Duke Deserves a Princely Valuation
The market underestimates this utility's ability to increase long-term earnings and the dividend.
Duke Energy (DUK) was once a darling of income investors, long trading at a premium to its peers. That premium faded with execution missteps during the 2012 Progress Energy acquisition, putting focus on Duke’s regulatory relationships in the Carolinas. Investors further soured when Duke Energy’s unregulated Latin American hydro assets and merchant generation went south, highlighting the assets’ cash flow volatility. Duke topped things off by paying the highest price/earnings multiple of any utilities M&A deal this cycle with its $6.7 billion Piedmont Natural Gas acquisition in 2016.
But CEO Lynn Good and her executive team have returned Duke Energy to its basic regulated utility business, and we think income investors should take another look. Duke’s regulatory relationships are on sound footing in the Carolinas, setting the stage for a five-year investment plan that can support 4%-6% earnings and dividend growth. We think this growth paired with a 5.0% dividend yield, narrow economic moat, and 4-star rating offers a compelling total return opportunity. The market is worried about weak near-term growth, but we think this ignores Duke’s growth in 2020 and beyond from three areas: greening up generation, meeting growing gas demand, and building out the electric grid. As Duke hits its growth objectives, we think income investors will again give it a princely valuation.
Duke’s 5.0% dividend yield is 130 basis points greater than the 3.7% industry median and 320 basis points more than the S&P 500’s 1.8% yield. Duke also trades at a discount to peers on a multiples basis. We think the market is focused on Duke’s near-term earnings weakness due to tax reform despite earnings growth that we estimate will accelerate in the latter half of our forecast. This should allow management to hit our 5% earnings growth target, in line with its 4%-6% target. As Duke continues to execute and deliver earnings growth, we expect its valuation will climb.
We think one of the market’s main concerns is Duke Energy’s ability to meet its five-year 4%-6% earnings growth target. Tax reform has accentuated this concern. A lower interest deductibility benefit from Duke’s corporate leverage and the need to issue $2 billion in equity leaves little room for earnings per share growth in 2018. The midpoint of management’s $4.55-$4.85 2018 EPS guidance is just $0.09 above 2017 results, less than 2% growth. We estimate the $2 billion share issuance will dilute EPS by $0.10 in 2018, and the lower tax shield from interest expense will knock $0.20 off EPS.
We expect earnings growth to pick up starting in 2019. Core to our forecast is Duke’s ability to execute its five-year $37 billion capital plan. During the next five years, we forecast 7% rate base growth. Slower accumulation of deferred tax liabilities due to the new lower federal tax rates should boost rate base growth. The long-term benefit of higher rate base growth and completion of the Atlantic Coast Pipeline will drive earnings growth to the high end of management’s 4%-6% target in 2020-22. Thus, investors trade modest 2018-19 earnings growth for higher growth in the latter half of our forecast.
Putting together our 7% annualized rate base growth forecast, earnings growth beyond 2019, and the near-term equity dilution supports our 5% annual average earnings growth estimate during the next five years. As Duke transitions from minimal growth in 2018 to 6% growth beyond 2019, we believe its valuation will rebound.
Importantly Duke’s capital plan is spread across the utility value chain, namely electric generation, transmission and distribution, and environmental investments. We highlight coal-to-gas transition, natural gas infrastructure, and grid modernization as thematic areas of growth in the utilities industry, in which Duke has a long runway of growth potential.
Growth Source 1: Duke Goes Green
Duke has made significant progress in transitioning away from coal. In 2005, it generated 61% of its customers’ energy needs from coal. That share of coal generation fell to 33% as of last year, but we still see substantial investment opportunities for Duke to further cut its reliance on coal generation and boost its share of gas-fired generation and renewable energy. Retiring coal plants and transitioning to gas and renewables should lower the share of coal generation in Duke’s energy mix to 16% by 2030. Renewable energy such as hydro, wind, and solar also will continue to play a larger role in Duke’s supply diversification, doubling to 10% by 2030.
