Rising Interest Rates and ROEs: Is There Upside for Utilities? Probably Not Until Rates Go Much Higher
Eric Selmon Hugh Wynne
Office: +1-646-843-7200 Office: +1-917-999-8556
Email: firstname.lastname@example.org Email: email@example.com
SEE LAST PAGE OF THIS REPORT FOR IMPORTANT DISCLOSURES
February 9, 2017
Rising Interest Rates and ROEs: Is There Upside for Utilities?
Probably Not Until Rates Go Much Higher
Markets are anticipating that President Trump’s economic policies will increase economic growth and drive interest rates higher. In this note we examine where returns on equity (ROEs) would likely be set in each state given current interest rates. We analyze the impact on utilities currently in rate cases and identify which utilities could be the first to benefit from higher allowed ROEs due to rising rates.
Portfolio Manager’s Summary
- Now at 2.35%, yields on 10-year U.S. Treasury notes have increased by 98 bps since bottoming at 1.37% on July 8, with approximately half that increase occurring since the presidential election. Corporate bond yields have risen by similar amounts over the same timeframe.
- As we have noted in prior research (see The Argument for an Overweight Position in Regulated Utilities, October 5, 2016), utility stock returns tend to move inversely to interest rates in the short term. Over time, however, higher Treasury bond yields should cause regulators to grant higher allowed ROEs to regulated utilities, raising utilities’ earnings potential and partly offsetting the impact of rising rates for individual companies.
- The average allowed ROE set by state regulators has moved closely with 10-year Treasury yields (0.915 R-squared) over the period from 1980-2016 (see Exhibit 3). At the current 10-year Treasury yield, this historical relationship would point to an average allowed ROE of ~9.76%, 23 bps above the 2016 average.
- Over the past 10 years, utility regulatory commissions in certain states have granted allowed ROEs that are consistently above or below the national average (see Exhibit 4).
- Since 2007, Georgia, California and Florida have on average granted the most generous ROEs, while Connecticut, South Dakota, New York and Rhode Island have been the least generous.
- At the current 10-year Treasury yield, we would expect the allowed ROEs granted by state regulatory commissions to range from a low of 8.8% in Connecticut to a high of 10.7% in Georgia, versus an expected national average of 9.76% (see Exhibit 5).
- Although rising rates will eventually drive ROEs higher, most utilities’ allowed ROEs are above what we would expect state regulatory commissions to grant today given the current level of 10-year Treasury yields.
- Among utilities currently in rate cases, almost all are likely to receive lower ROEs than that granted in their last rate case. Allete, OG&E Energy, and Westar Energy could see potential reductions of 50 bps versus their previous rate cases. Such a reduction in allowed ROE could reduce the earnings potential of these companies in 2018 by 4.0%, 3.4% and 2.8%, respectively, of 2018 consensus EPS.
- Among the few utilities likely to see their allowed ROEs increased in 2017 are Ameren (by an estimated 20 bps at its Missouri subsidiary and 25 bps at its Illinois subsidiary) and Great Plains Energy (by 30 bps at one of its subsidiaries). These ROE increases could raise the earnings potential of these companies by an estimated 1.6% of 2018 consensus for Ameren and 1.0% for Great Plains.
- The utilities best positioned to benefit from rising Treasury yields over the next 12 months are those that (i) are not currently in rate cases, (ii) whose 2017 earned ROEs will likely fall short of their expected allowed ROE by a significant margin (e.g. >100 bps), and (iii) whose allowed ROE in 2018 is likely to be higher than their current allowed ROE. Such companies could justifiably file rate cases in 2018 and as result realize improvements not only in their earned ROE but in their allowed ROE as well.
- At current Treasury bond yields, very few companies meet these criteria. Of the 13 companies whose utility subsidiaries we expect to under-earn their allowed ROEs by >100 bps in 2017, all could see their allowed ROEs cut in a rate case, assuming no change in Treasury bond yields.
- However, if 10-year Treasury bond yields were to increase by another 100 bps — which we estimate would raise allowed ROEs by ~50 bps — 10 of these companies (AEP, AGR, AVA, DUK, ETR, FE, GXP, POR, WEC, XEL) could receive higher allowed ROEs at currently under-earning subsidiaries.
- The increases in allowed ROEs could be significant for AVA and POR, raising the earnings potential of AVA by 5.4% of 2018 consensus EPS and of POR by 4.5% (see Exhibit 6).
