New arguments on a carbon tax

From Adam Ozimek, here are some very good points, which I had not previously pondered:

…what a carbon tax does is push the required cost threshold up. This would allow solar to become the more profitable source of energy in the US sooner and increase the speed of its dominance here.

However, a carbon tax would raise the threshold in the US relative to the threshold for developing countries.  In other words, the race for solar companies in the U.S. becomes to be cheaper than dirty energy + a carbon tax, which is a higher threshold than being cheaper than dirty energy alone, which is the threshold in many developing countries.

It is easy to see how this could cause downward march in solar costs to slow, and as a result solar would reach the threshold for China, India, and other developing countries perhaps much much later.

If this is true, it would suggest that for clean energy to become globally dominant faster it’s better for the U.S. to just subsidize solar innovation and let the untaxed U.S. market price of dirty energy stand as a strong incentive for solar to drive costs lower.

To see this, consider a world where solar was already dominant in the U.S. with current technology and costs, perhaps via a total ban of dirty energy. The supply curve of the installed base of solar technology would be much more price inelastic than the supply curve of today’s installed base of dirty energy due to higher fixed costs and lower marginal costs.  This means a steeper residual demand curve for marginal innovators that provides less market share rewards for marginal declines in price, and therefore lower rewards for marginal cost cutting.

In this way, a carbon tax could make global warming worse.

From Jerry Taylor at the new Niskanan Center, here is a paper on the conservative case for a carbon tax.


Tag team posting, Marginal Revolution style.

From Adam:

Overall, the downward march of solar energy costs is perhaps the most important factor determining how successful we will be at mitigating global warming.

Given the state of modern climate models we can argue for quite a while as to the size of projected impact of global warming. However cheaper energy is good for everyone. If and when PV becomes cheaper than other forms of power generation it will be a great day.

The problem is that I don't see a much of a role for government here. This is an incremental research problem. Perhaps funding in basic science, but that is somewhat scattershot and who decides who gets what funds? Anyway, the number of watts / dollar from PV has been following an exponential over the last 35 years across multiple administrations. My guess is that the government needn't do anything.

It really has nothing to do with basic science. At this point, it is a straight up engineering problem. How do you lower the cost of producing solar cells? How do you lower the cost of installation?

While there is an argument to make that government can fund basic science, government does not do engineering. They don't make things cheaper to build or install, quite the opposite. Subsidies raise costs, look at heath care and higher education.

There are basic research opportunities in energy storage, which is vital to make solar and wind useful as any more than a fringe energy source.

There are basic research opportunities in making nuclear fission power more economical, although I find it amusing that those who think that climate change is The End Of The World won't consider new nuclear.


In a market economy, innovation comes from a a large stochastic search space. We innovate better than others not because our leaders are smarter or because a few big thinkers have the ear of government, but because we try MANY different things, and test them against each other in a free marketplace where they can live and die on their merits. We tolerate failure and accept it as the cost of progress.

It's tempting for academics and government officials to believe that they know where research money should be applied, which technologies are 'promising' and which aren't, and therefore to think they can speed up innovation with a regulation book in one hand and a checkbook in the other.

In the real world, the effect of government subsidy is to reward large, connected firms and punish the smaller ones. I know this from personal experience. I was working in a small research lab when our government decided to 'stimulate research' with subsidies.

Of course, the subsidies were strangely targeted at specialties the largest players in the industry seemed to be working on, suggesting that they played at least an advisory role in 'helping' the government decide where the money should go.

But worse than that, for a small shop already overworked and lacking a legal and accounting department, it is very hard to meet the legal and reporting requirements the government puts on you. So the big firms with legal departments win, get the contracts and the subsidies, then use that to drive down their prices and force other players out of the market.

The result is consolidation and ultimately a reduction in innovation as top-down direction slowly replaces bottom-up discovery.

If you want to drive innovation, the best thing you can do is look at whatever regulations exist that act as barriers to new startups, and reduce them. Make the cost of trying a new idea as low as possible. Encourage many different avenues of research, rather than subsidizing the ones you think are the 'best'.

Ultimately, the reward for a new clean energy source is already high enough to attract massive investment in R&D. There's no reason to believe the government can or should speed up the process by dumping money on some of the players.

One of the major arguments from the environmental left around climate change is that there needs to be massive, massive amounts of state investment into clean, zero-carbon energy, on the scale of the WW2 mobilization efforts when the government ordered private companies to start producing things for the war effort--which resulting in the US becoming, in a very short amount of time, a massive industrial powerhouse of war-related infrastruture and equipment.

All of this really depends on the paradigm one takes. If one thinks the task is to "make solar panels cheaper so that the market can sort out climate change", then yes there isn't a whole lot the government can do outside of trying to figure out how to tweak incentives and prices. On the other hand, if one understands that the task is to "cut carbon emissions to zero", then suddenly the obsession about trying to tackle the market disappears and you have a whole mess of new options.

