Sean Casten writes, “The Waxman-Markey proposal to reduce CO2 emissions by 17 percent over ten years is constrained only by its ambition.”

Cutting CO2 emissions by 20 percent could be realized with our existing assets — by ramping up our nation’s gas-powered electricity generators and ramping down our coal-fired generators.

Read more about how we can shift our electricity production away from a dirty resource to a clean one.

What all agree on is that uncertainty is unacceptable.  And so, not surprisingly, we get policies like Waxman-Markey that are neither a pure cap nor a pure tax nor a pure subsidy, but a bit of certainty scattered hither and thither. Sausage making at it’s finest.

But do we really have that much uncertainty?  At least in the electric sector (which is, after all, responsible for over 42% of US CO2 emissions), we have a fairly high degree of certainty on two ponits: in the short term, we’ll shift from coal to gas.  And in the long-term, power prices will fall.

Which is probably sufficiently heretical to demand explanation.

Near Term

So why can we be certain of a near term shift to gas?  That’s fairly easy: because we don’t have any other choice.

The current US power mix is supplied by coal (49%), natural gas (22%) and nuclear (19%).  Everything else is piddly.  6% hydro, 2% petroleum and 3% from all other renewables combined.  Given the 24+ month timeline required to design, finance, build and commission any new power plant, the only near term response to GHG pricing is to shift the resource allocation amongst those generators that are already built.  And while nuclear is a low-carbon power source, it can’t generate any harder than it already is.  As noted here (see Fig 4), the nuclear fleet is currently running at a 90% capacity factor, and on historically trends, appears to have pretty well maxed out.  Which means that short of building new nuclear plants – hardly a quick, near term solution – there’s no way to swap coal-fired electricity for nuclear.

The gas fleet, on the other hand, hardly runs at all.  In 2006, the fleet had a 20% capacity factor.  Roughly speaking, this means that any given plant was shut down for four days out of every five.  Gas fleet capacity factor bounces a bit from year to year, but generally stays in the 20 – 30% range.  Thus, if we immediately put a price on carbon that immediately applies to all generators (color me politically naive if you wish), the immediate impact would be to shut some coal plants off and run some gas plants a bit harder.  It’s not a long-term solution, and its cost depends solely on the price spread between coal and natural gas.  But as noted here, it does have the potential to quickly and massively lower the CO2 signature of the US electric sector.

Long Term

Now to the heretical part.

Let’s extend our gaze sufficiently far into the future that new capital has been deployed, facilitating the retirement of the old dirty stuff.  What’s it likely to look like?

I’m not foolish enough to make technology-specific predictions.  But I will go out on one very small limb: power plants deployed in response to GHG controls will be less GHG-intensive than the ones we build today.  Wind, nuke, solar, CHP, biomass, geothermal… and probably lots of other things we haven’t thought of (not to mention lots of end-use conservation).

Here’s the unifying feature of all those technologies: they cost less to operate on the margin than the stuff we use today.  That’s not to say they’re all cheaper.  After all, many of the technologies we will deploy in response to GHG regulation are technologies that today are held back due to high capital costs (solar, nuclear, etc.)  But once a power plant is installed, the decision to run it one more hour isn’t based on capital cost recovery, but on the marginal cost of production.  If it costs me $2.50 to make one more widget and I can sell it for $2.51, I’ll make that widget regardless of how much the widget factory cost me.  That, in a nutshell is why our nuclear fleet today runs all the time and the gas fleet doesn’t.  Inclusive of capital recovery, the gas plants have lower all-in costs… but on the margin, the nuke plants make more sense to run.

This point is key, and too often overlooked.  We assume that new, low-CO2 technologies are held back by economics – but forget that those economics include both capital and variable costs.  And in the long-run, it is only the variable cost that matters.  Shifting to low-CO2 power is therefore a shift to low variable cost power.  Which in turn is a shift to low cost power.

I should emphasize that it may take a while to get to this point, as initial prices from high-cost construction have to be amortized.  A comparison with nuclear in the 1970s is instructive, when huge cost overruns put upward pressure on prices until the political will was broken and owners went bankrupt… but the plants kept running, and today form the low-cost base for much of our grid.  This will happen again with new low-CO2 sources, for the simple reason that CO2 sources (e.g., fossil fuel) cost money.  Cut the source, save the money.

I should also note that there is one exception to the low cost/low CO2 paradigm: Coal with CCS.  It’s low CO2 (if it works) but high cost.  Which is why it will never matter.  It won’t be built unless subsidized, and if it is built, it won’t run.  I’m certain.

