Shortly after finishing my Smart Power manuscript I sent a copy to Amory Lovins, Chairman and Chief Scientist at the Rocky Mountain Institute. Amory was extraordinarily kind to send me four pages of dense corrections and comments, some of which (unfortunately) could not make it into the final manuscript. For typos and numerical errors, see the Errata section of this website. Amory also made a number of interesting observations, most of which are reproduced here, some with further comment by yours truly.
PFP: In my capacity as a principal of The Brattle Group, I’ve studied the Western Power Crisis since 2003 and continue to serve as an expert witness on behalf of the California IOUs, Attorney General, and Public Utilities Commission in federal cases attempting to get refunds from sellers for sales made during the crisis. I’ve submitted too much testimony to summarize it very well, but in short, I have concluded that a bad market design, adverse conditions (sometimes called scarcity), poor reactions by regulators, market manipulation, and market power exercise all occurred during the crisis. Many other distinguished experts have also studied the crisis and/or offered testimony. Here is a partial bibliography of my testimony and other reports on the Western Crisis.
U.S. Federal Regulatory Commission, Docket No. EL09-56-000, May 22, 2009: Involves emergency purchases the State authorized the California Energy Resources Scheduling division of the California Department of Water Resources (CERS) to make when the California investor-owned utilities (IOUs) could not purchase the power needed to serve their customers.
U.S. Federal Regulatory Commission, Docket No. EL00-95 et al, May 22, 2009 Analyzes seller behavior in markets operated by the California Independent System Operator, the California Power Exchange and in bilateral markets in which CERS procured power on behalf of the IOUs, in order to draw conclusions concerning gaming and market manipulation and the existence of undue market power between May 1, 2000 to June 20, 2001.
U.S. Federal Energy Regulatory Commission, Docket No: EL02-71-004, April 21, 2008: Declaration before the FERC; demonstrating evidence of sellers’ exercise of market power, including evidence of gaming, withholding, and anomalous bidding, and explaining why “market-wide relief” is an economically appropriate way to compensate buyers who purchase power in markets where prices are profoundly affected by the exercise and abuse of market power.
“Clean Growth- A Balanced Energy Policy for the 21 st Century,” Progressive Policy Institute, October 2001.
“What Does the California Experience Tell Us About Fixing the Rest of America’s Power Markets?” with Joseph B. Wharton. Presented at the National Association Business Economics Regional/Utility Roundtable, April 24, 2001.
“What not to learn from the California Crisis,” (Op-ed) The Providence Journal, March 3, 2001.
“California’s Electricity Crisis and Implications for the West,” for the U.S. Senate Committee on Energy and Natural Resources, January 31, 2001.
“The California Crisis and its Lessons for the EU,” for Energy, special edition, 2001.
Testimony before the Senate Committee on Energy and Natural Resources, January 31, 2001.
Comment 2: Chapter 5: The Regulatory Mountain, Page 53
AL: … clarify … whether states still do calculate avoided cost to pay QFs (I thought that part of PURPA was dead — certainly many IOUs act as if it were);
PFP: Amory’s uncertainty regarding PURPA is understandable because the mandatory purchase provisions of the law were only conditionally repealed. Under the Energy Policy Act of 2005, a utility can petition the FERC to repeal PURPA’s requirement that the utility purchase all cogenerated or renewable power from facilities that meet certain standards (“qualifying facilities” or QFs). FERC can release the utility from its purchase obligation only if it concludes that there is a sufficiently well-functioning capacity marketplace into which QFs can sell their power capacity. This includes PJM and New England, but not many other regions as yet.
Comments 3 and 4: Chapter 5: The Regulatory Mountain, Page 60
AL: … somewhere in that Chapter 5: The Regulatory Mountain, perhaps pick up again the dropped point from the Sequim story that efficiency and DR providers should be paid a "Dristan benefit" (decongestant rent) for freeing up grid capacity;
and
… mid-page 60, note that the DG owner will still be entitled in principle to capture all the other distributed benefits she brought to the system, even if not high locational prices, so could still make ample extra money if properly compensated for those benefits;
PFP: Amory is right that DR can "decongest" parts of the distribution system. This is presently reflected in wholesale markets in the form of the locational marginal price (LMP). There is a major FERC rulemaking on this topic ongoing at this moment, Docket RM10-17-000. This docket contains volumes of discussion on the benefits of DR, including comments by the legendary Alfred Kahn and comments somewhat in opposition from my colleagues at The Brattle Group. The primary disagreement referred to here is whether this "decongestant benefit" equals the entire LMP or the LMP less the avoided retail rate. Kahn supports the former; we the latter, but in both cases there is no disagreement with Amory's fundamental premise, which is that the customer who drops load is entitled to (at a minimum) the value of power at the place it is reduced, that value reflecting congestion at that point in the system.
