The need for many hundreds of billions of dollars in capital expenditures creates huge opportunities and challenges, especially in a more challenging credit environment.
Roger Wood is head of Rothschild’s Power & Utilities Group in North America. Contact him at firstname.lastname@example.org.
The utility industry needs to prepare for a period of much higher capital expenditures. This results from the confluence of several factors:
• Shrinking generation reserve margins, as the glut of surplus capacity from earlier in the decade has been worked through much more rapidly than expected;
• Spending on compliance with NOx, SOx, and mercury requirements;
• Pressures to invest in lower-carbon—and more expensive—generation technology and associated infrastructure, including transmission;
• The need to replace aging transmission and distribution (T&D) infrastructure, much of which was put in place 30-40 years ago and is nearing the end of its design life;
• Continued robust rates of population growth and economic growth in many parts of the United States, resulting in the need for system expansion; and
• Technology spending on areas such as customer information systems and automated meter reading.
The numbers are huge. Cambridge Energy Research Associates estimates that $900 billion of direct infrastructure investment will be required by electric utilities over the next 15 years. That compares with $750 billion of plant currently in place. And experience, both in the utility industry and more broadly, suggests that it is rare for major construction projects to be completed on budget (or on time).
The high capital costs of nuclear generation are well-known, although exactly how high is not entirely clear. General Electric recently estimated the cost of the new reactor that it is marketing in combination with Hitachi at anywhere from $2,000-$3,000/kW. FPL has put the cost of potential new nuclear units at its Turkey Point facility at $4,000/kW.
For those deterred by the projected costs of nuclear, there is not much relief with alternative low-carbon technologies. Tampa Electric filed documents with the Florida Public Service Commission in July 2007 regarding its proposed new 630-MW Polk 6 integrated gasification combined-cycle (IGCC) project, which is expected to enter service by 2013. The filing indicated a total cost of $2 billion, or $3,175/kW. $1.6 billion ($2,540/kW) was for engineering, procurement and construction, and the balance of $400 million ($635/kW) was for other costs such as transmission infrastructure and environmental permitting.
Nor is the news much better with renewables. Following a long period of decline for wind-project costs, prices have started to increase again. According to the U.S. Department of Energy, average installed project costs for wind projects coming online in 2006 were approximately $1,480/kW, which represented an increase of almost 20 percent over projects coming online in 2005. Further increases are in the pipeline. For projects proposed in 2006 but not yet built, the Department of Energy estimates that installed costs will be $1,680/kW. Much of this has been driven by higher prices for the turbines, which have increased by approximately $400/kW over the last five years. Costs for solar and other renewable technologies are even higher than for wind.
TXU’s experience trying to build a fleet of pulverized coal plants suggests that, as a practical matter, it will be difficult for other developers of conventional coal-fired plants to follow. And there was in any event widespread skepticism regarding TXU’s cost estimates of $1,100/kW. By comparison, Duke’s projected costs for its Cliffside coal-fired project in North Carolina increased from $1,250/kW in May 2005 (based on a 2-unit plant) to $2,288/kW in March 2007 (based on a 1-unit plant, including interest).
This leaves combined-cycle gas turbines, with capital costs of around $800/kW to $900/kW, looking increasingly attractive from a capital cost standpoint, though total generating costs will be highly dependent on volatile natural-gas prices. And in common with other technologies, construction costs for combined-cycle gas turbines (CCGTs) may come under further pressure from increased steel, cement copper, and other commodity prices, a weak dollar, and a shortage of skilled labor.
The story is similar with transmission and distribution, although the numbers are even larger. Southern California Edison has estimated that its rate base will almost double in 5 years, increasing from $10.9 billion in 2006 to $20.4 billion in 2011, with over 75 percent of total projected spending on T&D. The industry as a whole has been through an extended period of underinvestment, influenced in part by rate freezes. In the May 2007 issue of Public Utilities Fortnightly (“Spending Capital as if it Mattered”), Tom Flaherty and Tim Gardner estimated that average annual T&D capital expenditure between 2005 and 2010 is expected to be more than 50 percent higher in real terms than in the 30 years previously.
