Gearing Up for Grid Modernization
Our grid investment incentive structures need some re-tooling.
The Biden administration is swinging for the fences with this new infrastructure plan. There’s a lot going on in this proposal. Electricity enthusiast that I am, I’ve been focused on the push to accelerate investments in grid modernization.
Says Biden: “Put simply, these are investments we have to make.. we can’t afford not to.”
As far as the power sector is concerned, study after study suggests he’s right. If we’re serious about deep decarbonization, we need significant investments in transmission and distribution system expansions and improvements.
The political equivalent of a rare bird sighting, there appears to be bipartisan support for grid infrastructure spending. Especially the kind that connects renewable energy projects in red states to urban electricity consumers in blue states.
This come-together spirit quickly runs into trouble when it comes to figuring out how to pay for it. Raising the corporate tax rate- the proposed revenue source behind Biden’s infrastructure plan- could be dead-on-arrival. Big increases in federal spending on needed grid infrastructure will be a heavy political lift. It seems pretty clear that private investment will have a big role to play. But this begs a thorny question: How to incentivize these investments?
Here in California, we’re learning the hard way that old-school financial incentives (embedded in rate-of-return regulation) are not the right tool for this job. Along with grid infrastructure, our grid investment incentive structure needs to be re-invented.
Rate-of-return 101
In the simplest of terms, utilities make money when they build new stuff. Utility profits are determined by multiplying the value of undepreciated assets (or the “rate base” in utility-speak) by an authorized rate of return.
Two of our newly minted EI PhDs, Karl Dunkle Werner and Stephen Jarvis, have recently joined forces to study trends in the authorized rates of return on equity (ROE). Their work is not done, but they generously shared this sneak preview…
The graph shows how the authorized ROE (inflation-adjusted) for utilities held pretty steady since 1990 while other market-based measures of borrowing costs have been in decline. What explains the widening gap? After all, utility debt/equity ratios have also been falling over this period. And utility credit ratings have been pretty stable.
I think part of the explanation could be that state and federal regulators are feeling pressured to authorize relatively high investor returns to entice needed (and often risky) investments in energy infrastructure projects. To see how this can work, consider this California example:
The $3B Tehachapi Renewable Transmission Project connects huge wind farms in Kern County to millions of L.A. customers. To help get this project off the ground, FERC approved an ROE “adder” to increase the base rate for this project by 1.25% (bringing the total ROE to 12.87 percent). Over the life of the project, the ROE adder alone will likely earn the utility (and cost ratepayers) between $450-$500 million.
I’m not suggesting that we shouldn’t be compensating investors for taking on risky grid infrastructure projects. And it’s clear that juicing up the authorized ROE is a powerful tool that regulators can use to get this job done. The problem is that there are some unfortunate side effects.
Side-Effect #1: Capital Fixation
Under standard cost-of-service regulation, utilities earn a rate of return on capital investments. But operating expenses (labor, fuel, coffee) are typically recovered dollar-for-dollar. In other words, utilities earn profits for their shareholders if they build new stuff, but they do not similarly profit from figuring out how to use the stuff they already have more efficiently.
When it comes to modernizing the grid, there are some highly cost-effective and cutting-edge innovations that involve the optimization and digitization of operations. Utilities are relatively well positioned- given the troves of data they hold and the assets they operate- to deploy targeted demand-response programs and grid enhancing technologies. But these kinds of solutions are at a relative disadvantage under standard cost-of-service incentives.
Side-Effect #2: Electricity taxation
Higher authorized rates of return mean higher costs for ratepayers. This typically translates to higher per-unit electricity prices. In other words, we are taxing electricity consumption to pay for grid modernization. This a regressive way to raise needed revenues. Also, it slows progress towards electrification –the most promising path to deep decarbonization.
California is currently wrestling with this investment cost recovery predicament. Retail electricity prices are on the rise and capital investments in power system infrastructure are a key driver. Taking stock of this situation, our public utility commission analysts have warned that, going forward, “it will be essential to employ aggressive actions to minimize growth in the utility rate base and protect lower-income ratepayers from bill impacts”.
Regulatory innovation
We’re hearing a lot about how hard it is to finance and build energy infrastructure projects. So it seems clear that we’ll need financial incentives (in addition to streamlined permitting, regional cooperation, etc.) to attract needed private investment. Given the sense of urgency and the political headwinds, it will be tempting to just lean into the incentives we already have in place. But when it comes to rate-of-return regulation, there are efficiency and equity drawbacks to consider.
