The Minnesota Public Utilities Commission (PUC) has invited public comments on their energy affordability investigation by 4:30 pm on Friday, October 16, 2026.
The formal title for this is: In the Matter of Commission Investigation into Rate-Regulated Electric and Natural Gas Utility Service Affordability. PUC Docket Number: E,G999/CI-26-341.
The PUC has asked for written comments on the following five questions, which I address further below on behalf of the Center of the American Experiment (CAE).
The data in the charts presented below were sourced from the Energy Information Administration (EIA) and Bureau of Labor Statistics (BLS). Upon PUC request, I am willing and able to share the relevant spreadsheets where my charts and the data originated from.
How are electric and natural gas costs affecting households, businesses, agricultural customers, and other customer groups in Minnesota?
Minnesota’s rankings amongst the 50 states, in terms of average electricity rates, were well above average in 2001 for total, residential, commercial and industrial customers. These have all significantly sunk since, as can be seen in Chart 1 below. Highlights include:
- Total customer electricity rates rank dropped 16 spots from #14 in 2001 to #30 in 2026.
- Residential customer rank fell 9 spots from #19 in 2001 to #28 in 2026.
- Commercial customer rank fell 10 spots from #15 in 2001 to #25 in 2026.
- Industrial customer rank dropped 15 spots from #18 in 2001 to #33 in 2026.
- Note that the worst rankings were #37 for total in 2021, #36 for residential in 2020, #36 for commercial in 2021, and #39 for industrial in 2021-2022.
Chart 1
Minnesota’s average electricity rates have skyrocketed from 2001 to 2026 for total, residential, commercial and industrial customers, as can be seen in Chart 2 below. Highlights include:
- Total customer electricity rates rose 117% from 5.97 in 2001 to 13.03 in 2026.
- Residential customer electricity rates rose 113% from 7.61 in 2001 to 16.25 in 2026.
- Commercial customer electricity rates rose 105% from 6.03 in 2001 to 12.47 in 2026.
- Industrial customer electricity rates rose 124% from 4.34 in 2001 to 9.75 in 2026.
- Note that total, residential and commercial rates growth were largely flat from 1983 to 2000 at 6%, 14% and 10%, whilst industrial was literally flat at 0% rate growth.
Chart 2
The electricity Consumer Price Index (CPI) for the Twin Cities (MSP), as can be seen in Chart 3 below, started to significantly rise above that of the Midwest (MW) from REO in 2001 then the whole of the US from RES in 2007. Highlights include:
- MSP electricity CPI rose 127% from 141 index points in 2001 to 321 in 2024.
- MW electricity CPI rose 91% from 126 index points in 2001 to 241 in 2024.
- US electricity CPI rose 103% from 138 index points in 2001 to 279 in 2024.
- Note that the BLS set the CPI base period to 100 index points for 1982-84.
Chart 3
Electricity CPI for the Twin Cities (MSP), as can be seen in Chart 4 below, has been rising at a higher rate than general inflation (i.e. all items CPI) throughout the 21st century, and overtaking it from 2011 onwards. Highlights from 2001 to 2024 include:
- MSP electricity inflation of 127% exceeded general inflation of 70% by 58%.
- MW electricity inflation of 91% exceeded general inflation of 68% by 23%.
- US electricity inflation of 103% exceeded general inflation of 77% by 25%.
- Note that the BLS set the CPI base period to 100 index points for 1982-84.
Chart 4
Chart 5 below makes Chart 4 above even clearer, by subtracting electricity CPI from inflation. That shows electricity prices deflating inflation from 1978 then fuelling it since 2002 onwards.
Chart 5
What aspects of utility bills, rates, or recent cost increases are creating the greatest affordability concerns?
The greatest concern is that electricity use has been in decline, not due to efficiency, but to unaffordability. As per Chart 6 below, average electricity use, for all three major classes of customers combined, grew 26% from 21,417 kWh per customer in 1984 to a peak of 26,876 kWh in 2008 then shrunk -17% to 22,375 kWh by 2025.
It is important to note in this context, the Khazzoom-Brookes postulate which is that: “Energy efficiency gains can increase energy use even more directly by increasing the economic growth rate, not only by decreasing the effective cost of energy.” This is sometimes called the rebound effect or take-back effect as well as the Jevons paradox.