I disagree. I have a background in semiconductors - the technologies being installed today are little changed from 2 decades ago. There are technologies which are tremendously better - and these are needed because solar PV capacity factors are 1/3 or less that of "dirty" energy.
Thus similar costs per watt of solar PV vs. "dirty" energy is tremendously misleading because the energy production capacity of said solar PV is actually far lower: A 500 Megawatt solar PV plan only produces as much power as a 166 Megawatt natural gas plant.
It is good that the recognition of the true alternative energy benchmark is finally discussed: emissions growth is not going to come from the manufacturing-declining 1st world, it is going to come from the 2nd and 3rd world nations as they get richer over time. Failure to provide competitive alternative energy production technology means emissions growth via the enrichment of the vast majority of the world's population will be unstoppable.

Sure, this violation of Econ 101 (where higher demand causes lower quantity demanded) is a logical possibility, but I think it's irresponsible to suggest it corresponds to the real world in any way.

1) The reason we're coming down the solar cost curve is because as more and more units are built, factories learn to build them more efficiently. The industry also benefits from economies of scale. Increasing demand will help both of these factors, which should help lower the price.

2) Demand for solar is NOT a step function.

2a) Power plants are built very slowly. In the US, with its mature economy, electricity demand is only growing about 1% per year [1]. As a result, most power plants are built to replace old units rather than to meet new demand. And because old power plants have high fixed costs and low marginal costs, even if solar becomes cheaper than coal, it will still take 50+ years until all the coal plants are replaced [2]. This 50+-year decommissioning schedule is one factor that spread outs demand.

2b) Solar has diminishing returns. Solar produces the greatest amount of power during the day, when electricity is (usually) the most valuable. However, as more solar gets added to the grid, the value of daytime electricity will go down, leaving the evening as the new daily peak. This will make solar less valuable relative to natural gas and other controllable sources.

2c) Solar has increasing costs by adding instability. As our grid stands, it can handle around 30% of its generation coming from solar without major stability problems [3]. However, as we go beyond this point, we'll need more natural gas or demand response or possibly even batteries. Because of this additional costs, it becomes marginally more expensive to add more solar to the grid, again spreading out the demand curve.

2d) Different regions have different demand. In California, which has sunny weather, expensive electricity, and large state incentives, solar is demanded much more highly that, say, Washington, where the weather is cloudy and electricity is cheap from hydroelectric. Region-to-region variance is yet another factor spreading out the demand curve.

2e) Building gazillions of solar fields will take decades. Even without the power plant aging issue, it's still going to take decades to install the vast quantities of solar required to transform our grid. Utilities are cautious. They will add solar gradually to minimize risk of problems. Also, permitting can take years, not necessarily for the solar itself but for transmission lines to connect the solar (e.g., San Diego's Sunrise Power Link) [4]. And because time can be traded off with money/price, the non-instant construction schedules also spread out the demand curve.

3) This logic applies more to monopolies than to competitive firms. Ok, so suppose the demand curve really is a box and there does happen to be a known innovation that happens to be profitable if everyone's demand is the same but is not profitable if everyone's demand is different (quite a coincidence of low probabilities, but let's assume it's true). If a monopoly was building all the solar, then sure, a carbon tax that spreads demand out would make that innovation unprofitable. However, if you have a competitive industry - say, ten firms with 10% market share each - then for each firm, the benefit from the innovation would not only help them enter 100% of the developing markets, but also eat into the remaining 90% of the developed markets.


Ted, here in South Australia we get 6% of our electricity from solar and none of it is utility scale solar, it's all on rooftops. And rooftop solar has the capacity to be very disruptive. Currently, before tax or subsidy, the average installed cost for the median sized system in Australia is about $1.61 US per watt and larger commercial rooftop installations are being done for about $1.21 US a watt. This low cost of installation means rooftop solar can pay for itself without subidy or a feed-in tariff. So even if all the electricity exported to the grid is simply stolen, many people will still find it worthwhile to install solar. Our installation cost is currently about half what it is in the US and since we're not doing anything magical here, and it's a high wage country, there's no reason why the US can't match us on installation costs. In Australia with our currently low to no feed-in tariffs, as solar comes down in cost, people have an incentive to increase the size of their systems in order to increase their self consumption and avoid using much more expensive grid electricity. This means more solar electricity is exported to the grid during the day, and so within a decade we could end up with most of our daytime electricity use being generated by rooftop solar.

Thank you, Ted, for making economic sense.