The leading indicators already are evident in increasing construction budgets. The graph shows capital spending by investor-owned-electric utilities on a trailing 12-month basis. (see Figure 1, p. 50)
Fortunately, the starting point for utilities to finance their future capital programs has improved a little over the last few years. Average equity ratios for investor-owned electric utilities have increased from 38.2 percent at year-end 2003 to 46.2 percent in March 2007 (source: EEI). Also, even though they are off recent highs, utility P/E ratios remain relatively robust, indicating that the timing for equity issuances may be relatively good.
But credit ratings for the industry remain weaker than they were earlier in the decade. At year-end 2001, 42 percent of electric utilities were rated A- or higher, and 68 percent were rated “BBB+” or higher, compared with 19 percent and 42 percent respectively at present. By contrast, 29 percent of the industry is currently rated “BBB-” or lower, compared with only 18 percent in 2001.
Also, free cash flows for electric utilities remain barely sufficient to cover current levels of capital expenditures, even before payment of dividends. And free cash flow after payment of dividends is expected to decline further in coming years (see Figure 3).
The scale of the projected investments is especially daunting relative to the size of the companies in the sector. A new coal or nuclear plant costing $3 billion to $4 billion or more represents a meaningful percentage of the equity market value of even the largest companies in the sector such as Exelon ($48 billion, as of the time of writing). If oil and gas supermajors with market capitalizations of $200 billion to $300 billion are concerned about the impact of funding deepwater exploration and production projects and LNG terminals costing similar amounts of money to large power plants, it is not surprising that utilities should be even more concerned about how they will finance their capital expenditure programs.
So how should utilities approach this challenge, especially during a period of more challenging conditions in credit markets? The magnitude of the numbers requires a multi-pronged approach.
It is not unreasonable for the ultimate beneficiaries of the investment, i.e., ratepayers, to bear the brunt of the burden, especially following a long period in which utility costs have represented a decreasing portion of total consumer spending.
The last couple of years already has seen a significant increase in rate-case activity, following a period of relative inactivity during much of the late 1990s and early 2000s.
Many more rate cases in the years ahead will be necessary to ensure that the new wave of capital expenditures can be financed.
Allowed returns for utilities remain relatively attractive, with the recent downward trend in average ROEs for both electricity and gas to their current level of approximately 10.3 percent (source: RRA) driven principally by a reduction in underlying Treasury bond rates. Figure 5 shows allowed ROEs for both electric and gas utilities, expressed as a spread over Treasurys (left scale), and actual Treasury bond rates (right scale):
Long-established laws of supply and demand indicate that federal and state regulators need to ensure that allowed returns stay at or above current levels if sufficient capital is to be attracted to the sector versus finding more attractive opportunities elsewhere. But companies and investors should be alert to the increasing burden on customer bills triggered even by large single projects, compounded by recent increases in commodity prices. In this context, the estimate provided by Appalachian Power Co. to the West Virginia Public Service Commission in August 2007 regarding the potential rate impact of its proposed 630-MW, $2.23 billion, Mountaineer IGCC project is noteworthy. APCo projected incremental rate increases between 2009 and 2012 of more than $100 million for West Virginia retail customers, equivalent to a total rate increase of 11.6 percent, from the $1 billion portion which is expected to be allocated to West Virginia.
Recent experiences in Illinois and Maryland, on top of the experience in California earlier in the decade, do not bode well for investors in terms of regulatory stability if end prices to customers are increasing rapidly. And price increases are inevitable, even without any further increase in commodity prices. Dan Ford of Lehman Brothers recently estimated that cumulative rate increases for his regulated coverage universe would amount to 66 percent by 2010 compared with 2005 levels.
In Illinois, the expiration of the rate freeze in January 2007, led to customers’ bills rising significantly. Even though customers had benefited from below-market prices for many years, and even though the structure for the generation auctions had been approved by FERC and by the Illinois Commerce Commission, the one-time shock of a step-up to market prices created a dynamic where politicians extracted a $1 billion settlement from Exelon and Ameren (with relatively modest contributions from Dynegy and Edison Mission) under the threat of retroactively extending the rate freeze and/or imposing additional taxes on generators.
