The Cost Disease of Mental Health Care

On rising costs, a system that will never catch up, and what to do about it (in an piece that unfortunately suffers from "too many words disease")

I didn’t plan or anticipate this but this piece is coming out just after Slingshot AI launched its much touted Ash therapy chatbot, which some very smart people think is a game changer in this space. Whether or not such tools can ever replicate the relational depth of human therapy is an open question, but the reason for their attractiveness is in large part due to recognition of the cost-disease problem I detail here.

In recent years, some of the most conservative state legislatures in the country (including Indiana) have made serious and long overdue investments in mental health. I’ve been hearing increasingly loud grumblings from some of the very champions who helped make those investments happen: When does the demand for more money stop?

This piece is an attempt to reckon with the honest answer to that question: “it doesn’t,” unless we confront some of the structural economics of the system itself.

Driven by rising rates of mental health challenges and greater public awareness, our collective response has been to increase funding and expand access to care. In other words, meet growing demand by trying to increase supply.

I agree with both the diagnosis and the prescription. But those pushing to expand clinical services often overlook a a major problem with an inflammatory name: the cost disease of mental health care.

Scarcity, Rationing, and Ever-Increasing Costs

I’ve written before about how our mental health system is one of scarcity, where demand consistently outpaces supply. That scarcity forces providers into a delicate dance of juggling increasingly complex reimbursement strategies just to stay afloat. At the center of it all is what I call “the formula”: the optimal mix of patients, payers, and services needed to keep the lights on. The formula always eats first, and it often devours the flexibility needed to truly meet the needs of patients and communities.

The standard response to this well-known problem is to call for increased funding and other policy changes to close the gap between supply and demand. But what if two things are simultaneously true: (1) many favored reforms (including more funding) often result in sharply increased demand, and (2) supply will never catch up, because the cost of providing care always keeps rising?

This is known as Baumol’s Cost Disease, a concept articulated by economists William Baumol and William Bowen in the 1960s. It is a rather simple idea: in labor-intensive service sectors like health care, productivity doesn’t increase much over time, yet wages still need to rise to keep pace with other parts of the economy. (Productivity here is defined in strict economic terms: units of output per unit of input.)

Take a traditional therapist who offers individual therapy. They can only see one patient at a time. That reality doesn’t change, no matter how skilled or efficient they become. Salaries have to go up, but productivity remains the same. Tech solutions can reduce administrative burdens (“patients, not paperwork”) and help providers operate more efficiently at the margins, but they don’t fundamentally alter the labor equation. Mental health care is a time- and energy-intensive human service, not a scalable commodity.

Put simply, this means mental health care will always cost more over time because it depends on skilled people spending time with patients, and no technology or trick can outrun that basic fact.

Unintended Consequences of Important Efforts

To understand the problem better, let’s look at three of the most prominent movements in the mental health space: stigma reduction, parity, and digital access.

• Stigma reduction has helped normalize mental health struggles and made it far more acceptable to seek help. This is an unequivocal social good, but when you reduce the social cost of seeking care, more people will understandably pursue it. Less stigma means demand for care increases, and supply will struggle (and likely fail) to keep up.

• Mental health parity is the fight to ensure that insurance covers mental health care on par with physical health care. It is a long overdue and necessary reform, but it is also subject to a basic rule of economics: when you reduce costs or friction for a service, demand spikes. That’s the entire point of the policy, and a major benefit to both individuals and providers. However, unless supply keeps pace, the result is increased system strain and rising costs, which will ultimately be borne by patients and taxpayers.

• Digital health tools and tech-enabled platforms promise greater convenience and lower friction for both patients and providers. And in many cases, they deliver on that promise. Virtual therapy, AI-assisted triage, and smarter EHRs all help clinicians do their jobs more efficiently and give patients more entry points to care. However, these tools don’t fundamentally alter the nature of the work. At the system level, the core challenge remains: costs rise, productivity stays flat. If a community mental health provider leverages tech to increase per-clinician billing, that means more costs for payers, not less.

In each of these cases—stigma reduction, parity, and tech—the result is the same: more people poised to access care in a system that remains structurally constrained in its ability to deliver it. This does not take away from the real and meaningful victories that advocates have fought hard for. Reduced stigma empowers people to seek help without shame. Parity makes mental health care more financially accessible. Tech solutions lower barriers for individuals who might otherwise fall through the cracks. These changes matter for the people they help.

The problem is that these individual improvements often come with system-level costs. In the short term, those costs can be absorbed by insurers or government payers, but over time, they add up and ultimately get passed on to all of us.¹

This is the structural reality of mental health care. The Niskanen Center has dubbed this “cost disease socialism”: an approach that boosts demand (often through subsidies) without addressing the underlying economics of the sector, resulting in a never ending supply shortage spiral.

