Utilities’ Operating Costs Are Rising: Which Can Control Costs Best and Which Are Most at Risk?

gcopley
Print Friendly, PDF & Email

______________________________________________________________________________

Eric Selmon Hugh Wynne

Office: +1-646-843-7200 Office: +1-917-999-8556

Email: eselmon@ssrllc.com Email: hwynne@ssrllc.com

SEE LAST PAGE OF THIS REPORT FOR IMPORTANT DISCLOSURES

______________________________________________________________________________

August 16, 2018

Utilities’ Operating Costs Are Rising:

Which Can Control Costs Best and Which Are Most at Risk?

In this research report, we compare the U.S publicly traded electric utilities based upon their absolute level of operation and maintenance (O&M) expense per customer and its rate of increase in recent years. We find that utility rankings on O&M expense per customer can be highly persistent over time (see Exhibits 16 and 17), possibly reflecting regional variations in climate, labor costs, and population growth. On the other hand, utility rankings on the rate of change in O&M expense per customer can persist for three to four years but ultimately are impermanent (see Exhibits 14 and 15), suggesting that cost cuts asymptotically approach a limit, and that persistent cost increases eventually trigger a management response.

In recent years, the rate of growth in O&M expense has shown a widening dispersion of outcomes across the electric utilities, with more companies achieving material cost cuts while at others expenses continue to rise (compare Exhibits 10 and 11 to 12 and 13). This trend would argue for a wider dispersion in valuation multiples among utilities than might have been appropriate over the last decade, when pressure on electricity rates was eased across the industry by the decline in fuel and purchased power costs, but the lack of persistence in above or below average growth rates means forecasting the winners or losers is difficult.

For purposes of stock selection, we recommend focusing on the ratio of O&M expense to revenues (see Exhibits 20 and 21). At a utility whose O&M expense per customer is rising, this metric can be used to assess the risk that future cost increases will put significant upward pressure on customer bills. This risk is lowest at utilities with a low ratio of O&M expense to revenues, and highest at those with a high ratio. Alternatively, at a utility that has announced a program of cost cuts, this metric can be used to assess their potential impact, and particularly whether future cost cuts might allow a utility to recover the expenses associated with rate base growth (depreciation and return on capital) without commensurate increases in customer bills. The opportunity to create headroom for rate base growth is greatest at utilities with high ratios of O&M expense to revenues.

  • Annual O&M expense per customer, excluding fuel and purchased power, varies widely among the vertically integrated electric utilities, ranging from $440 per customer at NEE to $1,730 at ALEThe five lowest cost utilities, in ascending order, are NEE, CMS, NWE, D and PNM. The five highest cost utilities, in descending order, are ALE, WEC, EVRG, LNT and XEL (Exhibit 2).
  • Utility rankings on O&M expense per customer can be highly persistent over time (Exhibits 16 and 17), possibly reflecting regional variations in climate, labor costs, and population growth. As O&M expense per customer tends not to converge to the industry mean, this metric may not be helpful in stock selection – if a utility’s ranking in the industry is unlikely to change, neither should its relative valuation.
  • In recent years, however, the rate of growth in O&M expense has shown a widening dispersion of outcomes across the electric utilities, with more companies achieving material cost cuts while at others expenses continue to rise (compare Exhibits 10 & 11 to 12 & 13). This trend would argue for a wider dispersion in valuation multiples among utilities than might have been appropriate over the last decade, when the pressure on rates was eased across the industry by the marked decline in fuel and purchased power costs. The vertically integrated utilities that over the last five years have achieved outright cost cuts in O&M expense per customer are AVA, HE, PNW, DTE, EE, D, PNM and DUK. Those at which O&M expense has risen most rapidly are NWE, EDE, LNT, POR, AEP.
  • Outperformance or underperformance with respect to the rate of increase in O&M expense per customer tends to be relatively short-lived, however. Utilities may occupy a top or bottom tier ranking on this metric for three to four years, but ultimately these ranking are impermanent (see Exhibits 14 and 15), suggesting that cost cuts asymptotically approach a limit, and that persistent cost increases eventually trigger a management response. Investors must be cautious, as utilities with a recent track record of cost cuts may soon approach an inflection point in this trend.
  • For purposes of stock selection, we recommend focusing on the ratio of O&M expense to revenues.
  • At a utility whose O&M expense per customer is rising, this metric can be used to assess the risk that future cost increases will put significant upward pressure on customer bills. The risk is lowest at utilities with low ratios of O&M expense to revenues, and highest at those with high ratios. (See Exhibits 20 and 21).
  • At a utility that has announced a program of cost cuts, this metric can be used to assess their potential impact, and particularly whether future cost cuts might allow a utility to recover the expenses associated with rate base growth (depreciation and return on capital) without commensurate increases in customer bills. The opportunity to create headroom for rate base growth is greatest at utilities with high ratios of O&M expense to revenues.
  • The five vertically integrated utilities with the lowest ratio of O&M expense to revenues, and thus the lowest risk that future increases in O&M expense will drive untenable increases in customer bills are, in ascending order, D, HE, SCG, CMS, and SO.
  • The five vertically integrated utilities with the highest ratio O&M expense to revenues, and thus the highest risk that future increases in O&M expense will drive untenable increases in customer bills, are, in descending order, WEC, ALE, AVA, XEL, PNM. Conversely, these utilities also have the greatest opportunity to cut costs and create headroom for rate base growth.
  • Particularly well positioned to manage future increases in O&M expense are four utilities that combine (i) low O&M expense relative to total revenues with (ii) cuts in O&M expense per customer over the last five years: D, DUK, HE and PNW. HE and PNW in particular have materially reduced their O&M expense since 2012. (See Exhibits 12 and 20).
  • Potentially well positioned are the vertically integrated utilities that combine (i) high O&M expense relative to total revenues with (ii) cuts in O&M expense per customer over the last five years: AVA, EE, and PNM. By cutting an expense equivalent to such a large share of revenues, these utilities have created headroom for future cost increases — including those associated with rate base growth, such as depreciation expense and the return on capital. (See Exhibits 12 and 20).
  • At high risk are the utilities that combine (i) high O&M expense relative to total revenues with (ii) rapid increases in O&M expense per customer over the last five years: ALE, EVRG, LNT and POR. The upward pressure these cost increases have placed on the revenue requirement of these four utilities diminishes their ability to push through the rate increases required to sustain rate base growth.
  • We have conducted a similar analysis of the transmission and distribution utilities (see Exhibits 13 and 21). For this purpose, we have calculated these companies’ O&M expense as a ratio of the total payments for electricity made by these companies’ customers, including payments to the T&D companies for the delivery of electricity as well as payments to the T&D companies or third-party suppliers for the electricity itself. Grouping these utilities into the three categories set out above, we find that:
  • PEG combines (i) a low ratio of O&M expense to total customer payments for electricity (14%) with (ii) low rates of growth in O&M expense over the last five years (0.9% p.a.);
  • ES combines (i) a high ratio O&M expense to total customer payments for electricity (22%) with (ii) low rates of growth in O&M expense over the last five years (1.2% p.a.); and
  • AGR and PCG combine (i) a high ratio of O&M expense to total customer payments for electricity (32% and 21%, respectively) with (ii) high rates of growth in O&M expense over the last five years (4.1% and 3.7% p.a., respectively).

