Neel Somani Explains Energy Markets in 80/20 Form

Most executives can describe what electricity costs. Very few can explain why. That gap between cost and market awareness matters more than it used to, because energy expenses are now a board-level concern for any organization running data centers, operating manufacturing facilities, or planning infrastructure at scale.

Neel Somani, a researcher and former quantitative researcher who covered markets at Citadel, has spent time breaking down the mechanics of regional electricity markets in plain language. His approach is characteristically Pareto-efficient: identify the 20 percent of concepts that explain 80 percent of what matters, then stop there.

Why Power Prices Are Not What You Think

The first thing to know about the markets is that the price you pay is not the average cost of generating power. It is the cost of the most expensive unit of power currently running.

This principle, known as marginal cost pricing, is the single most important concept for any leader trying to decipher a utility bill or an energy contract. As Somani explains, “the price is based on the last megawatt of power that’s produced.” If the cheapest sources, renewables and nuclear, are running at full capacity, but demand requires turning on an old, inefficient oil generator to meet the final increment of need, then every producer in the market gets paid at the oil generator’s high cost.

For executives, the implication is immediate. Energy cost volatility is not random noise. It is a structural feature of how electricity markets are designed.

The Generation Stack: A Hierarchy That Determines Your Bill

Every regional market has what traders call a generation stack: an ordered list of energy sources arranged from cheapest to most expensive.

Using California as his primary example, Neel describes a relatively clean stack. “There’s renewables, and there’s natural gas units, but in California, you don’t have any of that other junk, like coal or other dirtier units.” In this market, renewable energy, which carries essentially zero marginal cost, gets dispatched first. As demand rises, the grid operator activates progressively less efficient natural gas units until supply meets demand.

“The name of the game,” as Somani puts it, “is guessing how inefficient of a natural gas unit we have to turn on in order to meet the demand.” That framing is instructive for leaders. The question is never just how much power costs today. It is how close the system is to the expensive part of the stack, and what conditions would push it toward that.

Regional Topology and Why Location Matters

Electricity cannot be stored at scale or transported without loss, which means geography shapes pricing in ways that have no analogy in most other commodity markets.

In California, the state is divided into two main pricing zones: Northern California, called NP 15, and Southern California, called SP 15. They are connected by a transmission corridor known as Path 15. “When that line is congested, that’s what causes the price difference between northern and southern California,” Somani explains.

This is not a California-specific quirk. Every regional market has transmission bottlenecks that create locational price differences. For executives making infrastructure siting decisions, whether for a data center, a manufacturing plant, or a distribution hub, the local transmission node price is a more relevant number than the regional average. Organizations that ignore locational pricing routinely overpay or fail to capture savings available in lower-cost nodes.

The governance lesson here is about decision-making with incomplete information. Procurement teams that operate only at the level of regional averages are systematically leaving money on the table. Locational intelligence requires investments in energy market expertise, but the return on that investment can be substantial for high-consumption organizations.

Demand Dynamics and Cost Drivers

Supply-side mechanics explain a great deal, but demand-side dynamics are equally important for practical energy management. Neel identifies two primary demand drivers: temperature and time of day.

Extreme heat or cold pushes residential and commercial users to run air-conditioning and heating systems simultaneously, driving aggregate demand toward the expensive end of the generation stack. For organizations with flexibility in their consumption timing, avoiding peak demand periods during extreme weather events is one of the most accessible ways to manage costs.

Time of day introduces a more nuanced dynamic, particularly in markets with high renewable penetration. “During the daytime, you have the sun out, so you have this very cheap, zero marginal cost solar power.” But when the sun sets, that supply disappears just as residential demand climbs. “Everyone turns on their power all at once,” he notes, creating an evening price spike that now exceeds historical norms because the afternoon solar generation has to be replaced rapidly by less efficient peaking units.

This phenomenon, widely discussed under the name the “duck curve,” has real operational implications. Data centers with flexible workloads, manufacturing operations with shiftable production, and commercial facilities with smart building systems can all reduce exposure to evening peak pricing by shifting load earlier in the day or into overnight hours.

The New England Case Study

If California illustrates the interplay of renewables and natural gas, New England shows what happens when fuel markets intersect in ways that amplify price volatility.

New England has a winter problem that is structurally distinct from that of most other regions. Natural gas serves a dual purpose: it heats homes, and it fuels power plants. In the winter, residential heating demand creates a fixed floor of natural gas consumption. “There’s no longer enough natural gas to meet power demand,” Somani explains, which forces the grid to activate oil generators to fill the gap.

Once oil generators set the marginal price for electricity, natural gas suppliers face an unusual incentive. Any buyer of natural gas can use it to generate electricity and collect the high oil-equivalent power price. This creates upward pressure on natural gas prices themselves. As Somani describes the dynamic, natural gas sellers keep raising the price of natural gas until it’s basically the same cost to produce a megawatt of power from a natural gas unit as from an oil unit.

The result is that both power prices and natural gas prices spike simultaneously during cold New England winters. For organizations operating in the region, this dual exposure is a material risk. It cannot be hedged by focusing solely on electricity contracts; the natural gas procurement strategy matters too.

The 80/20 Takeaway for Leaders

Neel’s framework for power markets reflects the same instinct that has shaped his broader work: identify the structural mechanisms that explain most of the variance, then build decisions around those mechanisms rather than around anecdotes or averages.

For business leaders, that distillation produces a short checklist of questions that should be standard in any energy strategy conversation: What does the generation stack in our region look like, and how close are we typically to the expensive part of it? Where are the transmission constraints that affect our specific location, and what is our locational node price relative to the regional average? How does time of day affect our cost exposure, and do we have operational flexibility to shift load away from peak pricing windows? And in multi-fuel regions like New England, are we managing fuel market risk as part of our total energy exposure?

These are not technical questions reserved for commodity traders. They are governance questions with direct financial consequences. The executives who understand power market mechanics at this level are better positioned to evaluate energy contracts, assess infrastructure decisions, and hold their energy procurement processes accountable for performance.