of last year. In addition, he is coauthor of Beyond Spinoff: Military and Commercial Technologies in a Changing World and of New Rules for a New Economy: Employment and Opportunity in Postindustrial America, along with many shorter works dealing with technological change and its ramifications.