Last night I went to a presentation by the Renewables Yes technical and financial modeling team. They’ve put up a bunch of information about their modeling efforts on the web site. I’ve organized nine short videos of a previous iteration of the presentation into a single 90 minute playlist here if you want to see it yourself. It’s definitely worth watching if you use electricity in Boulder! This post is my attempt to digest and rephrase their conclusions.
A tiny bit of background first. Last year our 20 year contract with Xcel energy (an investor owned utility… or IOU) ran out, and we voted not to renew their franchise for another 20 years. We now have 5 years to figure out what we’d like to do instead. There were a lot of good reasons to take a look at alternatives to another 20 year contract. Xcel’s rates have increased rapidly in the past few years. They are heavily committed to coal generation, which precludes integrating renewables into the grid and emits the most CO2 of any other power source. Many people also feel that 20 years is just too long of a commitment given how rapidly the energy landscape is evolving. We would lose a great deal of future flexibility if we signed another 20 year franchise similar to the previous one.
Renewables Yes is an explicitly political non-profit citizens organization, which worked on the ballot measure last fall. After 2B passed, an all volunteer team started modeling the technical feasibility of building a coal-free grid for Boulder, that would allow the integration of large amounts of renewable power over time. They’ve also been looking at the finances. Fellow Caltech alum Sam Weaver from Cool Energy (which works on high efficiency, low cost, low temperature Stirling engines) headed the technical team, with most of the modeling being done by Tom Asprey. Frank Selto, a professor at CU’s Leeds School of Business, did the financial modeling.
The Problems With Coal
There are a lot of problems with coal. Both mining and burning coal are polluting in the traditional sense. The majority of the mercury found at dangerous levels in big, tasty fish comes from coal fired power plants. Mountaintop removal mines in Appalachia are watershed disasters. It’s also the most carbon intensive of fuels, putting out about 1000 kg of CO2 per megawatt-hour (MWh) of power produced. In the longer term, it’s also a limited resource. It will some day run out. However, there are also a couple of more subtle problems with coal fired power plants, which interact in such a way as to make a low-carbon grid an unthinkable business proposition for Xcel.
The first is that it takes a long time — a day or two — to turn a coal plant on or off. You can’t just flip a switch. You also can’t throttle them back without seriously impacting their fuel efficiency. Both wind and solar power — today’s most cost effective renewable options — are intrinsically variable on the timescale of minutes to hours. This means it’s very difficult to build a stable power supply that uses both coal and renewables effectively. You make a commitment to having a certain amount of coal power running in the background at all times (baseload) and if your renewable sources generate more power than is needed beyond that baseload, you have to “curtail” them — you have to throw away that free, clean energy. Wind and solar power are also forecastable, which mitigates their variability somewhat. The Research Applications Lab at UCAR works with utilities to improve their wind power generation forecasts, so that they can better predict how much fossil fueled power to idle. Better forecasting can only get you so far though, because you still can’t respond to rapid changes in available renewable power fluxes with coal, even if you know they’re coming.
The other subtle problem with coal has to do with how the utility’s acceptable profits are calculated. Xcel has a state-sanctioned monopoly, and so their profits are regulated. The formula that’s used to determine how much they can charge the rate payers includes a certain profit margin beyond whatever their capital expenses are. Coal plants are very capital intensive — they cost a lot to build up front. Because Xcel is entitled to a set percentage of profit beyond that expense, they have an incentive to spend as much as possible building plants, as it maximizes the dollar value of the profits they’re allowed to collect. For example, if they’re entitled to a 7% profit, and they spend $100M on a new coal plant, they get to recover $107M from the ratepayers ($7M in profit). If the same amount of natural gas fired generation only costs $50M to build, then they only get to collect $53.5M from ratepayers (for a $3.5M profit). It’s obvious which of the two is the better business decision. They’re also entitled to recover operational costs, such as the cost of the fuels that they burn, but they don’t earn a profit on them, so they’re in a sense fuel agnostic.
This arrangement creates a large financial incentive for Xcel to systematically underestimate the future cost of coal in its financial planning, because it allows them to justify building expensive coal plants. If they turn out to be wrong about the future price of coal, they simply pass that cost on to us, shrug their shoulders and say “Whodathunkit?”. And they’ll probably be in good company — historically nobody (including the US Energy Information Administration) has done a good job of predicting future fuel prices. Given how abysmally everyone fails at this game, it’s amazing to me that anyone keeps playing. In the meantime, they get to charge us “cost-plus” on the expensive power plants.
