Insull's Pyramid

On the evening of November 4, 1929, in Chicago, the newly built Civic Opera House premiered Aida. Four thousand seats were packed, and the golden curtain rose to thunderous applause.

The opera house was not a standalone building—it was embedded in the base of a 45-story skyscraper. The building’s silhouette, viewed from the east, resembled a giant armchair, with the back facing New York to the east and the front facing the west. Chicagoans had given it a nickname: Insull’s Throne.

Samuel Insull, the city’s power king, sat in a box. His holding company system supplied electricity to thousands of towns across more than 30 states, making his name synonymous with “safety” for Midwestern small investors.

Six days earlier, the New York Stock Exchange had just experienced “Black Tuesday.”

The people in the box did not know: this throne would collapse into one of the most famous ruins in American history within two and a half years; the savings of about 600,000 shareholders would vanish; and the man seen as the embodiment of American electricity would stand trial five years later—and be acquitted.

Acquitted. Remember this verdict. It is the true protagonist of this chapter.

I. Clarifying the Point: He Was Not a Fraud

Writing Insull as a fraud is the laziest and most inaccurate way to tell this story.

In 1881, at the age of 21, Insull arrived in the United States from London, serving as Edison’s personal secretary. Edison invented the light bulb, but it was Insull who developed the economics of turning electricity into a business: the economies of scale of large central power plants, time-of-use pricing, and load management (the person who used prices to move demand, as described in the book’s “Migration of Scarcity,” was him), extending credit to households to create load, and proactively accepting state government regulation in exchange for stable franchise rights. Under his leadership, Chicago’s electricity rates dropped over several decades.

The electrification of America was not a natural process. It was largely built by this man.

There was only one problem: a power grid consumes capital faster than almost any other industry. Power plants and copper wires take decades to pay back, while windows for expansion are measured in months. Equity financing could not keep pace. What was the answer?

Insull’s answer was a standard product on Wall Street’s shelf at the time: the holding company.

II. The Logic of the Pyramid Will Finish Building It for You

The structure itself is simple enough to be sketched on a napkin: place a holding company above an operating company, then another holding company above that. Each layer needs to own only a little over ten percent of the layer below to control it, and each can issue its own debt and preferred stock. The leverage multiplies layer by layer, with the top layer using extremely thin equity to control assets worth billions at the bottom.

As long as the assets at the bottom produce steadily growing returns, every layer of the pyramid amplifies them. Electricity demand happened to grow every year in the 1920s, making the “utility holding company” the sexiest financial product of the era. Insull’s shares were eagerly bought by Midwestern teachers, shopkeepers, and even his own electricity customers. The sales pitch was the phrase that would later prove fatal: as safe as the sunrise.

It is worth noting the origin of the pyramid’s final two layers. In 1928 and 1929, Insull created two top-level holding companies in succession—not out of greater greed, but in self-defense: financier Cyrus Eaton was quietly accumulating his shares in the market, and Insull had to build a fortress of voting rights. The top of a pyramid is often not the product of ambition, but of fear—once the structure begins, its own logic will finish building it for you.

Then came 1929, 1930, and 1931. On the asset side, share prices and demand fell; on the liability side, interest and preferred-stock dividends did not decline by a cent. Leverage amplified death on the way down by exactly as much as it had amplified returns on the way up. In April 1932, the core of the empire entered receivership. The investments of roughly 600,000 shareholders evaporated—many of them still used Insull’s electricity by day and counted his worthless paper by night.

There is a detail that must be recorded in the ledger: Throughout the entire process of the empire’s collapse, the lights in Chicago did not go out. The power plants continued to operate, streetcars continued to run, and most of the operating companies continued to make profits. What died was not the grid, but the holding structure above it.

Real assets, fake structures. Four words are the key to understanding every subsequent construction frenzy.

III. An Acquittal—and a Death Sentence Written into Law

In 1932, Insull left for Europe, moving between Paris and Athens. In 1934, he was detained in Turkey and extradited back to the United States. The media treated him as a “public enemy,” and Roosevelt singled out “the Insulls” in his campaign speeches as symbols of the old era.

Then came three trials: federal mail fraud, state embezzlement, and violations of federal bankruptcy law. He was acquitted in all three.

The jury’s dilemma was: the prosecution searched through the books but found no missing money. Everything was legal, everything was disclosed, and everything was industry practice. More troublingly, Insull had also lost everything—he had staked his entire fortune on his pyramid, living off his pension in his later years. He believed in his tower. But “belief” is not a proof of innocence; it is a source of danger: a fraud who does not believe will get off the bus early, while a true believer who builds the tower will collapse with it, pulling everyone who believed in him down with them.

