Increased fuel efficiency, however, is not free.... Any truly cost-effective increase in fuel efficiency would already have been made.

—Former Council of Economic Advisers Chairman N. Gregory Mankiw, 2007

N. Gregory Mankiw is the mirror image of Amory Lovins, the protagonist of the preceding chapter. Lovins knows that every energy measure we could possibly need will save more than it costs. Mankiw knows that all such measures will cost more than they save. Mankiw served as George W. Bush’s Chairman of the Council of Economic Advisers from 2003 until 2005 and is well respected within the economics profession.

The Mankiw-Lovins bipolarity highlights an important split in energy-policy circles. On one side, we find Mankiw and other “neoclassical” economists. They oppose not only fuel-economy standards, but all energy-efficiency standards and energy-efficient building codes. That is, they oppose all measures favored by members of the “physics camp,” as I called them in the last chapter.

On the other side of the split, the physics camp is less strident. Although they tend to believe efficiency standards are all we really need, they rarely take a strong stand against the policies favored by the neoclassicals.

The neoclassicals, being economists, favor policies which change the price of energy. They call this “sending a price signal” to the market. They favor sending the price signal by taxing fossil fuel. That would, of course, raise its price. But since taxes are unpopular they’ve come up with a stealth tax, which will not be recognized so easily. The stealth tax is called a “cap and trade” policy, and six or seven of these have now been proposed to Congress. Like a tax, a cap-and-trade policy raises the price of fuel or electricity. It “sends a price signal,” which pleases the neoclassicals.

The physics camp tends not to like either taxes or stealth taxes, both of which they see as unpopular because they are clearly costly, not cheaper then free. Instead they prefer to mandate more efficiency with a standard, which they have pre-calculated will save more than it costs. While working at Lawrence Berkeley National Laboratory, I helped make these pre-calculations for national appliance standards.

So where do “free lunches” fit into this controversy? “A free lunch” is what the neoclassicals call any policy that provides a benefit that is greater than its cost. The term is descriptive but it also conjures up the slogan “there’s no such thing as a free lunch,” which helps them win their point.

In summary, the neoclassicals say: the physics camp claims all its proposals are free lunches, but there’s no such thing. We need taxes. The physics camp says: call them free lunches if you like, but there are a lot of ways to save money and energy at the same time—who needs taxes?

The Energy-Policy War

The neoclassicals dismiss the efficiency programs of the physics camp even more vehemently than the “free-lunch” label suggests. If efficiency programs are not free lunches, then there is a net cost to the energy efficiency gains the programs provide. But there is also a net cost to the efficiency gains from the neoclassicals’ taxes, so why is one better than the other?

The neoclassical answer is that efficiency gains from government standards are always more costly than efficiency gains selected by the market. The neoclassicals propose sending a price signal to the market and letting the market choose how to improve efficiency. Since, by assumption, markets always do better, the physicist proposals are always worse.

Because the neoclassicals see taxes as a more market-based approach and markets as better than government, they actively oppose all efficiency standards. Under the administration of George W. Bush, the neoclassicals helped to derail appliance-efficiency and fuel-economy standards—government-run, free-lunch programs all.

Some neoclassicals are a bit less hostile to efficiency standards. William Nordhaus, a Yale economist who has probably spent more time studying energy and climate-change policy than any other economist, simply calls the policies favored by physicists “fluff.” Lovins, from the physics camp, is well aware of the neoclassicals’ view and enjoys talking about putting “several trillion dollars back in Americans’ pockets” and then saying, “That’s not a free lunch. It’s a lunch you’re paid to eat.”

Mainly, the physics camp spends its time defending its position rather than attacking the neoclassical position. But since the times when a Democratic Congress attacked President Gerald Ford’s $2-per-barrel tax on imported oil, the physics camp, by constantly downplaying the importance of energy prices, has lent support to those who oppose strengthening price signals. Many in the physics camp believe in so many free lunches that they think price matters very little.