Adam's argument relies on the following theory of firm behavior: "the innovative effort is proportional to the profit margin the the US", which is a substantial departure from the usual theory of profit-maximization i.e. we have a hurdle theory of innovation, where the hurdle is the price of dirty energy + any tax in the US. Even though solar technology has a market internationally, the manufacturers apparently don't care about any returns outside the US, according to Adam's argument. That the carbon tax doesn't change the profit margin abroad of any innovation should be clear.

Sounds like an argument by the coal lobby. In the U.K., the demand bump occurs at around 7pm, since solar is already offsetting the increased demand during the day. (The graph is showing demand minus distributed solar and wind production.)

For a long time I've been saying that solar is great for peak demand, but some countries have already passed that point.

I'm not in favor of a carbon tax, but this argument just seems self-evidently wrong.

Why would a higher price point for the viability of solar in the U.S. reduce downward pressure on solar costs? Even if I can manufacture a widget for $90 and sell it for $100, it is still in my interest to reduce the cost of manufacture as much as possible. Doing so will help me beat out competitors and/or increase profit per item. And the fact that I can already sell the item profitably gives me increased resources to work on lowering the manufacturing cost. If the market would only bear a price of $80, then I would have to take a lot of upfront risk to invest in research in how I can get the manufacturing cost below $80.

Is there such an example in a subsidized market? It is far easier and cheaper to buy a few politicians than engineers.

A subsidy would only harm solar innovation if certain companies were favored regardless of results. Some programs aiming to boost renewable energy might have that effect, but not a carbon tax.

A carbon tax would:

1) Harm carbon-related innovation.

2) Increase the incentive for non-carbon power sources to innovate.

It seems to me that a carbon tax applied in one region only has other effects that will counteract the positive gains from the tax, but I've never seen these quantified:

1. A carbon tax in the U.S. would create or increase other countries' comparative advantage in energy. That could mean energy-intensive industry would move to China or other locations that don't have the tax. A tax large enough to drive large-scale adoption of solar power is one large enough to cause massive shifts in the pattern of energy-intensive industrial development.

2. A carbon tax would reduce the consumption of fossil fuels in the U.S., which would drive down the price of oil even further on the world market, making it harder for solar to compete without subsidy elsewhere, and slowing the growth of alternative energy. The drop in the price of fossil fuels on the global market would increase the comparative advantage describes in my first point, further hurting U.S. manufacturing and increasing manufacturing in countries that may not be as energy efficient. This could cause the carbon footprint of many goods to actually increase.

3. To the extent that unilateral action on carbon emissions by the U.S. helps with the global warming problem, it lowers the social cost of carbon emissions for everyone else, further reducing the incentive to move to alternative energy in the rest of the world.

I believe these effects are real, but I don't have a good sense of how important they are. Does anyone have data or even informed thoughts on how much of an effect this would have on overall carbon emissions?

There is an extensive economic literature on carbon leakage, which is exactly what you're talking about. Models vary, but typical estimates put the amount of leakage in a range from +20% to -9% (that is, what share of US emission reductions is negated by general equilibrium effects in other regions).

Yes, that's not a typo, some models predict negative leakage--a tax in the US can actually reduce emissions elsewhere. Why? A carbon tax would increase investment in the US clean energy sector, drawing capital away from other sectors, including dirty, capital intensive sectors in other countries. This reduces the size of those sectors, and hence emissions. This "negative leakage" effect is proven in a 2014 paper by Baylis et. al. Of course, this is just one component of the many general equilibrium effects--its existence does not imply that net leakage will necessarily be negative.

Also, the forces pushing positive leakage are often weak. For example, your point (2) argues that increased solar will reduce the consumption of US oil, driving down its price. But the US does not use oil for electricity generation, so increased US solar would have no effect on oil consumption or prices. Solar would primarily reduce US natural gas consumption (the marginal generation fuel during the daytime), which is not currently exported abroad. And if the gas export plants get built, it will still be very expensive to ship abroad. So solar might reduce domestic natural gas prices, but it wouldn't have much of an effect on foreign natural gas prices, barring a technological breakthrough on LNG shipping.

The bigger concern would be that a carbon tax leads to fuel switching from coal to natural gas, reducing coal demand, and hence international coal prices. My guess is this effect might be small, since the marginal cost of coal mining is pretty much constant (a flat supply curve), implying that prices don't respond much to a change in demand.

The relocation issue (your point (1)) is certainly a concern, which is why (a) the recent US-China climate deal is so important and (b) border adjustments are necessary to equalize the location incentives across countries.

Ugh, a typo somewhere didn't close my hyperlink. Sorry!

Dan111, maybe the arguement holds up if we assume the following conditions: (1) Americans are the only ones who can make cost improvements, and (2) once they beat fossil fuels + carbon price they give up on account of how no American has ever had an interest in selling to a foreign market or in beating domestic competitors.

I am in favor of a carbon tax and agree with your analysis. His argument is extraordinarily weak - heavy on subjunctives, light on evidence. The very opposite of what he suggests is likely to be true - as solar is deployed on a wider scale, there is likely to be a higher chance of increasing innovation.

