Wholesale Market Studies
Academic Studies On Savings From Organized Regional Electricity Markets
To “ensure that electricity consumers pay the lowest price possible for reliable service,” the Federal Energy Regulatory Commission (FERC) in Order 2000 encouraged Regional Transmission Organization (RTO) formation. RTOs independently operate transmission systems over large regions and organize markets for wholesale electricity transactions. In doing so, RTOs facilitate equal access to electrical highways for power delivery and enable markets to select the least cost resources to serve customers. In addition, these markets reduce overall variability in wind and solar generation across broader regions as well as smooth demand across time zones and temperature ranges.
Regional Transmission Organizations. Source: FERC. FERC oversees the U.S. RTOs except for ERCOT. “RTOs” herein also include Independent System Operators (ISOs) which are not materially different for most purposes.
In discussions about whether utilities should join such markets, some have asserted that there are no unbiased studies of their benefits. Others have questioned wholesale market benefits because retail electricity rates in some market regions are higher than other regions that are not part of these markets. However, these assertions do not address the academic studies showing that RTO markets have produced cost savings and how other costs utilities include in customer bills erode these savings. Further, comparing rates in different market and non-market regions is an inappropriate counterfactual.
This post will summarize key independent studies on wholesale market savings and how retail-level issues can offset these efficiency gains. In doing so, it clarifies that there are at least two separate questions relevant to considering options for enhancing electricity sector competition:
Whether utilities should participate in organized wholesale electricity markets, even if these markets are imperfect; and
How to structure utility incentives to ensure they cost-effectively serve customers. Conflating these two questions may lead us to miss the benefits of greater market efficiencies as well as the opportunities to diagnose and fix the real problems.
As an initial matter, it is important to clarify why commonly made rate comparisons across markets and non-markets regions are unsound.
First, electricity rates are determined through factors other than the presence of a market, such as fuel, land, tax, labor, and regulatory compliance costs. These tend to be higher in more populated areas, and market regions include the largest U.S. metropolitan centers. The appropriate comparison for studying the impact of markets is thus not between different regions with and without markets, but to compare how a given region would look like with and without markets.
Second, rates as a metric do not fully convey information important to assessing social welfare. Customers ultimately pay bills, and utilities with energy efficiency or low-income assistance programs can have higher rates but lower total bills. Further, energy burden, by comparing bills to income, can better distinguish between regions with high energy consumption due to greater wealth from communities where energy bills can compete with essential household budget items. Notably, energy burden is not more significant in market regions.
Independent retrospective studies find RTO cost savings to be substantial
RTO savings largely come from more efficient use of existing resource fleets and reduced need for additional resources. While the compounded savings from avoided investments are projected to be an order of magnitude greater than the production cost savings from efficient use of existing assets, most studies have focused on the latter.
RTO markets have reduced production costs by increasing trade, better coordinating power plants, and driving efficiency improvements at plants, according to the most cited literature surveyed by U.C. Davis and Dartmouth researchers. Recent academic studies have quantified efficiency gains from wholesale electric energy trading.
A University of Chicago researcher estimated that wholesale markets nationwide saved about $3 billion per year in production costs, based on data from 1999-2012. The savings accrued from greater use of lower-cost plants and increased trading among utilities. The author noted that while market power was a concern, this was far outweighed by market efficiency savings.
A Dartmouth study found that 19 Midwest utilities joining PJM in 2004 produced efficiency gains of over $160 million annually, exceeding the one-time $40 million implementation cost. These Midwestern utilities already had been trading bilaterally with their eastern neighbors. After joining PJM, the energy traded between them tripled, and production shifted to lower-cost facilities as the market identified new trading opportunities.
An Oberlin study examining Texas’ transition from a bilateral market to a centralized auction found improved market efficiency that dominated any change in market power incentives. Following the transition, production shifted to lower-cost generators, leading to annual cost savings of about $59 million.
Retail rate mechanisms can erode RTO cost savings in end-use customer bills
Even though RTO markets reduce wholesale rates, costs incurred through utility rate mechanisms that vary by state can offset these savings. These expenses determined through state-jurisdictional processes can be half or more of the total bill.
Example Breakdown of 2018 Total Retail Rate in PJM for Average Customer. Source: PJM, using EIA, EEI, and Monitoring Analytics data. “Tied to state tariffs” costs are those that vary by state. Production costs (including constraints and losses) are in dark blue. Capacity costs keep plants in service, contribute to building or upgrading generation assets, or pay demand response to be available.
Utilities can participate in RTOs whether they are vertically integrated or restructured, and how utility incentives are structured impact costs tied to state tariffs. States can require their vertically integrated monopoly utilities to divest their generation and competitively procure it from the market, and some have further enabled retail competition by allowing end-use customers to choose suppliers. However, most utilities that were required to restructure were allowed to sell their generation assets to subsidiaries instead of functionally divesting them, and in most jurisdictions enabling retail choice, incumbents were allowed to become the default service provider. Such incomplete restructuring limits competition and can raise costs for customers.