There is also congestion on the local system, between the point at which wholesale prices are set and the point where you take power off your local wires. This is the area where the Smart Grid will introduce even more ability to estimate "decongestant value," but it will also be a big regulatory challenge for state regulators, as these local congestion prices will be state jurisdictional. Because local congestion value can vary much more than wholesale congestion price differences, regulators will want to consider carefully how much variability is best and how to control the variability without causing other unintended side effects.
Comment 5: Chapter 5: The Regulatory Mountain, Page 65
AL: … cite the Defense Science Board report: More Fight — Less Fuel, which recommended that the Department of Defense, now 99+% dependent on the vulnerable commercial grid, get its facilities off the grid and make its own power onsite or nearby in netted islandable microgrids to ensure mission continuity; note that these vulnerabilities, such as the classified Aurora phenomenon, exist in today's dumb grid and would only be augmented, not created, by the smart grid as implied by page 64.
PFP: I regret not discussing this subject a bit more in my transmission chapter. The U.S. military has indeed recognized that the military capability of the United States is heavily dependent on conventional energy supplies, including the power grid. Without sufficient backup or self-generating power capabilities, military installations are vulnerable to grid failures and sabotage. This, in turn, has encouraged the military to recognize the strategic value of energy efficiency and onsite power generation, primarily from renewable sources.
While I don’t think this will be a large enough force to change the industry’s structure or business model by itself, I think it will be another factor discouraging expansive high-voltage transmission projects that might otherwise claim to provide national security benefits. Mid-scale renewable projects are ideally suited for military installations, and it only makes sense for the military to rely on a balance of grid-independent onsite sources and some grid power.
Ceteris paribus, this will further reduce traditional utility sales to the military and substitute more military self-production. However, it is possible that local utilities or other renewable companies will own and operate these local generators. Military capital budgets are always constrained, and the military need not necessarily have renewable energy production core competencies “in house.” In short, the same vertical structure questions apply to the military as a (very) large commercial power customer with the potential for onsite distributed generation (DG): Who is best to own it? Operate it? How do we price backup power? And so on.
In addition to RMI’s paper, here are several more papers on the growing relationship between U.S. military security and domestic power supply options:
Comment 6: Chapter 6: The [Highly] Unlikely Future of Sales, Page 71
AL: … insert the January 2009 McKinsey study of efficiency potential and optionally mention RMI's calculations (still the most empirically based and detailed), which in the late 1980s found (see my joint RMI/EPRI Scientific American article 9/1990) a practical retrofit potential to save ~75% of U.S. frozen-1986 electricity consumption at an average cost of 0.6¢/kWh (1986 $) — the difference between that assessment, based on empirical cost and performance data for >1k end-use efficiency technologies, and EPRI's was almost all methodological, not substantive;
- importantly, try to insert a sentence or note here stating that end-use efficiency technologies are continuing several decades of rapidly improving performance and declining cost, they are being more widely and maturely marketed and delivered, and Lovins asserts that new "integrative design" techniques — a new layer of innovation beyond the technologies themselves — can often make very large energy savings in buildings and factories cost less than small or no savings (you could cite my 2005 white paper for Steve Chu at www.rmi.org/rmi/Library/E05-16_EnergyEndUseEfficiency, my 2007 Stanford Engineering School lectures at www.rmi.org/stanford, and our project to spread this sort of design, www.10xE.org) (in my opinion, efficiency policies will be left in the dust by these complementary improvements in technology and design! ... but Chapter 6 is really weak on efficiency, scarcely mentioning anything but policy and not mentioning the extraordinary market trends observable in real estate — important because buildings use ~70% of U.S. electricity; you mention the EPRI and ACEEE estimates on p 97, but I think they're both very conservative);
Comments 7 and 8: Chapter 7: The Aluminum Sky, Pages 88 and 89
AL: … you repeat a common fallacy that comes from drawing the map to stop at U.S. shores. NREL also does other maps (and you actually show and correctly describe one on page 90 without reconciling it with your text on page 88) showing that the wind resource in the Great Lakes (typically class 6+) and offshore roughly the northern halves of the lower 48 states' east and west coasts (often 6-7) are comparable to or stronger than those in the High Plains (page 260 note 40). One must pay for marine engineering and submarine cables, but the land is free, the wind is stronger and steadier (and less gusty due to less surface roughness), and the site is much nearer load centers, so it's not at all obvious that we need big transmission to remote wind. (You imply this for offshore in the middle of page 88, but miss the Great Lakes.) You get this point right later but not on this page. Similarly, annual insolation differs by only 2 times between Mojave and, say, Buffalo. (AZ = 1/4 more than KC = 1/4 more than Buffalo: see http://rredc.nrel.gov/solar/pubs/redbook/; in general, it's not worth shipping SW U.S. solar power to other regions.