Another major weakness is the author's arbitrary emphasis on solar. Without huge reductions in the cost of energy storage, variable renewables like solar and wind can only displace a fraction of fossil fuels. Total grid costs start to rise considerably as solar penetration exceeds its typical capacity factor of ~15%, so it is difficult to see solar penetration exceeding this by much. On the other hand, putting a tax on carbon would inspire many different efforts to reduce emissions, including nuclear power and carbon capture and storage.

Overall, I am surprised that this article made Tyler's cut.

True but i could see more adoption of solar in the west keeping the cost of fossil fuels lower.

Yes, but that could only diminish the effect of a carbon tax, not completely eliminate it or reverse it (as the author is claiming).

The cost of fossil fuels only goes lower if there is less demand for them, meaning that reduced carbon use is happening, just as the tax intended.

Right. Don't reason from a price change!

Solar is problematic for China as a source of domestic power but it hasn't stopped China from becoming the world's leader in the production of solar panels. U.S. made solar panels may be better quality but they are much higher in price, the much higher price in the U.S. sufficient to discourage conversion to solar. The theoretical points made by Ozimek are fine, but they ignore the practical differences, the real world differences, that impact solar, its development, and its use. Indeed, the import duties imposed on Chinese and Taiwan made solar panels do far more to discourage the conversion to solar in the U.S. than does theory.

I don't think that curve will ever cross. Natural gas is much more reliable and cheaper than solar, as well as cleaner. There is no way in hell that the dense cities of the eastern US could ever be provided with enough power via solar. Maybe the spread out suburban developments in Houston where the sun shines and there is adequate surface area per load.

Derek, rooftop solar here in Australia already has fossil fuels beat. Because it competes with the retail price of electricity rather than the wholesale price, even if the cost of generating electricity from coal or natural gas was zero cents a kilowatt-hour it still wouldn't be able to compete with rooftop solar. Looking at satellite photos of New York I see plenty of roofspace that could be used for solar power. And looking up New York electricity prices I see that even with a 10% discount rate, at Australian installation costs, electricity from rooftop solar would be cheaper than New York grid electricity.

"Derek, rooftop solar here in Australia already has fossil fuels beat". This is only true for low penetrations of solar. See the papers of Hirth et al. Also, comparing retail prices to wholesale prices means ignoring the cost building and maintaining the grid. Unless you advocate going totally off grid, which is generally not economical, it is incorrect to pretend that these costs don't exist.

Jdm, because households and businesses have in incentive to install larger systems as the cost of solar comes down in order to increase their own self consumption and avoid paying for more expensive grid electricity, we could end up with quite high solar penetration from rooftop solar alone. As for the costs of the grid, we have a weird system in Australia where the grid is expected to pay for itself by selling electricity. For example, they will take solar electricity generated by people's roofs or coal power plants and maybe pay 5 US cents a kilowatt-hour for it, or they may pay nothing, and they will sell it to people. If they sell it to me then all up on my last bill I paid 35 US cents a kilowatt-hour for grid electricity.

Do you use electricity at night and on cloudy days? If so, where does it come from? Is the generating capacity which supplies this electricity free? Is the grid which delivers this electricity and maintains a constant potential for anyone to draw power at any time free? Let suppose everyone in Australia were to put solar on their roofs, how much would the non-solar capacity required to maintain the grid decline? When calculating the cost of an energy system, say x% solar and y% wind and z% coal or whatever, you need to calculate the total costs of the system. It is a mistake to look at one small part in isolation.

Jdm, rooftop solar lowers transmission costs in two ways. Firstly by reducing peak energy demand, which in most countries occurs in summer and definitely occurs in summer in Australia. Secondly, it shifts demand to take advantage of the cheaper daytime electricity. That is, people use less electricity during the new peak in the evening and more electricty in the daytime. Rooftop solar also benefits consumers by putting downward pressure on wholesale electricity prices. And just to be clear, people use the grid to export solar electricity and the feed-in tariffs for new solar are now low, sometimes zero, and electricity distributers sell that electricity at the normal retail price. Which is about 35 US cents in my case. Now at some point, we may have so much rooftop solar capacity that the decrease in grid electricity use that results will cause grid costs per unit of electricity sold to increase by more than transmission costs and wholesale prices are lowered by rooftop solar. If that happens, then distribution costs would increase. Given the huge savings to consumers that rooftop solar results in, that doesn't look like much of a problem. It's the normal result of less demand for a product with high fixed costs and is something that happens quite frequently/ Over time we will see the grid adapt to the new situation. For example, with the decreased cost of energy storage grid operators are now looking at using battery storage to meet evening peak demand in some high growth locations, either with storage on grid or through subsidising energy storage in people's homes and businesses. And yes, this has costs, but they are considering it because it is cheaper than the alternative of increasing transmission capacity.