There are encouraging signs that DOE and FERC and state PUCs are pursuing more innovative alternatives. This is important work! We should be seriously considering alternative financing tools, creative uses of limited federal funds, and performance-based incentives. The regulatory incentives we put in place now will determine what investments get made, what it costs, and who ultimately pays the price.
Keep up with Energy Institute blogs, research, and events on Twitter @energyathaas.
Suggested citation: Fowlie, Meredith. “Gearing Up for Grid Modernization” Energy Institute Blog, UC Berkeley, May 10, 2021, https://energyathaas.wordpress.com/2021/05/10/gearing-up-for-grid-modernization/
Categories
Energy Institute Blog 
I strongly support both “theses” – that there is capital fixation and that utility rates are the most regressive means of financing a clean energy transition; however, I beg for clarity around the use of the term “grid modernization,” and suggest it has little to do with promoting clean energy. Dr. Fowlie implies that grid modernization should emphasize “targeted demand-response programs and grid enhancing technologies,” which are not supported by current rate base incentives. I’d like to address two issues:
1) I raise a plea to use the term “grid expansion” rather than “grid modernization.” Most “studies” re. grid modernization – at least with respect to promoting clean energy – really explain the need to expand the T system and renewable generation to achieve clean energy by building gen where there are renewable resources and moving the electricity to load; however, this is not at all what local utilities (such as SCE and PG&E in California) mean by “grid modernization.” Their GM strategies are all about installing sensors, reclosers, and communications equipment on the distribution grid, and making massive computer software investments in their operating control rooms. Most of the alleged benefits derive exclusively from reliability gains due to improved switching operations (switching load from one circuit to another). Tangentially, the utilities claim these investments are necessary to better integrate DERs, and create a “DER marketplace” where customers can eventually bid individual or aggregated services. My review of all this suggests that these alleged benefits are miniscule compared to the level of investments and have nothing to do with the “transmission and generation expansion” issues discussed by Dr. Fowlie.
2) I am not convinced by the transmission O&M example for adopting a “different model” than rate based incentive compensation in order to spur demand response or other clean technologies to optimize grid use. I find the case somewhat weak that current incentives are inadequate to foster the expansion of T and renewable G necessary to dramatically cut GHG emissions soon. It seems that the main barriers are the disparate jurisdictions that control siting and permitting of T that would access renewable resources in some states and transport them to others. Furthermore, the CA grid mod investments do little to accelerate deployment for distributed solar, storage or demand response on the distribution grid. The California IOUs separately fund distribution capacity upgrades to increase the hosting capacity for distributed generation. My impression is that the main barriers to promoting distributed storage and demand response have to do with integration into wholesale markets, though this is a big issue. Only at very high levels of DG penetration (aka Hawaii and a few circuits in CA), do you need much work on the D system. And the work is not “modernization,” it is simply installing bigger traditional assets to increase hosting capacity. The “modernization” – i.e. much more digital equipment and communications and software – provides little additional value added, at least based on my review of PG&E and SCE plans. It seems to me that the IOUs would like to have a cool operating center just like the CAISO’s, but with data on all local DERs. What problem this solves is unclear?
Obviously, your mileage in other states might differ.
“We should be seriously considering ….performance-based incentives.” Agree Meredith, and fortunately the Biden administration is doing just that:
“The White House has signaled privately to lawmakers and stakeholders in recent weeks that it supports taxpayer subsidies to keep nuclear facilities from closing and making it harder to meet U.S. climate goals, three sources familiar with the discussions told Reuters.
New subsidies, in the form of ‘production tax credits,’ would likely be swept into President Joe Biden’s multi-trillion-dollar legislative effort to invest in infrastructure and jobs, the sources said.”
https://www.reuters.com/business/sustainable-business/white-house-eyes-subsidies-nuclear-plants-help-meet-climate-targets-sources-2021-05-05/
No better use of infrastructure funding than extending the same production tax credit that currently (and arbitrarily) favors solar and wind to nuclear plants. Why? It avoids
• the expense, permitting requirements, and environmental impact of thousands of miles of new transmission, and maintenance thereof;
• foisting responsibility for providing a reliable supply of electricity on customers (via “demand/response” pricing);
• the manufactured need for grid “modernization” only to benefit gas and renewables developers;
• dependence on natural gas generation for the foreseeable future.