Note: This first appeared on Grist.

Broadly speaking, there are only three things we can do to lower CO2 emissions:  switch fuels, use energy more efficiently, or use less energy (conserve).

Our CO2 conversations too often focus on one of those three in isolation: Coal bad. Recycled waste heat good. Conservation isn’t an energy policy. Each assertion is both narrowly true and broadly incorrect, to the extent that each simplifies three prongs into one.

To understand why, try to answer a simple question: if we shifted our power generation fleet to preferentially dispatch natural gas plants instead of coal plants, how much would CO2 emissions fall?

That would seem to be an easy bit of math: just measure the CO2/MWh of each plant, multiply the difference by the MWh switch, and we have our answer, right? Turns out it’s a tad complicated, for the simple reason that the fuel switching strategy is also an efficiency strategy. Does a newly dispatched gas plant look like one of the old, 30% efficient, natural gas-fired Rankine “steamers,” or does a newly dispatched gas plant look like one of the new, 50% efficient combined cycle gas turbines? What about the coal plant that gets turned off?

Better still, let’s ask an easy question: how has the CO2 signature of our nation’s coal and gas-fired power fleet changed with time?

DOE/EIA keeps voluminous records of fossil fuel consumption and power generation by fuel type. On the following charts, I’ve divided total fleet fuel use by total fleet MWh and then multiplied by a consistent 0.06 tons of CO2/mcf of natural gas / 2.7 tons CO2/ton of coal to yield the following:

Interesting. From 1960-1990, there was no statistically significant change in gas fleet CO2 emissions, which held steady at 0.63-0.65 tons/MWh. Then all of a sudden in the 1990s, the fleet transformed itself, reducing its CO2-intensivity by 25% in just 10 years. What happened?

In a word: competition. The introduction of competitive access in the 1992 Energy Policy Act (and subsequent FERC rulings) brought forth a flood of natural gas plants, many of which were nearly twice as fuel efficient as the old junk that the grid had previously relied on. Prior to that point, costs were simply something that you passed along to customers. After that point—for much of the grid—cost control was a route to greater profits. Not surprisingly, generator owners suddenly got religion on cost-control. And when your number one cost is fuel, that means they got religion on fuel control. That’s good.

Now let’s look at what happened to the coal fleet during the same period:

From 1960-1970, the coal fleet holds steady at 1.17 tons/MWh, but then starts an inexorable upward trend. While the gas fleet became more efficient with time (after 1990, at least), the coal fleet is steadily less efficient. Way less in fact—to the point that the CO2 emissions associated with a MWh of coal-derived electricity are 18% higher today than they were in 1960.

What happened here? Two things:

1. Unintended consequences. 1970 saw the passage of the Clean Air Act,  a deeply flawed bill. It was good in terms of what it did for regulated pollutants, but lousy in terms of what it did for unregulated ones (e.g., CO2). By effectively mandating pollution control approaches that impose parasitic loads on coal plants, the CAA is directly responsible for lowering coal plant energy efficiency, so that we now burn way more coal per MWh than we did before passage. We therefore emit way more CO2 per unit of useful electricity. That’s not to ignore the beneficial elements of the CAA, from sulfur to particulate control, but simply to point out that an environmental regulation that encourages energy inefficiency leaves much to be desired.

2. Dispatch considerations. As noted here, the last 30 years have seen virtually no construction of new baseload power plants in the US, but have seen a steady increase in the annual load factor of currently existing baseload plants. In other words, plants that used to spend most of their life turned off now spend most of their life turned on. In the coal fleet, that means that the least efficient stuff runs more now than it used to. So in addition to the unintended consequences of the Clean Air Act, we also have the simple fact of steadily growing electricity demand that causes us to pull our power from ever-more-undesireable sources.

Why didn’t the competitive forces unleashed by the 1992 EPACT also drive up the efficiency of our coal fleet, like they did for gas? Again, it’s an easy answer: competition. Coal plants are lousy investments. No one builds them who has to put their own money at risk.

One last thing

Here’s the tragedy: If we had run the gas fleet at a constant fuel efficiency from 1960-present, we would have emitted an additional 1.3 billion tons of CO2 into the atmosphere. That’s 1.3 billion tons not in the atmosphere today thanks to energy efficiency.

On the other hand … if we had run the coal fleet at a constant fuel efficiency from 1960-today, we would have emitted nearly 9 billion fewer tons of CO2 into the atmosphere over the last fifty years.

1,300,000,000 steps forward, 9,000,000,000 steps back.

Note: This first appeared on Grist.