# Technology Policy and Climate Change

Date March 18, 2014 Adam B. JAFFE(Director, Motu Economic and Public Policy Research / Fred C. Hecht Professor in Economics, Brandeis University) YASUNAGA Yuuko(Deputy Director-General for Industrial Science and Technical Regulations, Standards and Conformity Assessment Policy, Industrial Science and Technology Policy and Environment Bureau, METI) NAGAOKA Sadao(Program Director and Faculty Fellow, RIETI / Professor, Institute of Innovation Research, Hitotsubashi University)

## Summary

I will talk about technology and climate change. I will start off briefly by saying that carbon policy, a direct environmental policy, is necessary but insufficient to deal with climate change. I'll talk about market failures in technology and comment on the current state of public support for energy research, and then talk about what we can learn from other technologies that governments have tried to develop in the past.

We want the world economy to continue to grow over the next three or four decades. If the world economy is to grow while we stabilize and then reduce the overall level of greenhouse gas emissions, the ratio of greenhouse gas emissions to world gross domestic product (GDP) will need to decline significantly. How significantly depends on exactly which scenario or trajectory you decide to follow. I would argue that it's almost not controversial that it has to decline by approximately 50% or more by around 2050. Over the next several decades, we will have to reduce significantly the greenhouse gas intensity of world GDP.

To illustrate how hard it will be to do that, I will look back in time. 2050 is about 40 years ahead, so I will look backward to about 40 years ago. I show here not the ratio of greenhouse gas intensity to GDP but instead the ratio of oil usage to GDP. We've experienced an increase in the price of oil by a factor of six or 10 over this 40-year period. The ratio of oil to GDP has fallen by about 40% over that period. Over the next 40 years, we will have to reduce the ratio of greenhouse gases to world GDP by more than that. I don't think many people feel that we're going to increase the price of carbon by a factor similar to that of oil. People talk about carbon taxes. A 50% tax on carbon would be big, but a 600% tax is not something that anyone is talking about.

The price of oil hasn't increased steadily—it increased significantly between 1970 and 1980 and then fell for a long time and then increased again. If it had risen to this level and stayed there, we might have reduced our oil intensity more than we have done. On the other hand, oil is a much narrower category of consumption than all fossil fuels, so it's much easier to reduce oil use than it is to reduce use of all fossil fuels. In fact, a significant factor in the reduction of oil intensity was switching from oil to gas.

While I'm a big advocate of carbon taxes and related measures, I don't think it's going to be enough to bring about the changes that we need over the next several decades. More generally, do prices spur innovation? After all, if we are going to decrease greenhouse gas intensity by 50%, we will need new technologies that are not available today. Can that come about just from raising the price of carbon?

There is economics literature on what's called "induced innovation"—the idea that innovation responds to economic incentives. However, there is little empirical evidence of this being a very strong effect. Another analysis with which I was involved asked whether a carbon tax could generate innovation in this area. We looked at U.S. cigarette taxes to see whether they had spurred the development of products to help people stop smoking cigarettes. The finding was that there was almost no relationship. This doesn't prove that it wouldn't work for carbon but should give some pause.

To prevent dramatic consequences from climate change, we will need a qualitative transformation in the economy—not just changing behavior on the margins but very significantly changing the organization of the socioeconomic system, particularly with regard to the use of energy. I would argue that the transformation needed over the next several decades is similar qualitatively to the transformation we have observed in information and communications technology over the last several decades. Today's technologies, such as the Internet and cellphones, weren't even imagined in the 1970s. A similar leap in technologies will be required by 2050 to get to where we need in terms of climate change. That won't be brought about by proper environmental policy, carbon taxes, or emissions trading. It will require significant technological change and whole new paradigms.

A recent Massachusetts Institute of Technology (MIT) paper concluded that, for a very long time, the use of carbon-based technologies will exceed that of non-carbon technologies. The incentive to engage in innovation to improve technology with respect to carbon is going to be greater than the incentive to improve non-carbon technologies, creating an uphill struggle in building a non-carbon economy.

