U.S. Surgeon General Begins Issuing “Income Prescriptions,” Citing Strong Association Between Adequate Income and Human Well-Being

WASHINGTON, D.C. — The U.S. Surgeon General has reportedly announced a controversial new public-health initiative allowing physicians to prescribe income corresponding to the local cost of living to willing and able individuals after decades of research repeatedly suggested that having enough money to afford housing, food, transportation, healthcare, and ordinary participation in society is associated with substantially better well-being.

The program, formally known as the Income Sufficiency Therapeutic Initiative, began after federal health officials spent years studying social determinants of health before reaching what one researcher described as an:

“Extremely inconvenient conclusion.”

A Surgeon General’s office representative summarized the evidence:

“Turns out being broke all the fucking time is stressful.”

Researchers reportedly checked again.

The finding persisted.

CLASSIFICATION: PUBLIC HEALTH — SOCIAL DETERMINANT REIMBURSEMENT AND FORMULARY ANALYSIS
DISTRIBUTION: Office of the Surgeon General, Department of Labor, Department of the Treasury, Internal Revenue Service, National Association of Health Plans, Anyone Currently Deciding Between the Prescription and the Rent
PREPARED BY: The Externality Research Division
DATE: August 2026

Our Research Division wishes to record at the outset that no part of the underlying finding is new. The association between income and health outcomes is among the most replicated results in the discipline, has survived every specification thrown at it for the better part of a century, and is taught to first-year students as an uncontroversial background fact. Nothing was discovered. What changed is that somebody wrote it on a form with a box for the dose.

The desk regards this as the entire event. A finding that sits in a journal is a finding. A finding that sits on a prescription pad is a claim on somebody, and the identity of that somebody is what the following ten months of proceedings were actually about.

Doctors Can Now Prescribe Dollars

Under the pilot program, clinicians treating qualifying patients may issue prescriptions specifying an income floor based on local living costs.

A typical prescription reportedly reads:

Patients are strongly advised not to discontinue income abruptly.

Possible withdrawal symptoms include:

and:

HAVING TO FIGURE SOME SHIT OUT BY FRIDAY

Federal researchers described the adverse-event profile as:

“Pretty fucking severe.”

Our analysts have reviewed the specimen and note that it is, as a document, unremarkable. Every field on it exists on every prescription written in the United States. The dose is specified, the route is specified, the frequency is specified, the duration is specified, and the instructions warn against a known interaction. Nothing about the form is novel. The only novelty is the substance in the first line, and the substance in the first line is the only one on the entire formulary that the patient could obtain himself if anyone would sell it to him.

The desk draws attention to the discontinuation profile, which is the single most technically defensible section of the pilot. Withdrawal, as the term is used clinically, describes a predictable cluster of adverse events following abrupt cessation of an agent to which a system has adapted. Eviction, utility shutoff, food insecurity, and loss of transportation follow the cessation of income with a reliability that no pharmaceutical agent in the American formulary can match. The onset is measured in days. The dose-response relationship is clean. The events are dose-dependent, reversible on reinstatement, and observable without instrumentation.

Our Research Division notes that a drug producing this adverse-event profile on discontinuation would carry a boxed warning, a mandatory taper schedule, and a patient assistance program funded by the manufacturer. Income produces it at scale, continuously, in a population of tens of millions, and is classified as an employment outcome.

The final listed symptom is the one the desk regards as substantively important, and it is not a joke. Having to figure some shit out by Friday is a description of an executive-function load — a deadline, a shortfall, and no reserve — imposed weekly on a person who is also expected to sleep, work, parent, and comply with a treatment plan. The clinical literature calls this scarcity-induced cognitive load and has measured its effect on decision quality repeatedly. The Surgeon General’s office did not invent the finding. It merely declined to translate it.

Dosage Adjusted for Local Cost of Living

Officials emphasized that income prescriptions cannot use a single national dosage.

A patient living in rural Mississippi may require a different therapeutic income than someone living in Miami, Boston, San Francisco, or New York.

The prescription therefore targets what researchers call:

LOCAL THERAPEUTIC INCOME

The calculation considers:

One patient asked whether his prescription could simply be:

“More money.”

His physician checked the guidelines.

“Technically yes.”

The desk has examined the dosing methodology and finds it to be the least controversial component of the initiative and the one that generated the most correspondence. The reason is that a locally indexed dose makes a comparison visible that a national figure conceals. A single national number invites the argument that the number is too high for some places, which is an argument about the number. A locally computed number invites the argument that a specific city is unaffordable to the people who staff it, which is an argument about the city.

Our analysts note that the seven inputs are the seven inputs. There is no eighth category doing hidden work. Housing, food, transportation, utilities, healthcare, taxes, and ordinary expenses is a complete description of what it costs to be a person in a place, and the Initiative’s critics have not disputed a single line of it. The dispute has been entirely about who is required to read the total.

