$500 idea · Pattern break · Edition 054

Medicare’s 2027 rule pays ACO clinicians double on a code 76% never bill.

Three days ago, on 4 November 2026, CMS issued CMS-1848-F. Every trade outlet led with the conversion factor: $32.63, a 2.31% cut. In the same rule CMS retires G2211 and replaces it with two modifiers — B1, which “would increase the payment of the associated E/M code by 16%, instead of a flat rate,” and B2, which “will increase the payment of the associated E/M code by 32 percent” and is reserved for practitioners in a Medicare ACO. The 16-point gap between them is 6.9264× the cut that got the headlines. And a JAMA analysis of 100% of traditional Medicare claims found that of 855,265 clinicians who billed a G2211-eligible visit in 2024, “24.0% billed G2211” — 14.5% of eligible visits, $394 million paid, and $2,196,068,518 left on the table at the realised rate. The repricing starts in 55 days.

The scorecard

6 fixed dimensions, scored for one person starting with $500

Real-world feasibility5 = highly feasible	3 out of 5
Operating difficulty5 = very hard	4 out of 5
Ongoing input after launch5 = heaviest	3 out of 5
Probability of failure5 = very high	4 out of 5
Legal and regulatory risk5 = very high	3 out of 5
Team and support needs5 = team required	1 out of 5

Why each score: the playbook

Jump to the ladder

The $500 Ladder Saturday, 7 November 2026 Edition 054

Context

When an American doctor sees a patient in the office, Medicare pays from a national price list that the government rewrites every autumn and publishes in early November. For the last three years that list has carried an extra payment for a doctor who is the patient’s continuing clinician rather than a one-off — an add-on claimed by ticking an extra code on the bill. Most doctors never ticked it, and almost nobody outside billing departments noticed. From January the extra payment stops being a flat amount and becomes a percentage of the visit, at one rate for most doctors and twice that rate for doctors who belong to one of the groups that share savings with Medicare. Today’s edition is about who gets paid to notice, and whether the arithmetic supports a one-person business doing it.

Start with what the rule actually does, because the thing being reported and the thing being changed are not the same. CMS issued the CY 2027 Medicare Physician Fee Schedule final rule, CMS-1848-F, on 4 November 2026. Its own fact sheet gives two conversion factors: $32.95 for qualifying APM participants, down $0.62 or 1.83% from $33.57, and $32.63 for everyone else, down $0.77 or 2.31% from $33.40. That is the number in every headline and it is a real cut, caused by the expiry of the one-year 2.50% increase under Public Law 119-21.

Buried in the same fact sheet is a different kind of change. The office/outpatient E/M visit complexity add-on, HCPCS G2211, stops being paid as a separate code on 1 January 2027. In its place CMS finalised a modifier, B1, that “would increase the payment of the associated E/M code by 16%, instead of a flat rate” — and a second modifier, B2, “to be used in place of G2211 for practitioners participating in a Medicare Accountable Care Organization,” which “will increase the payment of the associated E/M code by 32 percent instead of a flat rate.” CMS’s own fact sheet adds that B2 “can be billed for all beneficiaries they serve,” not only those assigned to the ACO, and that use of it is voluntary.

Set the two against each other. A conversion-factor cut of 2.31% falls on every service a clinician bills. A 16-percentage-point difference falls only on office E/M visits — but on those visits it is 6.9264× the size of the cut, and the full 32 points is 13.8528×. The rule that was reported as a pay cut is, for the office visit, a repricing in which membership of an accountable care organisation is worth more than the cut takes away. Nobody wrote that headline.

Now the part that makes it a business rather than an observation. In JAMA, 19 February 2026, Ganguli, Daley, Hicks, McWilliams and Rosenthal published a cross-sectional analysis of 100% traditional Medicare claims for 2022–2024. In 2024, “a total of 26 million G2211 codes were billed by 206 813 clinicians for 10.6 million patients,” “totaling $394 million in payment (1.1% of these clinicians’ total Medicare billing).” And: “Among the 855 265 clinicians who billed any G2211-eligible visit in 2024, 24.0% billed G2211.” Three quarters of the eligible population billed the code zero times.

The visit-level figures are starker. Billing rates were “24.1% vs 13.1% for specialist physicians and 8.9% for other clinicians” — 9,616,505, 10,789,868 and 4,206,416 visits respectively. Divide each by its rate and the eligible pool is 169,531,016 visits, of which 24,612,789 carried the code: 14.5182%. The other 144,918,227 eligible visits carried nothing. At the realised rate of $394,000,000 ÷ 26,000,000 = $15.1538 a code, that is $2,196,068,518.02 of a payment Congress authorised and clinicians declined.

