There are two possible future scenarios at the extremes, maximum central renewable energy (RE) from wind, solar, biomass, and hydro -- or from fantasy solar by satellite -- and maximum distributed generation (DG), mostly solar photovoltaics (PV) with storage and bio. Either, or the more likely blend, could lead to large dislocations and even a death spiral for the grid, especially if the price of needed backup gas generation is not as low as currently forecasted. Regulators need to be cautious to minimize the damage and maximize the benefits. This means applying age-old regulatory principles that require careful review of proposed investments. Who pays should be a function of who benefits. Low-income families are likely to be the last to benefit and to need protection.
To paraphrase Lewis Carroll in Alice in Woderland, "If you don't know where you're going, any road will get you there." Regulators need to keep on eye on goals, net benefits, economics, and how useful new technologies really are.
These are deliberately drawn extremes to illustrate the great uncertainties of very different futures, with high risks for both utilities and ratepayers:
Central renewables require transmission, which is particularly expensive under the current Federal Energy Regulatory Commission (FERC) policy of bonus rates of return. Other costs that amount to subsidies to RE include production tax credits, Renewable Portfolio Standards (RPSs), and mandatory Purchased Power Agreements (PPAs), all of which often require purchase of otherwise uneconomic power. Costs of central RE are likely to be socialized, causing all rates to rise. Rates that rise enough could encourage DG bypass, including uneconomic bypass, and ultimately a resulting death spiral, with the smallest consumers left holding the bag.
DG has been encouraged by such expensive regulatory strategies as net metering -- a direct subsidy by which lower-income customers subsidize those wealthy enough to invest in DG such as PV; also by rebates, time of use (TOU) rates, smart meters, and lack of standby rates that would otherwise pay the costs of having the grid available upon the failures of DG. Costs are also imposed outside utility rates by tax credits. In addition, DG imposes costs on the grid -- often unfairly socialized -- for monitoring, uneconomic incentives, control, and integration, as well as reserve requirements needed to respond to demand variability and the ability to handle two-way flows.
DG should not be confused with consumption reduction; rather, DG simply shifts consumption to subsidized services. Nor should DG necessarily be confused with demand reduction; the solar (PV) peak, for example, often does not occur at the time of system or local residential peaks. Customer adoption of DG can thus often be seen as successful regulatory arbitrage. DG also poses risks of stranded transmission and distribution investment, as well as stranded non-utility investment.
The success of DG, whether economic or not, could result in a death spiral for the grid as fewer utility kWh sales are left to pay the fixed costs of the grid; this leads to higher rates, which encourage more DG bypass, which leads to even fewer kWh to pay fixed costs... Perhaps this will begin with microgrids, but eventually, in the spiral scenario, who pays the most is the last one on the grid, likely to be low-income, thus raising a serious equity issue.
A hundred years ago, large electricity users had their own generators -- distributed generation -- until utilities of that day (such as Samuel Insul's Commonwealth Edison in Chicago) figured out that they could provide service to these DG users, and their existing customers, more efficiently by using central utility generators spread over more kWh. Now the process is reversing, encouraging a return to DG in part to again improve load factors, this time by clipping peaks instead of filling valleys. But we need to be aware that, just as 100 years ago, the presence of DG reduces the number of kWhs paying for fixed costs, so we need to investigate under what conditions the shift does and does not contribute to overall economies and affordability.
On the other hand, if the bet on DG is wrong, costs imposed include stranded metering investment, and uneconomic customer investments (which become stranded) due to lack of compensation for grid standby services and overcompensation for energy provided to the grid.
Current policy is to promote both divergent forks, DG and centralized RE.
We should not count on the low price of gas to bail us out of high technology costs. The current unrealistically low gas price projections distort decision-making by encouraging reliance on gas to back up RE when the wind does not blow and the sun does not shine. Gas will turn out to be much more expensive than now forecast, due to local pipeline constraints (especially in harsh New England winters), price volatility, and a national policy to export LNG and thus join the much-higher-priced world market. (Gas prices in Asia are about double those in the US.) Unrealistically low gas price projections could contribute to missing an opportunity for developing, for example, distributed battery storage technologies including fuel cells.
Telecommunications went down a similar road as the industry unravelled a web of cross-subsidies. Thirty-eight percent of American households are now wireless-only; Morgan Stanley projects this to reach as many as 60% by 2018. Wireline subscriptions are down 48% since 2002, network usage minutes of use about 60% since 2000. Prices are up (a review of local New Jersey prices shows prices up more than five times since 1982), cellphone replacements are not cheap, provide lower quality, and are less reliable. New communications technologies may prove worthwhile for large customers; for small residential customers, not so much. And this is obviously not a pretty economic picture for wireline telephone carriers, either.
So have we learned much from the sobering telecom experience? Or the similar experience of broadcasting disrupted by wires, then cable by wireless? Or from the erosion of music and publishing industries’ broad cross-subsidies of narrow content due to the mass appeal of the internet? Maybe not.
The combination of energy efficiency and net-metering-induced DG has already slowed load growth to close to zero in some parts of the West. With apologies to my friends who really are economists, I'll pretend for a moment to be an economist: assume an economic advanced battery. If and when someone invents a truly economic advanced battery, the entire electricity industry will change.