The patient who asked for more money is, in the desk’s assessment, the only participant in the entire proceeding to have stated the intervention correctly on the first attempt, and his physician’s answer is accurate. Technically yes is what the guidelines say. The Research Division notes that the word technically is carrying the difference between a clinical program and a raise, and that nobody in ten months of hearings has been able to specify what that difference consists of.

Physicians Initially Attempt Lifestyle Intervention

The clinical guidelines recommend that doctors first evaluate whether conventional interventions can address the patient’s symptoms.

A physician reportedly told one patient experiencing chronic financial stress to:

The patient stared at him.

“With what fucking money?”

The physician looked back at the chart.

He checked a box.

FIRST-LINE INTERVENTIONS FAILED DUE TO BEING BROKE

The income prescription was approved.

Our Research Division has obtained the counseling protocol and confirms that every item on it is correct, evidence-based, and priced. Regular exercise requires time that is not being sold to somebody else, and in most of the pilot’s counties it requires either a facility or a neighborhood in which walking after dark is a reasonable plan. Nutritious food is a grocery budget and a kitchen and the ninety unbroken minutes in which to use them. Eight hours of sleep requires a residence in which eight hours are available consecutively, which excludes a second shift and most of the third. Stress reduction, as written, is a description of the outcome rather than a method. Preventive healthcare requires coverage, a copay, and a workday that can absorb the appointment.

The desk notes that the protocol was not wrong. It was priced, itemized, and handed to a person who had already been asked for the money once that month by somebody else.

The checkbox is the item the Research Division regards as historically significant, and we wish to be precise about why. Clinical documentation does not record the world. It records the world in the categories the form permits. For decades the form permitted non-adherent, which attributes the failure to the patient, and counseling provided, which attributes the success to the clinician, and there was no third box. The addition of a box reading FIRST-LINE INTERVENTIONS FAILED DUE TO BEING BROKE did not change a single patient’s circumstances. It changed what the chart says, and therefore what the chart can be counted for.

Our analysts have reviewed early aggregate data from the pilot sites and note that the box is checked at rates the program’s designers did not anticipate and the physicians did. One participating clinician, asked whether the finding surprised her, said that it did not, that she had been writing it in the free-text field for eleven years, and that nobody had ever been able to run a query on the free-text field.

Pharmaceutical Industry Requests Clarification

Drug manufacturers reportedly became alarmed after learning that doctors could potentially address some health-related stressors by improving material conditions instead of prescribing another product.

One pharmaceutical representative asked:

“Have you considered medication?”

Federal researchers responded:

“Yes.”

“And?”

“We also considered $4,000 a month.”

The representative requested additional clinical trials.

Our Research Division wishes to state plainly that the industry’s position here is not absurd and that treating it as absurd concedes the wrong point. Medication for anxiety and depression works. It works in patients who are poor and in patients who are not. Nobody at the agency, and nobody at this publication, disputes that a person in financial distress with a treatable mood disorder should be offered treatment for the mood disorder.

The desk’s interest is in the request that followed, which is the only move available and was made immediately. The representative did not argue that income does not work. He asked for additional trials. A request for further evidence is unanswerable, indefinitely renewable, and costs the requesting party nothing, and it is the standard instrument for delaying a conclusion that has already been reached.

Our analysts note the asymmetry in what the request implies. A pharmaceutical agent enters the formulary after trials funded by the party that profits from the result. Income has been trialed repeatedly — in cash-transfer programs, in guaranteed-income pilots, in natural experiments produced by lottery wins, casino disbursements, and abrupt benefit changes — and the results have pointed the same direction for decades. What income has never had is a party with an incentive to fund the next study, a sales force to carry the finding to clinicians, and a budget for the dinner at which it is discussed.

The desk regards this as the substantive point of the exchange. The evidence base for income is large, old, and unrepresented. It has no manufacturer. When the representative asked for additional clinical trials, he was not disputing the data. He was correctly identifying that nobody is paid to produce more of it.

Insurance Companies Immediately Deny Coverage

Health insurers reportedly praised the Surgeon General for exploring innovative approaches to preventative medicine before clarifying that income is:

OUT OF NETWORK

One patient submitted a claim for his prescribed $62,000 annual income.

The insurer denied it.

Reason:

INCOME NOT MEDICALLY NECESSARY

His physician appealed.

The insurer requested prior authorization.

The physician submitted documentation demonstrating that the patient required money to pay rent.

The insurer responded that housing expenses should first be managed through:

STEP THERAPY

The patient must initially attempt:

Only after documented failure may income therapy be reconsidered.

Our analysts note that the denial is correctly reasoned within the framework that produced it, and that this is the difficulty. Medically necessary is not a clinical judgment. It is a contractual term, defined in the plan document, and the plan document defines it in terms of services covered by the plan. Income is not a covered service. Therefore income is not medically necessary. The syllogism is valid. It is also entirely self-referential, and the Research Division has been unable to identify any point in the chain at which the patient’s condition is consulted.