Taken together, these realities make it very unlikely that Xcel will get behind a transition to clean energy in a significant way unless we can price CO2 emissions or re-write the public utilities regulations, de-coupling the utility’s profits from the amount of power they sell, and creating the right incentives for efficiency and low emissions.
A Baseline Scenario
The initial citizen modeling effort isn’t meant to be a detailed grid design. It’s not meant to be highly optimized picture of just how good things could get. It’s a very conservative sketch of the kinds of systems we might consider if we did want to create a municipal utility. How much of what resources would we have to use to meet our demand? How much would it cost? How would it affect our emissions profile?
The very basest of the baseline scenarios is to simply use the renewable power we’ve already got installed — about 10 MW of rooftop solar and 12 MW of hydro — and get the rest of our power (~180MW) from natural gas. This would immediately reduce our CO2 emissions by 50%, and our power would cost the same or less than it does now. Switching to an all gas fired grid would be very reliable, and have low up front costs. Efficient gas turbines are cheap; the fuel makes up the overwhelming majority of the cost of ownership. With local small and medium scale gas generation, we’d also have the option of doing a lot of co-generation of useful heat along with the electricity. This can be used for district heating, cooling and industrial processes, and can push the overall thermal efficiency of the plant as high as 90%, vs. 55-60% for electricity-only generation, in which all that potentially useful heat is just thrown away.
More importantly, natural gas plays nice with renewables. It’s dispatchable, unlike coal. Which means you can essentially flip a switch to turn gas turbines on and off in a matter of minutes, and so use them as very effective backup power to firm up an intermittent renewable power supply. Natural gas certainly isn’t a panacea. It’s still a limited resource; it has serious pollution issues of its own (if you haven’t already, definitely check out Joshua Fox’s film Gasland, about hydraulic fracking); it’s too carbon intensive to avoid climate change; it will certainly have some price volatility. Nevertheless, it’s a big improvement over coal because it’s a much smaller up front commitment, and it will allow us to integrate a much higher proportion of renewable energy into our grid, until we can get scalable, low cost energy storage figured out.
We can get at least a 50% CO2 reduction for free by switching to gas, if we assume that natural gas prices stay where they are today, around $5/MMBTU. This is roughly the average price of gas over the last decade, but the price has spiked occasionally to around $10. What if that became the new normal? In that case, it becomes cheaper to integrate some renewable power into our grid. And unlike on the current coal fired grid, this would actually be a practical option. We can hedge against high gas prices with renewables, but we can’t use them to hedge against the possibility of high coal prices, because coal and renewables aren’t really grid compatible. In the model runs that the tech team did, the most cost effective non-coal option in case of a doubling of natural gas prices is 70% gas, and 30% renewables, and it led to about a 30% increase in the cost of power production ($0.114/kWh vs. $0.084/kWh). It also resulted in a 64% reduction in our CO2 emissions versus today. For a very modest additional increase (up to $0.117/kWh), renewables can come up to 39% of the mix, and our overall CO2 reduction is 68%. Because we can effectively integrate gas and renewables, there’s an upper bound to how much high gas prices can affect us. Should gas prices become very high, we would want to integrate more renewables. With doubled gas prices, at 61% renewables (and an 80% reduction in CO2 emissions) we’d be at around $0.142/kWh.
We’re going to be exposed to future fuel price volatility one way or another. If we stick with Xcel, it’ll be coal prices which seem low today, but are difficult to hedge against. If we strike out on our own, it’ll be gas prices, which are about the same cost as coal today, and are easier to hedge with renewables. Obviously $0.142/kWh is a lot more than our power costs today, but one has to imagine that if gas prices have doubled, coal prices are also likely to be changing. In that kind of uncertain energy future, having a grid with a very predictable long term cost of energy would become attractive to businesses, which will often pay a hedging premium to lock in a predictable price they can plan their operations around.
