The US power grid faces a perfect storm—aging, underfunded infrastructure, extreme weather fueled by climate change, and a changing electricity generation portfolio. Experts warned earlier this year that huge regions of the US could have shortfalls in the power supply during times of high demand, which could lead to rolling blackouts. While blackouts haven’t occurred thus far, the system remains vulnerable to extreme weather events.
Given the serious health and economic impacts of blackouts, it’s crucial to understand what’s driving the risk and how we can build a more reliable and resilient grid.
GPI experts Brian Ross, vice president, Renewable Energy, and Matt Prorok, senior policy manager, Energy Systems, answered some key questions from Jennifer Christensen, managing editor, about why we’re facing reliability issues and what’s standing in the way of available solutions, using the Midwest as an example.
Here are some takeaways from the discussion:
- We need to invest in infrastructure to bring clean energy generation to where it’s needed and meet changing energy demand if we’re to build a resilient, reliable, and affordable system.
- Our power grid’s aging infrastructure limits the amount of new generation that can come online.
- Aging, uneconomic power plants will continue to retire as the market demands cheaper, cleaner electricity.
- While clean energy sources are ready to come online, market barriers stand in the way of adding enough capacity as aging power plants retire.
- Climate-driven weather extremes coupled with more and new kinds of demand strain our aging system and power plants.
Why have so many parts of the US faced blackout risks?
Matt: Most times, people lose power to their homes because of a local issue on the distribution grid. Widespread blackouts happen when we don’t have enough electric power generation operating across the region to meet demand and keep the system running. Many factors are driving the risk today.
We have aging grid assets (like transmission lines that are 100 years old in some parts of the country) that limit the new power generation we can bring online to keep the system going. We also have aging power plants that need more frequent maintenance than in the past to keep running. That means they’re not available to provide electricity as often. And increasing occurrences of extreme weather drive load increases (think AC during heat waves) and put more stress on power plants.
Brian: Diving into those factors, we haven’t invested in the transmission infrastructure required for bringing electricity generation to where it’s needed. So, we haven’t been able to build the new capacity that’s waiting in the wings to come online.
Transmission investment would support new market-driven capacity (e.g., renewable energy projects) that can meet the increased demand on the system (including new demand like electric vehicles).
And as Matt said, we have aging power plants. That means we have older, uncompetitive power generation that will retire in blocky increments over time, creating large drops in system capacity. You don’t retire an uneconomic coal plant 10 megawatts at a time. You retire the whole thing.
Added to those factors, we have increasing climate-driven weather volatility, heat, and humidity, which have many impacts. These affect the demand side of the equation (increasing demand for cooling and dehumidification). They also affect the supply side (increasing the unreliability of power generation and greater risks to transmission and distribution systems).
The North American Electric Reliability Corporation (NERC) found that the Midwest is at “high risk of energy emergencies” this summer. Why has the region been at risk of a power supply shortfall?
Brian: This is primarily due to a lack of investment in transmission infrastructure. We have a lag in infrastructure investment relative to market demand for electricity and electricity generation that’s clean and less expensive.
The market demand will move faster than our ability to respond by building infrastructure. This is no different for electricity needs than for any other market demand (housing, manufacturing, etc.). Infrastructure supports the market, but it isn’t always profitable to build in anticipation of the market demand.
Escalating demand for new power sources also tests the capacity of our wholesale power infrastructure to accommodate new supply. Demand and supply need the connecting infrastructure. If it’s not there, we end up with demand that can’t be met with the existing system.
Matt: First, we need to recognize there is always some risk of rolling blackouts. We plan and build the grid to balance reliability and affordability. We can always compel utilities to build more generation capacity, but that would make electricity more expensive. Instead, we plan to a minimum reliability standard that sets the probability of a blackout equal to one day of outage in ten years.
The NERC reliability assessment and the Midcontinent Independent System Operator (MISO, the Midwest’s grid operator) annual summer reliability assessment found that a combination of high loads driven by hot weather and higher than normal generation outages could lead to reliability risk. There’s a risk—it’s not guaranteed to happen. The reports simply highlight the conditions under which the risk of localized blackouts is elevated.