The desk regards OUT OF NETWORK as the more revealing of the two determinations. A network is a set of parties with whom the insurer has negotiated a rate. The category has no application to income, because there is no party the insurer could contract with to supply it at a discount. The determination is not false. It is a correct statement that the insurer’s entire apparatus — the discount, the network, the negotiated rate — has nothing to attach to, and the apparatus responded by producing the code it produces when it has nothing to attach to.

Step therapy is the component our analysts find most instructive, because unlike the others it is not an accident of vocabulary. Step therapy exists to require a cheaper intervention before an expensive one, and it works as designed. What the pilot exposed is what the four required steps have in common. Roommates transfers the cost to a housemate. Moving farther away transfers it to the patient’s commute and to whichever employer absorbs the lateness. Asking family members transfers it to a relative who is likely to be in a similar position. Earning more money independently transfers it to the patient, in hours, at his existing rate.

None of the four steps reduces the cost. Each of them moves it to a party who is not on the claim, and the claim closes because the claim is the only thing the system can see. The Research Division notes that this is not a failure of step therapy. It is a precise account of what step therapy is for.

Prior Authorization Becomes Completely Insane

Doctors say obtaining authorization has already become burdensome.

One physician reportedly spent 47 minutes on the telephone attempting to secure an income prescription for a patient.

The insurer’s representative asked:

“Has the patient tried not being poor?”

The doctor paused.

“What?”

“Our system says that’s first-line.”

The doctor requested a peer-to-peer review.

An insurance-company physician came on the line.

“Have you considered generic income?”

“What the fuck is generic income?”

Nobody knew.

The claim was denied anyway.

Our Research Division obtained the call recording and reports that the representative was reading. She was not improvising, editorializing, or expressing a view about poverty. Her screen displayed a first-line alternative, and the field on her screen said not being poor, and she asked the question the workflow required her to ask before the denial could be finalized. She had asked forty-one questions that day from the same interface and had no more discretion over this one than over any of the others.

The desk notes that the sentence is grotesque and that the process that produced it is ordinary. Every prior-authorization system in the country contains a field for the cheaper thing that should be tried first. Somebody populated the field. There was no cheaper thing, and the field could not be left empty, so it was populated with a description of the desired outcome. The interface has no way to distinguish a therapy from an aspiration, because it was never asked to.

The peer-to-peer review is the exchange our analysts have spent the most time on. A generic is a chemically equivalent version of a compound whose patent has expired, supplied by a competing manufacturer at a lower price. Applied to income, every term in that definition fails simultaneously. There is no compound, no patent, no manufacturer, and no mechanism by which a competitor could supply a dollar more cheaply than a dollar.

The reviewing physician was not being obtuse. He was performing the review his role defines, which is to identify a lower-cost alternative, and the vocabulary available to him for that task is pharmaceutical. Asked to review a therapy that is not a drug, he asked the question he is paid to ask, and it produced a category error that neither party on the call could name.

Nobody knew is, in the desk’s assessment, the most honest line in the entire record. Two credentialed physicians spent forty-seven minutes inside a system neither of them designed, discovered together that it had produced a term with no referent, and were unable to do anything about it. The claim was denied anyway, which the Research Division notes was the outcome at every branch of the workflow, including the branches nobody reached.

We record one additional finding. The forty-seven minutes were unbilled. The physician’s time was consumed, the representative’s time was consumed, the patient waited, and no line item anywhere in the American healthcare system records the transaction. Our analysts estimate the cost of the call at several times the monthly increment under dispute, and note that it appears on nobody’s ledger because there is no code for it.

Employers Ask Whether They Are Pharmacies Now

Businesses have meanwhile demanded clarification after discovering that the most obvious delivery mechanism for therapeutic income already exists.

It’s called:

PAYROLL

One employer attended a federal briefing.

“So the patient works?”

“Yes.”

“And you want money administered every two weeks?”

“Correct.”

“Through direct deposit?”

“Preferably.”

Long silence.

“Big dawg, you’re describing a paycheck.”

Federal health officials acknowledged substantial overlap.

The Department of Labor suddenly became extremely interested.

Our Research Division regards this exchange as the decisive exchange of the entire pilot and wishes to be exact about what occurred in it. The employer was not mocking the program. He was performing an identification, correctly, under conditions where nobody else in the room could afford to. Route, frequency, and delivery mechanism had all been specified. Every one of them already existed, in his own accounting system, administered by his own staff, on a schedule he did not choose and cannot alter.

The desk notes that the identification runs in both directions, and that only one direction was discussed at the briefing. If a therapeutic income is a paycheck, then the pilot is proposing a wage floor with a clinical justification. But if a paycheck is a therapeutic income, then every payroll decision made in the United States is a dosing decision, made without a prescriber, without a monitoring protocol, and without any requirement that the dose be adequate for the indication.