Perhaps the sunny nature of Straya has something to do with it?

NP, the two maps you linked to measure two different things. One is hours of sunshine in Australia, which is the average number of hours a day that at least seven eights of the sky is free of cloud, while your map of the United States shows the average number of kilowatt-hours of sunshine on a one meter square panel fixed at an optimal angle to receive sunlight. They are two quite different things. And when you think about it, how is it possible for somewhere to be twice as sunny as Florida or California? In reality, the inhabited parts of Australia, where people actually live, are comparable to the sunnier half of the United States.

Title of map says Map above reflects the average amount of hours of sunlight received on a daily basis.

Is it that big of a mystery how that could be possible?

You are the one who stated Australia not a certain area of Australia, don't move the goal posts here.

N P, hours of sunshine in a location and kilowatt-hours of sunlight falling on a fixed panel are different things. Berlin gets an average of 1,626 hours of sunshine a year, which is an average of 4.45 hours a day. If they are the same thing, then it would mean that Berlin is sunnier than Florida. And this really is not the case:

But you get sun. Do you have to heat your homes? As I said, Houston texas wold be great for that kind of installation.

A localized installation without transmission losses, and the loads geared to solar generation with some kind of localized storage makes sense if there is enough sun. That doesn't apply everywhere or even many places.

Australians live in the most hospitable parts of the continent, which are the parts with the most clouds. As a result a very large number of US cities receive about the same level of sunshine as Australian cities do. For example, New York receives more sunshine than Melbourne does. It is roughly between Sydney and Melbourne with regards to insolation.

All this is beside the point. I think the goal of a carbon tax shouldn't be to encourage renewable vs carbon fuels, but rather to discourage energy consumption, period. Less driving, more fuel-efficient cars, smaller homes, more efficient air conditioning, etc.

If a huge carbon tax were to raise my home energy costs, I wouldn't run out and buy solar panels. It would make a lot more sense to change my habits and reduce use of (or upgrade) the energy hogs in my house.

Bingo. The trade-off is not between solar energy. The trade-off is between polluting and not-polluting. A carbon tax encourages not just solar energy but all forms of less-pollution intensive activities. I suppose the relevant counter-argument is that the US will not be the biggest polluter, that coal energy in developing nations will be, and that the most effiective way to reduce pollution is to encourage solar energy to fall below the cost of coal energy as quickly as possible, regardless of other "polluting" activities. Not sure I agree with that, as subsidies merely allow solar companies to operate at loss.

I thought the point of the carbon tax was to internalize the costs of externalities, not social engineering.

Hey at least this guy is honest, you don't see that too often.

"The main goal of a carbon tax is to have power over others and make their behavior conform to my beliefs."

Sounds like the core belief of liberalism to me!

And what if increased CO2 is contributing to a beneficial moderate warming of the planet, or preventing a natural cooling cycle?

Scientists really have no idea what is going on. The climate is too complex and our data too meager. The complete failure of the models have shown this quite clearly.

So any cost-benefit calculation must consider the potential harm that may come from reducing carbon, along with the equally important consideration of how much will spending x dollars on carbon reduction end up harming the millions of people who continue to die prematurely because of food scarcity and preventable disease.

Face it, that half a billion thrown at Solyndra could have saved many people who are now dead today.

No argument on carbon tax should be "waged" without factoring carbon subsidies.

never happen. That wouldn't advance the narrative.

So, re-title the post: "New arguments on national bankruptcy."

Soem point in the ugly future: "Oops, we (Big Brother, et al) didn't intend to destroy the economy . . . "

I gather you are in the free lunch economic camp where job killing destruction of capital especially by burning capital is the way to grow GDP in a time of labor surplus because the lower labor costs are, the higher the profits, and the higher the profits the greater the wealth that bankers can lend to people with little to no income or assets to buy increased GDP. Followed by periodic redistribution of wealth when the bubble bursts by Constitutional government technocrats giving the wealth of the wealthy to the poor: the bankruptcy judges.

When you have a labor shortage, destroying and burning capital becomes a common way to spur growth. Andrew Jackson engaged in slash and burn ag, so he wanted Jefferson's treaties broken so he could take the property of others to keep his farming going without having to invest in the land to keep it from being used up.

But by the 20th century, the capital of the United States that had made growth so easy had been extensively depleted. The new land for farming to replace the land that had been depleted was gone. The forests that could be clear cut were getting fewer and fewer. (Some regions did not follow that pattern largely because of the immigrant farmer cultures which treated the land with care as God's stewards. The rotated crops and use to keep the land in good condition and in some cases improved the farm land's productivity without adding stuff to it.)