I am particularly interested in this policy war, because I believe one key to recovering from oil addiction and reducing carbon dioxide emissions is a fuel-economy policy for cars and light trucks (including SUVs), something the neoclassicals dismiss out of hand. At the same time, I believe the neoclassicals are right that raising the price of carbon is the most important step, though I favor doing so with an Untax, not a tax. The policies of the two sides, in my opinion, actually complement each other.

Efficiency Measures Can Save Money

Economists are not all strict neoclassicals. Allow me to introduce Stanford economist Kenneth J. Arrow. A winner of the Nobel Memorial Prize in Economic Sciences, Arrow is one of the most respected of all economists and is a central figure in the development of mathematical neoclassical economics. In 2007, after the U.K. government issued a major report estimating that the cost of climate stabilization would be between 3.4 percent and negative 3.9 percent of the world’s total gross national product (GNP), here’s how Arrow responded: “Since energy-saving reduces energy costs, this last estimate [negative 3.9 percents] is not as startling as it sounds.”

If Arrow thought a cost of negative 3.9 percent of GNP was impossible, he would have called it startling. Instead, he said it was not so startling. Arrow is saying that a cost estimate of negative 3.9 percent of GNP just might be right. A negative cost means a net savings. Since this concerns the global economy, he is saying there just might be $3 trillion per year of free lunches from energy-savings schemes.

In other words, according to Arrow, quite a few of the physicists’ favorite policies might save more than they cost. This opens a door slammed shut by the neoclassicals’ extreme views, which are based on an assumption of completely rational consumers. However, it’s important to note that both Arrow and the author of the report believe that total cost is more likely to be plus than minus. That means they believe that, although some policies may save more than they cost, it’s most likely the policies on average will cost more than they save.

If it just might be possible to save a few trillion dollars per year instead of paying a few trillion extra, it seems foolish not to even try just because of some disputed economic theory (see Suspect Theory).

The Taste of a Free Lunch

When Art Rosenfeld looked into refrigerator efficiency, he didn’t need any fancy economic theory to tell him we were being charged way too much for “lunch.” Art Rosenfeld is a real physicist. He coauthored a text in nuclear physics with Enrico Fermi, who developed the first nuclear reactor and who won the Nobel Prize in physics in 1938. Rosenfeld also participated in the discovery of subatomic particles with Luis Alvarez, who won the Nobel Prize in 1968. In 1973, at the start of the first energy crisis, Rosenfeld noted that “if we Americans used energy as efficiently as do the Europeans or Japanese, we would have been exporting oil in 1973.” He’s been the country’s top energy-efficiency expert ever since.

By 1975, Rosenfeld was hard at work developing residential building standards, and in 1976, he recommended an efficiency standard for refrigerators and freezers to California’s governor, Jerry Brown. That’s how appliance standards got started.

When Rosenfeld looked into refrigerator efficiency, he found a wide range of efficiencies but no correlation at all between cost and efficiency. It appeared that a lot of money could be saved on electricity by buying an efficient refrigerator that didn’t cost any more—but people weren’t doing that.

Before standards, manufacturers skimped and used fiberglass insulation instead of rigid polyurethane. They made the walls thin to get more room inside. With thin walls and poor insulation, the outsides of the refrigerators got cold enough in spots to cause condensation. To prevent this, some manufacturers installed heaters in the outer walls of refrigerators! The heater uses energy and then the refrigerator uses more energy to cool the heater.

Apparently, because consumers paid no attention to efficiency, manufacturers saw no point in spending any money to make them efficient. This was reflected in the history of refrigerators. Between 1950 and 1974, energy use per refrigerator grew more than twice as fast as refrigerator size. While the size of refrigerators more than doubled, their energy use more than quadrupled.

