Surface lanes where HHI is collapsing and new entrants are arriving. The leading indicator of structural price erosion before the median falls.
Across the 680 molecules with three years of declared median USD/kg, the cross-sectional median price moved from $1.61/kg in 2022 to $1.18/kg in 2024, a 27% erosion at the median. The tail is steeper: ten molecules eroded more than 96% over the same window, led by tigecycline (-99.97%) and amodiaquine (-99.3%) on volume bases above n=60.
Type the molecule INN and the destination country. The AI agent pulls the HHI trajectory across the trailing 3 years and the entrant count delta.
Linear fit to HHI history projects the lane's concentration in the next year. Lanes projected to fall below 2,000 within 4 quarters are flagged as approaching commodity floor.
Citation-anchored brief: the trajectory, the new entrant count, the expected price compression at commodity floor, and a recommended action: defend, hold, or exit.
New entrants change the lane concentration months before the median price moves. HHI drops first; price erosion follows. Tracking HHI gives the user a leading window.
Trailing 3-year HHI fitted to a linear trend gives an expected HHI for the next 12 months. Lanes projected to fall below 2,000 are flagged as approaching commodity-floor erosion.
Each lane carries a recommended posture. Defend means raise minimum bid and shore up customer mix. Exit means deprioritise capacity. Hold means the erosion is manageable, reassess next refresh.
Lane competitiveness refreshes weekly. The trajectory reruns on every refresh; the brief carries the exact computation timestamp.
A real chat thread in PharmaDB. Type a question, the AI agent runs the tools, the answer lands as a saveable note.
pharmagraph_query Ten lanes · 2024 HHI on the 0-10,000 scale · falling HHI indicates erosion trajectory.
Price erosion forecasting is the leading-indicator lens for lane structure. PharmaDB tracks HHI per lane across the trailing 3 years, projects the next 12 months, and flags lanes approaching commodity floor. The table is lane_competitiveness. Refresh cadence is weekly.
New entrants change the lane concentration before they take material share. By the time market share has redistributed enough to move the volume-weighted median, the lane HHI has already fallen. HHI is the early-warning surface for price erosion.
HHI above 5,000 is concentrated and supports premium pricing. HHI between 2,000 and 5,000 is moderately concentrated; price discovery is active. HHI below 2,000 is competitive; the lane is approaching commodity floor. Crossing 2,000 typically corresponds to median dropping 30%+ within the year.
Linear fit to the trailing 3-year HHI per lane. Lanes with consistent negative slope and an extrapolated 2026 HHI below 2,000 are flagged as approaching commodity floor. The brief surfaces both the slope and the projected value.
Common pattern: HHI bottoms out as new entrants stabilise, then partially rebounds. The empagliflozin Brazil lane shows this with 2025 partial data ($1 from 2,491 to 3,181). The rebound rarely returns to the prior premium; the lane is structurally re-priced even after concentration stabilises.
When a lane is approaching commodity floor, the bid floor should be reset toward the new equilibrium price. The brief flags lanes where the bid-floor anchor (lane p25) is becoming stale because of HHI erosion.
Lane competitiveness aggregates refresh weekly. The trajectory reruns on every refresh; the catalog row carries the exact timestamp.
Bring the molecule, the lane, or the supplier you're sourcing this week. The AI agent runs it on PharmaDB in 30 minutes. You keep the brief.