In the Carolinas, Duke recently completed the $600 million, 750-megawatt W.S. Lee combined-cycle gas turbine in South Carolina and is nearing the completion of the $1.1 billion Western Carolinas. In Florida, Duke expects its $1.5 billion, 1,640 MW Citrus County combined-cycle gas facility to go into service later this year. Additionally, seven of Duke’s coal plants totaling 10.3 gigawatts of capacity have an average age of 46 years, making them ripe for retirement. We expect Duke will replace that capacity with either natural gas generation or renewable energy.
Duke’s plans to increase its share of gas generation are due in part to the proximity of two large interstate pipelines, Transco and the planned Atlantic Coast Pipeline, which run north and south through the Carolinas. These pipelines offer plenty of gas supply for additional gas generation tie-ins. Once completed, the ACP should be able to support Duke’s additional generation demand. The ACP could upsize its capacity by 33%, or 500 million cubic feet per day, with little additional investment if Duke and other utilities demand additional gas supply. ACP also offers Duke the opportunity to expand investment in dual-fuel generation similar to the company’s three planned dual-fuel projects along Transco, which runs through the western Carolinas.
Renewable energy also offers plentiful growth opportunities for Duke. In the Carolinas, Duke has just 101 MW of solar capacity. All renewable energy sources, including wind and hydro, produced 8.2% of North Carolina’s energy needs in 2017. Duke has the chance to increase its share of renewable energy to help meet North Carolina’s 12.5% renewable portfolio standard by 2021. We assume hydro and wind generation remain constant , given the geographic and economic advantage of solar in the state. We project the state will need to install nearly 3,600 MW of additional solar through 2021 to meet the state’s RPS based on our 1.2% long-term demand growth forecast and a 20% average solar capacity factor. Numerous states have subsequently increased their RPS after utilities met or approached the current requirements. We expect North Carolina will follow suit with an enhanced standard past 2021.
The economics of solar are improving, which is likely to lead to additional investment opportunities throughout Duke’s service territory. We estimate the levelized cost of electricity, or LCOE, for solar is $56/megawatt-hour compared with $45.36/MWh for natural gas today. In Florida, we think the solar LCOE might be even lower for certain projects Duke could develop. Fellow Florida utility NextEra Energy, the largest renewable developer in the United States, predicts prices for solar may reach as low as $30/MWh by 2020, well below the current LCOE for natural gas. Regulators increasingly are supporting solar investments because falling costs result in smaller rate increases for customers.
In North Carolina, HB 589 legislation recently signed into law reforms the Public Utility Regulatory Policies Act to speed the state’s progress toward its 12.5% RPS. The legislation allows for competitive bidding for 2,600 MW of planned solar development. Duke can build as much as 30% of the capacity and is allowed to acquire additional facilities from third parties beyond that threshold. Additionally, legislation allowing recovery through renewable energy rate riders reduces regulatory lag. We think Duke’s renewable energy growth opportunities will exceed the 2,600 MW initially planned, with an opportunity for 1,000 MW of solar to meet the state’s renewable portfolio standard and demand growth.
In South Carolina, a technicality recently sank H. 4421 legislation, which would have removed the state’s 2% cap on solar power in the state and presumably led to more solar investment. The measure passed the state House with a majority vote, but it wasn’t not large enough to reach the two-thirds vote requirement. Although Duke and the other South Carolina utilities opposed the legislation, their primary complaint was implementation of the plan and its potential impact on revenue. We think opposition was the best move for shareholders in the near term since the state still has usage-based rates. With a sustainable regulatory policy such as rate decoupling, we expect the utilities will align with the solar supporters in the state to invest in solar.