But the high energy prices of the first energy crisis changed all that. People starting thinking about saving energy, and that’s difficult when you have no idea how much energy an appliance uses. So in 1975, the federal government required energy-efficiency labels on some appliances. In 1978, California imposed efficiency standards, tightening them in 1981 and 1987. The federal government took over the process and set even tighter standards in 1990 and again later.

As a result, by 2001, refrigerators used 69 percent less energy than in 1974 even though they were 20 percent larger. Saving that much electricity saves $127 per year. Meanwhile, the cost of a refrigerator had dropped by half. If the extra efficiency had a cost it could not have been much because by 2001, the average price was only $850. Even if $400 of that was attributable to efficiency, which is highly unlikely, it would have been repaid in three and a half years by the reduced cost of electricity. For the next twelve or more years of the fifteen or more years that a refrigerator will last, the $127 per year of energy savings would be gravy. It looks like refrigerator standards are a lunch we’re paid to eat.

Are New Car Buyers 100 Percent Rational?

I am skeptical that Mankiw read any studies (if they exist) proving that car buyers are perfectly rational before he predicted in the New York Times that “any truly cost-effective increase in fuel efficiency would already have been made.” Neoclassicals usually rely on their theory for these sorts of pronouncements.

Cars now come with Environmental Protection Agency (EPA) mileage ratings, but these fall far short of telling consumers their total future gas costs if they buy the car. Neoclassical economics assumes consumers know this and much more. In particular neoclassical theory assumes consumers will

• Estimate the price of gasoline for the next ten to fifteen years.

• Estimate how many years they will keep their cars.

• Receive the full remaining value of gas savings when they sell their cars.

• Estimate how far they will drive their cars each year before selling them.

• Estimate how much their actual mileage will deviate from the EPA ratings.

• Discount future savings at a percentage corresponding to either the interest rate on their credit cards or the interest they earn on investments. (Even economists find this one confusing.)

If consumers make all of these estimates without bias and purchase their new car on this basis, Mankiw should be right. The conventional wisdom on car purchases is, however, that consumers take account of less than half of a cars’ future gasoline costs.

Limits to Free Lunches

So it looks like the neoclassicals are wrong. Call them free lunches if you like, but there are opportunities to save more on energy that it costs to gain efficiency. And, at least in the case of refrigerators, it looks like some of that opportunity was captured by a government regulation—an efficiency standard for refrigerators.

But all this really shows is that the extreme neoclassical position is wrong. Perhaps very few efficiency standards can save more than they cost, or perhaps the opportunities are enormous. Either the neoclassical view or the physics camp could be nearly correct.

Unfortunately, both sides are so sure they are right, that neither side documents their case carefully. Even the case for refrigerator standards is clear only because it is dramatic, not because it is well documented.

Generally, claims that the programs will save more than they cost omit four considerations, each of which can be quite important:

• Regulatory inefficiencies.

• The take-back effect.

• Consumer inconvenience.

• Consumer variability.

Regulatory Inefficiencies. Neoclassicals assume perfectly efficient markets. Physicists implicitly assume perfectly efficient regulation. This bias is the result of omitting any cost for regulatory mistakes, such as setting a standard incorrectly. I have not found energy regulators to be any more rational than new car buyers.

The Take-Back Effect. When an appliance is made more efficient it often becomes cheaper and more convenient to use. Consequently people use it more or buy a bigger one. This is a benefit to society and actually makes efficiency programs more valuable than the physicists claim. But it also means efficient appliances use more energy than estimated. (See Take-back numbers.)

Consumer Inconvenience. Some ways of gaining efficiency cost no money, but do cause inconvenience. For example, making the walls of a refrigerator thicker means it either takes more space in your kitchen or holds less food. The cost of such inconvenience is nearly always ignored.

Consumer Variability. If I run my air conditioner 1000 hours per year, any improvement in efficiency will be 100 times more valuable than if I run it 10 hours per year. It does not make sense for a low-use appliance owner to buy as much efficiency as a high-use owner. This means that even the best efficiency standard is likely to be a waste of money for the low-use owner. I have never seen this accounted for.

