GLP-1 and the Obesity Economy

The question

Is the incretin category priced as a late-stage specialty launch that tops out in the mid-2030s near $150B, or as a chronic metabolic franchise with cardiovascular, renal, hepatic, and neuro pull-through that re-prices the entire cost-of-care stack for the 42% of US adults who meet the obesity threshold?

Base rates

The reference class is prior chronic-disease categories that crossed from specialty into primary care: statins (1987-2005), SSRIs (1988-2004), and antihypertensives (1978-1995). In each case, the category took 8-12 years to reach terminal penetration, peak revenue landed 3-4x above consensus five years in, and the originator duopoly retained roughly 55-65% share through loss of exclusivity because formulation cycles and indication expansion extended the moat. Gross margins stayed above 80% until generic entry, and the picks-and-shovels sub-tier -- glass vials, elastomer stoppers, pre-filled syringes, cold-chain -- compounded revenue at a premium to the branded names in years four through eight because utilization scaled faster than price compressed.

The obesity category is roughly four years into a comparable curve. Injectable semaglutide and tirzepatide have reached roughly 12 million US patients; the addressable pool at current label is closer to 110 million. The imputed penetration embedded in forward consensus is around 18-22% by 2030. The SELECT and SURMOUNT cardiovascular outcome readouts have already moved the payer conversation from cosmetic to preventive; that is a discrete change in the base rate for coverage, not a marketing claim.

Healthcare has trailed SPY over the last twelve months by a wide margin -- the sector is down while the index compounded. That underperformance is the thesis. Multiples on LLY and NVO have compressed roughly 25-35% from peak on concerns about oral entrants, CMS negotiation, and compounding pharmacy substitution. The imputed-expectations gap is the setup.

Why consensus is wrong

Consensus treats the category as a pricing story. The binding variable is supply. Fill-finish capacity for GLP-1 injectables is still roughly 18-24 months behind demand; compounded versions exist only because the FDA shortage list created a legal window, and that window is closing as LLY and NVO bring new aseptic lines online. When the constraint shifts from access to script volume, the economics of the originators improve non-linearly, because marginal unit cost on a scaled aseptic line is a fraction of the gross price.

The second miss is the oral formulation cycle. The market reads an oral GLP-1 as a margin-dilutive substitute. A working reference class -- oral versus injectable insulin analogs, oral versus infused biologics in rheumatology -- suggests the oral wave expands the prescribing population by 3-4x while cannibalizing injectable volumes by roughly 20-30%. Net category TAM expands, not contracts.

Third, Medicare Part D repricing under the Inflation Reduction Act is already modeled in consensus. What is not modeled is the CMS coverage path for obesity as a standalone indication, which moves from denied to conditionally covered once cardiovascular outcome evidence is accepted. A coverage decision for the 65-plus population with established CV disease is a stepwise increase in addressable scripts that the sell-side has been treating as a binary rather than a probability distribution.

Position construction

The book has three 20% anchors, a catalyst book, a distribution sleeve, and a small-cap optionality tail.

Originator duopoly (~40%). LLY at 20% and NVO at 20% are the two molecules, the two cardiovascular outcomes datasets, and the two capacity build-outs. Owning both removes molecule-selection risk; the differentiation between tirzepatide and semaglutide matters for share within the category, not for category TAM.

Catalyst book (~46.7%). ABBV at 20% is the non-incretin adjacency -- Skyrizi, Rinvoq, and the immunology franchise that continues to compound while the market is distracted by the GLP-1 narrative; it is a ballast position whose multiple has been dragged down with the sector. AMGN at ~14.3% owns MariTide, the monthly dosing candidate whose Phase 3 readouts are the cleanest near-term catalyst outside the duopoly. PFE at ~12.4% is the oral danuglipron/next-molecule book plus the broader pipeline at a mid-single-digit free cash flow yield that requires almost no GLP-1 contribution to underwrite.

Distribution (~10.3%). CVS at ~9.3% is the PBM and retail pharmacy that dispenses the scripts, with Caremark's formulary position translating category volume into gross profit dollars regardless of which molecule wins. HIMS at ~1.0% is the direct-to-consumer cash-pay channel that captures patients outside commercial coverage.

Picks-and-shovels (~2.3%). WST at ~1.7% and STVN at ~0.6% are the elastomer stoppers, pre-filled syringes, and fill-finish capacity that every injectable dose requires. Their revenue scales with units, not price.

Next-wave and adjacency optionality (~1.0%). VKTX at ~0.3% is the oral dual-agonist and injectable VK2735 readout book. CRNX at ~0.4% is the oral somatostatin analog and broader endocrine pipeline. WW at ~0.02% is a de minimis behavioral-adherence call-option; the position is small enough that the binary outcome does not meaningfully affect the book.

Asymmetric payoff

If injectable capacity doubles by late 2027, oral formulations launch into a TAM-expanding rather than substitutive pattern, and CMS accepts obesity as a covered indication for the CV-established population, the weighted book returns roughly 22-32% annualized over three years. If the IRA negotiation cycle catches semaglutide earlier than expected and compounding persists longer, the book returns roughly -5% to +5%. If a second cardiovascular or neuro outcome readout moves the coverage conversation toward primary prevention, the right tail is 40-55% with material multiple re-rating on the originators.

At 55% base, 25% bear, and 20% bull, expected value is roughly +16 to +22% annualized against an SPY base rate near +8%. The payoff is asymmetric because the originators already trade at compressed multiples with growing cash flows, which truncates the bear case, while supply-constrained volume and TAM expansion keep the bull tail open.

Three things that would change our mind

1. A CMS proposed rule in 2026 that explicitly excludes obesity as a covered indication for the Medicare population with established cardiovascular disease, with language indicating the exclusion survives administrative challenge.
2. Phase 3 SURPASS-CVOT-style readouts for a non-duopoly oral incretin showing non-inferior CV outcomes at roughly half the gross price, with a commercial launch cleared for 2027 -- this would collapse the pricing premium before the originators have scaled capacity.
3. Real-world evidence at 24-month follow-up showing muscle mass loss, regain rates above 75% off-therapy, or a durable increase in pancreatitis or thyroid signals that moves the FDA label toward a black-box warning and caps script growth.

What we're explicitly NOT betting on

We are NOT betting on a specific CMS negotiation outcome for semaglutide or tirzepatide -- the book survives a bad print because the capacity and volume story operates on a different axis than single-molecule pricing. We are NOT betting on LLY over NVO or vice versa; the 20/20 sizing is deliberate. We are NOT betting on any single oral readout from VKTX, PFE, or AMGN; each is sized as optionality, not as a core position. We are NOT betting on compounding pharmacies being shut down or preserved. We are NOT betting on a specific weight-loss percentage in a future Phase 3. The thesis requires only that obesity coverage expands, injectable supply scales, and the oral wave expands rather than contracts category TAM.