Our analysts note that no participant at the briefing pursued the second reading, and that the Department of Labor’s sudden interest suggests at least one attendee understood it immediately. The Department has spent a century regulating the conditions of the transaction — hours, safety, overtime, the mechanics of withholding — under a framework in which the adequacy of the amount is a matter for the market. The pilot did not challenge that framework. It attached an outcome to the amount and put a physician’s signature underneath it, which is a different thing, and which the Department’s own general counsel is understood to have described in a memorandum as a jurisdictional question we would prefer to receive rather than raise.

The Research Division has reviewed the employer’s position on the merits and finds it to be substantially stronger than its reception suggested. He was told that a patient in his employ required $31.50 an hour for medical reasons. He observed that he had not been told where the difference was to come from, that the prescription specified no payer, and that the only party in the room with a direct-deposit relationship to the patient was him. Nobody at the briefing contradicted him. Federal officials acknowledged substantial overlap and moved to the next slide.

Treasury Develops Income Formulary

The federal government is reportedly considering a standardized formulary.

Insurance companies immediately classified Tier 4 as experimental.

Our analysts have examined the draft formulary and report that it is, considered strictly as a document, a competent piece of drafting. It sorts a continuous variable into five bands, attaches an access condition to each, and specifies which bands require review. This is what a formulary does. The Research Division’s interest is in what the bands turn out to describe once they are written down in one column.

Tier 1 is defined by survival and carries the lowest copay, which our analysts note is the standard structure: the cheapest option is the most accessible option. Applied to income, the structure inverts into a description of the labor market’s floor, and the phrase technically remain alive is doing the same work in the formulary that minimum does in the statute.

Tier 2 requires documented employment. The desk observes that this is the only clinical formulary in the United States in which a therapy’s tier is conditioned on the patient holding a job, and that no member of the drafting committee appears to have found this notable, because it is how the underlying system already works.

Tier 3 is where our analysts locate the substantive content of the entire draft. It is defined not by an amount but by a property: the capacity to absorb an unexpected expense without catastrophe. This is the standard definition of financial resilience in the household-finance literature, it is measurable, it is the threshold at which most of the documented health effects appear, and it is the tier at which the draft imposes prior authorization. The document is therefore structured so that review is triggered precisely at the point where the intervention begins to work.

Tier 4 was classified as experimental by insurers within the week. The Research Division notes that experimental is a coverage determination rather than a scientific one, that it means the payer will not pay pending evidence, and that the evidence in question concerns whether owning a house is good for you. Our analysts decline to summarize the literature on housing stability and health outcomes on the grounds that the reader can guess it.

Tier 5 is the only tier on the formulary that is accurate without qualification. It is not clinically indicated, it is not available by prescription, and the stated routes of administration — inheritance and private equity — are correct. The desk notes that this is the sole entry in the document that describes existing American practice with complete precision, that it required no drafting committee to produce, and that it is the only tier for which nobody has requested additional clinical trials.

Economists Warn of Side Effects

Economists immediately warned that large-scale income interventions could affect:

The Surgeon General acknowledged these concerns.

“Absolutely.”

He pointed toward the medical evidence.

“But y’all are the economists.”

Then toward himself.

“I’m telling you this motherfucker can’t sleep because rent is due.”

An interagency working group was established.

Nobody expects it to finish before rent is due.

Our Research Division wishes to record that the economists’ warnings are legitimate and were treated in the coverage as though they were not. A general income floor interacts with labor supply, with prices in supply-constrained housing markets, with the tax base that funds it, and with the incentives of every party who currently pays a wage below it. These are real mechanisms with real magnitudes, several of them are contested in the literature, and a program of any scale would need to answer for all of them.

The desk’s observation concerns the Surgeon General’s reply, which is more precise than its delivery suggests. He did not dispute a single one of the six side effects. He conceded all of them in one word and then made a jurisdictional claim: that the identification of a harm and the modeling of its remedy are different professional tasks, and that he was performing the first.

Our analysts note that this is the standard structure of a public-health finding. The epidemiologist who established that a water pump was killing people was not required to first model the municipal budget for a new one. The finding that a thing is harmful is not a claim to know what the correct policy response costs, and a discipline that could only report harms it had already costed would report very few.

What the Research Division finds significant is the second half of the reply, in which the Surgeon General reduced an aggregate to a person who cannot sleep because rent is due. This is a rhetorical move and it is also a methodological one. Every one of the six side effects is denominated in an aggregate — a market, a price level, a rate. The condition is denominated in a person. A dispute conducted entirely in aggregates has no term for the patient, and the party who introduces the patient is generally accused of being unserious by the party who cannot fit him into the model.

The interagency working group is the outcome our analysts predicted and the one the record supports. Its charter runs to eleven pages. Its first report is scheduled for a date after the pilot’s funding authority lapses. Nobody expects it to finish before rent is due, and the desk notes that rent, uniquely among the parties to this proceeding, has never once missed a deadline.