Background:

  • From 2006 through 2016, U.S. investor owned utilities doubled electric plant in service, increasing the industry’s regulatory rate base by over 80%. By contrast, utilities limited the increase in their aggregate revenue to just 10% over the decade, as the cost of fuel and purchased power dropped by 30%. (See Exhibit 5 and our research report of July 31stThe Challenge of Limiting Rate Increases in the Face of Rate Base Growth and Rising Costs: An Analysis of Capex, Opex and Their Impact on Customer Costs.)
  • On a per customer basis, electric utility revenues increased by only 0.5% p.a. over 2006-2016, well below the rate of consumer price inflation, which averaged 1.8% p.a. over the period. The industry’s outlook for the next five years is less benign; revenue per customer of the investor-owned utilities, we estimate, must expand at ~2.3% p.a., or over 4x the rate of 2006-2016.
  • Aware that industry costs will follow a less benign trajectory in the coming years than they have over the prior decade, utility management teams have increasingly focused on the opportunity to create headroom for rate base growth not by raising revenue but by carving out costs. The long-term track record of the industry in controlling non-fuel O&M expense is a poor one, however, suggesting that the objective the industry has set for itself may be a difficult one to achieve.
  • Increases in O&M expense per customer at the publicly traded, U.S. electric utilities have significantly outpaced inflation over the last 15 years. Over the period 2002-2017, vertically integrated utilities and transmission and distribution companies have registered increases in O&M expense per customer of 2.8% and 2.9% p.a., respectively, as compared with a 2.0% average annual increase in the consumer price index over this period (see Exhibit 6).
  • Over the last five years, however, the rate of increase in O&M expense per customer among U.S. electric utilities has changed in three important ways. (Compare Exhibits 10 and 11 to 12 and 13).
  • Among the vertically integrated utilities, the rate of increase in O&M expense per customer, which exceeded the rate of inflation by 80 b.p. p.a. over the last ten years, lagged the rate of inflation by 10 b.p. p.a. over the last five.
  • Over the last five years, eight of the 27 vertically integrated utilities have achieved outright reductions in O&M expense per customer, as well as two of the 15 T&D companies.
  • The gap between the best and worst utilities, assessed on their average rate of increase in O&M expense per customer, has increased markedly. Among the vertically integrated utilities, this gap increased from 640 basis points over the last ten years to 1190 basis points over the last five; among the T&D companies, it increased from 820 basis points to 1120.
  • That a substantial subset of the industry has been successful in cutting O&M costs per customer on a sustained basis has important implications:
    • To the extent that certain management teams have been successful in carving out headroom for growth in plant in service by reducing the portion of their revenue requirement that must go to cover O&M expense, these companies now have a material advantage over the others in sustaining the growth of rate base and regulated earnings.
    • Over time, moreover, the ability of less efficient companies to pass through their cost increases will be limited by regulators benchmarking their performance against the industry’s leaders.
  • These trends would argue for a wider dispersion in valuation multiples across the best and worst performers in the industry than might have been appropriate over the last decade, when the pressure on electricity rates was eased across the industry by the marked decline in fuel and purchased power expense.
  • While recognizing the increased importance of cost management in sustaining rate base growth, overall we remain positive on the ability of the U.S. electric utilities to achieve continued growth in rate base without overly burdening their customers with rate increases. We expect the publicly traded U.S. utilities to increase net electric plant in service at a 5.9% compound annual rate over the next five years — commensurate, we calculate, with average annual increases in retail electricity revenues per customer of ~2.3% p.a., or slightly above inflation expectations of ~2.0% p.a. over the next five years (as implied by the difference in yield between five-year U.S. Treasury notes and five-year TIPS).
  • Within the industry, our preferred stocks are listed in Exhibit 1. Assessing these stocks on the framework presented above, we note that:
    • Among the vertically integrated utilities that we favor, ETR is clearly the best positioned, with low O&M expense to revenues (27% vs. a sector average of 30%; see Exhibit 20) and expense growth close to the industry average (1.4% p.a. over the last five years vs. a sector average of 1.2%; see Exhibit 3). NEE’s low ratio of O&M expense to revenues (also 27%) is an advantage, but one that is being eroded by well-above average O&M expense growth (3.4% p.a.). AEP faces the toughest challenge: not only are its O&M expenses high as a percentage of revenues (34%), but it has experienced rapid increases in O&M expense per customer over the last five years (4.1% p.a.).
    • Among the vertically integrated utilities we do not favor, SO is the best positioned, with low O&M expense to revenues (26% vs. a sector average of 30%) and expense growth close to the industry average (1.6% p.a. vs. a sector average of 1.2%). ALE and POR face the highest risk, with a high ratio of O&M expense to revenues (40% and 37%, respectively) and above average growth in O&M expense per customer over the last five years (2.6% and 4.7% pa., respectively). We note, however, that this could provide more opportunities for a committed management to reduce O&M going forward.
  • Among the transmission and distribution utilities that we favor, EIX may be the best positioned: while beginning with a high ratio of O&M expense to revenues (21% vs. a sector average of 17%; see Exhibit 21), its success in reducing these expenses may be creating headroom for future rate base growth. EIX has aggressively cut its O&M expenses per customer over the last five years, reducing its O&M expense at an average annual rate of 2.2% p.a. vs. a sector average that has increased at a 2.4% average annual rate over the same period (see Exhibit 13). FE is also well positioned, with a relatively low ratio of O&M to revenues (15%), but its O&M expenses per customer have grown rapidly over the last five years (5.1% p.a.). EXC’s performance is almost indistinguishable from the sector mean. PCG faces the greatest risk that its rates may come under pressure from rising O&M expense: not only does it have a high ratio of O&M expense to revenues (21%), but its O&M expenses per customer have been rising rapidly over the last five years (3.7% p.a.). While we continue to find PCG attractive at its current valuation, we see O&M expense growth compounding the risk the company faces from wildfire liabilities.

