Long Term Service Agreements – past performance and future viability

This is Part 3 in a four-part series of mini blogs concerning Long Term Service Agreements in the power industry.

Power Generation Expert

By Bill Ray and Craig Nicholson

Previously we’ve discussed why parties have traditionally agreed to such maintenance contracts, followed by their basic make-up. In this blog we will review how LTSAs have held up, given the market dynamics over the last 15 years, and discuss whether LTSAs are still a viable option going forward.

Looking back over the last decade and a half, the vicissitudes of the energy markets — in particular the interaction between coal and gas plant generation — have fostered structural changes and shifts in the U.S. power markets. These factors, coupled with the acceleration of affordable renewable technologies, have created uncertainty for gas plant owners.  At a high level, the advent of the F class gas turbine can be categorized into the periods of rising gas price, low gas price and the entrance of renewables.


Many U.S. high technology gas plants were developed and installed between 1999 and 2004 with the expectation of stable natural gas pricing that would be competitive in a generating environment dominated by coal.  However, that assumption proved short lived. The year 2003 saw average Henry Hub natural gas prices at $5.47 / MMBtu, which increased some 60% to a peak of $8.86 / MMBtu in 2008.

To understand the ramifications of gas prices on the electricity market, we’ll look at how the power markets function. In many of the 10 power markets in the U.S., wholesale power prices are set through a competitive auction bid. Energy is bid the day prior to operation.  These auction-based markets determine the ‘clearing price’ for the next day based on the ‘marginal operator’. The marginal operator is the last unit called for dispatch to run, and, as the most expensive operator to clear the auction to meet demand, sets the price for all generation that day. From 2003 to 2008, coal-fired generation provided about 50% of U.S. power generation and would typically set the clearing price in many markets. All generators that cleared the auction (that is, bid under the clearing price) would get paid at that clearing price set by coal generators.  With approximately 75% of gas plant operating expense associated with fuel, as fuel price rose, profit compressed.

Some regulated utilities could shield exposure with the ability to rate case fuel costs as an expense to address rising fuel cost; however, such pass-through costs are subject to timing of rate cases and the regulatory procedures determined by the respective state’s public utility commission or the gas plant could fall to the bottom of the regulated utility dispatch stack.

The balance of power producers earn their money through bilateral power purchase agreements with larger utilities, and in some cases leveraging the larger utilities’ gas supply network and passing through gas costs in tolling type agreements. Nevertheless at some point in the supply chain, one of the entities is picking up the fuel cost risk or reward. In some cases, both entities could share the risk and reward through incentive mechanisms in the power purchase agreement between the parties.

The result of the fuel price rise was a sharp decline in gas plant operating hours and in some cases, plant insolvency.

“With approximately 75% of gas plant operating expense associated with fuel, as fuel price rose, profit compressed”


The U.S. fracking boom yielded a seven-fold increase in natural gas supply and an average Henry Hub monthly price of about $3.40 / MMBtu starting in 2009.  Sharply lower gas prices beginning with the end of 2008 through 2018, along with the closure or retirement of less competitive or environmentally challenged coal plants, drove the decline in U.S. coal generation from 50% to 30%.  Gas plants filled the gap and became the predominant marginal operator, with low gas prices and high efficiency gas turbines now directly affecting the price of electricity by lowering the wholesale price in competitive energy markets.  Wholesale electricity prices at major trading hubs on a monthly average basis for on-peak hours were down 27%-37% across the nation in 2015 compared with 2014, driven largely by gas plant cost structure.

Lower wholesale energy pricing has a negative effect on plant profitability by depressing revenue per unit of electricity relative to the cost of fuel and lowering gross margins.  Tighter gross margins led to greater scrutiny of operational and maintenance budgets and a desire for differentiating performance upgrades to increase gross margin in a market dominated by similar generating technology.


Although renewable technology has existed since the ‘60s and ‘70s, adoption of wind technologies started in states such as Texas, aided by favorable public utility commission programs supporting transmission infrastructure development.  In 2013 Texas was the first state to exceed 10,000 MW of wind power generating capacity and continued to expand its wind portfolio to 22,000 MW by 2017. Texas still leads the nation in wind power; however states such as Iowa, Oklahoma and California have all grown their wind portfolio over a comparable period.

California has led the way in solar technologies and became the first state with more than 5% of its annual utility-scale electricity generation from solar power in 2015. The state promoted solar power through a series of state policies, including a renewable portfolio standard (RPS), which today stands as one of the more ambitious renewable energy policies in the nation. California continues to lead the way, with the state committing to become 50% renewable by 2020 and 100% by 2032.  California is on track to achieve its commitment, meeting 32% of 2017 electricity needs from renewable energy. Solar portfolios continue to grow rapidly in multiple states such as North Carolina, Arizona and Nevada.

