BERKELEY - As the United States and the industrial world prepare to spend millions if not billions of dollars to reduce global greenhouse emissions to head off major climate change in the next century, they also should invest in more research and development to find the best course of action, according to two U.S. experts on energy policy.

For the benefit of the Third World, in particular, this research should focus more on small-scale technology and "mundane" problems like more efficient stoves or better lightbulbs, as well as renewable energy solutions for single homes and villages, said Daniel M. Kammen, an associate professor of energy and society in the Energy & Resources Group at the University of California, Berkeley.

"We're at an important time with global warming, where nations will be spending a lot of money cleaning up their emissions," Kammen said. "To get there, to greatly reduce emissions, we need to get much cleaner technologies in place, and that requires research."

The money spent worldwide on energy R&D has dropped 39 percent over the past 20 years, a dangerous decline, claim Kammen and his colleague, doctoral student Robert M. Margolis of the Science, Technology and Environmental Policy Progam (STEP) at Princeton University. Kammen chaired the STEP program at Princeton until June 1999, when he took up his current position at UC Berkeley.

This decline reduces the ability of the energy sector to innovate, and slows the search for solutions in the third world, which relies heavily on research done in the industrial nations.

In addition to spending more money on R&D, however, countries need to find better ways to measure the impact of the R&D they already do, Kammen stressed.

"It's easy to say, just put more money into R&D," he said. "But let's also get some sense of what the return has been on past R&D and use that as a guide."

Kammen and Margolis discuss the problem of underinvestment in energy R&D in this week's issue of the journal Science. The viewpoint piece is part of a larger section on global use of energy appearing in the July 30 issue.

Kammen and Margolis point to a disturbing two-decade decline in funding on energy technology in the industrial world. The decline in U.S. R&D funding - down 58 percent since 1980 - is exceeded only by the decline in Germany and the United Kingdom, both down more than 85 percent.

"The wholesale dismantling of large portions of the industrial world's energy R&D infrastructure could seriously impair our ability to envision and develop new technologies to meet emerging challenges," the authors wrote.

A soon-to-be-released report on how the U.S. can play a bigger role regarding energy technology in developing countries that was prepared by the Panel on International Cooperation in Energy Research of the President's Committee of Advisors on Science and Technology (PCAST) argues for somewhat increased investment in energy R&D.

Kammen would like to see even more. In their Science paper, he and Margolis show that, by several measures, investment in energy R&D really pays off economically.

"The more you invest in something, the more you get out of it," Kammen said. Considering our continuing decline in spending on energy R&D, he added, "that's a dangerous correlation."

They set out to find a reliable way of estimating the impact, or return on investment, of money spent on research and development of energy technology, whether that investment is in fossil fuels, nuclear power, renewable resources or other areas.

It's clear, for example, that money spent in the late 1970s to develop clean coal was largely wasted, Kammen said. There has been little return on that investment, which created a sizable bump 20 years ago in total U.S. investment in energy R&D, he said.

Equally clear is that a simple and inexpensive program called Green Lights, sponsored by the U.S. Environmental Protection Agency (EPA), worked very well in reducing electricity use. Through that program, EPA offered businesses a small subsidy to replace fluorescent light bulbs with more energy-efficient lighting. Another quiet but effective program was one by the motor industry in the 1980s to improve the efficiency of electric motors.

The difficulty comes when trying to assess other kinds of R&D investment, such as the savings from switching to renewable energy sources, such as wind. Cause and effect also can be unclear. Though an increase in investment on research on photovoltaics in the 1980s led to a steep rise in sales for U.S. companies making these devices, which convert sunlight to electricity, it is difficult to prove that the investment caused the increase in sales.

They found what they consider a good estimator, however: the number of patents filed per year. The total number of patents filed in the U.S. per year follows closely the ups and downs of total R&D investment in this country. Similarly, the number of energy-related patents fell steadily over the past 20 years, in perfect step with falling investment in energy R&D.
This indicates a significant link between R&D investment and innovation, the authors state, and suggests that patents may be a good barometer of R&D activity.

Another way of judging how much money should go into R&D is to look at the R&D budget in a particular sector as a percentage of sales in that sector. For example, in 1995 in the U.S., more than 10 percent of net sales of drugs and medicines was reinvested in the form of research and development. That same year in the energy sector, only one half of one percent of sales was reinvested in R&D.

"I don't know where we should be on this scale, but the energy sector should be investing more than half a percent in research and development," Kammen said. "That's a factor of 20 less than what is seen in medicine and biotechnology in level of funding, the lowest of any sector in the country."

The type of R&D Kammen prefers is what he refers to as "mundane science."
"We need to tackle projects that may not sound exciting first off, but which involve important energy-use problems," he said. "More people should work on small-scale and decentralized energy technology."

He has been involved in many small-scale projects throughout the Third World, ranging from the introduction of more efficient cookstoves in East Africa to the development of safer and more efficient batteries in south Asia. Today, his students in the Renewable and Appropriate Energy Laboratory at UC Berkeley are studying the efficiency of solar panels in Kenya and looking at effective ways of providing clean drinking water in East Africa and Mexico.

One problem with U.S. support of developing countries, Kammen notes, is that it usually is predicated on benefiting a particular U.S. industry.
"The best assistance is not to say, 'I'm going to help sell U.S. technology to another country,'" he said. "What works best is to develop a local industry to support that technology from inside the country."

The story of photovoltaics in Kenya is a perfect example. Kenya has a thriving industry selling solar panels and batteries to people of all income levels. Perhaps 10 to 20 percent of the developing nation market for photovoltaics is in Kenya, Kammen said.

Because the industry is local, it responds to the local market to the point of creating smaller and cheaper panels for the lowest income levels and improving the quality and performance of batteries. Because of this, the photovoltaic industry has been willing to listen to suggestions by Kammen and his colleagues on how to improve their products.

"My students coming back from fieldwork in Kenya have been talking with vendors and installers of solar systems, and they seem to be responding pretty quickly to new information on the efficiency and performance of the systems they sell," Kammen said.

This illustrates a crucial point, he said. It is not enough to provide a better technology for Third World countries, you must also work with the people and consider the social impact of that technology on their culture.

But new ideas are needed first, before you can attempt to change energy use in the developing world or even in the industrial world.

"It's hard to get changes in the energy sector, and even harder if you're not investing anything in it," he said. "But let's do a much more consistent analysis and put our money where it works."