The broader industry trends Brian mentioned are also present in the Midwest region. Over the last several years, we’ve seen older and more expensive power plants retire.
As wholesale markets have made the overall system more efficient, the result is a smaller overall number of installed megawatts of generation connected to the grid, a sign that the system has become more efficient and affordable. Now MISO and state regulators must work together to balance affordability and reliability under much more complex conditions.
One of MISO’s jobs is to ensure enough generation capacity is available. Why then are we still at risk?
Matt: Each year, MISO sets annual requirements for each member utility to own or contract a minimum power generation supply. But older power plants typically have higher forced outage rates, meaning they are more likely to be unavailable when needed than newer plants.
As the grid operator, MISO also has to continuously forecast demand and generation availability (both renewables and other generation types) days and hours in advance so they can coordinate the output of all the power plants across the region. They don’t always get it exactly right, and forecast errors can cause reliability challenges, in addition to generators not being available.
How is MISO set up to mitigate reliability risks to the system?
Matt: MISO runs a whole host of processes to ensure reliability. We can describe those as planning and operations.
First, there’s planning. As mentioned above, MISO sets an annual requirement for generation capacity for every utility. The annual requirement ensures that, as a region, there is enough installed capacity to meet load plus a reserve margin to account for generation outage and load forecast uncertainty.
MISO also conducts seasonal reliability assessments to identify reliability risks and their underlying drivers and communicates that to their member utilities.
Finally, MISO does transmission expansion planning for reliability needs every year in partnership with its member utilities. Building more transmission means more connectivity to move power to where it is needed when reliability challenges arise.
Regional and interregional transmission is notoriously challenging to plan and build. Yet it provides the most resilience benefit against large-scale extreme weather events by enabling the import of power from distant neighboring regions.
And then there’s operations. In the weeks and days leading into any challenging grid conditions, MISO’s operators and forecasters run continuous analyses to understand the risk and how to mitigate it. Up to a week ahead, they will start to bring on large, slow-moving power plants to be available when needed.
If pre-positioning power plants to be ready for extreme conditions is not enough, MISO has other tools they can call on within hours and minutes.
First, MISO can initiate a Capacity Advisory, which tells all its members that reliability risk is elevated, and all generation assets need to be available to the extent possible. If conditions worsen from there, MISO can enter into Emergency Operating Procedures. Those procedures allow them to call on additional “Emergency Only” generation and demand response capacity.
The last step in this emergency process is controlled load shedding or localized blackouts. Those are necessary to keep the whole grid from going down. If that were to happen, it could take weeks to regain power. That’s instead of hours or days that are typical when only a small area experiences controlled load shedding.
Stories on grid reliability issues have highlighted the role of coal plant retirements, with some describing such retirements as early. Is that accurate? And is it policy or market forces mainly driving coal plant retirements?
Brian: It’s the market at work. Coal-fired and nuclear power plants are retiring because they’re not competitive with other forms of generation (primarily renewable generation). Wind and solar are the least expensive ways to generate energy. If we can build and operate new wind energy plants at a lower cost than running existing coal plants, that’s not “early” retirement. That’s letting market forces work.
People want to buy the least-cost energy, which also happens to be the cleanest form of power. So, utilities, corporations, and institutions are all selecting renewables. Coal plants are closing because they are more expensive. Some states are providing policy support to keep nuclear plants online for their zero-carbon energy.
So, if the market is driving coal plant retirements, are other generation sources coming online to replace that capacity?
Brian: The problem is not a shortage of replacement power options. In fact, the power development market is overwhelmed with proposals to add power to the system.
Since 2012, the number of economically viable proposed projects waiting for interconnection approvals has skyrocketed to new levels. Nationally, projects totaling over 1,200 GW of new capacity are in line for interconnection at the start of this year, over four times the amount proposed in 2012. But the infrastructure to connect all this capacity is not yet in place. The wait time could be five to ten years before much of this market-driven development can be connected. In MISO, the capacity of proposed projects has risen almost eightfold from 2012 to 2021.
The resources to replace retiring coal plants are more than sufficient to meet needs, and the market is responding to the need with plenty of proposed projects. Connecting all this new capacity to the grid is taking longer than expected, and transmission investment has significantly lagged behind the market demand and supply.