Exhibit 1: Heat Map: Preferences Among Utilities, IPP and Clean Technology

Source: SSR analysis

Introduction

In this research report, we compare the U.S publicly traded electric utilities based upon their absolute level of operation and maintenance (O&M) expense per customer (Exhibit 2) and its rate of increase in recent years (Exhibit 3). We identify those utilities whose high and rapidly rising O&M costs may absorb the revenue increases allowed by regulators and render difficult the recovery of expenses, such as depreciation and return on capital, required for rate base growth. We also identify utilities whose success in cutting O&M expense has created headroom for the cost increases associated with rate base growth. Finally, we consider the usefulness of these criteria as stock selection tools.

Exhibit 2: Ranking of Vertically Integrated Utilities on Electric O&M Costs per Customer (Excluding Fuel and Purchased Power), 2017 (1)

____________________________

1. O&M costs have been calculated on a three-year trailing average basis to smooth the impact of one-off events.

Source: FERC Form 1, SNL, SSR analysis

Exhibit 3: Five-Year CAGR in Electric O&M Costs (Excluding Fuel and Purchased Power), Vertically Integrated Utilities, 2012-2017 (1)

____________________________

1. O&M costs have been calculated on a three-year trailing average basis to smooth the impact of one-off events.

Source: FERC Form 1, SNL, SSR analysis

Background

Exhibit 4 presents our estimates of the contribution of the different categories of electric utility costs to the revenue requirement of the industry. The left-hand chart presents a breakdown of electric utility costs based upon the 2017 FERC Form 1 filings of U.S. investor owned utilities, while the right-hand chart adjusts this breakdown to reflect the lower corporate tax rate prevailing in 2018. The cut in the corporate tax rate from 35% to 21% is reflected in a corresponding reduction in the pre-tax return on equity allowed electric utilities by their regulators, reducing the aggregate revenue requirement of investor-owned utilities and slightly decreasing the weight of utilities’ return on invested capital in total industry costs. Nonetheless, costs associated with the return of and on utilities’ invested capital (depreciation expense and the pre-tax return on capital) will continue to comprise an estimated 45% of the of the total revenue requirement of U.S. investor owned utilities. The other two principal categories of costs are fuel and purchased power, accounting for 30% of utilities’ revenue requirement, and non-fuel O&M, accounting for 28%. A final contributor to the revenue requirement of the industry, accounting for 7% of utilities’ revenue requirement, is the recovery of net regulatory assets.

Exhibit 4: Approximate Breakdown of the Aggregate Revenue Requirement of the U.S. Investor Owned Electric Utilities, by Category of Cost

2017 2018 Est.

 _________________________________

Source: FERC Form 1, SNL, SSR analysis

From 2006 through 2016, U.S. investor owned utilities doubled electric plant in service, increasing the industry’s regulatory rate base by over 80%. By contrast, utilities limited the increase in their aggregate revenue to just 10% over the decade, as the cost of fuel and purchased power dropped by 30%. On a per customer basis, the aggregate electric revenue of investor-owned utilities increased by only 0.5% p.a., well below the rate of consumer price inflation, which averaged 1.8% p.a. over the period. (See Exhibit 5 and our research report of July 31stThe Challenge of Limiting Rate Increases in the Face of Rate Base Growth and Rising Costs: An Analysis of Capex, Opex and Their Impact on Customer Costs.)

