The new Round of Competitive Bid proposal explained:

Step 1: Suppliers Place Bids

Let’s say we’ve got 5 suppliers bidding to provide oxygen concentrators in Miami:

  • Supplier A: $100

  • Supplier B: $110

  • Supplier C: $120

  • Supplier D: $130

  • Supplier E: $140


Step 2: CMS Chooses Enough to Cover the Area

CMS decides: “Okay, we need 3 suppliers to cover this Competitive Bidding Area (CBA).”

  • They take A, B, and C.

  • Supplier C’s bid ($120) becomes the clearing bid.


Step 3: How SPA Used to Work (Pre-2021)

Instead of paying suppliers their actual bids, CMS said:
“Hmm… let’s take the median of accepted bids (A 110, C $120).”

  • Median = $110.

So:

  • Supplier A (who bid 110** – more than they asked for. 🎉

  • Supplier B (who bid 110** – exactly what they bid. 😐

  • Supplier C (who bid 110** – less than they bid. 😡


Step 4: Round 2021 Change

CMS said: “Fine, we’ll pay 100% of the clearing bid instead.”
So now:

  • All three suppliers get $120.

  • No one is underpaid relative to their bid.


Step 5: CMS’s New Proposal (Going Forward)

CMS wants to bring the pain back:

  1. 75th percentile method: Pay less than clearing bid, plus cut the number of contracts.

  2. Median method below clearing bid: Pay even less, but keep more suppliers.

  3. 100% clearing bid with half the contracts: Pay fairer rates, but kick half the suppliers out.


👉 So the ridiculous part is: under the old system, you could bid low, win a contract, and then get paid more than you bid, while your competitor who bid slightly higher gets paid less than they bid.
It’s like showing up to a car auction, bidding 11,000.” 🚗💸


Let’s run the Miami example through this ridiculous CMS math machine. 🧮


📍 Setup: Miami Competitive Bidding Area

  • Previous awards: 29 contracts

  • “Meaningful contributors” (those who did at least 5% of the volume): 9 suppliers

  • CMS assumes 5% fewer patients in FFS (so they shave off a bit).


⚖️ Option 1: 50% SPA (Median Below Clearing Bid)

  • CMS says: “We’ll keep more suppliers, but we’re gonna pay them less.”

  • Calculation: 29 – 5% ≈ 27 contracts

  • SPA set at the median of bids BELOW the clearing bid (ouch).

  • Outcome:

    • Lots of suppliers survive (27).

    • Payment rates are suppressed downward.

    • Basically: “Congratulations, you get a contract! …But don’t spend it all in one place.” 💸


⚖️ Option 2: 100% Clearing Bid (Half the Contracts)

  • CMS says: “We’ll pay you fairly, but we’re cutting the party in half.”

  • Calculation: (29 – 5%) * 0.5 ≈ 13 contracts

  • SPA set at the clearing bid (100%).

  • Outcome:

    • Only 13 suppliers left standing.

    • Rates are fairer — nobody gets paid less than they bid.

    • But half the suppliers get locked out completely.

    • Translation: “The money’s good… if you’re one of the chosen few.” 🎲


⚖️ Option 3: 75% SPA (CMS’s Favorite)

  • CMS says: “We’ll do a little of both — fewer suppliers and lower rates.”

  • Calculation: (9 meaningful contributors × 2) – 5% ≈ 17 contracts

  • SPA set at the 75th percentile of bids, relative to clearing bid.

  • Outcome:

    • 17 contracts awarded.

    • Rates are discounted again (lower than clearing bid).

    • CMS gets its “downward trend” back.

    • Or in human terms: “We’ll starve you a little, but hey, at least we’re not starving everyone.” 🍽️


🎭 How It Feels

  • Option 1: Everyone gets a ribbon, but your prize money shrank.

  • Option 2: Half the players get kicked out, but the survivors actually get what they asked for.

  • Option 3 (CMS pick): A middle ground where fewer contracts exist and the rates are squeezed.


Here’s a toy numbers walk-through that shows exactly how the SPA would land under each option. same Miami spirit, smaller dataset so it’s easy to see what’s going on. 🧮

Setup (toy example)

Ten bidders for the same item (sorted low → high):

$95, $100, $104, $107, $112, $118, $121, $127, $133, $140

Assume CMS decides it needs 5 suppliers to cover the CBA.
That means the clearing bid is the 5th accepted bid = 95, 104, 112


How the SPA would be set under different rules

A) Pre-2021 method (median of accepted bids)

  • Median of the 5 accepted bids = the 3rd value = $104.

  • **SPA = 104).

Impact

  • 100, $104 bidders get paid more than they bid (🥳).

  • 112 bidders get paid less than they bid (😬).


B) Round 2021 method (100% of the clearing bid)

  • SPA = clearing bid = $112.

  • Everyone gets $112.

Impact

  • No one is paid less than their bid among the accepted suppliers.

  • Lower bidders (107) get paid more than they bid.


C) CMS’s new proposals (illustrative math)

These are illustrative using your summary of CMS-1828-P. The exact percentile mechanics in the final rule could be more fussy, but this shows the direction.

Option 1 — “50% SPA” flavor: Median of bids below the clearing bid, keep (roughly) all contracts

  • Bids below the clearing bid are: 100, 107 (4 numbers).

  • Median of those four = average of the middle two = (100 + 104) / 2 = $102.

  • SPA ≈ $102.

Impact

  • Keeps more contracts (Miami framing ≈ 27).

  • Pushes rates down: everyone gets ~112).


Option 2 — “100% clearing bid with ~50% contracts”

  • Slash contracts to about half (Miami framing ≈ 13).

  • SPA = 100% of the clearing bid.

In our toy, if CMS only “needs” 3 suppliers, accepted are 100, 104**
So **SPA = 112 anymore because fewer suppliers means an earlier clearing point).

Impact

  • Fewer winners, but the winners get their clearing bid (no discounting).

  • Rate level depends on where the new (earlier) clearing bid lands; here it’s $104.


Option 3 — CMS’s preferred: 75th percentile relative to the clearing bid, with fewer contracts

  • Keep fewer contracts than old rounds (Miami framing ≈ 17).

  • Set SPA near the upper quartile of accepted bids (discounted vs 100% clearing, but not as low as Option 1).

Using our 5 accepted bids (100, 107, $112):

  • 75th percentile position = (n+1)·0.75 = 6·0.75 = 4.5

  • Interpolate between 4th (112):
    112 − 109.50**

  • SPA ≈ 110 wouldn’t be shocking in practice).

Impact

  • Fewer contracts than Option 1, higher SPA than Option 1, lower SPA than 100% clearing.

  • CMS gets both levers: capacity control and some rate discounting.


Quick side-by-side (with this exact toy data)

MethodContracts (Miami vibe)SPA result in toyNotes
Pre-2021 median of acceptedmany$104pays some below their bids
2021: 100% clearing$112no one underpaid vs bid (given 5 accepted)
Option 1 (median below clearing)~27$102keeps many suppliers, lowest rate
Option 2 (100% clearing, ~half contracts)~13$104 (with 3 accepted)fewer winners, fair to bids
Option 3 (75th percentile, fewer contracts)~17**110)middle path: fewer suppliers + some discount

TL;DR vibes

  • More contractslower SPA (Option 1).

  • Fewer contracts ⇒ can keep SPA at clearing (Option 2) or slightly below it (Option 3).

  • The “ridiculous” bit is still that under discounting methods, what you bidwhat you’re paid.