Chart 6
As per Chart 7 below, the index of average MN electricity prices and bills both took off from 2002 by 118% and 87% respectively, whereas average MN electricity uses shrinks by -15%. Note that bills are revenues divided by customers and revenues are prices times uses. Bills appear to be rising less than prices, due to shrinking use, as a result of rising prices not efficiencies.
Chart 7
As per the Chart 8 below, average MN electricity bills, for all three major classes of customers combined, inflated 33% from $1,142 per customer in 1984 to $1,519 in 2002, then shot up 87% to $2,834 by 2025.
Chart 8
Under Minnesota statutes, section 216B.16, the Minnesota Public Utilities Commission is directed to primarily rely on cost-of-service regulation in setting rates. What additional measures or metrics should be considered to evaluate whether utility service is affordable for different types of customers?
Given the PUC is investigating natural gas affordability as well as electricity, both of which they regulate, it is very important to compare the two. As per Chart 9 below, the MN average electricity price per MMBtu soared by 118% from $16.99 in 2002 to $37.12 in 2025, whereas natural gas diminished by -8.8% from $5.33 to $4.86.
Chart 9
As per Chart 10 below, the ratio of those two average prices per MMBtu, of electricity divided by natural gas, hit a low of 227% in 2006, then has since risen by 3.4x to 763% in 2025.
Chart 10
As per the Chart 11 below, PUC-regulated MN electricity revenues per year exploded by 131% from $3.6 Billion in 2002 to $8.3 Billion in 2025, whilst natural gas revenues only increased by 27% from $1.9 Billion to $2.4 Billion.
Chart 11
In Chart 12 below, MN’s electricity generation mix of both renewables and natural gas grew from 3% each in 2004 to 31% and 27% respectively by 2024, whilst coal and nuclear both shrunk from 65% and 25% respectively in 2004 to 20% each by 2024. Renewables include hydro, solar and wind.
Chart 12
In Chart 13 below, MN’s renewables generation mix of wind grew from 2% in 2004 to 25% by 2024 and solar grew from 1% in 2017 to 4% by 2024, whilst both hydro and biomass stayed relatively flat from 2004 to 2024 at less than 2% and 1% respectively.
Chart 13
Chart 14 below shows renewables generation ratio growth for MN, MW and US:
- The renewables generation ratio for MN increased from 3% to 31%, for MW from 2% to 21%, and US from 7% to 21%.
- The nuclear generation ratio for MN decreased from 26% to 20%, for MW from 21% to 20%, and US from 20% to 18%.
- The fossil generation ratio for MN decreased from 67% to 47%, for MW from 75% to 58%, and US from 68% to 59%.
Chart 14
Electricity unaffordability in Minnesota starts and ends with wind and solar renewables. Chart 15 below shows a comprehensive Levelized Cost of Energy (LCOE), where nuclear at $37.18 per MWh is:
- 1.3x less expensive than combined cycle at $50.03;
- 2.4x less expensive than coal at $89.79;
- 3.1x less expensive than natural gas at $115.83;
- 7.9x less expensive than wind at $296.97; and
- 13.3x less expensive than solar at $496.16.
It is important to note that nuclear power is also the best when it comes to total life cycle CO2, as well as note that wind and solar are not without environmental costs, as documented in Shattered Green Dreams. Highlights from this landmark CAE report include:
- “A wide variety and large quantity of minerals are used in solar panels, wind turbines, battery storage, transmission lines, and more. The US currently sources most of its minerals from foreign countries that do not adhere to modern environmental or worker health and safety standards.”
- “The low electricity density of wind and solar generation means that they require at least 10 times as much land per unit of power produced as coal or natural gas-fired power plants. If the US were powered entirely by wind turbines, the land area necessary would exceed two Californias.”
- “Some components in wind turbines and solar panels are hazardous. Current recycling pathways are uneconomic and underutilized, which means that decommissioned wind turbines and solar panels often end up in landfills.”
- “The operating lifespan of wind turbines and solar panels is between 20 and 25 years at maximum, while natural gas plants may operate for 40 years, and nuclear plants operate between 40 and 80 years. Decommissioning and repowering wind and solar energy is required more often than other forms of electricity generation, compounding costs.”
Chart 15
As per Chart 16 below, Aii found in July 2026 that linear authorized ROE (returns on equity) declined over time from slightly below 11% in 2002 to under 9.5% in 2026. NYU published, in January 2026, PUC-relevant industry ROEs of: “Cable TV” 11.06%; “Coal & Related Energy” -2.51%; “Green & Renewable Energy” -7.36%; “Oil/Gas (Integrated)” 9.96%; “Power” 12.40%; “Telecom. Services” 19.82%; “Transportation (Railroads)” 29.07%; “Utility (General)” 10.42%; “Utility (Water)” 10.28%; and “Total Market” 17.21%.