Exhibit 5: Growth in Net Electric Plant in Service and O&M Costs vs. Growth in Retail Electricity Sales and Retail Electricity Customers (2006 = 100)

_________________________________

Source: FERC Form 1, SNL, SSR analysis

The industry’s outlook for the next five years is less benign. The revenue per customer of the investor-owned utilities, we estimate, will need to expand at ~2.3% p.a., over 4x the 0.5% annual increase seen over 2006-2016. Our estimate reflects the likely trajectory of key cost components of the industry’s revenue requirement:

  • Plant in service. Based on the disclosed capital expenditure plans of publicly traded electric utilities, we expect net electric plant in service to grow at a compound annual rate of ~5.9% over the next five years.
  • Non-fuel O&M expense. Over 2006-2016, non-fuel O&M expense at the investor owned utilities increased at 3.0% p.a., or ~40% of the rate of growth in net plant in service (7.2% p.a.). Were this ratio to persist, 5.9% forecast annual growth in net plant in service would imply growth in non-fuel O&M expense of ~2.4% p.a. going forward.
  • Fuel & purchased power. Forward price curves suggest that the prices of natural gas, coal and wholesale power will be broadly similar in five years’ time to what they are today; if so, utilities’ aggregate cost of fuel and purchased power would also remain broadly unchanged.
  • Other costs. A final contributor to the revenue requirement of the industry is the recovery of net regulatory assets. We assume that this will remain roughly constant over the next five years.

Weighting the expected growth of net electric plant in service (5.9% p.a.) by its contribution to the aggregate revenue requirement of the industry in 2018 (35%, including both depreciation and return on invested capital), would suggest a need for an ~2.1% annual increase in electricity revenues. Similarly, weighting the expected increase in non-fuel O&M expense (2.4% p.a.) by its contribution to revenues (28%) would suggest a need for a further ~0.7% annual revenue increase. Summed together, and assuming other cost categories remain roughly flat, we expect the electric utility industry to require ~2.8% average annual revenue growth to cover expected cost increases over the next five years.

Given the ~0.5% average annual rate of growth in retail electricity customers, utilities would need to secure from their regulators average annual increases in retail electricity revenues per customer of ~2.3% p.a., or slightly above inflation expectations of ~2.0% p.a. over the next five years (as implied by the difference in yield between five-year U.S. Treasury notes and five-year TIPS).

For the industry as a whole, this is not a daunting prospect, and we remain confident that U.S. electric utilities can secure the revenue increases required to ensure the return of and on their planned investments.

Nonetheless, such a trajectory in the required revenue of the electric utility industry over the next five years would force utilities to secure from their regulators increases in electric revenue per customer at over 4x the rate they did over 2006-2016 (~2.3% p.a. vs. 0.5% annual increases historically). Various factors, moreover, could drive industry costs higher and contribute to an acceleration of required revenue increases. Long term interest rates are expected to rise as the Federal Reserve raises short term interest rates and the supply of Treasury bonds is increased by the Fed’s unwinding of its bond portfolio and the federal government’s higher borrowing requirement following the 2017 tax cut. Higher long-term interest rates could require utilities to seek revenue relief for rising borrowing costs and increases in their allowed ROEs. Labor costs, which were held in check by persistently high unemployment in the years following the 2008-09 recession, are now coming under upward pressure. Finally, the cost of two key inputs, steel and aluminum, have increased by ~30% and 10%, respectively, as a result of recently imposed tariffs.

Aware that industry costs are likely to follow a less benign trajectory in the coming years than they have over the prior decade, utility management teams have increasingly focused on the opportunity to create “headroom” for rate base growth not by raising revenue but by carving out costs.

The long-term track record of the industry in controlling non-fuel O&M costs is a poor one, however, suggesting that the objective the industry has set for itself could be a difficult one to achieve. Increases in O&M expense per customer at the publicly traded, U.S. electric utilities have significantly outpaced inflation over the last 15 years. Over the period 2002-2017, vertically integrated utilities and transmission and distribution companies have registered increases in O&M expense per customer of 2.8% and 2.9% p.a., respectively, as compared with a 2.0% average annual increase in the consumer price index over this period (see Exhibit 6).[1]

Exhibit 6: Electric Utilities’ Operation & Maintenance Expense per Customer (Excluding Purchased Power & All Generation-Related Costs) vs. Consumer Price Index, 2002-2017 (1)

 

____________________________

1. O&M costs have been calculated on a three-year trailing average basis to smooth the impact of one-off events.

Source: FERC Form 1, SNL, SSR analysis

We believe it likely, therefore, that only a minority of U.S. electric utilities will be successful in achieving outright reductions in O&M expense. One objective of this research report is to assess which companies these are likely to be. A second is to identify those companies which, because of their relatively low levels of O&M expense or moderate pace of past increases, are well positioned to pass through increases in O&M expense to their ratepayers in the years ahead. Finally, we will identify those companies which, based on their current high level of O&M expense or its rapid rate pace of increases, may be most at risk of regulatory pushback when seeking rate relief for expense increases in future – including for the expenses associated with rate base growth (depreciation and return on capital).