Surprisingly, in Florida, solar represented less than 1% of total generation in 2017. Unlike other jurisdictions, Florida has no renewable portfolio standard. However, we think solar generation will be used to meet our forecast 1.4% demand growth beyond currently proposed natural gas projects. We expect our forecast for above-average population and economic growth in Florida will create a need for 8.9 GW of additional solar generation. Our national renewable forecast estimates that 13.3% of domestic energy will come from renewables in 2025. If Florida were to fall in line with our national electricity demand forecast, it could need roughly 20 GW of installed total solar, or 17.5 GW more than what is currently planned. While we don’t view this as achievable in the next five years, it highlights what we think is a very long runway of renewable energy investment opportunities in Florida.
Regulators are waiting for solar economics to improve before pushing renewable energy development. However, they did recently approve the state’s three investor-owned utilities--Duke, NextEra Energy, and Emera--to install as much as 2.5 GW of solar in Florida through 2021. Duke was allocated 700 MW in additions from 2019 through 2021. Importantly, Florida solar investments come with some of the best regulatory treatment in the country through the solar generation base rate adjustments implemented upon project completion. The investments require no rate case, eliminating regulatory lag, and Duke can earn within its 9.5%-11.5% allowed return on equity. We think the next rate case filing for renewable development for 2022 and beyond will offer similar solar investment opportunities. We don’t expect changes in the favorable regulatory recovery for solar investments in the state.
Further highlighting the recent push for solar development in Florida, NextEra Energy announced a supply agreement in late March with JinkoSolar. JinkoSolar will provide NextEra Energy with 7 million solar panels, enough to provide 2,750 MW of energy during the next four years. Also, JinkoSolar committed to building a production facility in the state that will be able to produce 1 million solar panels annually, or roughly 400 MW of generation capacity. This will aid NextEra’s plans to quadruple its solar generation in the state to 4,000 MW during the next 10 years. We think Duke may enter into a similar agreement to help shore up its panel supply needs for the next decade.
While the Dan River coal ash spill in 2014 was an unwelcome headline for Duke, we believe it highlighted a critical need to address the potential hazard from wet coal ash ponds. Despite the company having to pay fines for its role in the disaster, coal ash pond remediation should be a net positive for Duke as it invests a planned $2.5 billion in coal ash pond remediation during the next five years. This is almost three fourths of its planned $3.4 billion capital investment in environmental compliance. Duke estimates the total cost to remediate all of its coal ash ponds could top $6 billion, of which nearly 80% will be in North Carolina, where HB 630 state legislation mandates coal ash pond remediation. Investments like these required by law usually receive little pushback from regulators, giving us high confidence that Carolina regulators will allow a timely return of and on capital.
Growth Source 2: Duke Gases Up
With the 2015 Piedmont Natural Gas acquisition, Duke Energy significantly increased its leverage to gas infrastructure. We were initially critical of the $6.7 billion all-cash price, including debt. The equity value was double our $30 stand-alone Piedmont Natural Gas fair value estimate, 3.5 times book value, and 31 times our forward earnings estimate. These values were all well above already pricey local gas distribution company valuations and takeout premiums in the last four years.
At the time of the purchase, management asked for patience from the investment community to prove the long-term growth opportunities and strategic rationale for the acquisition. We think the company is on the right path two years into the combination. We think management can attain its 10%-12% earnings growth target at the unit over the next 10 years, more than 2 times Duke’s consolidated earnings forecast. We expect earnings contribution from the unit will be 15% in 10 years, up from 8% at year-end 2017. The investments at the utility are supported by above-average customer growth, which we estimate to be 1.5% long term, and an aging infrastructure needing much investment.
The acquisition also increased Duke’s ownership of the Atlantic Coast Pipeline, a gas pipeline with 1.5 billion cubic feet per day of capacity that will run from West Virginia to North Carolina, past several large gas plants and potential coal plant conversion sites. We expect the ACP’s total cost will be $6.25 billion, in line with Duke Energy’s estimate for a total cost of $6.0 billion-$6.5 billion. Duke will own 47% of the project. Dominion (48%) and Southern (5%) are the other ACP owners. Dominion will construct, operate, and manage the ACP.