Not today, but in a world of advanced batteries, Morgan Stanley calculates a future on-grid v. off-grid differential, with reasonable reliability and depending on regional differences, of 26 cents per kWh on-grid v. 12 cents off, or 18 cents v. 14 cents, depending on the state. These are not my numbers, and they are certainly not precise forecasts, but Morgan Stanley projects that, with California rates rising at 5% a year, solar reaching 20% penetration, and a policy under which solar pays 50% of the fixed grid fee, off-grid solar with storage will cost only 12 cents per kWh compared to 26 cents for grid electricity in 2020. In New York, according to Morgan Stanley projections, an off-grid Stirling engine with storage (also producing hot water and some heat) would produce electricity at 14 cents per kWh compared to grid electricity at 18 cents. These forecasts depend on many uncertain forecasts and assumptions, but they point in an ominous direction.
Rocky Mountain Institute offers similar projections – base case grid parity on average in New York and California by 2020-2030, sooner for a significant minority of customers and sooner with expanded energy efficiency and/or technological improvement:
This future is not hard to find. In Germany, my wife Theo and I have a house where we pay about 50 cents (US) a kWh (it was about 18 cents in 2000) while our friends who invested in PV are on a solar feed-in tariff that pays them nearly that much for their production. This is with Germany at about 25% saturation of PV; German policy is to get to 45%, although, as you might imagine, there has been something of a consumer backlash.
There are other price pressures, such as gas price volatility, the coming link to high world gas market prices, sharply reduced reliance on low-cost (and environmentally unsound) coal, storm hardening costs, cybersecurity protection, and the possible loss of nuclear power plants (whatever you think of the nukes, right now the existing ones are cheap to run compared to anything that might replace them). As grid prices rise compared to alternatives, those who can invest to economically leave the grid will do so. (For example, Massachusetts prices are projected to rise about 33%-50%.) This will leave behind fewer kWhs to support fixed costs, so prices will rise again... and so on. Ultimately, who would be left on the grid? Those who cannot afford to get off.
There are at least three strategies for subsidizing specific energy fuels. Generous tax and government grant policies subsidize fossil fuels, socializing the cost through the modestly progressive tax system. Utility subsidies of renewables – presumably for the benefit of all – are often spread across all ratepayers, effectively taxing on the basis of electricity consumption. (In many jurisdictions, low-income households on average use about 15 percent less electricity than do average households, thus making this form of cost-sharing very modestly progressive – if one ignores, for example, large low-income households with above-average consumption.) However, the cost of net metering is borne by all ratepayers but targets a subsidy only for those who can afford to invest in distributed generation (most often PV), who are rarely low-income; thus net metering constitutes a perverse tax on the poor to support the rich. So what is a utility regulator to do?
To minimize costly mistakes, regulators should move slowly and cautiously, hedging their bets. This includes retaining and applying tried-and-true regulatory principles -- the common law is the original crowd sourcing -- to guide decisions, minimize costly errors, and address monumental transition issues, the most serious of which is the potential for death spiral. Most of these boil down to cost-based ratemaking, leavened by modest affordability supports:
1. Cost recovery only for known, measurable, and prudent investments,
2. The requirement that investments be used and useful (economic),
3. The requirement that investments be cost-beneficial on the basis of benefits to ratepayers that are concrete and measurable,
4. Investments that are the least-cost means to an agreed-upon objective,
5. Affordability, and
6. Rate designs that allocate costs in accordance with cost-causation and receipt of benefits (Electric Vehicles (EVs) are the easy case: EV customers should pay the cost of the infrastructure that uniquely serves them. Similar allocation rules should apply, for example, to equipment to handle two-way flows to better integrate DG as well as to data-handling costs of time-of-use rates and direct load controls.)
Thus regulators might stop or reduce utility-based subsidies of RE, but even nonbypassable charges (e.g., stand-by charges) won't help if policies have already pushed customers off the grid. If the grid disappears, it will be too late for nonbypassable charges because there will be no regulatory ratemaking.
Indeed, ultimately, we may need to figure out how to finance and service low-income and consumer DG, which will be especially challenging for renters. We will need to reinvent weatherization (Wx) and energy efficiency programs, too.
On the other hand, the grid may prove to be valuable in the context of DG, at least in a long transition period (the length of which will vary by place) or more or less forever (at least in some places). The New York Commission Staff, for example, sees such a future:
It is technically feasible to integrate energy-consuming equipment, as well as distributed generation and storage, fully into the management architecture of the electric grid. … Such an architecture offers the potential of increased efficiency and reduced volatility in system management at both bulk and distribution levels, as well as reduced total consumption and greater penetration of clean and efficient technologies, with ensuing benefits in overall system costs, reliability, and emissions. It also offers the potential for customers to optimize their individual priorities with respect to resilience, power quality, cost, and sustainability. It is not intended to replace central generation, but rather to complement it in the most efficient manner, and to provide new business opportunities to owners of generation and other energy service providers.
Significant questions remain of scope and timing of change, not to mention the end state. Even if the grid does survive, or remains in place during a long transition period, costs could be much greater than they are today, so there will still likely be a serious financial problem for small customers, especially those with low incomes.
For now, to determine who should pay for new investments, follow the money (apologies to Deep Throat) – see who is lining up for the benefits:
* RE and DG sellers
* EV sellers and consumers
* Data miners
* Large customers with an economic need for increased reliability
* Large customers who have loads that can be easily shifted in time to take advantage of time-of-use rates
* Residential customers with the means to invest in PV or other forms of DG
* Notice who is missing? – small residential consumers, including low-income.
Mitigating climate change is and will be expensive. There are many – such as large customers as well as vendors of meters, transmission, and storage – with the economic incentive and ability (unless regulated) to transfer their costs to small residential consumers. But let's not address our climate problems on the backs of the poor, forcing them to pay 26 cents per kWh so those who can afford the investment pay 12 cents.