Here is the non-consensus reading, and the evidence for it is in the two documents above rather than in anyone’s commentary. The 2027 rule is not principally a pay cut; it is a conversion of an under-claimed flat add-on into a percentage, with a 2.00× multiplier attached to ACO membership. CMS’s own Shared Savings Program Fast Facts, as of 1 January 2026, count 511 ACOs, 12.6 million assigned beneficiaries, 15,353 participant TINs and 687,739 participating physicians and non-physicians. Against the 855,265 clinicians with an eligible visit that is 80.4124% — derived, across different years and different definitions, and treated below as an indicator rather than a rate. The point stands either way: the 32-point side of this trade is not a niche. It is most of the market, and most of the market was not billing the 16-point version.

Which brings the arithmetic to the operator, who is not the doctor. Per clinician with an eligible visit, the pool is 198.22 eligible visits a year and 169.44 of them unbilled. Price the 2027 add-on at the level that makes B1 worth exactly what G2211 realised — an implied average base visit of $15.1538 ÷ 0.16 = $94.71 — and the recoverable amount per clinician is $2,567.71 a year under B1 and $5,135.41 under B2. As a monthly fee a reviewer could charge and still leave the practice ahead, that is $213.98 and $427.95. Those two numbers are the whole business: below $213.98 a month any clinician can pay you out of their own recovery; between $213.98 and $427.95 only an ACO clinician can; above $427.95 nobody can at the average non-billing rate. $5,000 a month after 2026 federal tax needs $79,812.05 of fee revenue, which is 31.08 clinicians at the B1 line and 15.54 at the B2 line.

The curve is the operator’s problem; the two verticals are the client’s. $79,812.05 of fee revenue buys $5,000 a month after 2026 federal tax, so the number of clinician-clients needed falls as 79,812.05 ÷ 12f — 133.02 at $50 a month, 11.09 at $600. What a clinician can actually pay is fixed by the add-on being recovered: 169.44 unbilled eligible visits a year at the realised $15.1538 is $213.98 a month under B1, and exactly twice that, $427.95, under B2. The shaded band between them is the part of the price list that only an ACO practitioner can fund, and it is where the business is: 31.08 clients at the B1 line, 15.54 at the B2 line. Both verticals assume the average non-billing rate; a practice already billing the modifier on most eligible visits sits far to the left of both.

The evidence

8 findings

Agency documentation · the repricing CMS, Calendar Year (CY) 2027 Medicare Physician Fee Schedule Final Rule, CMS-1848-F issued 4 November 2026; read 7 November 2026

“increase the associated E/M code payment by 16%” — and 32% inside an ACO

CMS finalised replacing the office/outpatient E/M visit complexity add-on code G2211 with a modifier from 1 January 2027. The fact sheet describes modifier B1 as increasing “the associated E/M code payment by 16%” and modifier B2, for practitioners in Shared Savings Program ACOs or LEAD Model ACOs, as increasing “the associated E/M payment by 32%.” B2 “is voluntary, and can be billed for all beneficiaries they serve” — not only beneficiaries assigned to the ACO. Read what that sentence does. It detaches the premium from the assigned population and attaches it to the practitioner, so a single ACO participation agreement reprices every office E/M visit that practitioner bills to fee-for-service Medicare, assigned or not. The same fact sheet gives the conversion factors: $32.95 for qualifying APM participants (down $0.62, 1.83%, from $33.57) and $32.63 for everyone else (down $0.77, 2.31%, from $33.40), the fall driven by expiry of the one-year 2.50% increase under Public Law 119-21. Caveat. The fact sheet, not the rule text, is the source read on this run; the full CMS-1848-F text and its Federal Register citation were not read, and the modifier descriptors as they will appear in the HCPCS file are not verified.

Specialty-society summary · the modifier language, confirmed twice American Academy of Neurology, 2027 Medicare Physician Fee Schedule Final Rule summary read 7 November 2026

“instead of a flat rate” — the phrase that does the work

The AAN summary records that CMS finalised policy to “transition HCPCS code G2211 to a two-digit HCPCS modifier, B1, that can be appended to the associated E/M base code,” and that this “would increase the payment of the associated E/M code by 16%, instead of a flat rate.” Separately CMS is finalising a modifier “to be used in place of G2211 for practitioners participating in a Medicare Accountable Care Organization” which “will increase the payment of the associated E/M code by 32 percent instead of a flat rate.” Why this row exists. Two independent documents, an agency fact sheet and a specialty society’s own reading of the rule, give the same two percentages and the same ACO restriction. The phrase “instead of a flat rate” is the mechanism: a flat add-on is worth the same on a level-2 visit and a level-5 visit, a percentage is not, and a percentage that doubles on a membership test is a different instrument again.

