Windmills and Oil Wells

#3: The Trouble with Renewables (and How to Solve It)

When discussing energy policy and the topic of renewables, perspectives often fall along political or ideological lines. Renewables are either the essential replacement to fossil fuels that will save the planet from climate apocalypse, or a fairy dust illusion subsidized by the government and misguided politicians.

The truth lies somewhere in the middle. Wind and solar can indeed provide low carbon electricity at a lower cost than conventional fossil fuel generation. But there is a problem with renewables. It is not an issue of politics - left or right - but an issue of economics.

In fact, oil and gas and renewable developers are in the same underlying business - extracting energy from natural resources - with similar economic considerations and constraints. Understanding the similarities highlights the bottleneck that will stifle new solar and wind development until it is addressed. Renewables are in desperate need of pipelines.

Net Present Value

The economics of an oil well and a windmill or solar array are more similar than you may think. Both fossil fuel and renewable development involve substantial upfront capital expenditures which are paid back with a return over the life of the asset.

Before deciding to drill or construct, one must first make assumptions on the viability of the project. One needs land, a concentration of resource, and an outlet to sell the energy produced. From an economic perspective, one must estimate the initial capital expense, ongoing operating expenses, and of course revenue.

Before an oil company drills a well, it estimates drilling and completion (“D&C”) costs, oil production over the life of the well (a “type curve”), the ongoing maintenance and transportation costs (“lifting cost”), and finally, an assumption of oil prices. Combining these assumptions creates a series of cash flows used to calculate an expected Net Present Value (NPV) of the well. If the company finds this return attractive relative to its cost of capital and its portfolio of drilling opportunities, it will drill.

To build a windfarm, a renewable developer follows the same process. It will estimate the cost of construction (usually determined by the turbine manufacturer and construction contractor), the expected electricity production based on a scientific wind study, the cost of labor and maintenance of the turbines, and finally, the price of power. These factors determine the NPV of the windfarm. Positive NPVs will induce development.

Of all the assumptions, the most critical and difficult to predict is the long-term commodity price - be it oil, gas or electricity. This long dated tail of price dependent cash flows makes both businesses risky and cyclical. A great investment becomes a bad one if commodity prices collapse in the second year of the asset’s thirty year life.

Ultimately, higher NPVs are the key to self-sustaining development. Further, economically viable development is more socially beneficial than subsidized development, and subsidies merely rob Peter to pay Paul.

Natural Resource

Wind, solar, oil and gas, all harness a natural resource to generate energy. Natural resource is not evenly distributed, but rather, is highly concentrated regionally. While this point is generally appreciated with respect to oil and gas production, it is equally relevant to renewables.

Below is a map of shale basins in the United States produced by the U.S. Energy Information Administration (EIA).

Today, nearly all new U.S. oil and gas production comes from just a subset of these plays - the most productive and economic. The majority of new oil wells are located in Texas, in the Permian and to a lesser extent the Eagle Ford, while natural gas is produced from the Marcellus and Utica formations in Appalachia.

If you were to a bore a hole in the ground in Connecticut or South Carolina, you would be able to find hydrocarbons. They exist in the earth’s crust to some degree everywhere. Yet, you see no oil rigs outside of Charleston - for while some resource technically exists there, it is not economically viable to produce.

Instead, we concentrate our oil and gas development where it is most efficient to extract. The hydrocarbons produced in these regions are then transported through pipelines, rail and trucks, to meet energy demand nationwide.

Solar and wind resource is also not equally distributed, even though the sun shines and the wind blows everywhere. Below are maps of wind and solar resource in the U.S.

Left: Purple shading shows the highest wind speeds in the country. Right: Dark red shading shows the highest solar irradiance (source: EIA)

Wind resource is highly concentrated in the plains, while solar resource is concentrated in the desert Southwest. Conveniently, these are also locations with relatively cheap and abundant land for new development.

In these regions, wind and solar power are highly efficient, low carbon energy sources, with lower cost than conventional power generation from natural gas or coal. This is the great promise of renewables…

The Bakken Dilemma

Renewables today face The Bakken Dilemma, a term that may be self explanatory to oil and gas observers.

In the early 2010s, as new fracking technology opened up vast new regions for oil development, and with oil prices over $100 per barrel, there was a (black) gold rush. The Bakken, a tight oil play in North Dakota, was a land of promise in the early days of the shale revolution.

Small-time wildcat drillers and international oil companies flocked to the region, backed with enormous sums of capital to bid up land values and begin drilling. And produce they did. But the ramp in production far exceeded the takeaway capacity (pipelines, rail, trucks), and there was quickly too much oil stranded above ground in North Dakota.

Without adequate takeaway capacity, the difference in realized prices in the region sank far below the benchmark West Texas Intermediate (WTI) which is based on oil prices at Cushing, Oklahoma, a central hub of U.S. oil infrastructure. At peak dislocation, oil in North Dakota priced at a $30 per barrel discount to WTI. The regional price discount, known as “basis”, was devastating to producers in the region who correctly identified and extracted a great new resource; a tragedy of riches.

This is the challenge of resource extraction. New resource attracts development, increasing regional supply and reducing prices, particularly if takeaway capacity lags the production ramp. The gold rush crashes the market for gold. Drillers take a bath as their initial commodity price assumptions prove far too optimistic, stalling further production of a prolific resource.

This phenomenon was typified by the Bakken, but experienced in all of the major shale basins to some degree, even more acutely in the Marcellus. Production will be limited by the bottleneck in the system - and in the case of energy extraction, the bottleneck is often in takeaway capacity. When large amounts of capital are deployed towards production without corresponding capital focused on transmission, resource will be stranded and development will stall.