Precedents from the previous cycle of nuclear-plant construction are not encouraging with respect to allowing utilities a fair return on their invested capital. The ability of regulators to conduct after-the-fact reviews of prudence allows them to apply the perfect wisdom of hindsight. Therefore, the more that utilities are able to obtain cost recovery during construction, through construction cost riders and other mechanisms, the more financeable their projects will become. Where real-time recovery of construction costs is not possible, keeping regulatory lag to a minimum will become all the more important. On this front, there was an unfortunate example in Arizona recently where Arizona Public Service (APS) was denied relief from its under-earnings situation (the company expects 2007 ROE to be around 7.5 percent versus its allowed ROE of 10.75 percent) that arose because of growth-oriented capital expenditure requirements and regulatory lag in recovery. The ACC rejected APS’s request for accelerated depreciation, inclusion of construction work in progress (CWIP) in rate base, and the earnings “attrition adjustment” (to bring actual ROE toward allowed levels). As a result, APS is put in a position to require routine rate case filings and may need to cut back its planned capital expenditures. The rating agencies have concluded that the inability of APS to receive timely recovery of its large capital expenditures could pressure its future credit profile.
Regulatory certainty is important in other ways. Investor appetite to fund new generating capacity in deregulated markets will depend on market signals being provided by capacity markets. Once in place, it is important that markets be allowed to function freely. Any indication that regulators would seek to limit investor upside through the imposition of price caps, or through changes in the level of such caps once they had been established, while leaving investors fully exposed to the downside, would create an asymmetrical risk profile and would result in an increase in the cost of capital or—in extreme cases—in capital not being available at all on reasonable terms. It is interesting to note in this context that only a relatively small amount of new capacity has been added in California since the last power crisis, so it is likely only a matter of time before the state experiences another supply squeeze.
Market structures also need to accommodate the potential for long-term contracts in deregulated markets. Given their bad experience in the early years of this decade, it is unlikely that investors will fund large new investments in generation purely on a merchant basis. In this context, it is helpful to see the Illinois Senate’s recent passage of House Bill 3388, allowing utilities to enter into long-term contracts for IGCC projects. For nuclear projects that are not in rate base, it will be even more important for project sponsors to remove market risk for a significant portion of the plant output.
And tax regimes need to be predictable. As of now, the 15 percent tax rate on dividends is scheduled to expire in 2010. Any reversion to standard income tax rates would have a major negative impact on the stock prices of companies that pay high dividends, including utilities. This in turn would raise their cost of equity, and also likely would limit market access during any period of uncertainty. In a similar vein, the extension of production tax credits is clearly of fundamental importance to further investment in renewables.
The track record of utilities in managing large capital expenditure programs has been mixed. The last round of nuclear-plant construction showed the downfall of customizing designs to meet the preferences of individual utilities. If the experience of the next generation of nuclear plants is to be different, utilities need to embrace the benefits of standardization. Constellation Energy has talked of developing a fleet of new EPR reactors that are identical, down to the “carpeting and wallpaper.”
The July 2007 edition of Public Utilities Fortnightly (“Avoiding the Next Debacle” by Rilck Noel and Terrel LaRoche) also spoke of the need to apply traditional project finance and project-management disciplines. The effective identification of risks and their allocation to parties best able to manage them is critical to funding the industry’s capital needs at reasonable rates.
Consistent with this, utilities need to manage their balance sheets in a way that recognizes the political realities of rising customer bills. In particular, a strategy of pursuing a series of rate increases while increasing financial leverage and rewarding shareholders through above-trend dividend increases and share buybacks is unlikely to find favor with regulators and politicians. Furthermore, aggressive debt-funding of capital expenditure programs may trigger downgrades from credit rating agencies, which will raise the cost of debt and also may hinder market access in volatile credit markets. Conservative funding practices will both preserve balance sheet strength and increase the probability of regulated utilities achieving the rate increases that they need.
To reduce risks for their shareholders and other stakeholders, utilities should explore ways of teaming up with partners. Many of the large projects previously constructed by the utility industry involved two or more companies coming together as co-owners. Partnerships can apply either directly to the projects being financed, or to other non-core areas where much-needed capital is tied up.
One interesting example is the recent announcement by Constellation Energy and Électricité de France (EDF) of a joint venture for developing next-generation nuclear facilities in North America. EDF will contribute up to $625 million in cash to the joint venture, UniStar Nuclear Energy, thereby reducing Constellation’s future funding commitments. In addition, Constellation’s risk profile should be reduced via EDF’s experience as the largest operator of nuclear plants in the world, and its direct experience of new EPR construction from its plant at Flamanville in France. Rothschild acted as financial advisor to Constellation on the transaction.
