It is not an audited loss, and it may ultimately prove too high. Federal officials should have to document and defend it.

Before Minnesota’s Medicaid fraud became national news, it was an open secret among many of the people working inside the system.
We saw providers who never answered their phones. We saw homeless people enrolled in services they never received. We saw strangers walking through encampments with clipboards, collecting names, birth dates, and Social Security numbers in exchange for promises of housing.
We saw it. We reported it. And for years, very little happened.
That history matters now.
Federal officials have deferred approximately $199 million in Medicaid reimbursements while Minnesota produces additional documentation for claims across 14 high-risk service areas. Minnesota says it is revalidating thousands of providers and strengthening fraud controls. It also argues that Washington has not adequately explained how it calculated the amount being withheld.
Minnesota’s failures warrant serious scrutiny. But legitimate providers and Medicaid recipients should not have to pay for the state’s failure to scrutinize the system sooner.
My experience inside Housing Stabilization Services shows how difficult, and necessary, that distinction is.
Housing Stabilization Services was only one Medicaid program. But its rise and collapse offer a window into Minnesota’s much larger oversight failure.
Launched in July 2020, the program used Medicaid dollars to help people with disabilities, mental illness, substance use disorders, and other barriers find and maintain housing. I began working with it shortly after its launch. In 2023, my husband and I founded a small nonprofit that became an HSS provider.
Before becoming a provider myself, I regularly referred homeless clients to organizations offering the service. After completing an intake, the person would be connected with a provider and wait for someone from the agency to call.
The call almost never came.
Days became weeks. Weeks became months. The client would return to us asking what had happened, and we would try to reach the provider on their behalf. Occasionally, someone answered the phone. Meaningful help rarely followed.
Eventually, the client might receive permission to transfer to another agency. Then the process began again: another intake, another promise, another few months of silence.
The person remained homeless. We had no reliable way to know whether Medicaid had been paying the original provider while the client waited for help that never came.
People living in encampments told us about strangers arriving with clipboards and promises of housing. They collected names, birth dates, Social Security numbers, and whatever other information people were willing to provide.
To someone safely housed, that sounds like a scam. To someone sleeping outside, it sounds like hope.
My husband and I knew the program had serious problems when we decided to become providers ourselves. That may sound contradictory, but we believed the service could be valuable if delivered honestly. We had watched people fall through the cracks, and we thought we could do better.
We would return phone calls. We would answer emails. We would meet clients in person. In other words, we would do what Medicaid was paying us to do.
That should have been the bare minimum. In this program, it felt almost exceptional.
The barrier to entry was astonishingly low. We completed some DHS paperwork, paid a fee, and waited for approval. There was no rigorous examination of our experience, capacity, or plans for delivering services.
I remember feeling almost triumphant: We were going to be an honest provider in a system riddled with fraud.
Looking back, the fact that this felt revolutionary should have concerned me more than it did.
Some clients came to us after spending months assigned to providers who appeared to have done nothing for them. Other times, we would be actively helping someone when DHS abruptly informed us that the client had transferred to another agency.
Under the program, only the agency listed as the client’s provider could bill Medicaid for helping that person. Once the transfer took effect, we could no longer be paid, even if we were still doing the work.
Often, the client had no idea a transfer had occurred.
Sometimes the person vaguely remembered speaking with someone who had promised housing and requested personal information. They had not understood that the conversation could cause the state to replace the provider already helping them.
DHS would not tell me the name of the new agency. That agency often never contacted the client. I was left with two choices: stop helping someone because Medicaid would no longer pay us, or continue working for free.
The state had built a program in which an unknown organization could take over a vulnerable person’s Medicaid authorization without that person meaningfully understanding what had happened. Meanwhile, the provider actually doing the work could not even learn who had replaced it.
The system had turned a person sleeping in an encampment into a billing opportunity with a Medicaid number attached.
My attempts to follow the program’s rules did not inspire confidence that anyone was firmly in charge of the money.
Housing Stabilization Services included up to $3,000 for certain expenses associated with moving a client into a home. At one point, I contacted DHS with a question from an eligible participant: Would the program reimburse wall art?
The answer was essentially: Probably. Not a $1,000 painting, though.
So… was that a yes?
I was not asking whether a newly housed person deserved a framed print. I was asking what Medicaid would pay for, where the limit was, and what documentation providers needed. The agency responsible for enforcing those rules could not clearly explain them.
This was not unusual. Providers asked basic questions about billing and allowable expenses and received answers so vague that even those trying to comply could not be certain they were doing so.