The reason we think we need carbon taxes is because the private market doesn't create appropriate environmental protection incentives. The environment involves an externality that requires policy to internalize. There's also a completely separate set of externalities or market failures related to technology which are not solved by environmental policies. There are spillovers in the research process whereby benefits from research flow to parties not responsible for the research, spillovers from the learning curve whereby a technology improves and people invest in it but they don't gain all of the benefits, improvements in technologies that come from users who don't capture necessarily all of those benefits, and issues of information and path-dependence which affect the way technologies evolve, and prices don't necessarily deal appropriately with those.

One caveat to the notion that the government has an interest in increasing investment in technology because of these market failures is because technology comes from people. Scientists and engineers are needed to do more research and develop more technology, and, in the short run, the supply of them is relatively fixed. The government can't make more physicists immediately. If government engages in policy designed to advance technology, it will be held back in the short run by the inelastic supply of scientists and engineers. Increasing their numbers is a policy concern.

Data from the International Energy Agency (IEA) show the total IEA member countries' investment in energy research and development (R&D) from 1974 to 2010. It's currently about $15 billion per year worldwide, slightly lower than in 1980 after adjustment for inflation. The mixture of energy technologies has changed somewhat over time. There's been some increase in investment in renewable energy and some decline in research related to nuclear energy. Overall,$15 billion, given the scale of the problem we face—conservative estimates of what we might need to spend in mitigating climate change over the next several decades are on the order of 1% of GDP and more radical estimates are 4%-5%—this is a very small amount.

There are historical examples of government initiatives attempting to advance technology in particular areas. The first category is crash programs. The two big examples are the Manhattan Project and the Apollo Project. It's estimated that in today's dollars, the United States spent about $28 billion over two to three years to develop the atomic bomb and about$140 billion over 10 years for the Apollo Project. Could we do that for energy and spend \$200 billion over the next 10 years to build a non-carbon-based transportation infrastructure or the like?

It's not clear that the analogy applies. The objectives in those projects were not some commercially viable output; they had well-defined technical objectives that they were trying to solve and largely were independent of cost concerns. Climate change is not just a technical problem; it's a technical-economic problem. This type of crash program might be useful for a sub-goal of the overall initiative; to advance the technology for carbon capture and storage, for example.

In the 1970s in the United States under President Richard Nixon, we declared war on cancer and declared our desire to abolish cancer by 2000. There have been some very important technological advances in health. Can we learn anything from this example that would be useful in considering a war on greenhouse gases and say that within 30 years we want a world economy that doesn't rely on fossil fuels?

At the same time that the government was pushing the health-related technology, there was a market for the technologies as they became available. If there is a new technology that has been proven to save people's lives, people will buy it. One of the lessons from the war against cancer is that you have to worry about the market and whether people will be able to buy the technology. Also, there can be adjustment costs—a very rapid increase in the scale of a research effort can cause big dislocations in the system. A research enterprise can't change quickly. The National Institutes of Health (NIH) invested in both research and training. This addresses the issue which I mentioned regarding the supply of scientists and engineers. A sustained public program to advance technology not only requires support for research, but also it requires an increase in the number of scientists and engineers who can do research in the relevant areas.

What was the government role in creating the information technology and communications revolution, which may be a model for what we need to do in the energy sector? The main lesson is that the government's role was not limited to funding research. Specifically, much of the information and communications technology of the latter part of the 20th century grew out of either defense or space programs for which the government purchased technology without regard to cost. This was incredibly important in spurring technological competition in which cost was not the dominant element.

Those examples were positive, prescriptive examples, whereas the last example is negative. In the 1980s, the United States embarked on what was called the synthetic fuels or "synfuels" program and invested a large amount of money in building commercial-scale factories to produce liquid fuels from coal and other non-liquid fossil fuels. This program is viewed today as a huge waste of money. One of the main problems was that the government didn't say, "If you can make liquid fuel from coal, we can buy it." Instead, the government built the factories. As a result, it crowded out private investment. The government built the factories and got stuck with them. The government has an important role in creating demand for new technologies but probably not as the provider of the technology itself (creating and supporting the market instead).