… on page 88 … I think your statement that we need new balancing capacity for "balancing a grid with many large variable renewable sources" is unsound and undemonstrated. NREL's study: Eastern Wind Integration and Transmission Study, (NREL January 2010), shows you can put >30% wind into the Eastern Interconnect without new balancing capacity, and it's a worst case because they assume little efficiency, no distributed generation, and almost no demand response. Our own hourly simulations in ERCOT are well over 80% renewable without adding balancing capacity or worsening reliability. The balancing capacity already exists; indeed, it's probably less than is now required to cope with the intermittence of big thermal plants that the renewables gradually displace (assuming a reasonable siting and technological diversification of renewables, with forecasting). NERC's quote about "greater access to ramping and ancillary services" is not equivalent to needing "additional balancing plants."
… on page 89, similarly, you have not shown, nor has anyone else, that a low-carbon future will "require much more transmission grid per kWh" than now; the existing studies (including the Eastern Interconnect one) show this only with little or no efficiency, demand response, and distributed generation. Nobody has yet done the full integration of all these resources with renewables. That's part of RMI's task in Reinventing Fire. I think you'd be ill-advised to anticipate the conclusions in such firm language. It may turn out, as you state, that 30-40kmi of new transmission will be needed, but so far there is no evidence that much or any will be needed, at least in a big-grid sense (there may be local exceptions, of which West Texas wind is the most obvious). Interestingly, the Eastern Interconnect study seems to have assumed very little offshore or Great Lakes wind, though it's impossible to tell; they were interested rather in new overland transmission.
PFP: Here I respectfully part company with Amory. Although I agree that increased energy efficiency and distributed generation could supplant the need for all new transmission, I think this is very unlikely to occur. I would love to see the United States embark on a crash program that vastly increases our energy efficiency, but I don’t see much more happening in the present U.S. political climate beyond that which I forecast in Chapter 4: Smart Electric Pricing. Meanwhile, nearly 30,000 miles of transmission is already on NERC’s drawing boards, not counting many ambitious projects not yet in the numbers. Accordingly, I do predict transmission additions – not because I have proven them to be the absolute lowest-cost alternative, but rather because I think this is the direction the industry will take, despite the best efforts of energy efficiency and distributed generation advocates.
Comment 9: Chapter 7: The Aluminum Sky, Pages 92 and 254
AL: … you had made explicit the need to both integrate and compete new transmission with efficiency, distributed resources, and distributed generation. That's a big omission, and easy to fix.
… on page 254: I'm glad you added endnote 13 (a link to NREL’s resource maps). However, you have not made, and I doubt you can make, a sound economic or engineering case that "the highly desirable utility-scale sites" are a better or cheaper resource than "less concentrated resources". For example, Steven J. Strong, the best distributed photovolatics (PV) practitioner, makes a strong empirical case that rooftop PVs delivering low voltage to the retail customer to compete with retail rates will almost always beat utility-scale PVs delivering high voltage to transmission to compete with bulk wholesale rates.