Chart 16
Although the MN AG is challenging the PUC on electricity affordability, ROE is a relatively small numerator over the massive capital base denominator, the latter being driven by wind and solar renewables.
In summary, Minnesota’s electricity affordability has worsened due to the meteoric rise in renewables generation, especially that of wind, and to a much lesser degree solar. However, this has been mitigated, to some extent, by the almost parallel rise in natural gas generation. But natural gas is far better suited to peak load than base load generation, whereas wind and solar are suited to supplemental generation. Coal and nuclear are best suited for base load.
What actions by utilities, the Commission, the Legislature, or other entities could most effectively improve utility affordability?
Shorter term solutions, to be implemented in the first half of 2027, should include:
- lift the moratorium on nuclear power;
- impose a moratorium on banning natural gas;
- impose a moratorium on retiring coal;
- impose a moratorium on expanding renewables.
Longer term solutions, to be explored and pursued from the second half of 2027, include:
- making the four shorter term solutions above permanent;
- make PUC commissioners elected positions, with term limits triggered when rates are above inflation (e.g. CPI+X), or at least have a truly independent non-partisan chair along with 2 Democrats and 2 genuine Republicans who are not just former Democrats;
- move from disincentive regulation (i.e. rate-of-return) to incentive regulation (i.e. CPI-X);
- open up the unnatural franchise monopoly markets to genuine competition.
The Becker Friedman Institute for Economics at the University of Chicago updated their landmark paper in July 2024 under the title, “Do Renewable Portfolio Standards [RPS] Deliver Cost-Effective Carbon Abatement?” The original paper was published in November 2020. The conclusion, by the two authors of Michael Greenstone and Ishan Nath remains the same. This is from authors who are not climate change sceptics, at a university who are no longer free-marketers.
[E]lectricity prices increase substantially after RPS adoption. The estimates indicate that in the 7th year after passage, average retail electricity prices are 1.4 cents per kWh or 11% higher[.] … Twelve years later they are 2.3 cents, or 17%, higher.
[T]he majority of RPS programs were enacted after 2000, loosely corresponding with a break in the trend of national electricity prices, which declined from about 14 cents per kWh to 12 cents per kWh from 1990 through 2002 but returned to 14 cents per kWh by the end of the sample in 2015.
This increase is statistically significant and economically substantial. Such a large increase in the retail price of electricity is striking, given the modest net requirements 7 years after passage. Further, these estimates are much larger than LCOE differences alone would indicate, suggesting that the indirect costs of RPS mandates are an important component of their total costs.
Standard “Levelized Cost of Electricity” (LCOE) estimates, capturing the direct capital and maintenance costs of various generation sources, provide an incomplete measure of the impact of transitioning electricity production to renewable sources on consumer prices.
A key challenge is that renewables are generally intermittent, so their installation must be accompanied by sources that generate electricity when the renewables are unavailable (e.g., when the sun isn’t shining or wind isn’t blowing). Additionally, they tend to be located away from population centers and cause the displacement of existing sources, both of which also introduce system wide costs.
For simplicity, the model assumes a vertically integrated setting with a single utility responsible for both power capacity and retail provision [including] the assumption that ratepayers always pay the full cost of installed capacity.
What costs, benefits, and trade-offs should the Commission consider when evaluating measures intended to improve affordability?
In my response to the Request for Information (RFI) from the public by the Great Plains Institute (GPI) regarding the Minnesota Nuclear Energy Study (MNES), I stated that:
Economic impacts should be assessed in a balanced way using sound Cost Benefit Analysis (CBA). This includes not using Keynesian Multipliers nor the Precautionary Principle in a one-sided manner. Better yet, CBA usually precludes both. The ‘gold standard’ of guides is the Australian Government’s January 2006 Handbook of CBA.
The PUC should produce their own comprehensive CBA guide along the lines of that of the Commonwealth of Australia’s Handbook of Cost Benefit Analysis which states:
The Handbook provides guidance in the use of cost-benefit analysis for evaluation and decision-making. It is aimed at enhancing capacities and fostering good practices in the use of cost-benefit analysis. The Handbook also covers two alternative methodologies – financial evaluation and cost-effectiveness analysis – for the evaluation of projects and programmes.