When you talk about the “market,” what does that mean in the context of the Midwest power grid and reliability issues?
Matt: The market in the context of the Midwest is a bit complex. Let’s break it down briefly to understand what forces are driving investment and retirement decisions for utilities.
Most states in the Midwest have a vertically integrated electric utility regulatory structure. That means the electric utility owns the generation, transmission, and distribution infrastructure to produce and deliver electricity. This is a natural monopoly.
So, states regulate these investor-owned utilities to ensure their investments are appropriate and beneficial for their customers. Cooperative and municipal utilities are typically not subject to state regulation.
Many utilities in the Midwest then participate in an organized wholesale market, run by MISO, the region’s grid operator, that optimizes economics and reliability for the region.
Importantly, vertically integrated utilities largely use MISO’s markets as a pass-through. They both buy and sell electricity in the organized market, ultimately passing along cost savings and increases to their retail customers. Those same retail customers also pay for the capital costs of power plants and power lines their utility has built through their retail rates. This model delivers billions of dollars in savings to customers across MISO every year.
But we can and should also think about the market more broadly than organized wholesale electricity markets. Other tools like power purchase agreements enable customers to purchase the products they want through mechanisms other than state planning processes or organized wholesale markets.
Broadly, these create the market environment in the Midwest:
- Coal plants are retiring because they’re uncompetitive in this market environment.
- Organized wholesale market prices have dropped in recent years, driven largely by cheap natural gas prior to the Russian invasion of Ukraine in 2021, pushing coal plants largely out of the money.
- Power purchase agreements are dominated by renewable energy and storage.
- State regulators are pushing their utilities to invest in new technologies because older coal plants can’t compete economically and because of their greenhouse gas emissions intensity.
Our power grid is gradually changing, with cleaner and lower- or zero-carbon resources coming online. How does that impact planning considerations for ensuring reliability?
Brian: The risk of blackouts or other grid failures has been with us as long as the grid has been with us.
All individual power plants have reliability ratings, with the expectation that they’ll have forced (i.e., unscheduled) outages. They’ll also need scheduled downtime for maintenance and upgrades. These estimates of the statistical probability of losing a power source, particularly when those power sources are needed to meet peak demand, have been around long before we saw wind or solar energy on the grid.
All utilities are expected to carry a reserve margin of 10 percent or more. That margin is targeted at times of peak demand, assuming that some power plants will not be available even when they are most needed.
An example is when almost 900 megawatts of capacity in the Sherco coal plant Unit 3 went offline due to a complete generator failure in 2011. The plant was unexpectedly offline for almost two years. Yet the grid never went down and was not materially affected because the system was designed for such an unexpected loss of capacity.
Increasing levels of renewable power on the grid have not resulted in the need for more reserves. The reliability safeguards needed to make sure traditional fossil fuel plants are available when we need them do the same for renewable energy plants.
Matt: Brian’s example is a good one. Grid planners and operators have been dealing with the sudden loss of a large power plant or transmission line since the grid has been in existence.
In some ways, moving toward a more dispersed generation fleet of many more, smaller resources will inherently help reliability. These unexpected outages may be smaller in magnitude and easier to manage.
The retirement of older power plants, along with the integration of renewable generation, smart electric vehicles and thermostats, and other clean energy technologies, is already changing the way we need to think about grid reliability.
To date, the industry has kind of tried to fit a round peg into a square hole, so to speak, assessing how renewables perform during pre-determined hours of the year that are expected to be hot summer days like other resources.
Based on historical analysis, MISO heavily discounts the accredited capacity value relative to the installed capacity of the plant (e.g., wind gets ~15 percent capacity credit in MISO). But as the grid changes, reliability risk is shifting both time of day and time of year. MISO has seen these changes already and filed an overhaul of its Resource Adequacy construct with the Federal Energy Regulatory Commission (FERC) this year to move from an annual to a four-season capacity auction.
The primary driver for this shift was to help better manage power plant maintenance schedules to avoid too many plants being unavailable at once during the fall and spring, yet another of many changing dynamics on today’s grid.