Another interesting example, involving the freeing up of capital from “non-core” areas, involves the announced commodity trading joint venture between Sempra Energy and Royal Bank of Scotland. The transaction will allow Sempra to free up over $1 billion of working capital from its trading business, which can be used for reinvestment in the business and/or for balance sheet management, including share repurchases.
Many infrastructure funds actively are looking for opportunities to invest capital in long-term, low-risk asset classes that allow them to match their long-term liabilities effectively. The utility sector is a natural fit for them. Where regulatory frameworks allow, they are interested in investing directly in specific projects. Many of the funds also are interested in acquiring whole utilities (following Macquarie’s acquisition of DQE), with subsequent rate base investment opportunities allowing them to average down their effective purchase price as a multiple of rate base.
To meet the challenges of financing their multibillion dollar capital expenditure programs, utilities will need to thoroughly evaluate their existing use of capital to ensure that it is being allocated most efficiently. For example, utilities that continue to own international businesses may consider monetizing these to help fund investments in their core business closer to home.
Mergers & Acquisitions
The funding of large capital expenditure programs is especially daunting for small and medium-sized utilities, which typically find it difficult to attract attention from Wall Street and are the first to suffer when conditions in capital markets deteriorate, as they have done recently.
If Wall Street firms decide to reduce their exposure to the sector, whether because of “hung bridges” or a more general reduction in risk appetite, it is likely to be the smaller companies who feel the pain first. These same companies also will suffer from the fact that individual capital projects likely will be lumpier and have a bigger impact on their cash flow and leverage ratios than they would for a bigger company. It is noteworthy that both DQE in its sale to Macquarie, and NorthWestern in its failed sale to Babcock & Brown Infrastructure, cited access to capital for future investments among their reasons for pursuing sale transactions.
Utilities with lower credit ratings also will feel increasing pressure to merge. In the current credit downturn, as in previous cycles, the cost of debt has increased and access to capital markets has deteriorated more for companies with weak credit ratings than for those with strong investment-grade ratings. The prime consideration for stronger companies contemplating mergers will be to avoid dragging down their own credit ratings and impairing their access to capital through any such mergers.
The need to fund large capital expenditure programs should therefore accelerate consolidation among utilities, with large well-capitalized entities being the partners of choice. But even for mergers between smaller companies, credit rating agencies generally attribute some benefit to size. They also see benefit in the increased diversification and lowered risk profile that often come with mergers.
Even with progress on all the issues above, some of the largest and most complex projects—in particular new nuclear plants—are unlikely to proceed without government support. The risk/return tradeoff for the private sector is simply not attractive enough without such support, given the large sums of money involved, potential technology risks and extended construction timetables. From a public policy point of view, government support can be justified by the contribution that nuclear and other clean generating technologies can make to reducing greenhouse gases and improving the energy security of the United States.
Government support can come most tangibly through a well-structured loan-guarantee program, which does not leave a slice of lender exposure uncovered or subordinate commercial debt to the Department of Energy’s lien. But it also can come in other ways, such as through a streamlined licensing process, and protection against the risk of delays arising from a failure to obtain regulatory approvals in a timely fashion, both of which reduce the risk of a repeat of the experience of the last cycle of nuclear plant construction. From the perspective of appropriate risk allocation, private sector investors never will be able to evaluate, price or absorb such risks as effectively as the public sector.
In conclusion, the need for many hundreds of billions of dollars in capital expenditures creates huge opportunities and challenges for both regulated and unregulated companies. Governments and regulators need to create a supportive climate for investment, in which the rules are clear and in which risks are allocated to those who are best placed to manage them. Customers will have to deal with higher bills if they are to enjoy continued safe and reliable service and lower emissions.
And companies in the sector will need to find ways to manage the risks that are within their control, allocate scarce capital efficiently, and evaluate partnerships and mergers so that they have access to capital through bad times as well as good times. Successful execution of the program will result in a high-quality utility infrastructure for customers, earnings growth for utilities, attractive returns for investors, and a cleaner environment for all citizens.