PFP: I hope that it is clear that the most desirable size for renewable installations seemed to me to be what I call mid- or community-scale precisely because they do not tend to trigger significant transmission upgrades. They may require small reinforcements, but fewer long haul lines. That said, my information differs from Amory’s. Many solar developers have told me that their delivered costs of power are much cheaper for five to 200 MW facilities than for individual rooftops. These developers don’t count the incremental distribution system costs to deliver mid-scale resources to the customer, but as I show in Chapter 5: The Regulatory Mountain it is exceedingly difficult for regulators to credit these to any one customer who installs rooftop solar. Indeed, one of the main points of my book is that regulatory difficulties at the distribution system level are the single most important pacing factor for the industry’s structural evolution.
AL: FYI, a bench-scale Shell technology — whose scalability to GW size we should know within a few more months — was expected a year ago to cut carbon capture cost from flue gas by approximately ten times and to be extremely compact and retrofittable. This would cut nominal carbon capture and sequestration cost by approximately three times … I'll shortly catch up on its status.
PFP: This is fascinating. I do not think that anyone can disprove the possibility of major cost breakthroughs for carbon capture and sequestration (CCS) or even nuclear — another reason why I think it is foolish to count these energy sources out as of now. When I find the time, I’ll update the website on nuclear and CCS technology cost trends.
Comment 11: Chapter 8: The Great Power Shift, Page 104
AL: … as earlier, you don't seem to be linking the time required to install lots of demand response with the time required to install lots of variable renewable capacity that needs it. I'm not clear why you imply that the renewables will way outpace the demand response.
PFP: Amory asks why I think controllable renewable resources will enter the market more quickly than utility-controllable demand response. I think there are several reasons. First, the renewable industry is better financed and better organized (in the political economy sense) than the demand response industry. It maps more easily into the existing utility control paradigm and business model — a topic that occupies much of Smart Power. It requires fewer and less complex approvals from state and federal regulators, a complex and significant factor even before one adds the further complication that the question of how much consumers will allow utilities to control their power use, and under what terms, is not yet clear. To be clear, these are what we economists call institutional impediments or (at times) transaction costs.
Comment 12: Chapter 8: The Great Power Shift, Page 106
AL: I also strongly disagree with your implication that grid integration adds lots of cost to wind power (which you unfairly represent as an average or blending of wind with complementary-resource costs: the complementary resources were what ran before the wind came along, and now that it's there, you run the complementaries less and pocket their saved opex!). Even with >30% wind, little efficiency, and virtually no demand response or distributed generation, the Eastern Interconnect found wind integration costs <$5/MWh (2008 $), and that's gross — not net of the integration costs of central thermal plants they displace on the margin. On page 260 note 39, you even quote short-run wind integration costs of just 0.1¢/kWh when I'd have said ~0.3! You will get a lot of flak for this passage. Likewise for suggesting that "it will become prohibitively expensive and-or unreliable to rely on wind for much more than 30 percent of the system's total power needs, at least for now." I know the literature pretty well and am aware of no basis for that statement, which is flatly contradicted by European experience.
… same fallacy about blending costs. You should only count net marginal integration cost, not claim that some mythical average of PV with fossil output constitutes a "true average cost of PV power." Also, it's worth noting before the crosshead that world photovoltaic (PV) installations are around 7–8 GW/y; the U.S. is a relatively small player.
… you say "The commodity energy from small-scale options is still more expensive than large scale sources", but I doubt this is generally true — see comment on 10 above, for example. And as noted earlier, I don't think you can justify saying that onshore wind "requires substantial...backup power and is presently limited due to integration issues." Four German states run unproblematically on 38-47% annual and sometimes >100% wind power.
PFP: This is a critical and complex area, so I’ll devote some discussion to it. There are two cases to discuss. The first is one in which demand is constant and wind displaces existing conventional resources; the second is one in which wind is added along with integration resources, such as gas turbines or large (as yet, uneconomical) batteries.
In both cases, I first note that economics are based on what the brilliant but little-known Bell Labs economist Edward Zajac called the property rights of the status quo. Once a system of payment and property rights is set up, “costs” and “benefits” must be understood to be differences between the current system of equilibrium payments to the factors of production and payments under some alternative system of rights or policies.
A second point to make before starting is that the integration needs we are discussing are a function of just how intermittent the wind (or any other intermittent) renewable resource is. If winds are relatively steady, the variability is lower; similarly, if wind turbines are spread over a very wide area, such that winds tend to blow in one part of the area when they are still in the other, aggregate variability is lower. While the cost of integrating variable renewables is significant, it is situation-specific and can vary by quite a bit.