But MISO’s work is not done. Its Renewable Integration Impact Assessment study identified a number of additional reliability changes, including those that can be addressed via transmission planning (e.g., voltage stability, weak grid, etc.).
Furthermore, the Energy Systems Integration Group has put forward a set of principles that articulate the need to characterize reliability risk in much more detail than the industry has historically. Instead of planning to a standard that says, “we aim not to have more than one ‘loss of load event’ in ten years,” MISO and its stakeholders could develop a framework built around the magnitude, frequency, duration, and even location of anticipated reliability risks. MISO has also just kicked off a new effort to identify and incentivize needed generation resource “attributes” for reliability.
Experts have been calling for investment in infrastructure and other preparation for expected changes to the US electricity system for years, including in the Midwest. Why aren’t we more prepared?
Matt: First, let’s highlight a success story. During Winter Storm Uri, the MISO Multi-Value Project transmission lines, approved in 2011 and built over the last decade, helped move power from the east coast, through the Midwest, and into the central plains to keep the lights on.
Simply put, transmission provides resilience value. And because of MISO’s work in 2011, we saved lives and money during the extreme winter conditions in 2021.
That said, MISO had not planned another portfolio of regional transmission in a decade prior to its recent approval of a portfolio of 18 Long-Range Transmission Planning projects. Transmission lines are expensive and complex to plan and build. Also, transmission investment is not always the preferred investment option for a utility. Generation assets are less risky to build and thus provide a better potential return for the utility.
Lack of transmission build-out has also hampered the region’s ability to build more power plants. MISO has to run an interconnection study each time a new power plant or wind farm wants to connect to the grid. When transmission capacity becomes limited, these studies tell the developer they need to pay for massive transmission upgrades to connect reliably. The prospect of such upgrades often drives those projects out of the realm of financial possibility.
Moreover, MISO has experienced staffing challenges that have made running these studies more time-consuming than their rules require.
Taken together, these mean that while generation capacity is retiring due to age and economics, not enough new generation capacity is coming online quickly enough.
Brian: A host of barriers to new infrastructure and investment in the regional and national grid have limited or delayed needed improvements. While there is broad agreement that new grid investment is needed, the question of who will pay for the investment is a major barrier. Existing power plant owners do not want to pay for transmission capacity for new competing power plants (renewable or not).
Also, new facilities need to be located somewhere, and land owners along existing and proposed new transmission lines are reluctant to host new or expanded facilities. Lawsuits about impacts on property values, the ability to continue to farm, and threats to natural resources slow the process down or even stop projects. The use of eminent domain for such purposes, while legal, creates enormous contention and even legislative action against projects.
Communities being asked to host new transmission facilities often oppose projects because the benefits of transmission projects are seen to go to remote energy users while the host community bears the risks and nuisances of the projects. We see this for any infrastructure expansion (roads, rail lines, airports, pipelines, etc.).
Planning for such large infrastructure is generally accepted to take years or even decades. But we failed to commit to such projects years ago when we needed to. And then the markets moved much faster than some decision-makers believed possible.
Many clean energy facilities (solar, wind, and hybrid resources) are waiting to come online in the Midwest. What are the main challenges with bringing these resources onto the system?
Matt: This primary problem is a lack of transmission capacity to keep connecting projects. MISO studies each interconnecting project for reliability problems. The solution to those problems is to upgrade transmission facilities, the cost of which is assigned to the interconnection customer. Over the last decade, these upgrade costs have risen to the point that in some parts of MISO, few if any projects can afford them, and most drop out of the interconnection queue. In addition, MISO prohibits some technologies like battery storage from addressing this grid capacity need, limiting the tools available to reliably interconnect new projects.
Brian: The plunging cost of renewables has created a demand for these resources that far outstrips the supply. The development market is responding with hundreds of gigawatts of proposed projects—probably more than we have ever seen in history. But transmission planning and investment have never operated at such transformative speed.
There are alternatives to transmission projects that could enable these resources to come online. Yet the regulatory systems don’t accommodate many of these options. Battery storage is one that is poorly accommodated in the regulations and market rules. Options like diverting renewable energy production to produce hydrogen or ammonia are promising but have their own immature market systems.