Peer-reviewed claims analysis · who actually billed it Ganguli I, Daley NE, Hicks AL, McWilliams JM, Rosenthal MB. “Billing of Medicare’s G2211 Longitudinal Care Code Among Traditional Medicare Beneficiaries.” JAMA. 2026;335(11):1003–1006; doi:10.1001/jama.2026.0424 published online 19 February 2026; 100% traditional Medicare claims 2022–2024; read 7 November 2026

“Among the 855 265 clinicians who billed any G2211-eligible visit in 2024, 24.0% billed G2211”

In 2024 “a total of 26 million G2211 codes were billed by 206 813 clinicians for 10.6 million patients,” “totaling $394 million in payment (1.1% of these clinicians’ total Medicare billing).” Among clinicians who did bill it, the median share of their own eligible visits carrying the code was “30.1% (6.9%-61.8%).” By clinician type the billing rate was “24.1% vs 13.1% for specialist physicians and 8.9% for other clinicians,” on 9,616,505, 10,789,868 and 4,206,416 visits; specialists billed “43.0%” of all G2211 codes, primary care physicians 39.7%, other clinicians 17.4%. And the counter-evidence, which belongs in the same row. The billing clinician had “no observed prior or subsequent visit with the patient for 15.2% of all initial G2211-billed visits” — 8.2% for primary care physicians, 9.0% for specialists, 30.1% for other clinicians. Part of what is already billed does not look longitudinal. The authors call the uptake “robust early uptake” relative to earlier care-management codes and conclude the results “demonstrate challenges in steering payment via specialty-agnostic fee schedule changes.”

Agency documentation · the size of the 32% side CMS, Shared Savings Program Fast Facts — As of January 1, 2026 as of 1 January 2026; read 7 November 2026

687,739 physicians and non-physicians, 511 ACOs, 12.6 million assigned beneficiaries

The Shared Savings Program counts 511 ACOs with 12.6 million assigned beneficiaries, 15,353 participant TINs and 687,739 participating physicians and non-physicians, plus 1,539 hospitals, 8,840 Federally Qualified Health Centers, 3,109 Rural Health Clinics and 590 Critical Access Hospitals. 76% of ACOs are in two-sided risk; 325 (64%) are low-revenue and 186 (36%) high-revenue; 129 (25%) use prospective assignment and 382 (75%) preliminary prospective with retrospective reconciliation. That averages 1,345.87 clinicians and 24,658 assigned beneficiaries per ACO. Caveat, load-bearing. This count is as of January 2026 and counts participation in the programme; the 855,265 in the row above counts clinicians with a G2211-eligible visit in calendar 2024. They are different years, different populations and different definitions, so 687,739 ÷ 855,265 = 80.4124% is derived and is an indicator of scale, not a participation rate.

Agency documentation · the ACO door, and when it is open CMS, CY 2027 PFS Final Rule (CMS-1848-F) — Medicare Shared Savings Program; CMS, Application Types & Timeline fact sheet 4 November 2026; timeline page last modified 1 September 2026; both read 7 November 2026

Phase 2 RFI closes “Nov. 12-17, 2026, at 12:00 p.m. (noon) ET”

For the 1 January 2027 start date, applicants “will be able to change their final selection during Phase 2 RFI” in the window “Nov. 12-17, 2026, at 12:00 p.m. (noon) ET.” The 2027 application itself ran from 9 June to 23 June 2026 in Phase 1, with later phases through 16 December 2026. CMS also finalised raising the BASIC track Level E sharing rate “from 50% to 60%,” increasing “the prior savings adjustment’s scaling factor from 50% to 75%,” and “removing the prepaid shared savings payment option” with “no additional cohorts of ACOs” admitted after the 1 January 2027 cycle. Advance investment payments for PY 2028 move to “a quarterly payment of $45” for qualifying beneficiaries and $25 for others, capped at 10,000. What this fixes. Whether a given practitioner may bill B2 in 2027 is settled by an ACO participation decision made in 2026, not by anything the practitioner does in January. Caveat. CMS has not published a 2028 application timeline, and the exact mechanics by which a practitioner is confirmed eligible for B2 on a claim were not verified on this run.