Reliable evidence is needed to quantify costs and benefits. In practice quantification often depends on detailed technical studies – for example, the electricity output expected from a new power plant. In these circumstances, careful attention is needed to the pricing assumptions which underlie the studies: demand may increase if prices are lower than has been assumed, but decrease if they are higher than has been assumed.
Market prices, where they exist, provide a great deal of information concerning the magnitude of costs and benefits. In most markets, consumers at the margin are willing to pay no more or no less than the actual price in the market. Accordingly, that price can generally be taken as a measure of the value placed by society on the good or service. Similarly, prices of inputs usually reflect the value which alternative users of these inputs place upon them.
The 2024 University of Chicago study also made the following points of relevance here:
The model demonstrates how intermittency, transmission, and the displacement of existing capacity infrastructure interact to raise total costs. Notably, the model highlights the wide range of parameters and nontransparent data inputs that would be required to calculate these costs directly. The paper’s empirical procedure sidesteps this difficulty by summarizing the aggregate effect of these mechanisms through the reduced-form impact of RPS programs on retail electricity prices.
There are two primary points that emerge. First, there is no evidence of a meaningful difference in price trends, either upwards or downwards, among adopting states in the six years preceding RPS program passages. Thus, for example, there doesn’t appear to be any evidence that prior to RPS passage, adopting states were differentially passing unobserved policies that influence electricity prices positively or negatively or facing differential cost shocks.
Second, it is apparent that retail prices increased after program passages, but not all at once; the figure suggests that a model that allows for a trend-break describes the data well. It is striking that the trend in prices appears to very closely shadow the trend in net RPS requirements. The estimates indicate that retail prices are higher in RPS states post-passage, and rise by roughly 0.17 cents each year, with statistically insignificant pre-trends…or approximately $140 per person in additional annual electricity costs.
There are several caveats and implications of the paper’s results that bear noting. First, the analysis is “reduced form” so we cannot assign precise shares of the RPS programs’ full costs to differences in generation costs, intermittency, transmission, and stranded assets. For example, it seems plausible that any stranded asset costs decline over time while intermittency costs increase as net requirements grow.
Third, it is often claimed that policies encouraging renewable energy deployment contribute themselves to reducing the cost of future renewable generation. The coincidence of the global proliferation of policies that support renewable energy (e.g., China’s production subsidies and Germany’s embrace of renewable generation) and the decline in solar and wind prices over the last decade is consistent with the possibility of such spillovers.
A particularly striking finding is that RPS programs meaningfully alter electricity market equilibria. This effect, which has not been possible to comprehensively measure to date, appears to account for the majority of RPS program costs and benefits.
It seems likely that the indirect costs, including intermittency, transmission, and stranded asset payments, account for a substantial fraction of RPS program costs. This finding raises the possibility that indirect costs associated with grid integration could represent the more important barrier to substantially increasing renewables’ share of generation than the direct cost of renewable energy.
Perhaps this paper’s central contribution to projecting the costs and benefits of future policy is to highlight the importance of understanding the indirect effects of renewable energy and the viability of mechanisms to facilitate their grid integration.
Are there effective approaches or policies used in other jurisdictions that could help inform Minnesota’s consideration of utility affordability? If so, what lessons should Minnesota draw from those approaches?
The main problem with the PUC’s rate-of-return (ROR) regulation is that incentivizes a capital base bias. Economists call this the Averch-Johson (A-J) effect from 1962. Professor Kenneth Train of UC Berkeley summarized the A-J effect in 1991 as:
The [ROR] regulated firm will utilize more capital than if it were unregulated. That is, the firm’s output could be produced more cheaply with less capital and more labor. It is possible that the firm will produce less output and charge a higher price than if it were not regulated.
These inefficiencies are the natural result of [ROR] regulation, in that a firm that is attempting to maximize profits is given, by the form of the regulation itself, incentives to be inefficient.
Furthermore, the aspects of monopoly control that [ROR] regulation is intended to address, such as high prices, are not necessarily mitigated, and could be made worse, by [ROR] regulation.
CPI minus X (CPI-X) incentive regulation goes a long way to countering the A-J effect, as it focuses on capital base efficiency and not just operating cost efficiency (i.e. opex) as well as efficient returns of capital (i.e. capex and depreciation) and not just actual and requested returns on capital (i.e. CAPM and WACC).