Housing Stabilization Services was one program, not all 14 now under review. But it demonstrates how a Medicaid program becomes vulnerable to fraud: low barriers to entry, unclear billing rules, weak monitoring, and no reliable way for clients or responsible providers to know whether someone else is billing for work that never occurred.
Not every improper claim was necessarily fraud. Minnesota created confusing rules, provided inconsistent guidance, and enrolled organizations in a lucrative program without ensuring that everyone understood what could be billed. Even providers trying to comply could not always determine where the boundaries were.
None of that excuses stolen identities, fabricated records, or billing for services never provided. It does help explain how the boundaries between legitimate work, improper billing, and outright fraud became dangerously easy to cross.
Meanwhile, a cottage industry was forming around the program.
Self-described consultants contacted providers and boasted that they had made a million dollars from Medicaid-funded services in the previous year. For a fee, they offered to teach others how to do the same.
The product they were selling was access to Medicaid money.
We were not merely speculating among ourselves. Other providers and I contacted Hennepin County and DHS about clients receiving no services, suspicious transfers, and other apparent signs of fraud.
We received no meaningful answers.
The program launched in July 2020. Yet the state’s broad crackdown did not begin until 2025—after spending had exploded, media outlets had published exposés, federal investigators had become involved, and the problems had become impossible to contain.
Five years.
By the time DHS shut down Housing Stabilization Services in October 2025, the agency said it had found large-scale fraud it could no longer control. Federal prosecutors have since charged multiple providers with billing Medicaid for services they allegedly never performed. In July, four men pleaded guilty to stealing approximately $2.2 million from the program and using artificial intelligence to fabricate records when their claims came under scrutiny.
Minnesota did not first learn that its Medicaid programs were vulnerable when federal officials began withholding money. Clients, advocates, and providers had been describing the problems for years.
A federal prosecutor has estimated that fraud may have consumed as much as half of the approximately $18 billion paid through 14 Minnesota Medicaid programs since 2018.
His estimate: as much as $9 billion.
It is not an audited loss, and it may ultimately prove too high. Federal officials should have to document and defend it.
Minnesota’s recent provider revalidation shows why the response cannot simply treat every irregularity as fraud. Of 5,583 providers reviewed in high-risk service areas, only 2,061 were initially approved. Roughly 3,400 received disenrollment notices: 916 after failed site visits and 2,491 because of incomplete or inaccurate administrative information. More than 2,000 appealed.
Those numbers are alarming. They also describe very different potential problems. A provider that cannot demonstrate it operates at its listed address is not the same as a legitimate organization that submitted incomplete paperwork. A competent accountability system must be able to distinguish between them—and do so quickly enough that Medicaid recipients do not lose care.
The federal deferral is not a finding that all $199 million was stolen—or even improperly paid. Legitimate providers should receive fair notice and an opportunity to correct paperwork or challenge adverse decisions. People who depend on Medicaid should not abruptly lose essential services because state and federal officials are fighting over documentation.
Weak oversight and indiscriminate enforcement may appear to be opposing failures. In reality, the first eventually produces the second.
Both Minnesota and the federal government should have to show their work.
Minnesota should report how much money it prevented from being improperly paid, how much it recovered, how many providers were removed for substantiated misconduct, and how many were restored after correcting honest mistakes.
Federal officials should identify the claims in dispute, explain how they calculated the amount being withheld, and promptly release legitimate reimbursements once they are documented.
Vulnerable people cannot become Minnesota’s excuse for resisting scrutiny. They are the reason that scrutiny should have occurred years ago.
Accountability is not a dramatic estimate or a late crackdown. It is knowing where the money went, what services were delivered, and whether the person sleeping outside ever received the help billed in their name.

Thank you for this thoughtful comment—and for the work you’ve done trying to address these problems from inside the system. The gap between policy and what actually happens on the ground is exactly what I hoped to capture. It’s a mess, and I totally get the cynicism. 🙁
Cally, thank you. Seriously, thank you for saying out loud what so many of us inside this system have been living for years. As a provider and former regulatory bureaucrat at a sister agency to DHS, I have spent the past several years during this administration trying to fix these problems from the inside out. What has become painfully clear is that the gap between policy fantasy and on the ground reality is not just wide; it is a canyon.
We keep producing frameworks that sound great in press releases but have zero real world implementability. The solutions already rolled out or slated for 2027 are so misguided and detached from how disability services actually function that it feels like we are not just watching the car go over the cliff. We are flooring the accelerator and turning the wheel to make sure we get there faster. The same mid management staff steering DHS operations still do not understand what they are regulating, and they are not trying to.
None of this surprises me anymore. That is the saddest part. Every week and every month, cynicism replaces whatever hope I had left.