While organized markets can and should be improved, they have been generating savings, particularly as natural gas prices have trended downward. This has reduced the RTO cost component in end-use consumer bills. This cost reduction is most apparent where utility and electricity provider incentives align with competitively serving end-use customers, such as in ERCOT. There, utilities must fully divest generation assets, market prices better reflect the value of electricity, and customers can choose providers. Rice University researchers examined total bill data from Texas and found residential customers benefited from retail choice compared to non-restructured parts of the state. They attribute this savings to declining service provider costs, reduced price mark-ups, and the increased pass-through of declining wholesale market costs to end-use customers.
In regions with limited competition, wholesale savings have been correspondingly muted in end-use customer retail rates. A Lawrence Berkeley Labs study shows that despite wholesale cost declines from 2007 to 2016, only regions where most utilities have been restructured show retail rate decreases (corresponding to northern PJM, ISO-NE and NYISO). In contrast, retail rates in regions with mostly vertically integrated utilities (the Southeast, West, and Midwest) held steady even as wholesale costs decreased. Report authors noted that many utilities have offset some or all of the savings with greater capital expenditures that go into customer rates. Much of this was on state-jurisdictional plant-in-service additions.
Retail rates (blue, 2016$/kWh) somewhat decreased along with wholesale prices (red) in mostly restructured regions (northern PJM, NYISO and ISO-NE).
Source: Lawrence Berkeley Labs. In this study, NPCC is NYISO and ISO-NE combined. RFC here includes the northern part of PJM. SERC here includes the southeastern US and utilities that are now part of PJM and MISO but not the Florida peninsula. MRO roughly corresponds to the northern part of MISO, and WECC is the Western Interconnection. Increases in state-jurisdictional utility rates through increased capital expenditures, mostly in regions with vertically integrated utilities, have offset decreases in the wholesale rates (in red) so that the retail rate stays roughly constant.
Retail rates held steady despite wholesale cost declines in regions with mostly vertical integrated utilities.
Utility monopoly influence can also persist after restructuring through long-term contracts and common ownership of regulated distribution and competitive generation companies. Two recent studies illustrate this point.
Ohio State researchers examining total bills for Ohio customers noted that wholesale savings have generally been offset due to financial arrangements between affiliated generation and distribution companies. Like many other states, Ohio’s restructuring process allowed utilities to divest generation assets by selling them to subsidiaries. These utility-affiliated generation companies were predominantly legacy coal plants losing revenues to new gas plants. Ohio allowed distribution utilities to add nonbypassable riders to customer bills that forced all customers, even ones that switched suppliers, to subsidize losses from the shale boom. Some of these charges amounted to over 60% of the total bill. These customers saw rates increase despite falling wholesale costs. However, Duke Energy Ohio, unlike other Ohio utilities, functionally divested its generation assets and did not have the incentive to pass generation company losses to its distribution customers. These customers observed net savings in response to falling wholesale market prices. Separately, the researchers noted that any cross-subsidization between customer classes would likely arise from the state-regulated component (not the RTO part) of the bill.
A Harvard and Columbia study found that long-term bilateral power purchases outside of organized markets were marked up compared to fuel costs. Many of these contracts were between affiliated generation and regulated distribution utilities. These markups were greater in utilities that had more residential customers because they were less likely to search for better deals and switch companies compared to other customer classes. Aside from market power, there could be reasons for bilateral contracting that is not based purely on costs, such as hedging or local development. The authors did not see evidence that renewable portfolio standards were responsible for the higher costs. In conclusion, they stated that their findings do not imply that electricity markets should remain regulated, but emphasized the importance of careful oversight of markets.
Independent studies show that RTO markets produce efficiency gains, but incomplete restructuring can add avoidable costs to customers’ retail bills and diminish these savings. While a thorough discussion of reforms needed to address the problems identified in the studies is beyond the scope of this article, the authors and other academics offer several ideas.
First, the researchers recommended better oversight and revisiting permissible affiliate relationships to improve restructuring implementation. Decision makers could better ensure that restructuring results in robust competition between multiple suppliers as well as independent distribution utilities that have incentives to efficiently serve end-use customers.
Improving data access and accuracy could also help regulators and stakeholders gauge what costs are reasonable to expect in rates. The researchers expressed a need for total bill data for all utilities with line items including indirect costs, flow-through revenues, and nonbypassable riders. Researchers also wanted data to reflect ownership and operator information for generation facilities and holding company structures. For example, EIA rate data are artificially suppressed because they include total electricity consumption by the distribution utility, even from customers who switched to a competitive generation supplier, while excluding revenues that flow through to affiliated companies.
Lastly, regulators could mitigate market power issues by improving customer engagement. The extent to which consumers can capture some of the savings created from competitive markets depends on their ability to respond to prices and put pressure on sellers to deliver value. Indeed, customers active in market bidding have the potential to neutralize supplier market power. Improving price signals to customers and providing them with devices that automate response can be important complementary mitigation strategies.