If gas prices don’t rise and stay high, then we’d be getting that 68% reduction in CO2 and a lot of future flexibility for just an extra penny per kWh. The average household in Boulder uses about 650 kWh/month, which would mean the cost of a serious start on mitigating climate change would be all of $78/household each year. This is the equivalent to a CO2 price of less than $20/ton of emissions avoided, which seems like an awfully good deal to me. And again, this is all assuming that wind and solar continue to cost as much as they do now, and does not take into account any energy efficiency measures, or economies of scale in renewable generation. There’s every reason to believe that the actual costs will be lower.
An Island Model of Conservatism
By design the model scenarios considered were very conservative. One conservatism that was integrated was the assumption that the Boulder grid would be an island — that we would not be connected to the rest of the regional electrical system. This isn’t a design decision; nobody is suggesting that we actually disconnect. Keeping your power reliable in a disconnected system means having relatively large reserves you can lean on in case something goes wrong. It also means that whatever solution you come up with is scalable and can stand on its own. You could build a very robust interconnected grid out of a system of many potentially self-sufficient islands, or you can reduce the required reserves (and the system costs) by sharing resources across different jurisdictions.
Renewable power which is only functional if it’s connected to coal plants isn’t renewable at all, as a system. This is in essence what Xcel was suggesting with its wind farm proposal (which has just been nixed by the city). They were offering to hook up 200MW of new wind generation, and to assign the renewable energy certificates from it to Boulder. This would mean that renewable energy would ultimately be above and beyond the statutorily required renewable energy portfolio standard. The problem with this arrangement is, it can’t scale up. You can’t just keep adding (and utilizing) intermittent power on a coal baseload grid. So we might get clean energy, but if others want to do the same, eventually curtailment costs become large. Within the proposal Xcel made, it was clear that we would be the ones on the hook for those costs, as well as the cost of constructing the wind farm in the first place.
The inability of coal plants to decrease power production dynamically means that on a coal grid as you add more and more renewable generation, the marginal reduction in emissions becomes smaller and smaller, and the cost of those additional clean megawatts becomes more and more expensive. Coal simply can’t play much of a role in any system that is trying to do what the atmosphere requires, which is achieve at least an 80% reduction in our greenhouse gas emissions.
The Known Unknowns
As far as I can tell, most of the risks involved are legal and political. Fuel costs are uncertain, but that’s true regardless of whether we stay with Xcel or not. It’s just a choice between whether we want to be exposed to the price of coal and stuck with it, or the price of gas and able to replace a lot of it with renewables, placing a firm upper bound on our potential energy costs.
Politically, on the city’s side it would be tempting to use the municipal utility as a cash cow to pad the general fund. This is a bad idea, especially this early in the process. If people want city services (and I think we do) then we have to pay for them, and that means taxes of one kind or another. We just have to get comfortable with that. It’s okay to tax, especially if the tax is transparent and localized. Wrapping taxes for unrelated things up inside your utility bill obscures both the cost of energy and the costs of whatever it is those taxes are paying for. It also makes it more difficult to do straightforward comparisons — to show that you are in fact succeeding (or failing) in a cost context.
We don’t know how much it’s worth to Xcel to prevent us from municipalizing. It might be worth more to them to stop us, than it is for us to go through with it in a purely financial sense. They may want to prevent us from setting a good example. They may want to try and make a bad example of us to scare others off. It would be nice to get some kind of upper bound on how long they can make it take in court, and how much they can squeeze out of us for their potentially stranded assets. The city says there will be no stranded assets. Xcel says it could be as much as $600M. Those startup costs will have significant ongoing consequences (but keep in mind that the annual revenues for the city utility would be about $100M… which makes a startup cost of $300M seem less insurmountably huge.)
This risk — that Xcel is willing to spend more than we are — will first appear in the form of ads. Propaganda. Campaign messaging. We’ve already had some push-polling from Xcel (which they deny of course), suggesting that the poor and the elderly might get their power cut off, or that businesses might flee Boulder if we raise the cost of power (commercial users do consume 80% of our power). We have to get our messaging down. The vast majority of voters aren’t going to watch a 90 minute presentation or flip through 100 slides. This has to be sound-bite and poster compatible. Much though we might like folks to make their decision on the basis of a deep and subtle understanding of the issues, in the end we are mostly emotional beings, and we have to frame the truth in a way that makes it accessible to people who aren’t going to go on digging. Xcel will not play nice, or clean. We shouldn’t expect to have that luxury either.