The true verdict was not in the courtroom. The Securities Act of 1933, the Securities Exchange Act of 1934, and the creation of the Securities and Exchange Commission all bore the fingerprints of this case. The Public Utility Holding Company Act of 1935 went further, building in what Wall Street called a “death sentence clause”: multilayered holding-company pyramids had to be dismantled.

The jury acquitted him, but Congress sentenced his structure to death. These two verdicts are not contradictory—they are the complete verdict: the individual was innocent; the playbook was guilty. When the collapse of a structure can vaporize the savings of 600,000 families, society’s response is not to find the bad person, but to make that structure disappear from the legal world.

Legality is the gravest accusation one can level at a system.

IV. Each Frenzy Burns on the Financial Structure of That Round

Chapter 8 of this book discussed the general rule of bubbles: the foundation and fuel. The bubble burns the investors, leaving behind the infrastructure of civilization. This chapter adds a colder detail: Each round has its own new fuel formula. History does not repeat the method, only the fire.

In the 1920s, the fuel was the holding pyramid; RCA burned in the same furnace. The “shovel seller” stock of the radio frenzy rose from the teens to more than a hundred dollars at its peak, never paid a cent in dividends, and by 1932 had fallen more than 90 percent.

In the 2000 telecom bubble, the fuel changed its formula: vendor financing. Lucent and Nortel lent their customers billions of dollars—money to buy their own equipment. The shovel sellers financed the prospectors’ purchase of shovels (Chapter 8 put this case on record), and revenue replicated itself inside a circular pipeline until the pipeline burst. After the collapse, the script’s final act arrived on schedule: the Sarbanes-Oxley Act rose from the ruins of Enron and WorldCom. Once again, finance ended in regulation.

Cisco provided a case in the same year: once the world’s most valuable company, it lost 86% of its value—yet the company was real, profitable, and still alive. The assets were real, but the price structure was a mirage. Insull’s Chicago lights, Cisco’s routers, same sentence.

Thus, the mechanism can be written as a chain: The capital hunger of the infrastructure boom inevitably summons financial structures that sell “certainty”; these structures amplify glory in the ascent and amplify death in the descent; they vaporize public savings; legislators then arrive, precisely burying the previous game. Note the qualifier at the end—PUHCA couldn’t regulate vendor financing, and Sarbanes couldn’t regulate the next round of structural innovation. Regulation is always building fortifications for the last war.

The thesis narrows: “Sell shovels” leads to finance, and finance leads to regulation. An operational inference: When those selling shovels start lending to those buying shovels, the boom is in its final act. This signal appeared in 1929 (for defense, the top-tier holding company) and 2000 (vendor financing). Each time, the form was different, but the core was the same: genuine demand could no longer sustain growth narratives, and financial structures were used to book future demand.

V. The AI Era: Count Which Layer You’re On

Now, superimpose this film onto the present. (The numbers and cases in this section must be verified for timeliness before publication.)

The capital hunger of the compute boom has entered the financial-innovation phase, with an ever-lengthening list of forms: loans collateralized by GPU clusters; circular structures in which chip manufacturers invest in model companies and those companies use the financing to buy chips; data centers placed in SPVs and financed by securitizing long-term leases; and compute commitments and equity swaps among model companies. Viewed individually, every one is legal, publicly disclosed, and standard industry practice—

These sixteen words are precisely what the prosecution wrote in the Insull case file in 1934.

Criteria list:

Draw the cash flow loop. Map the money flow in your ecosystem: who pays, who the money passes through, and who ends up with it. Any part that loops back to the start should be deducted from “revenue” mentally. The growth in the loop is an echo, not a demand.
Count which layer you occupy in the pyramid. How many layers of promises separate you from the true end-user—the person paying with their own money for their own problem? The number of layers is your fall height. Operating companies survived in 1932; the ones that died were at layer N.
Beware of “as safe as the sun rising.” When high-leverage structures start using the inevitability of infrastructure (compute is like electricity) to justify themselves, remember the 600,000 shareholders: they bought the narrative word for word, which was “electricity is an eternal demand.” They didn’t buy the wrong industry, they bought the wrong structure. The inevitability of an industry does not guarantee the safety of its structures.
Read the next PUHCA in advance. After a collapse, there will be legislation, and that legislation will precisely bury the hallmark play of this round. Do a thought experiment: if Congress were to legislate this round of the boom in three years, what would be the first ban?—Don’t build your company on that answer.

In July 1938, in Paris, a seventy-eight-year-old man suffered a heart attack on a Métro platform and died. He left little behind. The newspapers noted in passing that he had once been America’s king of electricity.

Across the ocean, the grid he had built was supplying tens of millions of households, as it always had, without stopping for a single second.

Infrastructure does not remember the people who built it, and financial structures do not remember the people they buried. The architects of the next boom will visit the ruins and copy the blueprints of the foundations—then choose a completely new structure for their own tower, one that this time will surely never collapse.