That doesn’t mean we should stop pursuing these improvements. It means we have to pair them with ideas that account for the reality of cost disease.

Greed and Government Spending

Before we get to examples of such ideas, I should address two common rejoinders to this line of thinking.

First: Some say that if we could just eliminate greed and profiteering from the health care system, we could absorb the rising costs. Greed damages the system in countless ways: hospital monopolies that drive up prices, opaque billing practices, and yes, bad actors in the insurance industry.

The uncomfortable truth, however, is that even if we wrung every drop of profit and excess out of the system, we’d still be left with the core problem: a structural mismatch between rising demand and cost disease constrained supply. Insurance company profits, for example, are already capped by law. Their huge revenues come largely from the volume of care they process, not from sky-high margins. In a system defined by scarcity, someone still has to do the rationing. And when demand spikes without a proportional increase in supply, that rationing becomes more painful, no matter who’s in charge.

Second: Let’s just spend more public money: treat mental health like a public good, and fund it generously.² Government absolutely has a vital role to play, but we are no longer living in an era of cheap public money. Interest rates are up. Federal deficits are high. And every major potential policy priority—from housing to education to child care—is facing its own crisis and has a compelling case for the same finite resources.³

Hope

So… is there any hope? If you’re a mental health advocate (like probably half my subscribers are), is the takeaway here just to despair and give up?

Absolutely not.

The key is to refocus on mental health innovations that sidestep the cost disease trap: models that either reduce reliance on labor-intensive clinical encounters or dramatically multiply their impact. Here are three promising directions (there are many others, please offer examples in the comments):

1. Single Session Interventions (SSIs)

These are structured programs designed to deliver meaningful therapeutic benefits in just one encounter. SSIs can be digital, in-person, self-guided, or clinician-led. And increasingly, the evidence shows they work. For many common issues—especially among youth—SSIs can be as effective as longer courses of therapy. They offer a way to meet people’s needs quickly, efficiently, and at scale, without adding pressure to a workforce already stretched thin.

2. Family Mental Health

Rather than treating mental health solely as an individual issue, some of the most exciting models now work with families as the unit of care. A well-designed family intervention multiplies the impact of a single session across several people. It’s one of the only ways to actually increase “productivity” in a cost-diseased system:

Families at the Center

3. The Three Ps: People, Place, and Purpose

As former NIMH Director Tom Insel has argued, recovery is about far more than access to clinical treatment. People get better when they have:

• People – supportive relationships, family, and social networks.

• Place – stable housing, safe communities, and physical spaces where they belong.

• Purpose – meaningful roles, goals, or contributions to make.

Building environments of connection and belonging isn’t easy, but it’s a different kind of hard than scaling expensive mental health care.

Changing our Defaults

To be clear: this is not an argument against access to clinical care. Access to high-quality, clinician-delivered mental health services is essential, and we should continue fighting to expand it.

But it is an argument for something else: changing our defaults.

In one of the stronger arguments from the recent book Abundance, the authors make the case that we don’t need to abandon environmental protections or community input, both of which are often blamed for blocking new housing. Instead, we need to change the default setting by making the presumed answer to building “yes” rather than “no.”

We should consider a similar idea with mental health. Maybe employee assistance programs should begin with Single Session Interventions, rather than referrals to clinicians for multi-session therapy. Maybe the first responses to youth mental health challenges should focus on families, not just the individual child. Maybe community and connection should be seen as the foundation of mental health, not as things that supplement treatment.

Cost disease is chronic, not acute. It usually isn’t obvious to anyone working in the system, but over time, the system becomes weaker and more distorted. Like any chronic disease left unaddressed, it compounds.

The default instinct in an under-resourced system is to fight for more. But in the current fiscal and political climate, more isn’t coming, not at the scale and speed necessary to fill the gap. That doesn’t mean we’re out of options. It means we’re overdue to focus on something better.

1. This tension between individual wins and systemic limitations is at the core of the mental health field’s biggest dilemma, one that philanthropists, advocates, and policymakers wrestle with constantly. How do we celebrate and expand critical individual-level progress while also confronting the structural barriers that limit the system’s capacity to respond at scale?

2. There where the ROI argument usually comes in: that investing in mental health care will pay off through reduced costs elsewhere. I’ve made this argument and I believe in it. I think it is effective at getting people to the table and far less effective at keeping them there, for a variety of reasons that I will probably examine in another piece.

3. There’s also a third objection, which is the argument that things like efficiencies introduced by tech and better business practices can actually meaningfully change the cost disease equation. For example, two of the largest Serious Mental Illness treatment providers just announced a merger, creating a mega-entity. I think the people involved with that decision would argue that a main reason was to create economies of scale so powerful that they could overcome the cost disease problem. If any reader wants to make that argument, please let me know and I’ll gladly give you space to make it.