Natural Gas And The Electric Power Sector: What Are The Latest Trends?

by: Bluegold Research

Electric Power sector is the primary consumer of natural gas.

Coal prices plunge, natural gas prices follow.

Coal-to-gas switching outlook was revised higher but not by as much as was projected before.

The total stock of natural gas-fired power plants continues to increase.

Total natural gas balance in June will be looser than last year by around 4.0 bcf/day.

The injection season is in full swing. June contract expired last week and July contract is now the prompt month contract. It is down more than 17% y-o-y and is fundamentally undervalued. However, the whole energy complex has been under bearish pressure lately - partly driven by a classical risk-off trade. Additionally, a 20% drop in the price of Central Appalachian coal has dented natural gas competitiveness on a relative basis.

Will electricity-driven demand from the electric power sector prop up natural gas price during this injection season? In this article, we will look at natural gas usage in the Electric Power sector and will update you on the latest trends.

Why should natural gas traders care about the latest trends in the Electric Power sector (EP sector)?

The simple answer is because the EP sector is the primary consumer of natural gas. Its share in the annual demand structure is more than 31% (on a 12-month basis), while its share in the injection season demand is close to 50% (see the charts below).

Source: Energy Information Administration, Bluegold Research estimates, and calculations

Source: Energy Information Administration, Bluegold Research estimates, and calculations

Unlike heating-driven demand from residential and commercial users, the electricity-driven demand from the EP sector is much more sensitive to price changes. It is precisely this high price elasticity of demand that makes trading during the injection season more price-dependent and less weather-dependent. High price-elasticity of natural gas demand during the injection season is also the primary reason for another, subtler phenomenon. History shows that the lowest prices during the injection season have been higher than the lowest prices during the withdrawal season (see the table below).

Source: CME Group, Bluegold Research estimates, and calculations

As you can see, the average of injection season "price minimums" is higher than the average of withdrawal season "price minimums." Indeed, we have seen cases in the past when warm winters pushed natural gas prices as low as $1.611 per MMBtu, yet we have never seen cases (at least in the recent past) when cool summers pushed prices below $2.000 per MMBtu (for summer contracts). That is because cheap natural gas during the summer is much more supportive of consumption than cheap natural gas during the winter. By the way, it's also the reason why the highest prices during the injection season have always been lower than the highest prices during the withdrawal season.

What indicators should traders monitor?

Here's a list of four key indicators that we think should be monitored:

  1. The spread between natural gas and coal (NG/coal spread) - measures the competitiveness of natural gas usage in the Electric Power sector vis-à-vis its closest substitute, coal; specifically, Central Appalachian coal (bituminous) and Powder River Basin coal (sub-bituminous).
  2. Coal-to-gas switching - is a displacement of coal-fired generation by natural gas-fired generation due to short-term fuel price competition. Coal-to-gas switching has a positive impact on total natural gas demand. It occurs because of lower natural gas prices relative to coal. The lower the NG/coal spread is, the stronger is the impact of coal-to-gas switching and the higher is the consumption of natural gas in the EP sector.
  3. Power plants outages - electricity generation offline due to maintenance (either planned or unplanned).
  4. Generation capacity - the amount of natural gas-fired power plants measured in gigawatts.

Let's now look at each of these indicators in more detail.

NG/Coal Spread

Coal is the closest "substitute" to natural gas when it comes to electricity generation. The relationship between these two commodities becomes particularly important during summer months when heat drives air conditioning use higher. Because of the substitution effect, coal prices and natural gas prices should exhibit some correlation. In other words, natural gas prices cannot increase independently of coal prices (at least in theory and especially during the summer). Recently, this correlation has weakened due to structural changes in the market (retiring coal power plants, new additions of gas-fired capacity, environmental regulations, surging production of shale gas, etc.), but because coal power plants still generate around a quarter of all electricity produced in the U.S., the link between coal and gas remains relevant.