With this in mind, consider Case 1, which I think is what Amory refers to in his comment. In the status quo, the utility has a collection of baseload power plants that run constantly, cycling plants that run occasionally, and peaking plants that run very rarely (see Chapter 4: Smart Electric Pricing). The utility is paying the annual capital costs for all of these plants plus their operating costs, which vary with the level of output from each particular plant.
If a utility adds wind turbines to its system but experiences no greater demands, it must still maintain energy balance at every moment. System operators can only control the wind turbines by turning them essentially fully on or fully off. If they’re on, whatever energy they generate goes into the grid. Compared to how the utility ran the system prior to the wind turbines, the system operators have to turn off just enough of the cycling and peaking power plants to exactly match the energy coming from the wind generators.
The utility’s total costs change as follows. First, its capital costs are the same, since it still owns and must pay for the same group of plants it still owns (it may retire some of them, but let’s assume it doesn’t). Whatever energy is produced by the wind generators reduces the utility’s need to generate power from an existing plant; the utility pays the wind generator for the energy (say five cents) and saves the cost of producing power from its generator (three to ten cents a kWh).
This is why Amory says adding wind should lower utility costs. The utility has to pay for every unit of wind energy, but it saves the fuel it would have burned to create that same kWh. True, but the situation is often more complex than this under the current utility control paradigm. Right now, reliability rules require that utilities keep a certain amount of generating capacity idling to prevent sudden outages, which in this context includes the winds dying down suddenly. The costs of keeping these plants idling (spinning reserve) is significant, and it is a positive function of the amount of wind on the system. Since many power plants can’t just start up suddenly without warm-up periods, and have required periods of cooling off before they can start again, turning plants on and off to exactly match the energy generated by wind producers can add still more costs.
Also, wind resources require transmission (see Chapter 10: Energy Efficiency: The Buck Stops Where?), which adds to the cost of purchased wind. The conventional power plants owned by the utility need transmission too, but this transmission was added to utility rates long ago and is used today to run the system. Remember the idea of property rights in the status quo? The transmission embedded in current utility rates, much of which was built when it was cheaper and has already depreciated and thus is even cheaper, is part of the status quo costs. The costs of building new transmission to accommodate wind, at today’s transmission prices, are counted as part of the cost of change.
When all these costs are added up, the utility’s total costs of meeting system demand with wind resources often (not always) end up higher even if the average price paid by the utility for a unit of wind is five cents and the average cost of the fuel saved in not making that kWh of wind is the same or even seven to ten cents. This is a case where the devil is in the details, and comparisons of the total costs of system operation with wind can be higher than the costs with less wind, even when the unit price of purchased wind is cheaper than the unit price of fuel. Roughly speaking, the difference between the system operating cost with and without wind is what I refer to as integration costs.
Roughly the same logic applies to Case 2, but the calculations are more complex because the utility is going to have to add some sort of power generator to meet increased demand (my hypothetical case assumes that energy efficiency was not pursued aggressively enough to reduce demand growth to zero or below). In this case, the cost comparison is between two power systems, both larger than today, with larger and smaller amounts of wind (or other variable renewables). With growing systems, the cost of integrating variable renewables can be reduced by blending different types of generation together and by integrating demand response and storage technologies, as explained in the book. Ultimately, it is sort of intuitive that the ability to control an energy resource has value, or alternatively, the inability to control it has a cost.
Comment 13: Chapter 8: The Great Power Shift, Page 110
AL: … you discuss district-heating combined heat and power (CHP) as if it all used steam. Only obsolete systems do. Modern ones use pressurized hot water, which is the standard method in all European systems for over 20 years now. Also, although you mention "industrial processes" in two words on page 110, plus the word "factories" on also page 110, your whole discussion on p 143 is about buildings. It should be easy to redress this building/industrial imbalance.
Comment 14: Chapter 8: The Great Power Shift), Page 112
AL:… you suggest "pollutants" (presumably NOx and CO2) are about the same for natural gas-microturbines as for natural gas-reformer-fuel-cell systems. Huh? Microturbines' efficiencies are around 28-29% last I looked; a decent microreformer like Sandy Thomas's H 2Gen is ~62-65% and a good fuel cell is ~60-70%. As you might expect, I also think fuel cells will ultimately be formidable competitors with other distributed generation and other automotive propulsion (slowed, however, by Steve Chu's having zeroed out the latter without talking to any of the stakeholders, who could have filled his knowledge gaps).
