Don’t view Texas as an isolated event

State’s woes show a new sense of interdependency

THE RECENT MAJOR power outages in Texas can be viewed as isolated to the particular political and electric power system design in that state, but that would be a mistake.

Let’s first put aside Gov. Greg Abbott’s unfounded blame for the problems on the use of renewables in the state. In fact, of the 46,000 megawatts of generation that were offline on one day, 28,000 were from coal, gas, and nuclear, and 18,000 were from solar and wind. In all, 185 generating plants tripped offline, an astounding and unprecedented cascading failure.

Texas, therefore, provides an important object lesson that is relevant to other power grids. The lesson has to include, too, how to reach out to customers in helpful and vital ways when a system failure occurs. And, by the way, the object lesson is likely relevant in other sectors as well, such as healthcare.

A January 2021 paper, “Exploring the Impacts of Extreme Events, Natural Gas Fuel and other Contingencies on Resource Adequacy,” by Richard Tabors and colleagues hits the nail on the head. Here’s the summary:

The power industry currently relies on planning methods that systematically understate the probability and depth of high impact events. The methodologies used today to calculate resource adequacy typically assume that outages and reductions in generator output are independent and uncorrelated. This assumption of independence is no longer valid.

In other words, electric power system operators are currently required by regulatory agencies to consider contingencies that might occur on their grids. For example, a transmission line might fail because of a lightning strike—a single contingency event. A power plant along that transmission line might also trip off line because of a turbine problem—a double contingency event. The system operators must have plans in hand to put in place to restore power, voltage, and frequency to customers within very short periods of time—minutes or hours—depending on the degree of contingent failures.

But what happens when a high impact event occurs, like when the polar vortex sweeps across the entire grid in Texas? At that point, the discrete probability of any part of the system failing is no longer an independent event. Those independent failures not only occur at the same time, but the fact that they are occurring on a physically linked system sends waves through the grid that make all those independent events become interdependent.

Tabors and his colleagues make recommendations for dealing with this problem on the power grids. For example, they suggest that power system operators should develop scenarios in their regions of high impact, common mode events, and estimate the probability distributions of the scenario’s physical impacts and associated economic costs. And they say we should have a better understanding of the economic cost and benefits of different levels of reliability.

They also recommend that we look at the relationship between fuel supply resources and generating output, specifically combining an operational model of the natural gas pipeline network with the electric power production cost model. Finally, they suggest that we should develop “a disruptive weather classification system including intensity, geographic coverage, and duration directly targeted for use by the US electricity market.”

These are sound suggestions, and you can bet that this prescient paper by Tabors and colleagues is now being read and considered all across the country.

But another lesson from Texas is seen in the frozen pipes that are now disrupting people’s water supplies. Unwary households who have never experienced power outages during freezing weather of this magnitude probably never learned that the way you keep water from freezing and cracking your pipes is to maintain a stream of flow through them. This suggests that infrastructure suppliers also need to plan to serve their public education obligation to help people weather a crisis.

But now let’s turn to other fields in which “supply disruptions are correlated and no longer limited to the outage of independent units and may be due to widespread or long-duration events with significant economic impacts on consumers.” The most obvious, as we’ve seen over the past year, is healthcare.

Here, too, we’ve tended to plan for the single contingency or double contingency event. Every city has a backup plan for hospitals to share emergency room procedures in the event there is a natural disaster, terrorist act, or the like. For example, we saw how well the Boston hospitals performed together when the Boston Marathon bombing took place, arguably saving many lives that could have been lost.

And sure, with the governor’s leadership, the Boston hospitals also have done their best to coordinate care for COVID patients. But it does not take much imagination to consider what kinds of disruption at a single hospital might also be occurring at many hospitals in our healthcare network.

For example, early on, there was a shortage of personal protective equipment (PPE) that was thrust upon all the hospitals by worldwide events. Remember how Gov. Charlie Baker was betrayed by the federal government when it intercepted PPE that he had arranged for distribution to Massachusetts health facilities.  All of a sudden we had correlated supply disruptions that were no longer limited to the problems of single hospitals.

Rosemary Gibson in her book China Rx has also documented the utter dependency of the US healthcare system on medications produced in China. Think through the potential that exists for mischief by a foreign power who can limit the export of antibiotics or who could nefariously interfere with the effectiveness of the exports it does permit. They might do so for political or strategic reasons, just as Russia and other bad actors engage in cyber attacks on our corporations, institutions, and infrastructure.

Meet the Author

Paul F. Levy

Former hospital executive and chairman, Massachusetts Department of Public Utilities
I’m not suggesting we be alarmist, but let’s take a look at our key physical and service infrastructure providers and think through the interdependencies that underlie seemingly independent players in our society.  Let’s develop some plausible scenarios and estimate their social and economic impact and decide how much we are willing to spend now to create plans to inoculate ourselves from the worst that could happen.

Paul F. Levy was chair of the Massachusetts Department of Public Utilities from 1983-1987, executive director of the Massachusetts Water Resources Authority from 1987-1992, and CEO of Beth Israel Deaconess Medical Center from 2002-2011.