Solar and wind are harvested by those who invest in the land as a capital asset. Many farmers and ranchers want to install wind turbines to add to the diverse revenue crops they harvest, but can't because of a lack of power lines.

Coal depends on rail that was heavily subsidized by government in the 19th century. And the "free market" will not invest in rail without special government subsidies even when the rail is required for the venture, like mining in Africa or SA. If the "free market" was able to build railroads on their own, then SA and Africa would be crisscrossed with rail like the US is. Coal power electric is not a solution in Africa because the coal can't be delivered to the coal power plants. The same is true of most of the electric grid - the US grid is the result of Federal and some State policies, not the free market.

The "free market" prefers pillage and plunder to building capital assets, killing jobs and slashing consumer spending capacity in its reach for profit, and then blames government and demands government give consumers money to spend to generate the profit growth the "free market" wants but refuses to pay for by paying its labor a fair price so consumers have more money to spend.

Just remember, wind and solar power is more expensive than the pillage and plunder of burning fossil fuels and causing massive destruction of the land because a sustainable solar and wind economy needs to pay a far larger labor force than that required to pillage and plunder existing capital assets.

Capitalism requires sacrifice: productive capital assets are not free, and they don't even grow on trees as rapidly as human labor can build assets. The solar efficiency of plants is roughly 0.1% for the conversion of solar into capital assets that can be burned for energy. Solar cells and wind are converting incoming solar energy at the 5-10% range, two orders of magnitude higher than mother nature. Fossil fuels are even less efficient given a century of fossil fuel burning of the West required mother nature working for several hundred million years to build those capital stores.

Easy solution: Enact a carbon tax now, so that we get the expensive solar first; but promise to reduce the carbon tax slowly over time, so that solar technology is forced to become more efficient. (The promise is necessary to generate long-term research.)

The hard part is figuring out what the timeframe should be, and how high the carbon tax should be; the really hard part is having the markets trust the promise.

carbon taxes are extremely regressive like consumption taxes. Increasing energy costs now in *hopes* of bring solar costs down will hurt poor people now. Why would we want to hurt poor people in an attempt to develop your solar pet project?

Tyler's premise implies that he wants a carbon tax, but not in its present form. The question is about the precise design of the carbon tax, not about whether it hurts poor people.

the comment was in response to your "easy solution" to enact the carbon tax *now*.

Artificially increasing energy costs in any kind of tax system is going to hurt people especially the poor.

Still besides the point.

Put a $30/ton tax on carbon to raise $192B in revenue. Use that money to write a $1,500 check to every household in America. That would offset all federal taxes (income and payroll) paid by the poorest 20% of Americans as well as half the federal taxes (income and payroll) paid by the second poorest quintile.

There are plenty of ways to make a carbon tax friendly to the poor.

The regressiveness can be offset by using the revenue raised to lower taxes at the bottom of the income bracket. For example, use a carbon tax to boose the ETIC and make the first $6000 immune from payroll and income taxes.

Also I'm not sure a carbon tax would be extremely regressive. Are you sure that the poor use the same carbon as the rich? Carbon intensive activities include doing a lot of driving, taking trips that require a plane, etc. It seems to me you do more and more of these things as you move up the income ladder.

doing a lot of driving and heating/cooling a home are very carbon intensive which the poor do an awful lot of.

the poor will also not have a lot of capital to install solar panels, or other expensive "green" energy products. In fact on the air conditioning side, the switch from R-22 to R143a refrigerant increases the energy costs for the same capacity AC because R-143a is less efficient than R-22, but has a lower "Greenhouse Gas" score.

The poor are less likely to have a car, or have a smaller one if they do. They are also less likely to do a lot of 'pleasure driving'. Homes/apts also tend to be smaller if you are poor. While a rich person may put solar panels on his roof, the fact remains it still takes more energy to heat and cool a 5+ bedroom house than a 2 bedroom apt.

May not be doing much "pleasure" driving, but may have to commute farther just to get to work. Even if they have a smaller home, they are much more cost sensitive than rich people.

'Cost sensitive' is not the same as regressive. If someone spends $500 a month on utilities and pays $50 in taxes while a poorer person pays $100 a month and gets taxed $9 that would be a progressive tax even if the poorer person says they really feel the $9 pinch and the richer person says he doesn't notice the $50.

AndrewL, in Australia the lower income half of population installs more rooftop solar than the higher income half. This is because lower income people spend a larger portion of their money on electricity and so have more of an incentive to save money in this area. (In case you're not aware, rooftop solar is a clear money saver in Australia.) The poorest Australians tend not to have a roof to put solar on, but they benefit from the downward pressure on electricity prices rooftop solar causes and from reduced pollution.

"the race for solar companies in the U.S. becomes to be cheaper than dirty energy + a carbon tax, which is a higher threshold than being cheaper than dirty energy alone ..."

