By now, the evidence is overwhelming. Utility-industry construction costs have risen and will remain elevated for some time.
Some of the factors underlying these trends are straightforward. For example, costs for steel, copper and concrete have risen sharply due to high global demand, as well as production and transportation costs (in part owing to high fuel prices), and a weakening U.S. dollar. Other drivers are less transparent. Labor costs generally have tracked inflation rates, but shortages in skilled workers have driven costs higher for utility equipment and construction services.
Moreover, constraints in component-manufacturing capacity as well as engineering, procurement and construction (EPC) services exacerbate cost pressures. In January 2007, for example, OG&E executives reported that the cost estimate for EPC services for building the company’s proposed Red Rock coal-fired power plant increased by more than 50 percent in just nine months, from $223 per kilowatt to $340/kW.1
Although customers will not see the full rate impact associated with construction cost increases until infrastructure projects are completed, these increases now are affecting industry investment plans and presenting new challenges to regulators.
The recent rise in many utility construction cost components follows roughly a decade of relatively stable (or even declining) real construction costs, adding to a growing sense of sticker shock among power companies and state regulators.
Moreover, these increased costs are largely absent from the capital costs specified in the Energy Information Administration’s (EIA) 2007 Annual Energy Outlook (AEO), leading to a substantial divergence between EIA’s data assumptions and market evidence. For example, the AEO estimates construction costs for advanced nuclear plants at just over $2,000/kW, but a recent report from Moody’s Investors Service forecasts costs between $5,000 and $6,000/kW—three times the EIA figures.2
To provide reliable indicators of current or future capital costs, the Edison Foundation commissioned the Brattle Group to study recent increases in the costs of building utility infrastructure—including generation, transmission and distribution facilities. The study also identified the causes of these increases and explained how these increased costs will translate into higher consumer rates.3
The overall effects will be borne out in various ways, depending on how utilities, markets and regulators respond to these cost increases.
Construction-cost inflation during the past several years has reached every corner of the electric utility industry.
Infrastructure costs were relatively stable during the 1990s. But between January 2004 and January 2007, prices increased rapidly. Costs for steam-generation boilers, transmission facilities and distribution-grid equipment rose by 25 percent to 35 percent, compared to an 8 percent increase in inflation, expressed by the GDP deflator4(see Figure 1, “National Average Utility Infrastructure Cost Indices”).
The cost of gas turbines increased by 17 percent during 2006. Similarly, prices for line transformers and pad transformers increased by 68 percent and 79 percent, respectively, between January 2004 and January 2007, with increases during 2006 alone of 28 percent and 23 percent.5
These rapid cost increases have raised the price of recently completed infrastructure projects. To the extent services and materials were acquired before the most recent inflationary trends, the effect has been mitigated somewhat. Rising prices have a more dramatic effect on the estimated cost of proposed projects, which fully include the recent price trends (see sidebar, “Ballooning Project Costs”).
As a result, utilities and regulators increasingly are worried the next wave of utility investments might cause rates to increase significantly. Rising construction costs and recent increases in wholesale power prices have motivated industry participants to more actively pursue energy-efficiency and demand-response initiatives, to reduce future consumer-rate increases. Nevertheless, economic growth and the need to replace aging infrastructure will necessitate major new investments during the next two decades.
According to EIA’s most recent projections, U.S. electricity sales are expected to grow by about 1.4 percent each year through 2030, and the North American Electric Reliability Corp. (NERC) forecasts peak demand will grow by 19 percent, or 141 GW, from 2006 through 2015. EIA predicts power companies will need to build 258 GW of new generating capacity by 2030 to meet demand growth and replace plants that will be retired.
Likewise, the high-voltage transmission grid requires significant investment. After a long period of decline, transmission investment began a significant upward trend starting in the year 2000. Since then, the industry has invested more than $37.8 billion in the nation’s transmission system, and a recent Edison Electric Institute (EEI) survey suggests its members plan to invest $31.5 billion in the transmission system from 2006 to 2009. NERC anticipates nearly 13,000 miles of new transmission will be added by 2015, an increase of 6.1 percent in the total miles of installed extra high-voltage (EHV) transmission lines (230 kV and above) in North America between 2006 and 2015.
Similarly, distribution-system investments began rising in the mid-1990s, preceding the corresponding boom in generation, and the flow of distribution investments shows no sign of diminishing. In 2006, utilities invested more than $17 billion to upgrade and expand distribution systems, a 32 percent increase over investments in 2004. EEI estimates distribution investment during 2007 will again exceed $17.0 billion.