Exhibit 1: Heat Map: Preferences Among Utilities, IPP and Clean Technology
Change to Heat Map Preferences
We are removing Calpine from our list of favorite names. Our thesis when we placed them on the list in September was that (i) they were best positioned among the IPPs for the upcoming trends in the industry, including the implementation of the Clean Power Plan and higher capacity prices in PJM due to the capacity performance rules, and ii) their modern gas-fired generating fleet that should be better insulated from low gas prices than the other IPPs, particularly those with higher percentages of coal fired generation. However, a number of factors have changed: i) the Trump administration will likely limit the extent and possibly eliminate the Clean Power Plan; ii) the outlook for capacity prices has worsened, given the recently released parameters for the PJM capacity auction, the continuing announcements of new power plants receiving financing and going into construction in PJM and around the country, and the weak New England capacity auction results released last night; and iii) the impact of increased electric energy storage capacity on generating margins in California, and potentially other markets, which we believe will be larger and be felt sooner than we previously expected.
The last two items are also important factors in why we have kept NRG on our list of least favorite names in spite of activist investor involvement and discussion in the press about a potential sale of the company.
Now at 2.35%, yields on 10-year U.S. Treasury notes have increased by 98 bps since bottoming at 1.37% on July 8, with approximately half that increase occurring since the presidential election. While bond yields have been relatively steady since peaking in December, the economic policies supported by President Trump during his campaign – a combination of tax cuts, large increases in infrastructure spending and no offsetting reduction of Medicare or Social Security spending — could raise deficits and push interest rates higher, particularly if they succeed in boosting economic growth and inflation.
Rising rates can be a sign of strength in the economy, but for regulated utilities, a sharp increase in Treasury bond yields is generally predictive of material underperformance versus the broader equity market. A regression analysis of regulated utilities’ total returns relative to the S&P 500 since the recovery began in 2009 shows that for every 100 bps increase/(decrease) in 10-year Treasury yields, regulated utilities have underperformed/(outperformed) the S&P 500 by 11 percentage points over the next twelve months. Over the long run, however, higher Treasury bond yields should also result in regulators granting higher allowed ROEs to utilities, raising utilities’ earnings potential and partly offsetting the impact of rising rates for individual companies.
In determining a “fair return” for utility shareholders on their equity investment, state utility regulators apply a number of methods, including the capital asset pricing model (CAPM), that use government or corporate bond yields to estimate the cost of equity. As a result, average allowed ROEs for regulated electric utilities have declined, along with long term Treasury yields, since the early 1980s, reaching a low of 9.53% in 2016 (see Exhibit 2). The historical relationship between 10-year Treasury bond yields and allowed ROEs is a remarkably tight one: a regression analysis for the period 1980-2016 of the relationship between (i) the average ROE allowed by state regulatory commissions in electric utility rate cases and (ii) the yield on 10-year U.S. Treasury notes results in an R-squared of 91.5% (see Exhibit 3). This regression analysis suggests that for every 100 basis point move in 10-year Treasury yields, allowed ROEs should move by ~53 bps. Using the current 10-year Treasury yield of 2.35%, we would expect an average allowed ROE of ~9.75%, 22 bps higher than the average for 2016.
|Exhibit 2: Average Allowed ROE Set in Electric Utility Rate Cases by State Utility Commissions, 1980-2016||Exhibit 3: Relationship Between Average Allowed Electric Utility ROEs and 10 Year Treasury Yields, 1980-2016|
|Source: US Federal Reserve, SNL, SSR Analysis||Source: US Federal Reserve, SNL, SSR Analysis|
The historical record, however, shows that the allowed ROEs set by the various state regulatory commissions can vary widely in any given year. To better understand the variation in allowed ROEs among the states, we calculated for each year since 1980 the gap between (i) the national average of the allowed ROEs set by the state regulatory commissions in electric utility rate cases and (ii) the allowed ROEs granted by individual states in the same year. We found that over the last 10 years a number of state regulatory commissions have been consistent in granting allowed ROEs either above or below the national average. In Exhibit 4 we lay out the average premium or discount versus the national average allowed ROE for each state. Georgia has been the most generous state with an average premium of 88 bps. A number of states offered premiums between 30 and 45 bps, including California, Florida, South Carolina, North Dakota, Virginia, Tennessee and North Carolina. The least generous states have been Connecticut, with an average discount to the national average ROE of 97 bps, South Dakota with 92 bps, and New York with 78 bps. Rhode Island, Washington D.C., New Hampshire and Illinois follow, with discounts between 46 to 62 bps. Applying these premiums or discounts to our forecast of a national average allowed ROE of 9.76%, given the current 10-year Treasury yield, we estimated the allowed ROE likely to be set by each state regulatory commission (see Exhibit 5. On this basis, we would expect allowed ROEs to fall in a range from a low of 8.8% in Connecticut to a high of 10.7% in Georgia.