It's easy to get depressed about the huge challenge that climate change poses. On the other hand, it is a very long-run problem. That tells us we have some time. From that perspective, conceiving technology as an important part of the solution makes sense. Technology could be greatly advanced within the timeframe we have to solve this challenge. I hope that over the next 10 years we can launch a worldwide sustained endeavor to increase our capability and build the human capital along the way. Based on the IT example, it's not enough for the government just to support research. It needs to pull the technology along by being a buyer or creating markets for the new technology as it develops. There will be failures, but timidity won't breed success. We need to invest in this area and have both successes and failures. If we succeed, it almost surely will be a result of technologies that we haven't envisioned today.

Nothing should be off the table. In the 1990s when people first started talking about climate change, it was taboo to talk about adaptation to climate change. People thought that doing so would give people an excuse not to worry about avoiding it. We've now realized that we can't avoid it completely so we had better think about how to adapt to it even as we attempt to reduce greenhouse gas emissions to mitigate the impact. But now, some people talk about geoengineering in the same way, which I think is a mistake. It is very important to reduce greenhouse gas emissions, but, at the same time, we should be studying geoengineering. The size of the challenge is so large that we can't afford to disallow any technology or possible approach.

Finally, the fact that this is a long-run problem means that if we can determine which policy interventions are most effective, 10 years from now, we could have better government policies to encourage technology than what we have today. This means systematically and continuously evaluating policies. Each new program should have a program to evaluate it and assess its effectiveness.

### Commentary

YASUNAGA Yuuko

Your presentation provided many insights and implications. Innovation has been a buzzword for the Shinzo Abe administration in boosting the Japanese economy. It has a tendency to be overused. My first comment is about the effectiveness of our policy. My question is: how should the effectiveness of greenhouse gas reduction programs or policies be defined? For basic research, the amount of new knowledge created may be observed and examined. For application-inspired basic research or applied research, the main areas for the Ministry of Economy, Trade and Industry (METI), the possible contribution to industrial development may be observed and must be examined to determine the effectiveness of the policy. In the case of climate change, by contrast, the future or possible mitigation of global warming which may be realized should be examined, but we'd like to know how. My second question is closely related. What technology should be invested in, especially by the government? We need to build an optimal portfolio combining some incremental technologies while also stressing disruptive technology. It is obvious to many employees that diverse disruptive technologies could reduce greenhouse gases more intensively. However, we need to examine the type of technology. We have learned that the U.S. Department of Defense (DOD) seems to spend 80% of its R&D budget on incremental technologies and 20% on disruptive or riskier innovation with the Defense Advanced Research Projects Agency (DARPA). The third comment is closely related to the government's role. How can we be creative enough to avoid technology lock-in without causing unnecessary confusion? We may need to transform completely the entire transportation system to tackle greenhouse gas emissions. What decides the optimal life of today's dominant technologies? Large-scale infrastructure needs a long time and large investment to change. My fourth comment relates to the government role in creating new markets. I agree with your idea, however, the Japanese government is not a big procurer. For example, we use regulation to encourage greater use of new technology or deregulation to discourage use of old technology, or to provide economic incentives for the use of newer and better technology. Those kinds of policy instruments may work fine. However, we need to explore further the best policy for the global warming issue. Thank you very much.

I have a few comments. You've raised a number of very difficult practical considerations. With respect to the question of defining effectiveness, clarity is necessary with regard to the ultimate goals and the intermediate objectives of your programs and the steps to achieve them. This will allow you to come up with metrics for effectiveness.

On the issue of portfolios and incremental versus disruptive technologies, because this is such a long-run problem requiring large changes, disruptive technologies are important. Rather than requesting specific systems, it would be better to point out promising directions—zero-emission vehicles, better building envelopes—and set targets to encourage technological development in those directions.