PFP: Amory is no one to mess with when it comes to technology attributes but my information is a little different than his. First off, if the hydrogen used in a fuel cell is made from renewable sources then he is flat out right — fuel cells have essentially zero GHG and NOx emissions while micro-cogeneration does not. If, however, the H2 is made from fossil fuels, as I suspect will be the case for some time to come, then my understanding is that it's roughly a draw, which is what I say in the book. Similarly, micro-cogeneration reportedly gets efficiencies as high or higher than the best full fuel cell cycle according to my data.
More importantly, Smart Power doesn't pretend to be a book that predicts the future mix of generating technologies by size, fuel type, or conversion technology. Its conclusion stops at saying that my best guess is that: a) all three size ranges will continue to be used through the mid century — current utility scale, “community scale”, and building-scale, with the lowest cost size flattening out around community scale, not building scale. This reflects my belief that many of the benefits of avoiding grid construction simply cannot be practically monetized, so smaller distributed generation will have to compete largely on avoided generation cost and other measurable costs and benefits. And b) I conclude that many different fuels and fuel cycles are going to be around, rather than one dominating. I can't come close to declaring whether fuel cells will beat out microturbine cogeneration; my only observation is that I don't see either one winning big shares for a long time to come in the U.S. Amory, who watches this more closely, could well be right — or perhaps not.
Comments 15: Chapter 8: The Great Power Shift, Page 120
AL: … you say large-scale onshore wind "is close to being cost-competitive now"; so how do you account for Lawrence Berkley National Laboratory (LBNL's) repeated finding in its annual wind power status reports that wind prices actually come in around or below the same regions' wholesale power prices (by a sufficient margin that you could take away the 1¢ levelized production tax credit and get the same answer ... especially if you also desubsidized the competitors!).
PFP: When I wrote that wind was “close to competitive now,” natural gas was so cheap at $3.00/MMCF that gas-generated power was four cents a kWh, displacing coal-fired power at five cents/kWh. This is without any costs for backup or integration (see comment 13 above). Note that my numbers predict that wind will be among the cheapest power sources in 2030 (see Appendix B: Summary of Selected Large-Scale Generating Technologies).
Comment 16: Chapter 8: The Great Power Shift, Page 120
AL: … shouldn't you note that you haven't counted cogeneration or end-use efficiency — both very large resources?
PFP: See my response to Amory’s Comment 8 above regarding whether energy efficiency can provide more resources; similarly, I explain why cogeneration doesn’t seem able to penetrate more quickly in Chapter 8: The Great Power Shift.
Comment 17: Chapter 8: The Great Power Shift, Page 120
AL: … commodity costs were less important to wind power cost escalation than supply-chain bottlenecks, e.g. in gearboxes. These have now worked out, and 1H10 turbine contracts are ~18% below 2009 levels. EIA's $1,923/kW is above typical 08 costs per Wiser & Bolinger. Also, your note on wider turbine spacing note may leave the inaccurate impression that wind power is land-intensive. It's actually orders of magnitude less land-intensive than nuclear: see my "Four Nuclear Myths."
Comment 18: Chapter 8: The Great Power Shift, Page120
AL: … my friend Joe's book contains some good sense and much nonsense (The Hype About Hydrogen, 2005). I think you'll find my 2003 "Twenty Hydrogen Myths" a useful antidote. It is, for good reason, the industry-standard white paper even today.
Comment 19: Chapter 9: Billion Dollar Bets, Page 127
AL: … your comment: "Let’s start with Small Scale Wins, the scenario in which all large scale plants and lines are no longer needed. In spite of the hype surrounding the smart grid, the higher costs and regulatory impediments to rapid innovation make this scenario exceedingly unlikely," merits a fuller conversation (as does this whole interesting Chapter 9!) in view of the spectacular global distributed generation installation rates: our soon-to-be-published update to our micropower database figures micropower now provides ~2/3 of the world's new electricity, up from 1/3 in 2006.
PFP: In the coming months Amory and I have agreed to have this discussion and I’ll report back in the topical updates section of this website — although realistically the timetable is well into the autumn. Although I firmly believe that regulation, not technology, is the gating factor, technologies that are extremely cheap and efficient force regulation to move more quickly.