Exhibit 4: Average Difference Between State Allowed ROEs and National Average, 2007-2016
Source: SNL, SSR Analysis
Exhibit 5: SSR Forecast of Allowed ROEs by State, Given the Current 10-year Treasury Yield
Source: SNL, SSR Analysis
The Impact of Higher Rates on Utilities’ Earnings
The likely level of allowed ROEs is of little practical significance unless utilities have filed, or soon expect to file, rate cases that will cause their regulatory commissions to re-assess their allowed ROEs. Unless a utility has automatic annual rate reviews, as is the case for the two largest utilities in Illinois, owned by Ameren and Exelon, or it is called in for a rate review because it is overearning, a rarity in the electric utility industry, utilities generally can control the timing of new rate cases. As a general matter, utility managements will only file a rate case if the utility is significantly under-earning (usually by more than 100bps) the allowed ROE they expect the regulatory commission to grant in the rate case. Under these circumstances, there is a relatively high probability that regulators will grant an allowed ROE that materially exceeds the utility’s actual earned ROE, and therefore that the associated increase in allowed revenues will materially enhance earnings.
Generally, absent strong demand growth (which most U.S. utilities have not enjoyed over the last five years), utilities need to file rate cases periodically to ensure that allowed revenues keep pace with the increase in operating expenses, as a result of inflation and customer growth, and that earnings keep pace with the growth in rate base and shareholders’ equity investment in the company. Our expectation that utilities will file periodic rate cases to protect their earned ROEs is a key reason for our optimism with respect to regulated electric utility earnings over the next five years; promptly filed rate cases will allow utilities to translate the forecast growth in rate base into commensurate growth in regulated earnings. (See our research note from November 29, 2016, How Will the Trump and GOP Tax Plans Impact Rate Base Growth? The Opportunity and Risk Facing Utility Investors for our most recent forecast of utility rate base growth).
In a period of rising interest rates, however, the need to improve earned ROEs is no longer the only incentive to file a rate case: utility managements could also begin to file rate cases if they believe the allowed ROE they would be granted would be meaningfully higher than their previous allowed ROE, thereby enhancing the long-term earnings potential of the company. At the current level of Treasury yields, however, we do not expect utilities to be successful in pursuing higher allowed ROEs; rather, we would expect the primary driver of utility rate cases to continue to be the need to rectify under-earning.
Specifically, we have reviewed the allowed ROEs of all of the investor owned electric utilities and compared them to the allowed ROE we would expect each utility to be granted by their state regulatory commission, given the current level of Treasury bond yields. Our analysis suggests that, at current Treasury yields, very few utilities would receive an increase in allowed ROE in a rate case. Indeed, unless interest rates move much higher (at least 100bps and probably more than 150bps), we do not expect to see any utilities filing rate cases in order to obtain higher allowed ROEs.
On the contrary, for the utilities that have filed rate cases recently, we expect that all but a handful to receive reduced allowed ROEs versus their previous rate cases. Today there are 25 regulated utilities (the subsidiaries of 17 companies) that are engaged in rate case proceedings. Most have allowed ROEs from previous rate cases that are equal to or greater than what we expect the utilities to receive in the current rate cases. Three companies, Allete, OG&E Energy, and Westar Energy, could see potential reductions of 50 bps versus their previous rate cases. For these companies, the expected reduction in allowed ROEs will offset the other components of the rate case and reduce the rate increase regulators will permit. The cut in allowed ROE can thus be thought of as a reduction in the earnings potential of the company. To give a sense of the scale of this loss, we translated the revenue impact of the change in allowed ROE into earnings per share. As a percent of consensus 2018 earnings estimates, we calculate the impact to be 3.6% for Allete, 3.2% for OG&E and 2.6% for Westar. See the Appendix A for a table with the current allowed ROE and our forecasted allowed ROE for each company currently in a rate case.
Among the few utilities likely to see their allowed ROEs increased in 2017 are Ameren (by an estimated 20 bps at its Missouri subsidiary and 25 bps at its Illinois subsidiary) and Great Plains Energy (by 30 bps at one of its subsidiaries). These ROE increases could raise the earnings potential of these companies by an estimated 1.8% of 2018 consensus for Ameren and 1.0%.
The utilities best positioned to benefit from rising Treasury yields over the next 12 months are those that (i) are not currently in rate cases, (ii) whose 2017 earned ROEs will likely fall short of their expected allowed ROE by a significant margin (e.g. >100 bps), and (iii) whose allowed ROE in 2018 is likely to be higher than their current allowed ROE. Such companies could justifiably file rate cases in 2018 and as result realize improvements not only in their earned ROE but in their allowed ROE as well.