Although there are many imperfections in markets, that does not mean there are many free lunches. There’s only a free lunch if the problem can be fixed at a cost that is less than the savings. All four of the considerations just discussed either raise the cost of fixing the problem or reduce the benefit. Since they are generally ignored, the claims of free lunches are frequently overstated.

A Pricing vs. Efficiency Compromise

The physics camp wants many efficiency regulations and cares little for price signals. Neoclassical economists want only price signals and no efficiency regulations. The resolution of this conflict flows from the first principle of fossil philosophy, as explained in chapter 1: Treat the problem not the symptom.

The two camps focus on two different problems. The price of fossil fuel is too low, so we need the neoclassical solution of higher price signals. Consumers are short-sighted when evaluating future energy savings, so some efficiency standards can help them save money. Many if not most economists favor both approaches when each is used to solve the matching problem.

This compromise rejects the extreme neoclassical position, but it also requires two changes in the tendencies of the physics camp. It requires taking fossil-fuel prices far more seriously and it requires backing away from the notion that physicists know how to fix literally hundreds of market imperfections while saving money.

I believe economists are right to be suspicious of large numbers of “market-fixing” efficiency regulations. Their skepticism is not based on an implausible assumption of consumer rationality, but on the four realistic concerns listed above, which are consistently ignored. Moreover the design of even major efficiency standards is poor and fails to use modern economic tools. This has resulted in such fiascoes as fuel economy standards that remained at their initial 1975 setting for over thirty years, and that reward designs that kill more people while using more fuel. I am referring to the requirement that cars be aggressively redesigned so they can be reclassified as trucks and qualify for a lower fuel efficiency.

It would accomplish far more to design the major standards well and evaluate them carefully, rather than to charge ahead with hundreds of smaller measures that ignore economic concerns. But the real challenge for the physics camp is to accept the importance of price and to realize that their entire campaign is at risk without the proper price signals.

Having worked in the physics camp for years, I have heard many excuses for ignoring the take-back problem, but have never actually seen it taken into account. This is too bad because if the problem were faced squarely, the natural conclusion would be that the neoclassicals’ price signals do not substitute for efficiency measures but rather are a necessary complement.

The physics camp’s excuses for ignoring take-back are all short-term considerations, but think about the history of take-back. Ordinary bulbs are 150 times more efficient than candles. But we don’t uses 150 times less energy for light, we use more energy than in colonial times. Scientists of the past have provided us with enormous efficiency gains, but never enough to reverse our increasing use of energy and fossil fuel.

So the physicists are taking a real gamble. Compact fluorescent bulbs may save energy this year, but ten years from now, people may have discovered they can afford to light their gardens at night as brightly as the sun lights them in the daytime. And the less energy a light uses the less it pays to switch it off. If history is a guide, increasing wealth combined with the take-back effect will eventually win out over the energy savings of increased efficiency.

There is a simple way out of this dilemma. Raise the price of electricity, and refund the extra cost. This is again the Untax, and exactly why this works will be explained in chapter 16, but here is the outcome. The higher price of electricity reduces or reverses the take-back effect. The price increase should not be tied to the efficiency gain, but implementing the price signals of the economist greatly reduce the risks of take-back inherent in the physics approach.

The compromise then between the extreme neoclassical camp and the physics camp is simple and positive. The most important efficiency programs, especially fuel-economy standards, should be accepted and perfected. Neoclassical economists should stop arguing against these on the basis of untested theory.

The physics camp should recognize that there are real problems with “fixing” markets and that blind faith in regulatory fixes is no more appropriate than blind faith in markets. Physicists and economists should join forces to make the big efficiency programs work better, and on implementing better fossil-fuel prices. This will help protect the energy efficiency gains from the energy-consuming take-back effect.