Source: CME Group, Energy Information Administration

As you can see from the chart above, natural gas has almost always been more expensive than coal (on USD per MMBtu basis), but it does not mean that generating electricity by burning coal is more economically viable. Some types of coal are exceptionally cheap - particularly, Powder River Basin type (also known as Western coal or "PRB") simply because they are very low in energy content. Furthermore, some types of coal are notoriously dirty and have high "environmental costs". Conversely, commercialized natural gas is practically sulphur free and produces virtually no sulphur dioxide (SO2) - see the table below.

Source: EIA, International Gas Union, World Nuclear Association

However, the prices for two of the most energy intensive coal types - notably, Central Appalachian and Northern Appalachian - have recently plunged. In the week ending May 31, Central Appalachian coal lost as much as 20% in value, The EIA reported a weekly drop of $17.65 to $58.80 per short ton, which was the largest weekly decline that we have seen in the past seven years.

Unfortunately (for natural gas bulls), cheap coal is denting natural gas competitiveness on a relative basis. Therefore, while our coal-to-gas-switching outlook was revised higher (after Friday's massive drop in natural gas futures), it was not revised as high as we previously estimated.

When estimating the spreads between natural gas and coal (NG/Coal Spreads - see the charts below), it is important to remember that natural gas-fired power plants tend to be more efficient than coal-fired power plants. In other words, the heat rate (measured in BTU per kilowatt-hour) is lower for natural gas-fired power plants than it is for coal-fired power plants. Finally, when adjusting for transportation and environmental costs, it becomes evident that producing electricity from natural gas is a more efficient, a more economically viable and a more environmentally-friendly business. That is why there are no plans to build new coal-fired power plants in the U.S.

Source: CME Group, Energy Information Administration, Bluegold Research estimates, and calculations

Despite cheaper coal, we have calculated that NG/Coal Spread (when adjusted for all the factors mentioned above) is still below historical averages. Currently, the benchmark spread is as much as 40% below 5-year average and 22% below last year's level, which is rather unusual for this time of the year when cooling degree-days are rising. In the simplest of terms, it means that:

  • natural gas is still cheap vis-a-vis its closest substitute
  • power plants will increasingly turn to burning natural gas (as opposed to burning coal), thereby driving coal-to-gas switching higher and, therefore, pushing natural gas consumption in the EP sector higher

Coal-To-Gas Switching

Below are three key points to remember about coal-to-gas switching:

  • Lower natural gas prices (relative to coal) lead to higher levels of coal-to-gas switching (and vice versa)
  • The lower the price > the higher is the level of coal-to-gas-switching > the greater is total consumption (specifically in the Electric Power sector) > the greater is the total demand > the stronger is the "bullish pressure" on the EOS storage index
  • The economics of fuel-switching is an important element in natural gas trading, but only during the injection season (roughly, April - September)

Source: CME Group, Energy Information Administration, Bluegold Research estimates, and calculations

Our analysis shows that the level of coal-to-gas switching is already running at no less than 7 bcf/day, which is 1.6 bcf/d above 5-year average. At current natural gas and coal forward prices, this level of coal-to-gas switching could potentially rise to 8.0-8.5 bcf/day this summer. History shows that natural gas price rarely drops below $2.500 per MMbtu during the injection season whenever coal-to-gas switching is at least 1 bcf/d above 5-year average. However, as we have seen last Friday, history does not always repeat itself.

Power Plants Outages

Power generation facilities, both conventional fueled steam-generation and nuclear-powered plants require vigilant, well-organized operation using meticulous maintenance management to stay online and produce energy safely and efficiently. When the plant is undergoing a maintenance, it does not produce any electricity. This is known as "outage". Nuclear outages are often replaced by natural gas-fired generation. Therefore, there is a positive correlation between natural gas outages and natural gas consumption in the EP sector.

U.S. Nuclear Regulatory Commission reported that as of today, there are a total of 8,500 MW of nuclear power generation offline (+2,900 MW from Friday, +18.9% vs. 5-year average). Above normal nuclear outages should provide an additional boost to consumption in the Electric Power sector.