We have identified 21 subsidiaries of 14 companies that we expect to be under-earning in 2017. We calculate that none of these companies would receive a higher allowed ROE at the level of Treasury yields prevailing today. However, in the event of a 100bps increase in 10-year Treasury yields over the next 12 months — which we would expect to drive an increase in allowed ROEs of ~50bps — 10 of these 14 companies have subsidiaries which could expect to receive an increase in allowed ROEs (see Exhibit 6). We estimate that the ROE increases would range from ~10 to 70bps.
For a number of these companies the impact on such an increase would be limited because the utility subsidiaries affected are few among many; this would be the case, for example, for the utility holding companies AEP or FE. However, for single utility companies, such as Avista or Portland General, the impact of an increase in ROE due to higher interest rates could have a large impact on parent company earnings. Expressed as a percent of 2018 consensus EPS, we calculate that the higher allowed ROE associated with a 100bps increase in 10-year Treasury yields could increase potential earnings by 5.4% for Avista and 4.5% for Portland.
Exhibit 6: Increase in Earnings Potential Due to the Forecasted Increase in Allowed ROEs Resulting from a 100 Basis Point Increase in 10 Year Treasury Yields (% of 2018 Consensus EPS)
Source: SNL, SSR Analysis
In conclusion, the recent increase in Treasury yields is unlikely to have a meaningful positive impact on utilities’ allowed ROEs. On the contrary, at current Treasury yields, we would expect the allowed ROEs of the utilities currently in rate case proceedings to be cut.
For investors expecting further increases in bond yields, the utilities best positioned to benefit from rising Treasury yields over the next 12 months are those that (i) are not currently in rate cases, (ii) whose 2017 earned ROEs will likely fall short of their expected allowed ROE by a significant margin (e.g. >100 bps), and (iii) whose allowed ROE in 2018 is likely to be higher than their current allowed ROE. Such companies could justifiably file rate cases in 2018 and as result realize improvements not only in their earned ROE but in their allowed ROE as well. We calculate that there are 13 companies whose utility subsidiaries are likely to under-earn their allowed ROEs by >100 bps in 2017. In the event of a 100 bps increase in 10-year Treasury yields over the next 12 months — which we estimate would raise allowed ROEs by ~50 bps — 10 of these companies (AEP, AGR, AVA, DUK, ETR, FE, GXP, POR, WEC, XEL) could receive higher allowed ROEs at
currently under-earning subsidiaries (see Exhibit 6). The increases in allowed ROEs could be significant for AVA and POR, raising the earnings potential of AVA by 5.4% of 2018 consensus EPS and of POR by 4.5%.
Appendix A: ROE Expectations for Utilities Currently in Rate Cases
Source: SNL, SSR Analysis
©2017, 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.
- Rate-regulated utilities are allowed to recover their prudently incurred cost of service in rates, including all costs to procure fuel and purchased power, operation and maintenance expense, depreciation expense, income and other taxes, and a fair return on rate base. Rate base represents the capital invested by a rate-regulated utility monopoly in the supply of a public service (e.g., electricity or gas) and is roughly equivalent to the net depreciated historical value of the utility’s plant, property and equipment. Rate base may be funded by common and preferred equity, long term debt and net deferred tax liabilities. On the debt portion of rate base, utilities are generally allowed to earn a return equivalent to their embedded cost of long term debt. A similar approach is to taken the recovery of the cost of preferred equity. On the portion of rate base funded with equity (a proportion set by regulators at a level deemed adequate to sustain an investment grade rating on the utility’s long term debt, and referred to as the “equity ratio”) utilities are allowed to earn a fair return (the utility’s “allowed ROE”) as determined by regulators in periodic rate cases. Given this regulatory framework, it is common for investors to estimate future utility earnings as the product of rate base, the utility’s equity ratio and its allowed ROE. ↑
- In our assessment of Virginia’s allowed ROEs over the past several years we have excluded the incentive adders granted Dominion’s Virginia Electric Power Company subsidiary for its new generation projects, and have focused rather on the decisions of state regulators in setting base ROEs. Including the incentives, which are required under Virginia law for certain higher risk generation projects, Virginia’s average allowed ROE over the past 10 years has been 82bps above the national average. ↑
- Ameren and Exelon’s Illinois subsidiaries have annual rate cases with automatic adjustments in their ROE based on a formula (the average of the 30 year Treasury yield over the course of the year plus 5.8%). We used the formula in determining these companies’ allowed ROEs rather than applying our historic analysis of the Illinois average gap versus the national average allowed ROE. While neither of these two companies is in a rate case right now, we have included them in this group because they will in the next few months. ↑