Instead of paying utilities “to build stuff”, it rewards them for keeping existing zero-carbon resources on line – for the benefit of both customers and the environment.
I testified on behalf of the Environmental Defense Fund in the 2019 rate of return cases at the CPUC. I used a similar approach to show that IOU returns were out of line with corporate market returns. You can read a summary of our case in these two blog posts:
https://mcubedecon.com/2019/11/12/utilities-returns-are-too-high-part-1/
https://mcubedecon.com/2019/11/13/utilities-returns-are-too-high-part-2/
The maintenance of these high returns appears to have nothing to do with trying to manage returns to attract investment. Rather the problems are a sole focus on comparisons within the utility industry without looking at the broader market, and reliance on what is an outdated valuation methodology.
Almost all of these cases involve comparing the proposed return on equity to the ROE of other utilities. This endogeneity creates self perpetuating standards that are difficult to dislodge. FERC perpetuates this problem because it doesn’t question the ROE requests from transmission owners, and that becomes a reference point for state regulators.
Further, the two standard methods of capital asset pricing models (CAPM) and discounted cash flow (DCF) rely on a relationship between equity prices and bond returns that appear to hold no longer. I looked at the relationship between S&P 500 price to earning (P/E) ratios and Treasury Bond rates since the 1950s. The two held the expected relationship with a 0.89 correlation until 2002. And then something happened, and the correlation plummeted to -0.11. Without that standard expected relationship, the CAPM and DCF approaches are no longer valid.
Unfortunately, regulators don’t have a sufficient understanding of finance markets, much less theory, to be able to critique and reject the cases put forward by the utilities standard team of consulting experts. I’m not sure what it will take change this situation. I calculated that the ROE for the California utilities should be as low as 7%, but the CPUC just held them constant at about 10%. (The utilities had asked for an increase to 12% or more.)
Meredith what I have seen of the DC backbone transmission system proposed is that it’s not workable electrically. If all the lines are DC then the flows have to controlled on each line from a central point by a computer setting the flows. Unlike AC lines the DC line flows have to be dialed in. If the controlling computer were to go down the entire backbone would go dark and the US would have a major blackout. Another problem is that the few DC lines proposed are too high powered. You cannot interconnect a 5 GW DC line into ERCOT. The maximum power for a single DC line connected to ERCOT now is 2 GW. This is to insure that instant loss of the line will not disrupt the ERCOT system, i.e. we have enough reserve generation on line to pick up the loss. If ERCOT were to try to import say 40 GW of power from outside ERCOT we would be looking at 20 of these lines, not just a few, and they would be running west, north, and east from ERCOT, a much more complicated system than being contemplated by the New Green deal folks.
Finally there is another problem. ERCOT has a lot of solar and wind resources still inside the state. Its going to take a lot of new transmission inside the state before we can tap into all those resources. Texas will be spending money for its internal lines before running lines out of state. But the Texas legislature doesn’t want to spend money at this time on more new CREZ type lines. So the issue of who pays for the new lines inside the state remains unknown. I suspect this is the same situation across the US. States with new high powered new green deal lines crossing their state and the state not being able to tap the line will be opposed to those lines. This is where this backbone system differs from an interstate highway system where everyone along the way gets a benefit. The citizens will oppose these lines to the point they will never be built is my prediction.
Thank you, Gene. There is no economic argument to build the “national transmission superhighway” envisioned by renewables activists – but it helps to have a grid engineer with a lifetime of experience at ERCOT to explain why it makes no technical sense, either!
Well done Meredith! Especially the point about inadequate financial incentives for demand response.
David Wooley
I’m not sure I see the risk in the example $3B transmission project. It looks like the project with few moving parts first was: Deploy $3B into stationary assets and get to rate-base it and collect it over time plus 11.62% ROI on book value. And then the “kicker” of 1.25% jump in allowed ROI was applied to make it extract 12.87% from the rate payers on book value. Is the risk that the 80 year working life transmission project might be later disallowed and expunged from the book value in less than 8 years and then no more ROI? (8 years being how long it takes to collect a cost at 12.87% per year.). Perhaps a guarantee of not disallowing transmission projects collection of ROI in the first 20-30 years of useful life would lower the risk and allow a lower ROI. Doesn’t that already exist? I assume any O&M to keep the project operational is simply collected from ratepayers as an expense item on this investment. Unless I’m missing something, I see the risk for how useful or effective the project physically is… resides on the ratepayer already, just like the lack of the project had operational costs and value losses that resided on the ratepayers already.