The Independent Pricing and Regulatory Tribunal (IPART) of New South Wales (NSW) in Australia published in July 2011 “Incentives for cost saving in CPI-X regimes” which stated:
This paper defines the inherent incentives for cost or efficiency savings in the standard CPI-X regime. They arise principally because the regime disconnects prices from cost movements after the price path has been set in relation to forecast costs for the period in question. This delinking of costs from prices creates an incentive for the regulated firm to cut costs early in a determination period because it retains those cost savings until they are passed through to consumers in lower prices.
CPI-X regulation involves setting a price-path (price-cap regulation) for a utility, allowing for changes in inflation (the CPI factor) and expected efficiency improvements (the “X” factor). The “X” factor may incorporate other aspects in addition to the expected improvement in efficiency, such as rewards for improvements in output quality, service levels or demand management actions.
Regulators generally aim to set prices/revenues along a path that will generate sufficient revenue to recover the efficient levels of opex and capex, with a reasonable after-tax return on capital. The price path is set for a specified number of years, commonly referred to as the ‘regulatory control period’ or the ‘determination period’.
As both economics and history have demonstrated time-and-time again, neither ROR regulation nor PUC regulators are as effective and efficient as free-market competition. The best case study for this was Australia’s National Competition Policy (NCP) from 1995 to 2005. I didn’t simply study this from a distance, I was there and involved as a young economist.
In conclusion, in my May 2020 paper for the Heartland Institute of Chicago IL, entitled “How to Fix America’s Crumbling Infrastructure: Lessons from Australia,” I wrote the following:
NCP reforms involved: (1) removing anti-competitive regulation and legislation; (2) separating the non-monopoly from the monopoly parts of large state-owned enterprises (SOEs) and government-sponsored enterprises; (3) providing competitive neutrality for non-monopoly SOEs; (4) formalizing oversight of pricing for monopoly SOEs; (5) creating a third-party access regime for monopoly infrastructure, whether state-owned or not; (6) extending antitrust laws to all SOEs; and (7) having states and territories apply these reforms to local government.
Or, put more simply, NCP was aimed at removing or minimizing the monopoly power of the numerous federal, state, and local government-owned businesses in the economy by injecting competition (in an actual, potential, or proxy sense) and, thus, shifting infrastructure systems away from monopoly and oligopoly towards competition.
NCP was directly aimed at the following infrastructure industries: electricity (federal and state); natural gas (state); roads (federal, state, and local); and water and sewage (federal, state, and local). It was also designed to indirectly target airports (federal and local), ports (state), postal services (federal), rail (federal and state), and telecommunications (federal).
Policymakers implemented NCP through three agreements between the federal, state, and territorial governments. Annual performance-based payments were provided to the states for nine years, some of which the state of Queensland passed on to local governments for five years to incentivize NCP reform at that level. A small “watch-and-guide dog” called the National Competition Council (NCC) was established to assess progress and make payments on an annual basis
Cost-benefit analyses (CBA) were embedded in every aspect of NCP, including in three major assessments of the economic impacts in 1995, 1999, and 2005 by the Productivity Commission (PC). The 2005 PC assessment found there was a massive net-benefit (benefits over costs) from NCP, i.e. competition performance above competition payments. The sowing of AU$5.5 Billion in taxpayer payments during the NCP decade conservatively reaped more than AU$1 Trillion in additional benefits for Australian families, gained “from productivity and price changes observed over the 1990s in the electricity, natural gas, urban water, telecommunications, urban transport, ports and rail freight sectors.” This economic return was of Biblical proportions, well over a hundred-fold.
The key to putting downward pressure on prices from the late-1980s to early-2000s [in terms of the Australian Consumer Price Index (CPI), for the major Australian utility industries of electricity, natural gas, telecommunications, water, and sewage] through greater supply and choice, was NCP and its related micro-economic reforms (pre- and post- NCP).
Australia’s NCP reforms show market competition is the only high-quality regulator. This is true even in industries that have been considered for many decades to be natural monopolies, like infrastructure. Telecommunications continued to embrace these pro-competition reforms in the years that followed, and even more so after 2014. As a result, prices dropped substantially.
But, unfortunately for electricity, gas, and water/sewage customers, a new form of infrastructure socialism took hold in the mid-2000s: government-mandated and subsidized renewable energy sources, which have also significantly impacted US electricity prices, driving up costs in both countries.
Source: NASA, “50 Years Ago: 1968 Welcomed 2001“