BUT in a competitive market, solar producers also need to be cheaper than other solar producers. If a carbon tax increases demand for solar then one would expect a more competitive market. Instead of increasing prices to a level just below (fossil fuel + carbon tax) prices, I'd expect market competition to push prices closer to the margin, even as that larger market caised marginal production costs to fall due to economies of scale.

In any case, not all solar energy is photovoltaic, or even electric. Using solar energy to heat domestic hot water is low-hanging fruit in areas where air temperatures seldom fall below freezing (for example), as systems that combine solar water heating with natural gas backup are simple enough to be inexpensive and reliable. (In colder climates one needs anti-freeze and a heat exchanger, which reduce efficiency and significantly increase complexity and capital cost.)

Cheaper is always better. If I can produce solar for $10 per unit of energy while you can do it for $15, I will always have an advantage on you. But suppose carbon based energy costs $12 per unit. You are out of business while I am still in the game.

Suppose we have a 50% carbon tax so carbon based energy costs $18 a unit. You sell your solar for $16 a unit and I also sell my solar for $16 a unit. Carbon users have no choice since $16 is cheaper than $18. I, of course, enjoy a much higher profit than you do.

The argument here seems to be for the developing world, this will delay the introduction of solar. Absent the carbon tax, they could buy my cheap solar for $11. In the alternative world both you and I are in business but the developing world cannot afford either of our products.

My quibbles:

1. A cap-n-trade system could benefit more. Instead of snowy Maine trying to put up solar panels, they could pay to install them in some sunny tropical country thereby reducing carbon emissions there much more then they could do at home. I have this same beef with people who claim solar is a failure in Europe because Germany is dark and dreary while Spain is nice and sunny....errr Germany should just pay to set up their panels in Spain and run a power line from there to themselves.

2. 'no tax but subsidize solar' sounds to me like a carbon tax. In essense everyone gets a higher tax, but if you happen to be solar you get a subsidy. Isn't this the same as a carbon tax?

3. While solar is exciting sounding, the fact is the cheapest and lowest hanging fruit of all is simply demand management. Consider someone who works at home 1 day a week. He cuts the gas he uses to drive to work by 20% automatically and he hasn't brought a new car, hasn't invested in a Tesla, hasn't forced society to make gas from corn or gras or build a charging station in the parking lot at work, hasn't required anyone to invent a new way to drill for oil or build a rig in the middle of the artic. Quite a bit, not all but quite a bit, can be accomplished by conservation and price signals are the most efficient way to spur conservation (as opposed to nagging people about turning off lights and such).

The real issue that will prevent solar from replacing fossil fuel is not cost, but system operational issues. On a cost basis, (subsidized) solar is already very competitive with fossil fuels. However, solar energy is not dispatchable and its generation profile does not match load curves, especially in the winter. This means that at higher penetration levels you have too much energy on the grid in some hours and not enough in others. Further, sudden fluctuations in output as a result of cloud cover make it difficult for system operators to balance load and generation. Since there is no large-scale, economical storage, this becomes a reliability issue. If increased penetration of solar (or wind) generation is the goal, our focus should be on the development of reliable, cost-effective storage technologies.

I object to the blanket characterization of non-wind-or-solar energy as "dirty energy." First of all, energy is neither clean nor dirty. It is just energy. Secondly, as some have noted above, the devices and processes that capture and transmit solar and wind energy are not without environmental consequences, as though generated by magic. Factories are built, materials are mined, transmission lines multiplied. Birds are fried by large solar installations, and cut to pieces by windmills. Carbon ought not be the only externality considered.

Allow me to draw your attention to a recently published paper of mine.

I argue that a small multinational carbon tax should be used strategically to achieve several ends. First to greatly expand energy research and development. We need to find non carbon energy sources that will decrease rather than increase energy poverty around the world: sources less costly and more accessible than today's fossil fuels. Second to provide a tax framework if and as global warming concerns become more compelling. And third, to provide an easier first step for multinational collaboration on these issues..

Another problem with government deciding to push a specific technology like solar power (or create a tax system against prevailing energy sources that force adoption of the next-best thing) is that we are forcing adoption of a specific technology that may or may not be the right one for the future. I do not trust government to make good choices in this regard. We already did that with bio-fuels, with disastrous results.

Won't we look foolish if we spend hundreds of billions of dollars on massive solar fields, only to find in ten years that Thorium power or small fusion reactors are much better? Ask France how forcing widespread adoption of Minitel in the 1990's worked out for them when the internet took off.

Solar power is excellent for remote locations, for point-source power in specific applications, for trickle-charging batteries, and many other specialized applications. If the price comes down more, you'll see it start to make a real dent in the grid simply because so many things that used to be connected to the grid will go offline part or all of the time.