While much of the recent increase in distribution investment reflects expanding physical infrastructure, a substantial portion of this investment reflects the increased input costs of materials and labor. Cost estimates likely will increase further if market trends persist.
Using commercially available databases and other sources, such as financial reports, press releases and government documents, the Brattle Group collected data on installation costs for natural-gas-fired combined-cycle generating plants brought into service in the United States between 2000 and 2006, and found the average real construction cost was approximately $550/kW in 2006 dollars, with a range of costs between $400/kW and about $1,000/kW. Statistical analysis confirmed real installation cost was influenced by plant size, the turbine technology, the NERC region in which the plant was located, and the commercial online date.
Notably, the data showed a positive and statistically significant relationship between a plant’s real construction cost and its online date, meaning that, everything else equal, the later a plant was brought online, the higher its real installation cost.6 The average installation cost increased gradually from 2000 to 2003, followed by a fairly significant increase in 2004 and a very significant escalation—more than $300/kW—in 2006. This provides vivid evidence of the recent sharp increase in plant-construction costs.
Another major class of generation development during this decade has been wind generation, the costs of which also have increased in recent years. The Northwest Power and Conservation Council (NPCC) issued its most recent review of the cost of wind power in July 2006.7 The Council found the cost of new wind projects rose substantially in real terms in the last two years, and was much higher than assumed in its most recent resource plan. Specifically, the Council found the construction cost of wind projects, in real dollars, has increased from about $1,150/kW to $1,300-$1,700/kW in the past few years, with an unweighted average capital cost of wind projects in 2006 at $1,485/kW. The average cost of wind power plants now being developed is still higher, with construction costs estimated from $1,700/kW to $2,000/kW.8
Broadly speaking, four factors are driving rising costs for utility infrastructure: 1) material costs, including such commodities as steel and cement, as well as manufactured components; 2) limited shop and fabrication capacity for manufacturing major components; 3) costs for construction field labor, both unskilled and craft labor; and 4) the market for large construction-project management and EPC services.
Utility construction projects involve large quantities of steel, aluminum and copper (and components manufactured from these metals) as well as cement for foundations, footings and structures. All these commodities have experienced substantial recent price increases, due to increased domestic and global demands as well as increased energy costs in mineral extraction, processing and transportation (see Figure 2, “Raw Materials Costs”). In addition, since many of these materials are traded globally, the recent performance of the U.S. dollar affects domestic costs.
In particular, various sources point to the rapid growth of steel production and demand in China as a primary cause of the increases in both steel prices and the prices of steelmaking inputs.9 Today’s steel prices remain at historically elevated levels and likely will remain high for the near future.
Other metals important for utility infrastructure display similar price patterns: declining real prices over the first five years or so of the previous 10 years, followed by sharp increases in the last few years. These price increases also were evident in other metals—such as nickel and tungsten—that contribute to steel alloys used broadly in electrical infrastructure. Prices for wire products have spiked compared to the inflation rate, highlighting the impact of underlying metal price increases (see Figure 3, “Electric Wire and Cable Price Indices”).
In addition to metals, large infrastructure projects require huge amounts of cement as well as basic stone materials. And the price of these commodities has risen substantially in the past few years, for the same reasons cited above for metals. Cement in particular is an energy-intensive commodity that is traded in international markets, and recent price patterns resemble those displayed for metals.
Likewise, prices for plant components, such as large pressure vessels, condensers, pumps and valves, have risen sharply since 2004. While equipment and component prices reflect underlying material costs, some price increases and delivery lags are driven by manufacturing capacity constraints. Recent orders largely have eliminated spare shop capacity, and delivery times for major manufactured components have risen as a result.
To the degree delays in component deliveries cause construction schedules to lengthen, financing costs likewise will increase—with commensurate effects on overall plant costs.
A significant component of utility construction costs is labor—both unskilled (common) labor as well as craft labor, including pipefitters and electricians. Labor cost increases—while less dramatic than those experienced by commodities—nevertheless have exceeded the general inflation rate (see Figure 4, “National Average Labor Costs”).
Specifically, between January 2001 and January 2007, overall inflation caused general prices to rise by about 15 percent. During the same period, the cost of craft labor and heavy construction labor increased about 26 percent, while common labor increased 27 percent.10
Although labor costs have not risen dramatically in recent years, utilities increasingly are concerned about an emerging gap between demand and supply of skilled construction labor—especially if the anticipated boom in utility construction materializes. The average age of the current construction skilled workforce is rising rapidly, and high attrition rates in construction are compounding the problem.