The rapid pace of growth in O&M expense per customer among U.S. electric utilities is in large degree a function of underlying increases in net plant in service. In our research report of July 31stThe Challenge of Limiting Rate Increases in the Face of Rate Base Growth and Rising Costs: An Analysis of Capex, Opex and Their Impact on Customer Costs, we presented the results of several regression analyses of (i) electric operation and maintenance expense per customer (excluding fuel and purchased power costs) against (ii) net electric plant in service, across the universe of investor-owned, U.S. electric utilities. We measured both variables in constant 2017 dollars. As can be seen in Exhibit 7, among U.S. vertically integrated utilities, a regression of O&M expense against net plant in service results in an r-squared of 42% and high and statistically significant t-statistics for both the coefficient and intercept of the regression equation (4.1 and -4.2, respectively).

Similarly, a regression of (i) the ten-year CAGR in electric operation and maintenance expense per customer (excluding fuel and purchased power costs) against (ii) the ten-year CAGR in net electric plant in service, both measured in constant 2017 dollars, results in an r-squared of 32%, again with high and statistically significant t-statistics for both the coefficient and intercept (3.4 and -3.5, respectively). This equation suggests that every 1.0% increase in net plant in service has historically been associated with an ~0.3% increase in O&M expense. (See Exhibit 8).

Exhibit 7: Vertically Integrated Utilities: Exhibit 8: Vertically Integrated Utilities:

10-Year Change in Net Electric Plant in 10-Year CAGR in Net Electric Plant in Service vs. 10-Year Change in Electric Service vs. 10-Year CAGR in Electric O&M

O&M Costs, Excluding Production Costs, Excluding Production

(Constant 2017$ per Customer) (1) (Constant 2017$ per Customer) (1)

 

____________________________________

1. Both net electric plant in service and non-fuel electric O&M cost are calculated on a per customer basis and expressed in constant 2017 dollars using industry-specific PPI indices to eliminate the impact of inflation and customer growth. Exhibits excludes the outliers ALE and HE.

The relationship between growth in O&M expense and net plant in service has been changing over time, most notably over the last five years. As illustrated in Exhibit 9, the pace of growth in net plant in service has accelerated in recent years among both the transmission and distribution and vertically integrated utilities. However, at the vertically integrated utilities, even as growth in net plant in service per customer has accelerated, the pace of increase in O&M expense per customer has decelerated. Over the last five and ten years, the same holds true at the transmission and distribution utilities. Put another way, the ratio of (i) the average annual rate of increase in O&M expense per customer to (ii) the rate of increase in net plant per customer, has been declining over time. In the next section, we will seek to identify the leaders of this trend.

Exhibit 9: Growth in Utilities’ O&M Costs per Customer (Excluding Purchased Power and All Generation-Related Costs) vs. Growth in T&D Plant in Service per Customer, 2002-20171

____________________________

1. O&M costs have been calculated on a three-year trailing average basis to smooth the impact of one-off events.

Source: FERC Form 1, SNL, SSR analysis

A Note on Methodology

Before discussing the results of our analysis, it is appropriate quickly to review our methodology and its implications for the interpretation of our results.

Because a principal objective of our analysis is to assess the impact of increases in operation and maintenance expense on customer bills, we use operation and maintenance expense per customer as our primary metric. We exclude from operation and maintenance expense, however, the cost of fuel and purchased power. These costs are unpredictable, except by reference to forward price curves –which suggest that the prices of natural gas, coal and wholesale power will be broadly flat over the next five years. Excluding these costs from our analysis free us from the vagaries of commodity price movements, facilitates comparisons across utilities with differing generation fleets, and focuses on those costs over which utility managements have the most control.

To smooth the impact of one-off events such as storms or wildfires, we have calculated each utility’s O&M costs on a trailing three-year average basis. Thus, to calculate the compound annual rate of growth in O&M costs over the period 2007-2017, we compared the average of O&M costs per customer for the three years ending in 2017 with the average of O&M costs per customer for the three years ending in 2007.

Because vertically integrated utilities incur O&M expense related to their generating fleets that transmission and distribution companies do not, we have analyzed these two classes of utilities separately. In cases where a multi-state utility holding companies owns both vertically integrated utilities as well as transmission and distribution (T&D) utilities (e.g., AEP), we have compared the performance of the vertically integrated subsidiaries to other vertically integrated utilities and the transmission and distribution subsidiaries to other T&D companies.

Finally, our analysis is focused on the operating utility subsidiaries of publicly traded, U.S. electric utilities and relies for data upon these companies’ Form 1 filings with the Federal Energy Regulatory Commission.

Company Comparison & Analysis

In this section, we compare the U.S publicly traded electric utilities based upon their absolute level of operation and maintenance (O&M) expense per customer and its rate of increase in recent years. We identify those utilities whose high and rapidly rising O&M costs may absorb the revenue increases allowed by regulators and render difficult the recovery of expenses, such as depreciation and return on capital, required for rate base growth. We also identify utilities whose success in cutting O&M expense has created headroom for the cost increases associated with rate base growth. Finally, we consider the usefulness of these criteria as stock selection tools.

The two charts below present the compound annual rate of increase in O&M expense (excluding fuel and purchased power) for the vertically integrated utilities (Exhibit 10) and transmission and distribution companies (Exhibit 11). Both charts show how the average rate of increase in O&M expense per customer (2.5% p.a. at the vertically integrated utilities over the last ten years, 3.5% p.a. at the transmission and distribution companies) has significantly exceeded consumer price inflation over this period (1.7% p.a.). Note that none of the 27 vertically integrated utilities achieved outright reductions in O&M expense per customer, and only one of the 15 transmission and distribution utilities did so.