The ACP signed 20-year customer contracts for more than 90% of its capacity with customers using the gas to generate electricity or to sell to retail customers. Typically, pipelines have anchor contracts with gas producers, which can come with higher credit risk. However, the majority of ACP’s contracts to move gas out of the Marcellus and Utica shale regions are with more creditworthy utilities.
The ACP is expandable to 2 bcf/d, offering Duke Energy a source of wide-moat growth potential beyond its late 2019 in-service date if the region needs more gas to accommodate increased demand from new power plants and retail customers. Duke alone has seven coal plants in the Carolinas, with an average age of 46 years, which provides for natural extensions upon gas plant replacements. We believe the Marcellus and Utica shale plays will remain some of the lowest-cost natural gas in North America, so we think it is likely that this expansion could occur in the not-too-distant future. In addition, the expansion can be added at much lower marginal cost than a competitor building a new pipeline, demonstrating the efficient scale wide moat of the ACP.
The earnings from the ACP are critical for Duke to meet our 5% earnings growth estimate. While Duke hasn’t disclosed the project’s return profile, we use a 9% return on invested capital in our forecast. We estimate this is below returns that Duke earns on its other pipelines but in line with returns for similar pipelines in the region. In 2020, we forecast a pretax earnings contribution from ACP of $211 million, assuming the pipeline is operating at 80% capacity to allow for potential production delays or regulatory approval delays. The pretax earnings contribution increases to $265 million for 90% capacity in 2021 and $293 million in 2022, which assumes that the line is fully subscribed.
Growth Source 3: Strengthening the Grid
In the American Society of Civil Engineers’ 2017 Infrastructure Report Card, America’s energy infrastructure received a D+. Aging infrastructure and severe weather are leading to an increase in the numbers of outage reports across the U.S. Strengthening the grid, accommodating distributed resources, and preparing for widespread electric vehicle penetration provides a long runway of grid investments.
Duke’s investments in grid modernization address these nationwide concerns. The company plans to spend $10 billion over the next five years on reliability (storm hardening, targeted undergrounding, resiliency) and smart grid (advanced metering, advanced systems and communications, and self-optimizing grid).
Roughly two thirds of Duke’s planned $10 billion in grid investments are for reliability, with the largest opportunities in protecting the grid from severe storms. During Hurricane Irma in 2017, nearly 1.3 million Duke customers lost power. Following the storm, Duke restored power for 75% of the customer outages within three days and 99% within eight days. While this was a dramatic improvement from Hurricane Wilma in 2005, further investments in storm hardening and undergrounding of wires would probably improve restoration times after large storms. Regulators often support these investments because one of their primary responsibilities is to maintain a high level of customer reliability. We have high confidence that regulators will continue to support these investments, given the proven record of tangible customer benefits, which supports Duke Energy’s positive spread between return on invested capital and weighted average cost of capital and its narrow moat.
Duke’s remaining capital spending on its electric grid is designated toward a smarter grid. The largest portion is for smart meter adoption, which provides consumers numerous benefits, such as more control over home energy usage, potentially saving the end user money; allowing automatic meter readings, reducing utility overhead; and higher response times and more efficient storm restoration. Outside of Ohio, Duke Energy has deployed few smart meters across its service territory, with the significant investment needed in its largest service territories, the Carolinas and Florida.
Constructive Regulatory Environment Supports Duke’s Moat
When assessing utility moats, one of the main factors we analyze is the relationship utilities have with regulators in their service territories. In most cases, regulators set an allowed return on capital to set utilities’ rates such that utilities’ earned returns are sustainably higher than their costs of capital. Utilities like Duke that are investing more than depreciation will face an earnings headwind and lower earned returns if regulators set customer rates based on low allowed returns. We also consider rate-making mechanisms such as rate riders or automatic base rate adjustments that allow a utility to reduce regulatory lag, thus narrowing the time between a utility’s investment and its cash recovery from customers. And we look for regulatory consistency, as regulators are typically appointed by an elected official or elected directly, making them a political representative of the rate payers in the jurisdiction they serve. Overall, we believe Duke operates in average to excellent regulatory jurisdictions across its subsidiaries, boosting earned returns relative to other utilities.