IRS Announces Income Remains Taxable

The IRS reportedly issued emergency guidance reminding recipients that merely writing Rx before income does not necessarily make it tax-free.

A patient receiving therapeutic income expressed outrage.

“The doctor prescribed this.”

The IRS representative nodded sympathetically.

“Congratulations.”

He handed him a form.

“We prescribed withholding.”

The agency clarified that dosage calculations should therefore consider taxes.

Doctors responded by increasing prescribed gross income.

The IRS reportedly did not anticipate this development.

Our Research Division notes that the guidance is correct as a matter of law and that the agency’s speed reflects institutional experience rather than hostility. Income is taxable when received, characterized by its substance rather than by the document accompanying it, and no amount of clinical framing alters that. The agency has seen a great many attempts to recharacterize receipts and has never lost interest in any of them.

The desk draws attention to the sequence that followed, which is the most efficient feedback loop documented in this report. The agency stated that the dose would be taxed. The clinical guidance already specified taxes as one of the seven inputs to Local Therapeutic Income. Physicians therefore did the only thing the methodology permitted, which was to gross up the prescription so that the net dose still cleared the therapeutic threshold. This took under three weeks and required no negotiation with anybody.

Our analysts observe that the IRS did not anticipate this because the parties it usually deals with are optimizing for the after-tax amount they keep. Physicians were optimizing for the after-tax amount the patient needs, which is a different objective function and produces the opposite response to a withholding change. Nobody at the agency had modeled a counterparty who responds to taxation by increasing the taxable amount.

The Research Division notes one structural consequence that has attracted no attention and should. Grossing up moves money through the payer, through withholding, and back to the Treasury on the same schedule as the dose. Whoever is ultimately assigned to fund the prescription is therefore funding both the therapeutic amount and the public revenue it generates, and the Treasury’s net position depends entirely on who that party turns out to be — a question the pilot has now deferred for ten months, and the only question that has ever been at issue.

Researchers Discover Money Works Best When Patient Can Actually Use It

Early clinical trials have produced another remarkable finding.

Researchers compared two interventions.

Group A received financial-literacy pamphlets explaining budgeting.

Group B received enough money to pay their bills.

Group B displayed substantially greater enthusiasm for the study.

Researchers subsequently tested:

financial coaching without money

versus:

financial coaching with money.

Again, the group containing money performed surprisingly well.

One researcher reportedly stared at the results.

“Holy shit.”

“The financial literacy works better when there’s finances.”

The paper has been submitted for peer review.

Our Research Division has reviewed the trial design and wishes to defend the pamphlets before disposing of them. Financial literacy is a real skill, the material in the pamphlets is accurate, and households at every income level make expensive errors that better information would prevent. The intervention is not fraudulent. It is an intervention aimed at a decision, delivered to people whose problem is the size of the set the decision ranges over.

The desk notes that the second trial is the methodologically serious one and has been the least discussed. Comparing a pamphlet to money answers a question nobody was asking. Comparing coaching alone to coaching plus money isolates the contribution of the coaching, and the result is that coaching is substantially more effective when there is something to coach. This is not a repudiation of financial education. It is a specification of the conditions under which financial education has an object.

Our analysts observe that budgeting is an allocation procedure and that an allocation procedure requires a surplus over the constraint in order to produce different outcomes. Applied to a household whose obligations exceed its receipts, the procedure returns the same answer every month, correctly, which is that some obligation will go unmet. The literature has documented at length that households in this position are not making arithmetic errors. They are performing the arithmetic accurately and arriving at a result the pamphlet has no instruction for.

The Research Division wishes to record the finding that the trial did not measure and every participating researcher reported informally. Group B’s enthusiasm was not a treatment effect on mood. It was a treatment effect on retention: Group B returned for follow-up appointments at rates the study had powered for optimistically, and Group A did not, because attending a follow-up appointment costs a bus fare, a shift, and a parking fee that the study reimbursed on a thirty-day cycle.

Our analysts note the implication for the entire evidence base. Every study of this population that reimburses expenses on a delay is measuring, in part, the ability to float the reimbursement. The desk regards the enthusiasm gap as the study’s most portable result and notes that it appears in no abstract, because there is no field for it.

Surgeon General Rejects Claims Program Encourages Laziness

Officials stressed that the pilot specifically targets willing and able individuals whose earnings fall materially below locally adequate income.

The program is not premised on the idea that employment lacks value.

Quite the opposite.

One official explained:

“If somebody works forty hours and still can’t afford the basic shit required to continue showing up for those forty hours…”

He pointed toward the economic model.

“…we may have discovered a systems problem.”

The Department of Labor quietly requested a copy.