Exhibit 10: Ten-Year CAGR in Electric O&M Costs (Excluding Fuel and Purchased Power), Vertically Integrated Utilities, 2007-2017 (1)

____________________________

1. O&M costs have been calculated on a three-year trailing average basis to smooth the impact of one-off events.

Source: FERC Form 1, SNL, SSR analysis

Exhibit 11: Ten-Year CAGR in Electric O&M Costs (Excluding Purchased Power and All Generation-Related Costs), Transmission & Distribution Utilities, 2007-2017 (1)

____________________________

1. O&M costs have been calculated on a three-year trailing average basis to smooth the impact of one-off events.

Source: FERC Form 1, SNL, SSR analysis

The two charts also illustrate the extremely wide range of variation in utilities’ performance. Among the vertically integrated utilities, the range between the highest and lowest rates of increase in O&M costs per customer is 640 basis points per annum over the decade from 2007 through 2017. Among T&D companies, the gap is even larger, at 820 basis points.

Exhibits 12 and 13 present the rate of increase in O&M expense at the vertically integrated utilities and T&D companies, respectively, over the last five years. These charts are noticeably different from the ten-year comparisons above in three important ways. First, real increases in O&M expense per customer have been much lower over the last five years than over the last ten. In the case of the vertically integrated utilities, the gap between the average rate of increase in O&M expense per customer and the rate of consumer price inflation was 80 basis points p.a. over the last ten years; by contrast, over the last five years, the rate of increase in O&M expense per customer among this group lagged the rate of inflation by ten basis points p.a. Among the T&D companies, the rate of increase in O&M expense per customer exceeded the rate of inflation by 180 basis points p.a. over the last ten years, but this gap narrowed to 110 basis points p.a. over the last five.

Second, the gap between the best and worst performers is markedly wider over the last five years than over the last ten, having increased from 640 basis points to 1190 basis points among the vertically integrated utilities and from 820 to 1120 among the T&D companies.

Finally, we attribute both of the above trends to a marked increase in the number of companies that have succeeded in achieving outright reductions in O&M expense per customer; over the last five years, eight of the 27 vertically integrated utilities, and two of the 15 T&D companies achieved this.

Among the vertically integrated utilities, those that have succeeded in cutting costs per customer are AVA (-4.6% p.a.), HE (-2.7% p.a.), PNW (-1.3% p.a.) and DTE (-0.8% p.a.), followed by EE, D, PNM and DUK (see Exhibit 12). Among the transmission and distribution utilities, Duke Energy Ohio and Edison International achieved cuts in O&M expense per customer of 5.4% and 2.2% p.a., respectively (see Exhibit 13).

That a substantial subset of the industry has been successful in cutting O&M costs per customer on a sustained basis has important implications. To the extent that certain management teams have been successful in carving out headroom for growth in plant in service, by reducing the portion of their revenue requirement that must go to cover O&M expense, these companies now have a material advantage over the others in sustaining the growth of rate base and regulated earnings. Over time, moreover, it suggests that the ability of less efficient companies to pass through their cost increases will be limited by regulators benchmarking their performance against the leaders in the industry. These trends would argue for a wider dispersion in valuation multiples across the best and worst performers in the industry than might have been appropriate over the last decade, when the pressure on electricity rates was eased across the industry by the marked decline in fuel and purchased power expense.

Exhibit 12: Five-Year CAGR in Electric O&M Costs (Excluding Fuel and Purchased Power), Vertically Integrated Utilities, 2012-2017 (1)

____________________________

1. O&M costs have been calculated on a three-year trailing average basis to smooth the impact of one-off events.

Source: FERC Form 1, SNL, SSR analysis

Exhibit 13: Five-Year CAGR in Electric O&M Costs (Excluding Purchased Power and All Generation-Related Costs), Transmission & Distribution Utilities, 2012-2017 (1)

____________________________

1. O&M costs have been calculated on a three-year trailing average basis to smooth the impact of one-off events.

Source: FERC Form 1, SNL, SSR analysis

How durable are these distinctions likely to prove, however? In Exhibits 14 and 15 we provide a quintile ranking of the vertically integrated utilities and T&D companies, respectively, based on the annual change in their O&M costs per customer. As can be seen there, it is unusual for companies to sustain top or bottom quintile rankings for long periods of time. With the occasional exception (e.g., EVRG), most companies seem to cycle through the quintile rankings, suggesting that continuous reductions in O&M expense are very difficult to sustain over long periods of time and that continuous increases are ultimately countered by management action.

However, over periods of three or four years, it is not uncommon to see utilities maintain consistently strong or poor performance in the control of O&M costs. Focusing on the last five years, for example, consistent underperformers have included AEP, EDE, LNT, NWE and POR, among the vertically integrated utilities, and AEE’s Illinois subsidiaries and FE among the T&D companies. Outperforming over the last five years have been AVA, DTE, DUK, EE, HE, and PNW, among the vertically integrated utilities, and Duke Energy Ohio, ED, EIX and ES among the T&D companies.

It is clear, however, that investors must be cautious in attempting to capitalize on these trends, as utilities with a recent track record of cost cuts may soon approach an inflection point and drop in comparative ranking.