There’s a line near the end of the excellent animated short Wake Up, Freak Out! that is particularly apropos here:
It is now very clear that in order to actually win the fight against climate change, making big changes to the way we each live our own lives is not going to be enough; we’re also going to have to actively confront powerful vested interests who will stop at nothing to prevent the changes we need from taking place. We have to be more than just consumers.
Other cities will be watching us, especially early on in this process. Let’s make it easy to see what’s happening, and set a good example that others can follow. Ultimately this isn’t just about us having our own source of ethical, reliable, cost-effective power. It’s also about showing others how they can get there too.
20 thoughts on “Boulder’s Energy Future Is Bright”
If you want to get involved in this directly, come to the community bike ride and sunflower planting at the Valmont coal plant this Saturday, July 16th. We’ll be meeting at the band shell near the farmer’s market downtown at 10am.
City council will be considering what to put on the ballot Tuesday, July 19th. The meeting will start at 5pm (earlier than normal… this is going to be a marathon) in the municipal building at 1777 Broadway. You can grab a PDF of the agenda on this page.
Great article. I have linked to it in ReYES. I do want to re-emphasize that the numbers you cite and the model are meant to be a realistic worst case. Not only were the assumptions mostly Xcel numbers with some margin added, we also did not include some of the best ways to reduce the costs such as efficiency because they were not easy to quatify reliably without being debatable. Now that we have a worst case, we will be modeling a more realistic expected case that attempts to predict where things are headed and technologies we might use to bring costs down and renewables higher.
Thanks for clarifying so much that wasn’t quite clear to me before!
Question: Do I understand correctly that when Xcel builds a new coal plant, it can charge 7% profit on that expense? If so, for every year that it’s in operation?
What have I missed?
Thanks! And thanks for working with Renewables Yes!
I don’t know what the actual state mandated profit is (7% was just an example). It’s not every year. It’s just once, over the course of the plant’s lifetime. This is true for any capital construction they do.
Does that make more sense?
Privates, like Xcel, traditionally operate at what is called an ROE, or return on equity, of 10-12% depending upon the public commissions they serve throughout the country. “7%” may be a layer which works up to this amount, after fuel and administrative stuff is included.
Yes, they get a rate of return based on their cost of capital, which is the weighted average of the interest rate they pay on the debt they issue, and the ROE for the portion of their funding which comes from the shareholders. I can’t remember the exact number, but overall it’s between 8% and 10% I think. 7% was just a number I pulled out of thin air for illustration.
Well, it looks like city staff is recommending to Council that we start moving toward municipalization… This is going to be interesting! Here’s the memo, as a PDF.
very good in depth article. my sense, and i could be wrong, is that there are some who are ostensibly opposed to municipalization who would be open to facts and data such as that in the article, and who would not automatically thin or act from an ideological bent.
it might be important to reach out to those people.
also, these same people need to be more exposed to some of the meetings that have been called, in which political and business leaders from other locales have pointed out the successes of other such efforts and the way in which to go about it.
Well by all means feel free to forward the link around! 🙂
This is a really encouraging summary. It’s wonderful that there are communities that are trying to move into the future responsibly. However, it’s depressing that there isn’t a source for baseload power that doesn’t involve brutalizing West Virginia. 🙁
I agree that it’s an encouraging situation — the barriers to responsible power aren’t intrinsic to the electrical system… however, the “power” system is different. It remains to be seen whether we can win against Xcel in the propaganda department. It’ll be sad if we fail. I think the real take away is that we don’t need “baseload” power at all. In fact, the traditional always-on baseload providers (coal, nukes) are an impediment to using renewable (variable) power sources.
Another weird thing here… I don’t understand why we would ever agree to another franchise agreement. Xcel is legally required to supply us with electricity regardless of whether we’ve got a franchise agreement. In theory, we get a revenue stream from Xcel in exchange for a long term purchasing commitment… but in actuality, they just pass that fee right through to the ratepayers. So in exchange for our 20 year contract, all we really get is taxation via corporation instead of taxation via our local government, which seems like a lousy deal to me!
Very lucid explanation, Zane, of the variables in the vision and plan for a muni in Boulder – Thank you! One additional step: talk with Leslie about constraints on coal supplies. Cost projections will be nothing like the computer model Xcel took to the PUC.
Great, great summary Zane! It was good working on this with such a great team!