Our Research Division notes that the official’s formulation is narrower than the coverage reported and that the narrowness is the point. He did not assert that anyone is owed a living. He asserted a maintenance condition: that an arrangement requiring forty hours of attendance must, at minimum, fund the inputs that attendance consumes — transport to the site, food adequate to perform the work, sleep sufficient to perform it safely, and treatment for the conditions the work produces.

The desk observes that this is an accounting claim rather than a moral one. Every other input to production is expected to be maintained by whoever consumes it. Equipment is serviced, premises are heated, vehicles are fueled, and none of this is framed as generosity toward the equipment. Labor is the only input in the American ledger whose maintenance is treated as a matter for the input to arrange privately, on its own time, out of the residual.

Our analysts note that when maintenance is not funded from the transaction it is not thereby avoided. It is deferred and then billed elsewhere: to emergency departments that absorb the untreated condition, to public programs that supplement the wage, to the household member who covers the shortfall, and eventually to the employer in turnover, error, and absence. The pilot did not create these costs. It proposed to write them on the same page as the wage, which is why the Department of Labor requested a copy quietly rather than loudly.

The Research Division wishes to address the laziness objection directly, because it is the objection that has dominated the hearings and it is answerable on the pilot’s own terms. The eligibility criteria are willing and able individuals whose earnings fall materially below locally adequate income. A person meeting those criteria is, by construction, working. The objection asserts that paying such a person adequately will cause him to stop, which is a prediction about a population defined by the fact that it has not stopped, under conditions substantially worse than the ones the objection warns against.

First Patient Reports Dramatic Improvement

One participant reportedly entered the program while working full time and continuously struggling with rent, transportation, groceries, and bills.

His physician prescribed a local therapeutic income.

Several months later, researchers asked whether his well-being had improved.

“Yeah.”

“Are you exercising?”

“More.”

“Sleeping?”

“Better.”

“Eating better?”

“Yeah.”

“Stress?”

“Way down.”

Researchers asked which component of the intervention had been most effective.

The patient looked confused.

“The fucking money.”

Researchers wrote this down.

Our Research Division notes that the patient’s confusion is the appropriate response to the question and that the question was nevertheless required. Component analysis exists because most interventions are bundles, and a bundle whose active element is unidentified cannot be improved, cheapened, or reproduced. Asking which part worked is standard practice. The difficulty in this instance is that the intervention was not a bundle. It was a single agent, administered alone, and the patient was asked to decompose it.

The desk draws attention to the four intermediate answers, which the coverage skipped in order to reach the last one. He is exercising more, sleeping better, eating better, and his stress is down. These are, item for item, the first-line interventions his physician recommended at the start of this report and which he was unable to perform. Nobody counseled him a second time. The counseling had always been correct; it had been unfunded, and the funding was the missing input.

Our analysts wish to state the mechanism carefully, because it is the pilot’s only real finding and it is modest. The money did not improve his health. It purchased the conditions under which the recommended behaviors became available, and the behaviors improved his health. This is an ordinary causal chain of a kind the clinical literature handles without difficulty in every other context, and it is the reason the desk regards the program as unremarkable science and a remarkable administrative event.

The Research Division has one further observation about the interview. Every question in it was closed-ended, and every answer was one word. He was asked whether he was exercising, sleeping, eating, and stressed, and he was not asked what he had stopped doing. Our researchers later put that question to him. He said he had stopped taking the shift that ran until one in the morning on the day before his other job started at six. He said that nobody had ever asked about that shift, that he had worked it for three years, and that it had never appeared on any of his charts.

What the Prescription Actually Costs

The Research Division has attempted an independent reconstruction of the amounts in dispute, using the pilot’s own dosing methodology and publicly reported figures for a representative participant. The reconstruction is illustrative rather than audited, and the desk publishes it only to establish the order of magnitude of the argument.

Our analysts wish to draw attention to the structure of the reconstruction rather than to its figures. The increment in dispute is a single number, denominated in dollars, payable by an identifiable party on a known schedule. Every item below it is diffuse, denominated in different units, and payable by parties who do not appear on the transaction. This is the entire asymmetry, and it decides the outcome before any hearing begins.

The desk notes that the three unbilled lines are unbilled in the strict sense: the resources were consumed, the labor was performed, and no invoice exists anywhere. Physician time spent obtaining authorization is not a covered service, does not generate a claim, and is absorbed by the practice, which recovers it by shortening appointments. The cost is real and lands on the next patient in the schedule, who is not a party to the dispute and will not be told.

The three billed lines are billed to somebody, which is why they are visible, and the Research Division notes that in each case the party billed is not the party that generated the cost. The emergency department bills a public program. The public program bills a general fund. The replacement hire bills the employer who did the replacing, which is frequently a different employer than the one whose wage produced the departure.

The final line is the one every party to this proceeding has spent ten months not writing down. The prescription specifies a dose, a route, a frequency, and a duration. It does not specify a payer. Our analysts have reviewed the specimen, the guidelines, the draft formulary, and the interagency charter, and can confirm that no document produced by the Initiative contains a field for the payer, which is the only field anybody has argued about.