Exhibit 14: Quintile Ranking of Vertically Integrated Utilities on Annual % Change in O&M Costs per Customer (Excl. Fuel and Purchased Power), 2002-20171

____________________________

1. O&M costs have been calculated on a three-year trailing average basis to smooth the impact of one-off events.

Source: FERC Form 1, SNL, SSR analysis

Exhibit 15: Quintile Ranking of Transmission & Distribution Utilities on Annual % Change in O&M Costs per Customer (Excl. Purchased Power and All Generation-Related Costs) (1)

____________________________

1. O&M costs have been calculated on a three-year trailing average basis to smooth the impact of one-off events.

Source: FERC Form 1, SNL, SSR analysis

Exhibits 16 and 17 rank the utilities on a different basis, sorting them into quintiles not on the rate of increase in O&M costs per customer but on the absolute level of these costs. While it is exceptional in the industry to sustain a very high or very low rate of increase in O&M expense per customer over a long period of time, it is not at all uncommon for utilities to maintain very stable quintile rankings when sorted on the basis of absolute O&M expense. Among the vertically integrated utilities, companies such as ALE, AVA, EVRG, HE, SO and WEC, rank in the lowest quintiles on O&M expense per customer (signifying highest cost) and seem never to improve. Others, such as CMS, D, NEE, NWE, OGE and POR, are consistently among the lowest cost. Among the transmission and distribution utilities, AEP’s Ohio and Texas subsidiaries, AGR, and ES are persistently among the poorest performers on O&M expense per customer, while AEE’s Illinois subsidiaries, FE and PEG among are among the best.

Exhibit 16: Quintile Ranking of Vertically Integrated Utilities on O&M Costs per Customer (Excluding Fuel and Purchased Power), 2002-2017 (1)

____________________________

1. O&M costs have been calculated on a three-year trailing average basis to smooth the impact of one-off events.

Source: FERC Form 1, SNL, SSR analysis

Exhibit 17: Quintile Ranking of Transmission and Distribution Utilities on O&M Costs per Customer (Excluding Purchased Power and All Generation-Related Costs), 2002-2017 (1)

____________________________

1. O&M costs have been calculated on a three-year trailing average basis to smooth the impact of one-off events.

Source: FERC Form 1, SNL, SSR analysis

Importantly, if O&M expense per customer tends not to converge to the industry mean, this metric may not be helpful in stock selection: if a utility’s ranking in the industry is unlikely to change, neither is its relative PE.

A closer look at the absolute level of O&M expense at each of the electric utilities suggests that there may be structural factors that tend to preserve their relative rankings versus their peers. Exhibit 18 presents the average O&M expense per customer of the vertically integrated utilities over the three-year period 2015-2017. Note that of the twelve companies with O&M expense per customer below the industry average, nine have service territories in the southern or Pacific regions of the country: NEE, D, SCG, ETR, and DUK in the southeast, PNM, EE, PNW in the southwest, and POR on the Pacific coast. By contrast, of the 14 companies whose O&M expense per customer exceeds the industry average, eleven are located in the Midwest or Rocky Mountain regions. It is possible, therefore, that the relative level of O&M costs is driven by geography and associated demographic trends. It may be cheaper to maintain transmission and distribution networks in the southeast and southwest than in the northern states, reflecting both less severe winter weather as well as lower average labor costs. In addition, the rate of population growth in the southeast and southwest has been higher than the national average over the last several decades, implying a more recent build-out of electricity grids and thus a lower average age of plant in service. Finally, the southern regions of the United States still enjoy more rapid current customer growth than the Midwest, allowing fixed O&M costs such as SG&A to be spread among a growing group of customers. To the extent these factors are expected to remain constant over the next five years, the relative ranking of these utilities may also remain stable.

Exhibit 19 presents the average O&M expense per customer of the transmission and distribution utilities over the three-year period 2015-2017. The geographic distribution of the transmission and distribution utilities is less diverse, reflecting their origins in the deregulation of generation in California, Texas, the Midwest and Northeast. With only one company’s service territory in the southern region of the country (CNP), we find northern utilities at both the top and bottom of our sector rankings. Given the lack of regional diversity, the dispersion of O&M expense per customer among these utilities must be attributable primarily to management decisions; note the dispersion of costs among the Californian utilities (EIX, SRE, PCG), those in Ohio (Duke Energy Ohio and AEP Ohio), as well as the utilities in the mid-Atlantic and northeastern states (PEG, PPL, EXC below the average and ED, ES, AGR above).

Exhibit 18: Ranking of Vertically Integrated Utilities on Electric O&M Costs per Customer (Excluding Fuel and Purchased Power), 2017 (1)

____________________________

1. O&M costs have been calculated on a three-year trailing average basis to smooth the impact of one-off events.

Source: FERC Form 1, SNL, SSR analysis

Exhibit 19: Ranking of Transmission and Distribution Utilities on Electric O&M Costs per Customer (Excluding Purchased Power and All Generation-Related Costs), 2017 (1)

____________________________

1. O&M costs have been calculated on a three-year trailing average basis to smooth the impact of one-off events.

Source: FERC Form 1, SNL, SSR analysis

Investment Conclusion

In this research report, we have compared the U.S publicly traded electric utilities based upon their absolute level of operation and maintenance (O&M) expense per customer and its rate of increase in recent years.

We find that utility rankings on O&M expense per customer can be highly persistent over time (see Exhibits 16 and 17), possibly reflecting regional variations in climate, labor costs, and population growth. As O&M expense per customer tends not to converge to the industry mean, this metric may not be helpful in stock selection: if a utility’s ranking in the industry is unlikely to change, neither is its relative PE.