Enforcement documentation · how the operator may be paid HHS Office of Inspector General, Advisory Opinion No. 98-4; 42 CFR § 1001.952(d); Social Security Act § 1128B(b) issued 15 April 1998; read 7 November 2026

“a longstanding concern that percentage billing arrangements may increase the risk of upcoding”

The OIG reviewed an arrangement paying a management and billing company “a percentage of Company A’s monthly net revenues” and concluded it “may constitute prohibited remuneration” under the anti-kickback statute. It did not fit the personal services and management contracts safe harbour at 42 CFR § 1001.952(d) because the compensation was not a fixed aggregate amount set in advance; there was “at least a potential technical violation of the anti-kickback statute”; and the Office recorded “a longstanding concern that percentage billing arrangements may increase the risk of upcoding,” since the contractor “has an incentive to maximize Company A’s revenue.” Why it decides the business model. A reviewer paid a share of what the practice recovers is, on the agency’s own published reasoning, in the shape it has warned about for twenty-eight years — and the thing being recovered here is a modifier that pays 16% or 32% more per visit. A flat fee fixed in advance is the structure the safe harbour contemplates, and it is also what sets the break-even lines in the figure above.

Reported · the one price on the entry path AAPC, Certified Professional Coder (CPC) certification read 7 November 2026

$399 for the exam, “Proof of education or experience isn’t necessary to sit for these exams”

The CPC exam costs $399 ($325 for AAPC students) and the price includes “one free retake.” No education or experience is required to sit it. The designation, however, is gated: without proof of at least two years of medical coding experience the credential issued is CPC-A, the apprentice form. Membership must be maintained and 36 continuing education units submitted every two years. Caveat, and it matters to the verdict. The page does not state the membership fee, and it does not say the credential is legally required — because it is not. No federal licence governs reviewing a practice’s own claims under a business associate agreement. The credential is a commercial signal, which is why it sits in the entry cost rather than in the scarce input.

Derived · the arithmetic, shown From the CMS CY2027 fact sheet, the JAMA figures, the Shared Savings Program Fast Facts and the 2026 federal rates in IR-2025-103 announcing Rev. Proc. 2025-32 computed 7 November 2026

$2,196,068,518.02 unbilled, 31.08 clients at the B1 line, 15.54 at the B2 line

The add-on. $394,000,000 ÷ 26,000,000 = $15.1538 a code. Eligible visits by group: 9,616,505 ÷ 0.241 = 39,902,510; 10,789,868 ÷ 0.131 = 82,365,405; 4,206,416 ÷ 0.089 = 47,263,101; total 169,531,016, billed 24,612,789 (14.5182%, against the paper’s stated 14.5% on 24,610,168 — a rounding difference of 2,621 visits). Unbilled 144,918,227 = 85.4818%, worth $2,196,068,518.02 at the realised rate. Clinicians billing none: 855,265 × 0.76 = 650,001. Per clinician. 169,531,016 ÷ 855,265 = 198.22 eligible visits; 144,918,227 ÷ 855,265 = 169.44 unbilled. The implied base visit. The base E/M at which B1 is worth exactly what G2211 realised is $15.1538 ÷ 0.16 = $94.71; at that base B1 = $15.1538 and B2 = $30.3077, ratio 2.0000. This is an implied figure, not a published rate — no 2027 per-code payment amount was read on this run. Recoverable. 169.44 × $15.1538 = $2,567.71 a year ($213.98 a month) under B1; $5,135.41 ($427.95) under B2. The spread. (32 − 16) ÷ 2.31 = 6.9264×; 32 ÷ 2.31 = 13.8528×. Conversion-factor change computed from the fact sheet’s own pairs: ($32.63 − $33.40) ÷ $33.40 = −2.3054%; ($32.95 − $33.57) ÷ $33.57 = −1.8469%. The income model. $5,000 a month after 2026 federal tax, single filer, sole proprietor, standard deduction $16,100, brackets 10/12/22% at $12,400/$50,400/$105,700, self-employment tax 15.3% of 92.35%, § 199A excluded: profit $78,212.05 − SE $11,051.01 − income tax $7,161.04 = $60,000.00; taxable income $56,586.54; total federal tax $18,212.05, 23.2855% of profit. With an assumed fixed line of $1,600 (AAPC membership and CEUs $220, HIPAA-compliant storage $300, errors-and-omissions cover $600, software $480 — all assumed), fee revenue needed is $79,812.05. Clients. N(f) = 79,812.05 ÷ 12f: 133.02 at $50, 66.51 at $100, 44.34 at $150, 31.08 at $213.98, 26.60 at $250, 16.63 at $400, 15.54 at $427.95, 11.09 at $600. Entry. $399 exam (verified) + $50 registration (assumed) + $50 storage (assumed) = $499. Dates. 7 November to 1 January 2027 = 55 days; to the Phase 2 RFI close on 17 November = 10. Plot coordinates. x(f) = 110 + 540f ÷ 600, y(N) = 330 − 250N ÷ 140. B1 line at x = 302.58, N = 31.08, y = 274.49; B2 line at x = 495.16, N = 15.54, y = 302.25; band width 192.58 px. Curve plotted from f = $50 to $600 in $10 steps, first point 155.0,92.46, last 650.0,310.21.