Another barrier is the contracting norms for purchased power agreements. Traditionally, PPAs just offer a price for production ($/megawatt hour) for 20 years. But that doesn’t allow renewables to be used for load balancing and ancillary services. More creative contracting is needed so that renewables can be used to address grid issues.
How does transmission increase the power grid’s reliability, including as it transitions to a generation portfolio with more zero-carbon resources?
Matt: More transmission, specifically long-distance, multi-state, high-voltage transmission, provides optionality. It helps import power when there is not enough available locally to meet demand (like in the example of Winter Storm Uri mentioned earlier). It enables exports to other regions when they are short on generation, enabling the local utility to sell excess energy at a premium price, helping affordability for their local customers.
The grid of the future will be inherently more disbursed. Generation will not be located as close to load centers, and the average size of a power plant will be smaller (think a 100 MW wind farm vs. a 1000 MW coal plant).
Transmission connects it all together to keep electrons flowing to people’s homes and businesses.
Brian: Geographic diversity has a huge impact on the availability of renewable energy. The sun is always shining, and the wind is always blowing somewhere. The bigger the area that your clean energy fleet is stretched across, the more you smooth out intermittency (the variation in renewable energy production timing) and extend shoulder production periods (time periods of mid-level production, for instance, when the sun is nearing sunset).
The Minnesota Solar Pathways project showed a substantial reduction in intermittency just from the geographic connectivity across Minnesota. When we extended that to the MISO region, there was even more improvement. Across the Eastern Interconnect, even more.
What are the main barriers to building infrastructure on the scale we need?
Matt: Transmission is a “lumpy” investment. It takes nearly ten years to plan and build regional transmission. They are also expensive projects that will be in service for over 40 years.
This means we need to be fairly certain any investment in transmission lines will pay off. The only way to know or project that is by making assumptions about the future. That includes capital costs for new power plants, fuel prices, load growth, etc. This is complex work, complicated further by different state policies and utility goals across MISO’s footprint.
Fortunately, MISO just approved 18 transmission projects in its Northern region due to its Long-Range Transmission Planning process. This will be the single most significant investment in transmission in MISO history and help bring large amounts of renewable energy onto the region’s grid. It’s expected to provide an estimated two-to-one payback over the next 20 to 40 years.
Brian: The cost of the needed transmission improvements is enormous. But since the improvements are distributed across a whole lot of kilowatt-hours over a 40-year period, the cost per kilowatt-hour is generally pretty small. Moreover, it’s almost always offset by the savings from being able to use more inexpensive generation and increasing the efficiency of the system.
But the costs are all upfront, and the savings are down the road. While we like to think our planning and investment decisions are rational, this temporal mismatch is very hard for planners, regulators, and market participants to accept.
Not only are we moving to cleaner electricity sources, but we’re also going to see more electrification, including of transportation. How will that increased electric load impact grid reliability and resilience?
Matt: This is one of the most interesting questions facing the industry right now. If all this new load comes onto the grid uncontrolled, reliability risks may rise. This is because reliability risk would increase if we add more load during peak load periods (when grid stress is the highest).
However, policies could incentivize these loads to come onto the grid overnight or during other times of low risk. In that case, they may help grid reliability and affordability.
It’s all a question of planning and whether the utilities and MISO are anticipating this load growth sufficiently now and for the next ten years.
Brian: Electrification of transportation can pose significant challenges to the grid, or it can help make the grid more reliable. We need to make the correct choices as we roll out electric vehicle charging infrastructure to avoid challenges and capture reliability benefits.
Most car charging will occur while parked at home. With Level 2 charging, the charging time is likely to be only a few hours and can occur anytime overnight. If charging is managed to move it to off-peak times and when other cars are not charging, the impact on the grid is minimal.
Managed charging could actually put downward pressure on costs due to increased utilization (spreading the costs of the grid across a greater number of kilowatt-hours). If a small portion of car batteries can be used to supply the grid as a resource when needed, electric vehicles will improve grid reliability. New system design should take advantage of these options.
Stay up to date with the latest news and insights from GPI by signing up for our monthly newsletter.