But solar power is not baseload power. It's variable and unreliable, as Europe discovered this week when a solar eclipse threatened to crash the grid. A power grid still needs a substantial capacity for providing power on-demand 24/7. One of the problem with mixing high-variability power with baseload power is that you wind up with power gluts during peak power production and then shortages when the variable power goes offline. Germany has to import power when it's dark or cloudy, and it exports power when it's sunny. That only works because its grid-connected neighbors haven't followed it down the same path to high variability power sources. If everyone on Germany's grid was doing what Germany is doing, there would be regular brownouts and other power issues for everyone.

So if you have to maintain the baseload power capacity, supplementing it with variable power means running the baseload at a lower duty cycle, which is less cost and energy efficient. Engineering is a bitch, and the central planners and academic big idea people should have more respect for it..

In addition, solar power requires huge amounts of land area and raw materials. Keeping all those panels clean and running efficiently is going to soak up a lot of labor and lead to a lot of injuries and deaths from falls. It's not the energy panacea some people seem to think it is.

As for the 'price curve' of solar, I think too many people these days have grown up watching the effects of Moore's law and think that price reductions will continue for solar the way they do for computer hardware. But solar isn't like computers. It's more like building cars. Yes, you can get better at it and continue to shave costs, but at some point you run up against hard limits of material costs, labor, installation, and the sheer amount of hardware required.

We don't know where the cost per watt floor is for solar power, but there's no reason to believe that it will continue to drop dramatically in the future. On the contrary, if demand spikes too fast because of government pressure it could cause shortages in material and labor that drive the price up.

this. this this this. the whole point of a carbon tax is that it puts a price on the externality and allows the market to decide what the best way to deal with it is - solar, wind, nuclear, or heaven forbid conservation. straight subsidies to solar is, as they say, picking winners and losers.

The purpose of a carbon tax is not just to reduce the CO2 produced by electricity generation. It is also to spur CO2 reductions elsewhere from building heating and cooling, transportation, food production, to general product consumption and production. And, perhaps most important, get everyone to factor their energy costs into their economic and lifestyle choices. Only a carbon tax will accomplish this. How else are we going to get the idiots who leave their cars idling for 1/2 hour to warm up via remote starters to think about the consequences of their behavior?

In this way, a carbon tax could make global warming worse.

That's true. But far more important is that carbon pricing is highly unlikely to succeed. This explains why:

'Why carbon pricing will not succeed, Part I':

'Why The World Will Not Agree to Pricing Carbon, Part II':

Summary of the two main reasons:

1) for carbon pricing to succeed, around 80% of global emissions need to be included in the scheme, otherwise the cost to the participants to achieve a given amount of emissions reduction is prohibitive, so the scheme cannot be sustained even if started. But to 80% participation rate across the whole world is impracticable; even the EU, the most advanced and longest running of all carbon pricing schemes includes only 45% of EU emissions, the USA monitors 49% of emissions, and the short lived Australian carbon pricing scheme included 53% of emissions. So there is no realistic probability of getting all countries to include 80% of global emissions.

Carbon pricing and other regulatory approaches are the wrong approach. Instead we should be deregulating to allow markets to select the least cost technology. Appropriate deregulation of the nuclear industry is what is needed. Allow the industry to compete. Allow innovation and competition to do what it does. If we deregulate there is the opportunity for enormous cost reductions, accelerating rate of roll out leading to an accelerating rate of competition and innovation. Costs will come down faster. As nuclear advocates know, there is enormous opportunity for costs reductions; e.g. there is the potential for up to a factor of 100 improvement in efficiency of using the nuclear fuel. With faster rollout we save more lives and get electricity to more people faster. Cheaper electricity will also mean electricity more quickly substitutes for fossil fuels for heating and transport fuels (e.g. unlimited transport fuels from seawater and nuclear power: )

2) The cost of carbon pricing greatly exceeds the projected benefits for all this century and beyond: )

The problem with solar panels isn't their price vs fossil fuel energy. Even if solar panels were totally cost free (gratis - which they are - given government subsidies) - they would not be capable of supplying a significant part of our energy needs. The solar panels produce energy - in the best case and best climates - only 20% of the time (typically only 10-15%). This is a fact. We need energy 100% of the time. No amount of carbon taxes will change this simple fact.

I think where the Government needs to get involved is to remove the
net metering laws, and the laws that require utilities to pay extra for
solar produced Kwh.
This may sound counter intuitive, but grid assist is a requirement for most
photovoltaic setups. The energy from solar is too irregular and the density is too low. To be practical, the systems need the grid as a backup/supplement.
The Net metering and high fixed price for solar Kwh, discourage utilities
from allowing grid connections, and we are seeing push back.
If we want photovoltaic power to expand, it needs to be favorable to all parties involved. We also need a viable storage methodology.

Comments for this post are closed