After its initial ramp in the early 2010s, oil production in the Bakken had stalled by 2014 and fell 25% over the next three years.

Today, there is a flood of private capital, boosted by generous tax subsidies, seeking to build the green energy future. That capital is naturally focused on the most productive wind and solar regions in the plains and desert Southwest - regions with limited electrical transmission and little local demand for electricity. And so, renewable developers learn The Bakken Dilemma first hand.

Congested Ducks

Oil and gas is a natural battery - potential energy that can be released with combustion. These hydrocarbons can be shipped around the country or world where their stored energy can be released on demand in your car’s engine or at a power plant.

By contrast, wind and solar produce electricity. While electricity can in theory be stored in manmade batteries or other physical energy storage mechanisms like gravity storage, the vast majority of the production is simply injected onto the electric grid where it is consumed in real time.

While oil and gas well volumes decline over time, the wells produce a generally predictable flow on a day-to-day basis. Wind and solar, however, have extraordinary variability or intermittency. If wind picks up dramatically from a standstill, a windfarm can go from producing no power to maximum capacity in the course of minutes, and vice versa. Weather patterns also vary on a much longer time horizon - long-term atmospheric conditions can create significantly higher or lower resource as measured across months and years. Even the “average” or “expected” production fluctuates throughout the course of a day (for example, there’s no sunlight at night), and seasonally throughout the year.

When the wind blows or sun shines, it typically does so throughout the region. The result is that all the turbines or solar panels in the region produce, or don’t produce, at the same time.

Since these facilities produce electricity for immediate consumption, rather than stored energy, The Bakken Dilemma is sent into extreme overdrive. In regions with significant renewable development, there is often far too much power when it isn’t needed, and not enough power when it is.

When the wind blows and the sun shines, power prices are low or even negative. When the sun sets and the wind stops, prices spike. In California, power prices can go negative at the daily peak in solar production (a result of the famous “duck” curve shown below), and spike as the sun sets and natural gas ramps up to offset the solar declines. In the Texas panhandle, many windfarms are curtailed (shut off) at moments of peak wind speed because the grid simply cannot handle the output and power prices go deeply negative, even when the populated demand centers downstate are in desperate need of power. The picture below shows local power prices in Texas on May 9, 2022. In the panhandle, power prices hovered at zero even as power prices exceeded $1,000/MWh in the south, due to a lack of transmission capacity. Negative power prices have been a fixture of the Texas panhandle since wind development accelerated after 2015.

Left: Solar production dramatically reduces net load mid-day, resulting in sharp evening ramp (source: CAISO). Right: Transmission congestion results in negative power prices in windy regions on a hot summer day (source: ERCOT).

The result is that realized power prices almost always underperform initial expectations, destroying the NPV of the existing projects and incentive for further development.

The local region may benefit from cheaper power prices on average, but also suffer from drastic fluctuations in power price and grid stability - particularly as the average lower prices compress the margins of reliable coal and natural gas assets and force their retirement. While not the only factor, Texas and California, the states with the highest penetration of wind and solar, respectively, have suffered from increasingly frequent blackouts in recent years.

These problems only become worse as more renewables are brought online. The result, if the market is left to its own devices, is destruction of the capital invested to date and the stalling of new development.

Electric Avenue

The United States does not have one integrated electric grid. Instead, the country is comprised of three distinct grids (Eastern, Western and Texas), that operate independently. These grids are further divided into wholesale electric markets each with their own mix of generation assets and prices.

The U.S. electric infrastructure was not made for renewables. Rather, it was designed for transportable fuel that can be distributed across the country and consumed locally for power generation. Divides in the physical grid and wholesale power markets mean that supply and demand of power and the resulting prices are fairly localized.

Power cannot flow from New York to Houston, or Houston to Los Angeles. More critically – the wind power generated in the plains cannot capture the premium power prices of the East Coast markets with dense population, high demand for power, and limited local resource.

With development in high wind and solar regions far outstripping transmission, regional basis will push realized power prices in dense renewable regions towards zero as the power is stranded. There is little reason to build more generating capacity if power can’t get to market.

Imagine if oil and gas could be burned only in the state in which it was produced. Gasoline and power prices in Texas and Pennsylvania would plummet, and prices everywhere else in the country would spike. Further, production of oil and gas in Texas and Pennsylvania would decline dramatically to match to the much lower local demand. This is the constraint that renewables face today.

What we need are electric pipelines; both gathering capillaries and long haul arteries. We need greater connectivity between grids and more effective methods of balancing power and price incentives between wholesale markets. If this could be accomplished, renewable generation would blossom, without the need for further subsidy.

Not only would this help the economics of developers, but it would also alleviate the grid reliability issues that will only become more pronounced if ignored. If renewable electricity could be more easily spread throughout the country, the intermittency and variability would be much less problematic. There may be no need for power in the Texas panhandle on a windy day, but there is always ample demand for power on a national scale – we just need to build the arteries to get it there.

There would be tradeoffs – areas that have enjoyed cheap power would lose this benefit but gain from further development activity (good local jobs, royalty payments, tax revenues), and a more reliable grid. Meanwhile, the dense demand centers would see cheaper prices overall driven by low carbon sources. As a country, we would better utilize our natural resources.

This idea is not novel, but it remains practically challenging. It would require significant physical infrastructure construction and cooperation between grid operators, public utility commissions and utilities – bureaucratic, snail-paced, risk-averse entities with non-aligned incentives. It would likely require significant coordination and agreement at a federal level. And of course, engineering an actual system is much harder than suggesting the concept on paper.

Yet, if we want to maximize the true benefits of renewables, this is the way. Otherwise, they will go the way of the Bakken.