Source: U.S. Nuclear Regulatory Commission, Energy Information Administration, Bluegold Research estimates, and calculations

Generation Capacity

The total stock of natural gas-fired power plants continues to increase. As of March 2019, it stood at 461.15 GW, which at that time accounted for 42.9% of total generating capacity in the U.S. In June of this year, natural gas share in the total fuel mix is expected to rise to 43.1% (+0.5 percentage points, y-o-y) and equal just around 464.70 GW (see the chart below).

*Total annualized net effect on gas usage from changes in generation capacity = natural gas net additions + coal retirements - natural gas retirements - coal additions - nuclear additions - wind, hydro and solar additions + retirements of renewables and nuclear = +2,922 MW of natural gas-fired generation in June 2019 (vs. June 2018).

Source: CME Group, Energy Information Administration, Bluegold Research estimates, and calculations

Indeed, major additions are planned for June this year when 2.1 GW of new natural gas-fired capacities will be commissioned - most notably, St. Charles Power Station (1,000 MW) in Louisiana. Another important development this year is the planned retirement of Pilgrim Nuclear Power Station (680 MW). Overall, no less than 11 GW of nuclear capacities is expected to retire in the next 5 years while only 2 GW of nuclear capacities will be added over the same period. Other sources - particularly, natural gas and renewables - will have to replace the loss of nuclear generation.


Notice also that the share of "other renewables" (wind and solar) has already overtaken hydro and nuclear (see the chart above). Previously, in an attempt to estimate the levels of potential natural gas consumption in the EP sector, analysts would look at the schedule of power plants outages to try to figure out how many nuclear and coal megawatts will be replaced by natural gas. They would also study the level of snowpack to estimate hydro inflows and eliminate it from total calculations.

Now, however, analysts must also study wind speeds and the levels of solar radiation since the influence of "other renewables" can no longer be ignored. In this regard, please note that out of 12 calendar months, June has historically been one of the strongest months for renewable power (see full ranking in the chart below). Also, notice that electricity generation from renewable sources normally declines during the injection season.

Source: Energy Information Administration, Bluegold Research estimates, and calculations

Net Generation

The latest data indicates that the weight of the Electric Power sector in the natural gas market continues to grow. EIA has recently released its regular Electric Power Monthly Report. It shows that in March 2019, the share of total electricity supplied by natural gas-fired power plants increased by more than two percentage points y-o-y (from 32.69% to 34.83% - see the chart below). It is the highest share of natural gas-fired generation for the month of March ever recorded. At the same time, the share of coal has declined by just under one percentage point over the same period to 24.28%.

On a twelve-month average basis, net electricity generation from natural gas has reached a new all-time high - 35.75%. At the same time, the share of coal has dropped to a new all-time low of 26.88%.

Source: Energy Information Administration, Bluegold Research estimates, and calculations

Total Supply/Demand Balance

Overall, the fuel substitution element in our consumption models remains bullish for natural gas prices (ceteris paribus). However, the net effect on natural gas consumption should be smaller because there are other elements within the EP natural gas consumption model, which have both positive and negative implications.

Electric Power natural gas consumption model = NG-Coal spread + coal-to-gas switching curve + nuclear outages + coal outages - gas outages - hydro/wind/solar generation.

In addition, when we factor in other market variables such as production, imports, exports, and consumption by other users, we estimate that total natural gas balance in June will be looser than last year by around 4.0 bcf/day. In absolute terms, this is a bearish development, but natural gas price is already down more than 17.00% y-o-y, so a lot of bearishness has been priced in.

Total supply-demand balance in June 2019 = dry gas production (90.20 bcf/day) + imports (7.10 bcf/day) - consumption (69.80 bcf/day) - exports (11.30 bcf/day) = 16.20 bcf/day vs. 12.40 bcf/day in June 2018. Please note that we update our forecast for all market variables on a daily basis.

Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.