On technology lock-in, which relates to the systems I just mentioned, I would draw an analogy to IT. No government decided with respect to mobile technology that we would go through certain generations and use certain frequencies. The governments reacted to the technology as it developed, and I think this will be the case with regard to climate change.

You didn't talk about procurement policy.

You mentioned regulations earlier, and I agree that there's a role for regulation where the government procurement role is not as large. This would shape the private market to create demand that might otherwise be created directly by government purchases. It's important that it be performance-based rather than technology-based. It would be a mistake for the government to say, "Electric utilities need to get 20% of their power from windmills." By instead saying, "By some year, 20% of the power needs to be from non-carbon sources," you decide which non-carbon sources to use so that they would be analogous to using government procurement in a sector where the government actually isn't the buyer.

In the case of defense contracts, the DOD clearly specifies a technological target, so it invites competition to achieve technological progress. Simply buying carbon-free energy may not really induce technological competition. Combining regulation and R&D or technological competition may be very important in designing the system.

This is a global problem, and different countries will make different contributions. It may be that there are aspects that the Japanese economy and government cannot address due to their nature, but there may be others more appropriate for Japan to tackle.

### Q&A

Q1: On your point that nothing should be off the table, you cited geoengineering as an example. How about other traditional technologies such as coal? What is your view on the current U.S. government policy to end public financing for new coal power plants and encouraging other countries to adopt this policy?

What I meant by my earlier statement was that nothing that could potentially mitigate climate change should be off the table. I would not advocate abolishing the use of coal tomorrow, but it's clear that we need to be moving toward a world in which coal is not burnt. Certainly, policies that encourage the use of coal are undesirable, and policies to move us away from coal are an imperative step. Depending on how much climate risk is acceptable, there is some budget of carbon that is currently underground that we can allow to get into the air. Exceeding that budget will create worse climate change consequences. Coal uses more of that budget for the benefit it gives us than does any other energy source. As well, the faster we stop using coal, the more flexibility we will have to manage the transition in the transportation sector where the technological challenges are the greatest.

Q2: I have two simple questions. First, how do you evaluate the usefulness of nuclear energy? It is difficult politically but rational economically. Second, my impression was that you are rather pessimistic about the effectiveness of price mechanisms. However, while suddenly increasing the carbon tax to 600% would be bad, I think a policy to increase it gradually to that level would be a good policy.

I think carbon taxes are essential, although not sufficient. Supporting technology, for instance, would be wasted if the price of carbon was not raised. It is necessary to have consumers choose to use carbon-free technologies. I don't believe that raising the price of carbon and doing nothing else would bring about the transformation we need, but it is absolutely essential to raise the price of carbon predictably, permanently, and significantly.

For your first question, nuclear falls in the "don't take anything off the table" category. Climate change is a huge challenge to humanity. Nuclear represents a significant current non-carbon source of energy. If we as a society or a set of societies were to abandon nuclear power, it would be setting ourselves back from an already-distant goal. If we were to turn off all of our nuclear plants, that would take a huge chunk out of the carbon budget. I don't think we have the luxury of saying that we don't need nuclear.

Q3: One of the intensive discussions has been on patent policy. One extreme opinion is that patents related to emission reduction should be in the public domain. I'd like to get your comment on that.

I don't think a policy of not having patents on technologies to reduce or eliminate greenhouse gas would work. We need to recognize that we don't know where the technology that will solve this problem is going to come from, but we want as many incentives as possible for these new technologies to be developed. The problem of the cost of then acquiring that technology in poorer countries has to be addressed through other mechanisms. If there is going to be a global solution to climate change, resources will have to be transferred from the wealthier to the poorer countries to help finance the transition.

Q4: RIETI is not conducting natural science but rather social science, especially focusing on economic research. Do you have any ideas how to improve the productivity of economic research?

I can only say that research is a global enterprise, so maximizing the interactions of your economists with other economists elsewhere is probably the one suggestion I would have in terms of increasing productivity.

How about integration of theory and empirical data collection and empirical analysis?