Florida regulators and legislators have turned the state’s regulatory environment into one of the best operating jurisdictions for utilities. Investors need to look no further than Duke Energy’s most recent base rate adjustment settlement, which allows for $67 million in annual base rate increases over 2019-21, primarily to recover $1 billion in recent grid modernization investments. Additionally, regulators allowed Duke to collect a solar generation base rate adjustment, which allows the utility to recover costs for 700 MW of new-build solar capacity over 2019-21 with automatic rate adjustments and a 10.5% allowed return on equity. Thus, there is no regulatory lag between the in-service date and when Duke earns a return on its investment. Overall, the subsidiary is allowed an earned ROE of 9.5%-11.5%, compared with the national average allowed ROE of 9.6%. Duke has consistently earned in the upper half of the earned ROE band.
We view the Carolinas’ regulatory environment as less constructive than Florida, although Duke historically has received mostly fair regulatory treatment. Allowed returns and capital structures have typically been in line with the national averages. Adjustments for fuel clause expense are updated annually, a common time frame among utilities. Only South Carolina puts a limit to how often a utility can file a rate case, currently once every 12 months.
Unlike in Florida, where rates are set on a forward-looking basis, the Carolinas use historical test years adjusted for future known and measurable changes. This is less favorable than a forward test year since a historical test year increases the regulatory lag between a utility’s investment and its recovery of that investment from customers. Duke management previously mitigated this lag by managing operating expenses to earn its allowed returns on equity regularly. However, with the company boosting investment in state infrastructure, management is running out of room to reduce operating expenses. We expect management to file more frequent rate cases to reduce regulatory lag. Management continues to work on regulatory reform to reduce regulatory lag in the state. One notable success was its annual grid investment rider, currently about $600 million annually.
Regulatory proceedings support our view that the regulatory environment in the Carolinas will remain constructive. Duke Energy has recently filed two substantial rate cases at both of its subsidiaries: Duke Energy Progress and Duke Energy Carolinas. Regulators recently awarded Duke Energy Progress a 9.9% allowed return on equity. Although this is 30 basis points lower than its previous allowed return on equity, it is in line with other states’ average allowed returns, which also are falling as a result of persistently low interest rates. Recovery for coal ash remediation investments will be filed in future proceedings. This would add to the unit’s $8.15 billion rate base.
Duke Energy Carolinas also reached a preliminary settlement with the regulatory staff. The agreement includes a 9.9% allowed return on equity, consistent with the Duke Energy Progress ruling. Both sides continue to discuss the time frame for the recovery and amortization of coal ash basin costs and additional costs allowed in the annual grid investment rider. We forecast that Duke will file a rate case approximately every other year at both subsidiaries in our five-year outlook to support our $23.4 billion capital investment forecast in the Carolinas during this time frame.
Regulators in the Carolinas are not prone to overly punitive damages. Duke Energy paid a relatively small $6 million fine for its 2014 coal ash spill. We think management’s proactive involvement in admitting and addressing the coal ash spill helped lessen the penalty. Investments to transition to dry coal ash ponds is an investment area for which recovery is mandated by state law under the Coal Ash Management Act of 2014.
Recent state legislative action points toward an improving political and regulatory environment for utilities in the Carolinas. In North Carolina, HB 589’s competitive energy solutions law reforms the Public Utility Regulatory Policies Act and allows for competitive bids for 2.6 GW of utility-scale solar generation. Duke Energy can participate and build up to 30% of the cap amount. Duke is allowed to acquire additional solar generation from third parties that would not count toward the cap.
