The watchdog took a smaller bite, and private placement fraud doubled its share

I. The Number Nobody Announced

The fiscal year closed on September 30, 2025. One week later, Chairman Paul Atkins stepped to a podium and said something that had not been said out loud by an SEC chairman in a long time.

If you reward staff only for bringing enforcement actions, he said, you discourage them from deciding not to bring one.

He was talking about metrics. About the danger of measuring the wrong thing. About how counting cases can distort judgment.

What he did not say, at that podium or anywhere public in the months that followed, was what the final enforcement count for the year actually was.

The SEC has not publicly announced its enforcement statistics for FY2025. That is not typical. The annual numbers have historically been a point of institutional pride, released with charts and context and a press release. The silence itself is a data point.

Economists at the Brattle Group filled the gap. Using publicly available records, they counted 506 enforcement actions in FY2025. That is a 13 percent decline from FY2024. And within that year, the second half, which coincides with Chairman Atkins taking office on April 21, 2025, showed what the Brattle Group described as "an unprecedented decline in enforcement activity."

506 enforcement actions.

That number is not inherently good or bad. The chairman made a reasonable argument about incentives. Fewer cases brought carefully may do more than more cases brought sloppily. That argument deserves to be heard on its merits.

But here is the fact that sits next to it.

In the same fiscal year that the total number of enforcement actions dropped 13 percent, offering fraud cases, the category that most directly covers the private placement pitches landing in people's inboxes and voicemails right now, doubled as a share of what the Commission was bringing.

Not declined. Doubled.

The wall got thinner. The pressure on the other side of it went up.

That is where this story lives.

II. What Got Smaller

Start with the staff.

The SEC's Enforcement Division is the agency's largest division. In FY2024, it ran on 1,424 full-time equivalent positions. In the FY2026 budget request submitted to Congress, the Commission asked for funding for 1,178 positions. That is a reduction of 246 positions, roughly 17 percent, attributed to attrition from early retirement offers and buyout programs.

The full agency shrank too. Chairman Atkins told his own staff in a May 2025 town hall that headcount had dropped approximately 15 percent from October 2024 levels, from roughly 5,000 employees and 2,000 contractors to approximately 4,200 employees and 1,700 contractors.

These are not abstractions. An enforcement division that processes complaints, investigates tips, issues subpoenas, takes testimony, and prepares cases for litigation runs on people. It runs on investigators who have worked the same category of fraud for eight years. It runs on staff attorneys who know what a fraudulent private placement looks like in its third month before the collapse because they have seen seven of them close. When those people leave, the knowledge leaves with them.

The new FY2026 budget request describes "a focus on returning to the core mission that Congress set for the agency." That phrase appears in the document without further elaboration. What it means in practice will be tested over the next two to three years as cases that would have been opened in 2025 either surface in enforcement actions or do not.

Nobody outside the division knows which pile most of the incoming complaints went into.

Not really.

But the complaint volume did not shrink. The number of people reaching out about private placement pitches, unregistered securities, investment fraud, did not decline because the staff did. The ratio of incoming harm to available response just changed.

III. The Person the Number Did Not Reach

Picture a man in his late fifties. Retired from a warehouse job in the Southwest. His 401(k) rolled over when he left, and the broker who handled the rollover mentioned, once, that he knew of something that might work harder than a money market.

A private placement. That is an investment offered outside the public stock markets, sold directly to investors without the full disclosure requirements that govern a public company's shares. The rules restricting who can participate in these offerings have loosened over the years, meaning more people qualify for them. The pitch language around them has not changed. "Exclusive." "Not available to everyone." "Institutional-grade returns."

The term sheet on his table is one page. There are no audited financial statements attached. The broker explained that audited financials are not always required at this stage.

That last part is technically correct.

He put $60,000 in.

Now look at the agency that is supposed to catch the fraudulent version of that offering before the money is gone. In FY2025, the most frequently brought enforcement actions in the private placement category were offering fraud cases. That category doubled as a proportion of what the Commission pursued under the new administration.

That doubling happened inside a declining total.

The Commission was bringing fewer cases overall. Within that smaller number, offering fraud was more prominent than before. That suggests two possible readings. Either the Commission sharpened its focus on the fraud that most directly harms retail investors, which is the charitable reading. Or the fraud was so widespread and obvious that even a smaller staff working a narrower agenda could not miss it, which is the other reading.

The man with the term sheet does not know which reading is correct.

He has no way to know. That is not his failure. The machine was designed so that he would not be able to tell.

IV. The Staff That Left and What It Knew

In twenty years working fraud cases, I watched what happens when an agency loses institutional memory fast.

It is not that the incoming staff is less capable. Often they are sharper lawyers. The problem is pattern recognition. The veteran investigator who has worked the private placement space for a decade does not look at a new complaint and read it from scratch. She matches it against everything she has seen before. The nominee director on a Wyoming shell company. The unregistered broker taking commissions in cash. The promissory note with a three-year term and no security interest. She sees the shape of the thing before she reads the details.

When she takes an early retirement buyout, that template leaves with her.