The Externality

Our Research Division’s assessment of the Income Sufficiency Therapeutic Initiative is that it discovered nothing, proposed nothing new, and produced ten months of institutional conflict anyway. The following sections set out why.

A Prescription Is a Reclassification

Nothing in the pilot changed a patient’s income, a physician’s knowledge, or the underlying evidence. What changed is the category the shortfall is filed under. Before the pilot, a person unable to afford housing was a person with a housing problem, which is a matter for a market. After the pilot, the same person is a patient with an untreated indication, which is a matter for a system that is required to respond to indications.

The desk regards this as the whole of the initiative and the reason for the reaction it provoked. Categories carry obligations. A market has no duty to any individual participant; a clinical system has a documented duty to the patient in front of it. Moving a fact from one category to the other does not alter the fact. It alters who is required to have an answer about it, and every party in this report responded by attempting to move it back.

Coverage Determines What Is Real

The insurer’s denial stated that income is not medically necessary. Our analysts note that medically necessary has never described a relationship between a treatment and a body. It describes a relationship between a treatment and a contract, and the contract was drafted by the party who pays when the answer is yes.

This produces a system in which the effective definition of medicine is the set of things that can be billed. An intervention that works but has no billing code is not rejected on the evidence; it is not evaluated at all, because there is no field in which the evaluation could be recorded. The Research Division notes that the pilot’s designers understood this perfectly, which is why they built the program around a prescription pad rather than a policy paper. A prescription is the one document in American life that forces a payer to produce a determination in writing.

The determination the payers produced was no. The desk observes that this was the predictable outcome and possibly the intended one, since a refusal in writing is a record, and there had previously been nothing to record.

Step Therapy Is a Cost Transfer

The four required steps — roommates, moving farther away, asking family, earning more independently — were treated in the coverage as absurd. Our analysts find them entirely coherent and note that each one satisfies the only criterion step therapy applies, which is that the alternative be cheaper for the payer.

Each step relocates the cost to a party outside the claim: a housemate, a commute, a relative, or the patient’s remaining hours. The total cost of the underlying condition is unchanged by all four. The payer’s share falls to zero, the claim closes as resolved, and the resolution is accurate within the boundary the claim draws. The Research Division notes that the boundary is the mechanism. Everything outside it is not disputed, denied, or contested. It is simply not visible from inside the file.

The Delivery Mechanism Already Exists

The employer identified in one sentence what the briefing had spent an hour not saying. A dose administered every two weeks by direct deposit to a working adult is a paycheck. Our analysts have confirmed that the identification is exact and that it has an implication the Initiative has never contested.

If the delivery mechanism already exists, then the mechanism was never the constraint, and the prescription is not proposing a new apparatus. It is proposing a different number in an existing field. Every institutional actor in this report — the insurer, the manufacturer, the Treasury, the working group — has spent ten months on questions of mechanism, coverage, tier, and code, none of which is the constraint, because the field is already there, on the employer’s system, and the argument is about what goes in it.

The desk notes that this is why the Department of Labor’s interest was immediate and quiet. The Department is the only agency in the proceeding whose statutory subject is that field.

The Dose Was Always Being Set

The most common objection to the pilot is that it introduces a clinical judgment into a market outcome. The Research Division’s reading of the record is that it introduces a prescriber into a dosing decision that was already being made, continuously, by parties with no clinical training and no obligation to consider the indication.

Every wage is an amount, and every amount produces the health outcomes associated with that amount. The literature establishing this is not in dispute and was not disputed at any point in ten months of hearings. What the pilot changed is that a person with a license was required to write the number down and sign it, and the objection is not that the number is wrong. The objection is to the signature, because a signature creates a party who can be asked why.

The Unpriced Party

Our Research Division’s conclusion concerns the patient in the final interview, who reported that he had stopped working the shift that ended at one in the morning before a shift that began at six.

That shift existed for three years. It appears in no clinical record, in no claim, in no formulary, and in no economic model produced by any party to this proceeding. It was funded entirely by the patient, out of his sleep, and the sleep was the input the counseling protocol had instructed him to protect. The cost was real, continuous, and borne in full by the only participant who never appeared as a line item.

The desk notes that the pilot did not measure this and could not have. There is no code for a shift that should not exist. When the money arrived, the shift ended, and the improvement the researchers recorded as a treatment effect was in substantial part the discontinuation of an arrangement that nobody had ever priced, on either side of the ledger, in either direction, for three years.

Stakeholder Perspectives

Patients

Participants interviewed for this report were markedly less impressed by the initiative than its designers. Most understood the finding before it was explained to them and several were irritated by the explanation. Asked what he made of the pilot, one respondent said that he was glad to have the money and that he did not understand why it had required a doctor. Asked whether the clinical framing had helped him, he said it had helped him get it, and that he could not think of another route by which he would have gotten it, and that he considered both of these facts to be answers to different questions than the one he had been asked.