The influx of renewable technologies — both centralized utility scale and decentralized commercial scale — has changed the way the power grid operates, translating to a change in demand on thermal generators. This is driven primarily by the intermittent nature of renewable technologies, which now have a material impact on the grid and the need to match demand with supply in real time. Thermal generators are taking the role of balancing supply and demand, as well as absorbing renewable energy new capacity additions in a relatively slow load-growth environment.  The rise of renewable energy in this newly functioning grid will pose both technical and financial challenges on existing thermal capacity.


Have LTSAs in this period served their intended purpose? Have they stood the test of time or have they compounded issues as markets have changed, resulting in out-of-sync maintenance costs that were negotiated and agreed upon during prior market conditions?

The efficacy of an LTSA will depend on the unique circumstances of the organization, power plant and local power markets, as well as the timing and substance of the LTSA itself.  Initial LTSAs contracted during the F class build-out tended to be quite beneficial to the OEM with simplified assumptions of plant operation. Conversely the owner’s plant operations tended to benefit from the support while learning the idiosyncrasies of the plant.

Typical LTSA payment terms included some combination of a time-based fixed payment such as a monthly or quarterly fee, a fee per hour of operation and a milestone payment typically based around a planned maintenance event.

Such payment structures for a power plant operating regularly at full load, achieving in excess of 6500 hours a year, is acceptable, given power generation revenues offset the LTSA payments. Revenues are in sync with O&M costs, and O&M costs are covered by the gross margins made in the time period they are paid.

If; however, a plant changes its operational profile during the term of the LTSA to run at various partial loads with a lower number of hours per year and higher number of starts, this payment scheme may become financially problematic for the owner.  

In a typical combined cycle F class gas plant that changes its operational profile as described above, each start contains non-electricity producing start up time until reaching minimum load.  Minimum load would be the break point at which the plant starts to make money by producing enough electricity to offset costs. Based on plant configuration, a start may take anywhere from 10 minutes for a fast start-capable unit to several hours.

It is worth noting that typical startup costs are in the region of $15,000 to $20,000, which must be recouped in the subsequent run before turning a profit.  Additionally, O&M charges may be increased for each given start in the form of a multiplier on the hours of operation to produce an ‘equivalent operating hour’ — the theory being failure mechanisms related to starts correlate with failure mechanisms related to run time. Such hourly calculations are used for billing based on a gas turbine equivalent of an odometer.  The resulting starts are costly, and an LTSA may compound this issue by charging a premium. Starting costs and increased O&M charges will ultimately erode plant profits for such a given operational profile compared to continuous operation.

The power generation landscape has transformed over the last 15 years, and the pace of change continues to accelerate

Moreover, the plant operates at partial loads during start-up and, at times, during sustained operation due to grid requirements. At part load, the plant owner garners less revenue per hour of operation as the plant is producing less energy (MWh). Compounding the issue is the fact that the plant is less efficient at part load, so the fuel bill per unit of energy increases. This creates additional downward pressure on profits on a reduced revenue profile during part- load operation. Typically, the O&M payment per fired hour does not account for partial load and stays a constant fee for each hour of operation. As a result, the hourly payment consumes a proportionately larger portion of the owner’s profits.

Additionally, if the plant runs less frequently and has a time-based payment scheme, the plant may not generate revenue in a given payment cycle; and the LTSA creates a negative cash flow for the owner.  The owner is paying for maintenance costs on a plant that is not running.

The above example describes a common operational profile for today’s combined-cycle utility gas plant.  Asset owners are subject to markets dynamics that are directly correlated to how they run their power plants and make money.  Markets and grid requirements shift; however, LTSA agreements tend to stay static which, over time, may become burdensome, as occurred during the period of rising gas prices.  The operating and market assumptions of the LTSA at its outset may become stale or inaccurate over time, leading to O&M costs becoming out of sync with the market and a financial burden to the owner.

There are lessons to be learned from prior agreements that can help determine the best maintenance strategy going forward. The power generation landscape has transformed over the last 15 years, and the pace of change continues to accelerate. This has a material impact on how owners should look at long term operations and maintenance planning. They should rethink and reconstruct the building blocks of the future LTSA, bringing innovation and best practices into agreements to counter the innovation and subsequent disruption the power industry faces.  

The LTSA landscape is ripe with various vendors and agreement options available to owners.  If they deem an LTSA a viable option, a well-crafted agreement can be beneficial to both parties, providing the necessary flexibility to navigate the future while providing a framework that is financially prudent, mitigates operational risks and anticipates a competitive future.

Part 4 in the mini-blog series addresses next generation LTSAs.

For additional discussion




Gas turbine photo attributed to pro-per energy services / CC-BY-SA-3.0

Leave a Reply