The ladder

From $500 to $5,000 a month

Six rungs from $499 to $5,000 a month, built on a repricing that starts in 55 days — and on the 76% of eligible clinicians who never billed the old version of it.

Read the two sentences that the pay-cut coverage left out

CMS-1848-F, issued 4 November 2026: modifier B1 increases “the associated E/M code payment by 16%”; modifier B2 increases it by 32% and is reserved for Shared Savings Program and LEAD Model practitioners, “is voluntary, and can be billed for all beneficiaries they serve.” The reported cut is 2.31% on the conversion factor. On an office visit the gap between the two modifiers is 6.9264× that cut and the full B2 is 13.8528×. Everything below follows from reading the rule rather than the coverage.
B1 = +16% · B2 = +32% · (32−16) ÷ 2.31 = 6.9264× · effective 1 Jan 2027
Size the unbilled pool from the claims paper, not from a vendor deck

JAMA, 100% traditional Medicare claims: of 855,265 clinicians with an eligible visit in 2024, “24.0% billed G2211.” Reconstructing the eligible pool from the three published rates gives 169,531,016 visits, 24,612,789 billed, 144,918,227 not — 85.4818%, worth $2,196,068,518.02 at the realised $15.1538 a code. That is the market. It is not a forecast; it is what already happened in one calendar year.
$394M ÷ 26M = $15.1538 · 144,918,227 unbilled · $2,196,068,518.02
Convert the pool into one number a practice can act on

Per clinician: 198.22 eligible visits a year, 169.44 of them unbilled. Price the 2027 add-on at the base visit that makes B1 worth what G2211 realised — $94.71, implied rather than published — and the recoverable amount is $2,567.71 a year under B1 and exactly twice that, $5,135.41, under B2. Monthly, $213.98 and $427.95. Do not quote a practice a figure above its own line.
169.44 × $15.1538 = $2,567.71/yr = $213.98/mo · B2 = $5,135.41 = $427.95/mo
Take the flat fee, because the agency published its view of the alternative twenty-eight years ago

OIG Advisory Opinion 98-4: a percentage of collections falls outside the personal services safe harbour at 42 CFR § 1001.952(d), which requires compensation “fixed in advance,” and the Office recorded “a longstanding concern that percentage billing arrangements may increase the risk of upcoding.” When the thing being recovered pays 16% or 32% more per visit, a contingent fee is the arrangement the agency has already described. Charge a fixed monthly amount per clinician, in writing, before the first chart is opened.
42 CFR § 1001.952(d) fixed in advance · AO 98-4, 15 April 1998
Pick the price, and the price picks the client

$79,812.05 of fee revenue buys $5,000 a month after 2026 federal tax — profit $78,212.05, SE tax $11,051.01, income tax $7,161.04, plus an assumed $1,600 fixed line. N(f) = 79,812.05 ÷ 12f: 44.34 clinicians at $150, 31.08 at $213.98, 26.60 at $250, 15.54 at $427.95. Below $213.98 any clinician can fund you out of recovery; between $213.98 and $427.95 only a B2 clinician can. The price you choose is a decision about which half of the market you are allowed to sell to.
N(f) = $79,812.05 ÷ 12f · 31.08 at the B1 line · 15.54 at the B2 line
Sell the implementation deadline, not the recovery — and refuse the upcoding sale

Every practice billing Medicare office visits must change its billing on 1 January 2027, because G2211 stops being paid that day. That is a dated, compulsory, universal event 55 days out, and it is a far easier first conversation than a recovery pitch. The refusal matters as much: the same JAMA paper found the billing clinician had “no observed prior or subsequent visit with the patient for 15.2% of all initial G2211-billed visits.” Some of what is already billed does not look longitudinal. A reviewer who finds documented eligible visits has a business; one who manufactures them has a False Claims Act problem.
55 days to 1 Jan 2027 · 15.2% of initial G2211 visits had no other visit · document first

Interested in the actual playbook?

More like this

Same series or same scarce input

Browse the catalog — every edition, by verdict, scarce input and window