Physicians

Participating clinicians were the group least surprised by any part of the pilot and the group most affected by it administratively. Every physician our researchers spoke to reported that the underlying finding had been obvious in practice for their entire careers. Every one of them also reported that the pilot had added between forty minutes and three hours per patient in authorization work, none of it reimbursed. One noted that she now spends more time documenting that a patient cannot afford food than she spends treating the conditions the shortage causes, and asked our researcher to record that she intends to continue doing it, because the documentation is the only part that generates a record anybody else can read.

Insurers

Health plan representatives declined to be interviewed on the record and supplied written statements reiterating that the initiative raises coverage questions properly resolved through established medical-necessity review. Our analysts note that the statements are accurate and that they answer a question about process with a description of the process. Presented with the observation that every branch of that process terminates in denial, one representative said that the process is designed to determine whether a service is covered and that income is not a service. Asked what the patient should do, he said that the patient should discuss it with his employer.

Employers

Employers were the most candid parties in the proceeding and the most consistent. Several volunteered that they already knew which of their employees could not afford to live near the site, that they know it from the attendance data, and that the data has been unambiguous for years. Asked why the wage had not moved, respondents gave the same answer in different vocabularies: that no competitor’s wage had moved, that moving alone is expensive, and that the wage is set by what the position can be filled for rather than by what the person requires. The desk notes that this is a correct description of a labor market and that not one employer disputed the clinical finding at any point.

The Treasury and the IRS

Fiscal officials treated the initiative as a characterization question and resolved it within days, which our analysts regard as the most efficient work product in this report. Asked whether the agency had a view on the underlying public-health finding, a Treasury official said that the agency’s view on public-health findings is that they are taxable when received. He then noted, unprompted, that the withholding tables had been ready since before the hearings began, and that this had required no policy decision of any kind, because withholding does not depend on why the money moved.

Economists

The economists our researchers contacted were substantially more divided among themselves than the coverage indicated, and substantially less dismissive of the pilot. Several regard the labor-supply and price effects as serious and unresolved. Several others noted that the current arrangement is not a control condition, that its own effects are large and mostly uncounted, and that comparing an intervention against an idealized status quo is a specification error. One respondent, asked to summarize the disagreement, said that the profession is being asked whether the cure has side effects, that the answer is yes, and that nobody has been asked to price the disease since roughly 1978.

Closing Statement

The Surgeon General emphasized that income prescriptions remain experimental and should not be interpreted as suggesting that money alone determines health or happiness.

Human well-being is influenced by relationships, physical health, community, environment, behavior, purpose, genetics, healthcare access, and countless other factors.

Officials merely acknowledged one additional variable:

BEING ABLE TO AFFORD YOUR FUCKING LIFE

At the program’s first congressional hearing, a senator demanded to know whether the federal government was seriously proposing that doctors prescribe money.

The Surgeon General responded:

“Not exactly.”

The senator relaxed.

“We’re saying decades of public-health research keeps pointing toward adequate material resources as one component of well-being.”

“So what’s the prescription?”

The Surgeon General slid a document across the table.

The senator stared at it.

“Who’s supposed to fill this?”

The Surgeon General looked toward the Department of Labor.

The Department of Labor looked toward employers.

Employers looked toward Congress.

Congress looked toward the Treasury.

Treasury looked toward the IRS.

The IRS had already prepared the withholding tables.

Our Research Division notes that the senator asked the only question in ten months of proceedings that goes to the substance, that he asked it in four words, and that the room answered him by turning around.

The desk observes that the rotation is not evasion in the ordinary sense. Every party in the sequence was pointing at the party with the authority to act, and every one of them was correct. Labor regulates the transaction but does not set the amount. Employers set the amount but not the market. Congress can move the market but requires a vote. Treasury implements what Congress passes. The IRS collects on whatever moves, whenever it moves, for whatever reason, and had therefore finished its work before the hearing convened.

The Research Division wishes to record the composition of the room. Present were the Office of the Surgeon General, the Department of Labor, the Department of the Treasury, the Internal Revenue Service, counsel for three health plans, two trade associations, and eleven members of Congress. The patient was not present. The prescription written in his name was entered into the record, read aloud, photographed, and returned to the Surgeon General unfilled, and our researchers confirm that at the moment the hearing adjourned it remained the only document in the room with a specified dose.

Editor’s note: Our researchers requested cost-of-living dosing calculations for the fourteen counties in the pilot and received them within a day, from three separate agencies, in formats that agreed to within a few percent. We then requested the name of the party responsible for funding a prescription once written, and were referred, over six weeks, to five agencies, two congressional committees, and a trade association, none of which claimed the question and all of which knew precisely where to send it next. The Research Division notes that the government of the United States can compute what a person needs to live, to the dollar, by county, in under twenty-four hours, and has no office at which that figure can be presented.