Earlier this decade, relationships between Indiana regulators and Duke were strained as cost overruns piled up at the 618 MW Edwardsport integrated-gasification combined-cycle facility, which converts coal into syngas to fuel the plant’s combustion turbines. The plant was supposed to be an environmentally friendly way of maintaining coal generation for the future, but it became a costly proposition for consumers and Duke shareholders. Ultimately, Edwardsport cost $3.5 billion, of which $2.6 billion was recovered through customer rates. Shareholders had to pay the remaining $900 million. After initial difficulties, Edwardsport’s operating performance has improved substantially.
Since then, it appears that regulators and Duke have made amends. State legislation SB 560 has been approved by regulators for $1.3 billion in transmission and distribution investments in aging infrastructure over 2016-23, allowing for recovery of these investments through riders, reducing regulatory lag. Combined with Duke’s 10.5% allowed return on equity, Indiana has proved itself to be an above-average regulatory jurisdiction.
Generally, regulation is more constructive for gas utilities than for electric utilities, and this is also true for Duke. The regulatory environment for Duke’s gas local distribution companies is particularly attractive because of the numerous rate riders within its jurisdictions. Rate riders significantly reduce regulatory lag, turning capital investments into earnings quicker. For Duke, this is particularly important as the company is planning $4.2 billion of investment at its gas LDCs to drive 10%-12% unit earnings growth during the next five years.
Overall, Duke’s attractive regulatory environments have supported returns on invested capital greater than the company’s weighted average cost of capital. The downward trend in ROICs over 2013-17 is a result of the earnings volatility from the unregulated merchant generation and Latin American units, which Duke has since sold. We don’t forecast material changes to the regulatory environments in which Duke operates, resulting in a sustainable healthy spread between ROIC and WACC through 2022 and supporting Duke’s narrow moat.
Strong Management Team Supports Strategy Execution
Since taking the helm in 2013, Lynn Good and her executive team have done an excellent job refocusing Duke on utility basics. Management successfully integrated the $26 billion acquisition of Progress Energy, exceeding internal and our own expectations for cost reductions and fuel savings. We think management’s ability to exceed its savings targets for customers and work proactively with regulators was key to restoring regulatory relationships that were strained immediately after the Progress Energy close. Since the transaction’s close, management has reduced nonrider operations and maintenance from $4.7 billion to $4.6 billion, which is impressive because the company has invested $28 billion in capital over that four-year time frame. We expect nonrider O&M will remain flat throughout our five-year forecast period.
Management has also made significant progress in reregulating Duke’s operations. After several divestments and acquisitions during the past decade, the only nonregulated business remaining is commercial renewables. This segment represents approximately 5% of total earnings and has proved to be a steady contributor to earnings.
We think management has displayed good crisis management. In 2014, it faced an environmental coal ash spill at its Dan River facility. In our opinion, management handled the situation extremely well. Duke Energy immediately accepted responsibility, shouldered the cleanup costs, and proactively worked with state legislators and regulators to create a plan to clean up Duke’s remaining coal ash ponds in the state. Coal ash spending over the next decade has turned into a growth opportunity, helping support our 5% earnings growth target. We think that because of management’s handling of the spill, penalties were limited, with Duke Energy receiving a relatively small $6 million fine.
Another benchmark for utility executives is the ability to earn allowed returns. By earning its allowed returns in each jurisdiction, management is maximizing shareholder return. Since 2014, Duke management has consistently earned near or greater than the allowed returns of equity across its regulated jurisdictions.
Overall, we assign Duke’s management team a Standard stewardship rating. Strong corporate governance, constructive regulatory relationships, and an astute leadership team support our rating. Seeing continued ability to integrate and drive value at Piedmont Natural Gas, which we viewed as an expensive acquisition, would probably lead us to raise our stewardship rating to Exemplary.
Andrew Bischof does not own (actual or beneficial) shares in any of the securities mentioned above. Find out about Morningstar’s editorial policies.