On the other hand, utility rankings on the rate of change in O&M expense per customer can persist for three to four years but ultimately are impermanent (see Exhibits 14 and 15), suggesting that cost cuts asymptotically approach a limit, and that persistent cost increases eventually trigger a management response. It is clear, therefore, that investors must be cautious in attempting to capitalize on these trends, as utilities with a recent track record of cost cuts may soon approach an inflection point and drop in the comparative ranking.

For purposes of stock selection, we recommend focusing on the ratio of O&M expense to revenues (see Exhibits 20 and 21).

  • At a utility whose O&M expense per customer is rising, this metric can be used to assess the risk that future cost increases will put significant upward pressure on customer bills. The risk is lowest at utilities with low ratios of O&M expense to revenues, and highest at those with high ratios.
  • At a utility that has announced a program of cost cuts, this metric can be used to assess their potential impact, and particularly whether future cost cuts might allow a utility to recover the expenses associated with rate base growth (depreciation and return on capital) without commensurate increases in customer bills. The opportunity to create headroom for rate base growth is greatest at utilities with high ratios of O&M expense to revenues.
  • The five vertically integrated utilities with the lowest ratio of O&M expense to revenues, and thus the lowest risk that future increases in O&M expense will drive untenable increases in customer bills are, in ascending order, D, HE, SCG, CMS, and SO.
  • The five vertically integrated utilities with the highest ratio O&M expense to revenues, and thus the highest risk that future increases in O&M expense will drive untenable increases in customer bills are, in descending order WEC, ALE, AVA, XEL, PNM. Conversely, these utilities also have the greatest opportunity to cut costs and create headroom for rate base growth.

Exhibit 20: Ranking of Vertically Integrated Utilities on Electric O&M Expense (Excluding Fuel and Purchased Power) as a Percentage of Total Electric Revenues, 2017

____________________________

Source: FERC Form 1, SNL, SSR analysis

Exhibit 21: Ranking of Transmission & Distribution Utilities on Electric O&M Expense (Excluding Fuel and Purchased Power) as a Percentage of Total Customer Payments for Electricity, 2017 (1)

____________________________

1. For the transmission and distribution utilities, we have included customer payments for power purchased from third parties in total revenues.

Source: FERC Form 1, SNL, SSR analysis

Particularly well positioned to manage future increases in O&M expense are four utilities that combine (i) low O&M expense relative to total revenues with (ii) cuts in O&M expense per customer over the last five years: D, DUK, HE and PNW. HE and PNW in particular have materially reduced their O&M expense since 2012. (See Exhibits 11 and 19).

Potentially well positioned are the vertically integrated utilities that combine (i) high O&M expense relative to total revenues with (ii) cuts in O&M expense per customer over the last five years: AVA, EE, and PNM. By cutting an expense equivalent to such a large share of revenues, these utilities have created headroom for future cost increases — including those associated with rate base growth, such as depreciation expense and the return on capital. (See Exhibits 11 and 19).

At high risk are the utilities that combine (i) high O&M expense relative to total revenues with (ii) rapid increases in O&M expense per customer over the last five years: ALE, EVRG, LNT and POR. The upward pressure these cost increases have placed on the revenue requirement of these four utilities diminishes their ability to push through the rate increases required to sustain rate base growth.

We have conducted a similar analysis of the transmission and distribution utilities. For this purpose, we have calculated these companies’ O&M expense as a ratio of the total payments for electricity made by these companies’ customers, including payments to the T&D companies for the delivery of electricity as well as payments to the T&D companies or third-party suppliers for the electricity itself. Grouping these utilities into the three categories set out above, we find that:

  • PEG combines (i) a low ratio of O&M expense to total customer payments for electricity (14%) with (ii) low rates of growth in O&M expense over the last five years (0.9% p.a.);
  • ES combines (i) a high ratio O&M expense to total customer payments for electricity (22%) with (ii) low rates of growth in O&M expense over the last five years (1.2% p.a.); and
  • AGR and PCG combine (i) a high ratio of O&M expense to total customer payments for electricity (32% and 21%, respectively) with (ii) high rates of growth in O&M expense over the last five years (4.1% and 3.7% p.a., respectively).

©2018, SSR LLC, 225 High Ridge Road, Stamford, CT 06905. All rights reserved. The information contained in this report has been obtained from sources believed to be reliable, and its accuracy and completeness is not guaranteed. No representation or warranty, express or implied, is made as to the fairness, accuracy, completeness or correctness of the information and opinions contained herein.  The views and other information provided are subject to change without notice.  This report is issued without regard to the specific investment objectives, financial situation or particular needs of any specific recipient and is not construed as a solicitation or an offer to buy or sell any securities or related financial instruments. Past performance is not necessarily a guide to future results.

  1. At the transmission and distribution utilities, increases in O&M expense per customer have accelerated relative to inflation in recent years, with the gap widening from 90 basis points p.a., on average over the last 15 years, to 180 basis points p.a. over the last ten years (see the right-hand chart of Exhibit 6). The pace of increase has moderated slightly over the last five years, but nonetheless exceeded the rate of inflation by 110 basis points p.a. over this period.

    By contrast, among the vertically integrated utilities, the pace of increase in O&M expense has moderated relative to inflation over time. While exceeding the rate of inflation by 80 basis points p.a. over the last ten and 15 years, over the last five years O&M expense per customer at the vertically integrated utilities has lagged the rate of consumer price inflation by 10 basis points p.a. (see the right-hand chart of Exhibit 6). 

Print Friendly, PDF & Email