The new hire, good lawyer, excellent credentials, starts from page one.

This is not a criticism of the people coming in. It is an observation about what fraud operators know that you may not. The operators running private placement frauds have been doing this for years. Some of them have been doing versions of the same play since before the investigator now assigned to their case was working this category. They know what the warnings look like. They know how to route around the detection they have already encountered.

The wall did not just get thinner. The people on the other side of it got a head start.

V. The Cases That Came to Court Anyway

The ones that did get pursued tell you something about what the floor looks like.

On April 15, 2026, a federal court entered final judgment against Clarice Saw, a former registered representative. The SEC's case alleged she misappropriated approximately $2.4 million from an elderly customer between December 2021 and March 2022. She was ordered to pay disgorgement of $640,587.30, prejudgment interest of $98,144.04, and a civil penalty of $640,587.30.

The math on that judgment: the total ordered is roughly $1.38 million on a theft the complaint alleged was $2.4 million. The gap between what was taken and what was ordered paid back is over a million dollars, before you account for whether the judgment can be collected.

And the timeline: the conduct alleged ran through March 2022. The final judgment came in April 2026. Four years.

That is not unusual. Fraud cases take time. The civil process is slow. Collection is harder than judgment. None of that is unique to this case.

But read the sequence slowly.

An elderly customer loses access to $2.4 million in early 2022. The case works its way through the system. The final order comes in April 2026. In between, the agency that brought the case reduced its enforcement staff by 17 percent, announced it would measure success differently, dismissed categories of cases as a matter of discretion, and has not yet published its annual enforcement statistics.

The elderly customer waited four years for a court order that may not fully recover what was taken.

The man with the term sheet on his folding table does not yet know he is waiting.

VI. What the Machine Looks Like When It Runs

On March 26, 2026, the SEC filed settled charges against Krish Kumar for allegedly making false representations to investors and misappropriating approximately $7 million.

The complaint is in the public record. Read the structure, not the name.

A person with apparent credentials presents an investment opportunity. The pitch is credible enough that money moves. The money, once moved, does not go where the pitch said it would go. By the time the complaint is filed, the money is elsewhere and the investors are waiting for returns that are not coming.

That is the structure. The name on the complaint changes. The structure does not.

The SEC's new enforcement leadership has said its philosophy is quality over volume. Fewer cases, but better ones. More focused on getting money back to harmed investors rather than counting actions for statistical effect. That is a coherent position. A well-chosen case brought hard can do more deterrence than ten cases brought carelessly.

But deterrence requires visibility. The operator running a private placement fraud in a secondary market, pitching to retirees without a large institutional profile, is not watching the Commission's high-profile fraud cases in New York and recalculating his exposure. He is watching whether anyone ever knocked on the door of anyone who looked like him.

The wall being thinner does not mean there is no wall.

It means the gaps got wider.

VII. What Is Still Moving

The Supreme Court heard oral argument on April 20, 2026, in Sripetch v. Securities and Exchange Commission.

The question before the court: does the SEC need to show that investors actually lost money in order to pursue disgorgement? Disgorgement is the legal mechanism that forces a fraudster to give back the profits from the fraud. It is one of the primary tools for getting money back to victims.

If the Court rules that the SEC must demonstrate direct, quantifiable harm to each investor before it can seek disgorgement, certain categories of fraud become significantly harder to prosecute on the financial recovery side. Cases where harm is diffuse. Cases where the money moved through multiple accounts before investors realized something was wrong. Cases where the structure of the fraud made individual loss difficult to trace.

Those cases include private placement fraud.

The argument was four days ago. The decision will come before the Court's term ends. When it does, it will change the shape of the wall in ways that will not be visible until the next wave of cases hits the filing system.

Nobody knows which way it goes yet.

But the operators running the pitches will read the opinion the day it comes out.

VIII. The Trail

Here is what I want you to take away from this.

The SEC is not gone. The enforcement function still exists. Cases are still being brought. The Saw judgment, the Kumar charges, the new Enforcement Director named in early April, the Cross-Border Task Force launched in September 2025, all of it is real, all of it is operating.

But the protection that agency provides is not uniform and it is not guaranteed. It has never been uniform or guaranteed. What changes in periods like this one is the margin for error.

When the agency was larger, a fraud that generated a handful of complaints had a better chance of hitting a desk with capacity to investigate it before the money was fully dispersed. When the agency is smaller, that same fraud may wait longer, may be deprioritized, may settle below the threshold of what a reduced staff can reasonably pursue.

The fraud does not wait.

I have walked into offices after a search warrant where the shredder was still warm. The people who ran the office knew we were coming. They had enough time to destroy some of what was there. Not all of it. Never all of it. The trail always exists somewhere.

But the warm shredder tells you something about the timing.

The man with the term sheet on his table has one tool available to him right now, before the SEC gets to his complaint, before the Sripetch decision reshapes disgorgement, before a case with his broker's name on it works its way through four years of process.

He can read the term sheet as if the wall behind him is thinner than it used to be.

Because it is.