The industry always has suffered high attrition at entry-level positions, but now many workers in the 35 to 40 year-old age group are leaving the industry for a variety of reasons. As a result, the construction industry must recruit 200,000 to 250,000 new craft workers each year to meet future needs. Both demographics and a poor industry image are working against the construction industry as it tries to address this need.
Similar issues might affect the supply of electrical lineworkers who maintain the electric grid and perform labor for T&D investments. DOE forecasts qualified candidates might fall short of requirements by as many as 10,000 lineworkers, or nearly 20 percent of the current workforce.11 Such shortages likely will place upward pressure on the wages earned by lineworkers.
Finally, conditions in the market for EPC services are driving major cost increases. While the Brattle Group was unable to obtain specific information from the major EPC firms on their worldwide backlog of electric utility infrastructure projects, these companies’ financial statements specify the financial value associated with their backlog of infrastructure projects.
The cumulative annual financial value associated with the backlog of infrastructure projects at four major EPC firms—Fluor Corp., Bechtel Corp., The Shaw Group and Tyco International—rose sharply between 2005 and 2006, from $4.1 billion to $5.6 billion, an increase of 37 percent (see Figure 5, “Annual Backlog at Major EPC Firms”). This significant increase in the annual backlog of infrastructure projects at EPC firms is consistent with the data showing an increased worldwide demand for infrastructure projects in general, including utility generation, transmission, and distribution projects.
Growth in construction project backlogs likely will dampen the competitiveness of EPC bids for future projects, at least until the EPC industry is able to expand capacity to manage and execute greater volumes of projects. Although difficult to quantify, this lack of spare capacity in the EPC market undoubtedly will inflate the price of new bids for EPC services and contracts.
These factors, as well as the other inflationary pressures beyond the utility industry’s control, have contributed to an across-the-board increase in the costs of investing in utility infrastructure—and those higher costs show no immediate signs of abating.
As a result of the undeniable need for additional infrastructure, utilities and non-utility developers will continue investing in baseload generation, environmental controls, transmission projects and distribution systems. However, rising construction costs will put additional upward pressure on retail rates over time, and may alter the pace and composition of investments going forward. For example, the increasing fixed costs of base-load coal and nuclear facilities have reduced the cost savings the industry anticipated from expanding the solid-fuel fleet.
The overall impact on the industry and on customers will be borne out in various ways, depending on how utilities, markets and regulators respond to these cost increases. In the long run, customers ultimately will pay for increasing construction costs. Most directly, these costs will result in higher rates to recoup asset investments, and less directly, higher energy-market prices to attract new generating and transmission capacity in organized power markets. And customers will pay indirectly, when rising construction costs inevitably defer investments and delay expected benefits, such as enhanced reliability and lower, more stable long-term electricity prices.
1. Testimony of Jesse B. Langston before the of the Oklahoma Corporation Commission, Cause No. PUD 200700012, Jan. 17, 2007, p. 27 and Exhibit JBL-9.
2. “New Nuclear Generation in the United States: Keeping Options Open vs. Addressing an Inevitable Necessity,” Moody’s Investors Service, Oct. 10, 2007.
3. Chupka, Marc W. and Basheda, Gregory, Rising Utility Construction Costs: Sources and Impacts, Edison Foundation, September 2007.
4. The GDP deflator measures the cost of goods and services purchased by households, industry and government, and is a broader price index than the Consumer Price Index (CPI) or Producer Price Index (PPI), which track the costs of goods and services purchased by households and industry, respectively.
5. Ibid. Chupka.
6. To be precise, the authors used a “dummy” variable to represent each year in the analysis. The year-specific dummy variables were statistically significant and uniformly positive; i.e., they had an upward impact on installation cost.
7. “Biennial Review of the Cost of Windpower” July 13, 2006.
8. Annual Report on U.S. Wind Power Installation, Cost, and Performance Trends: 2006, U.S. DOE, May 2007, pp. 15-16.
9. Steel: Price and Policy Issues, CRS Report to Congress, Congressional Research Service, Aug. 31, 2006.
10. These figures represent a simple average of six regional indices. However, local and regional labor markets can vary substantially from these national averages.
11. Workforce Trends in the Electric Utility Industry: A Report to the United States Congress, Pursuant to Section 1101 of the Energy Policy Act of 2005, U.S. Department of Energy, August 2006, p. xi.