Explore the model

Don't take our word for it. Move the levers yourself.

This is a live, in-browser version of the NSWMSC workforce model. It uses the same formulas set out in our methodology paper, running on the Commonwealth's own 2026 Compendium figures. Change the assumptions and watch the training queue, competition ratio, and specialist supply respond in real time. It's illustrative, not a precise forecast. It shows the direction and plausible magnitude of effect under assumptions you can see and adjust.

§4–5 · Backlog clearance (transient queue)

How fast does the training queue clear?

Each year the waiting pool of prevocational doctors grows by new domestic graduates, plus any grandfathered or tenured-re-entrant IMGs still arriving. It shrinks by the training places the gate can process, plus attrition. Registrar intake already exceeds domestic graduate supply. So once new IMGs are de-prioritised, the queue drains. The question is how fast. And whether it stays cleared.

Try a scenario
The central case: existing cohort grandfathered, future IMGs earn priority after five years' service (θ = 0.3). The backlog clears and stays cleared.

Levers

Scenario A (prioritise) parameters - shared with tab ②.
Share of arriving IMGs re-admitted to the priority pool from 2031, after 5 years' Australian service. θ = 0 is "hard prioritise".
Held constant, anchored to AHPRA first-time international registrants (4,947 in 2024–25).
Prevocational doctors leaving the queue each year (career change, emigration, etc.).
Annual growth in accredited training places. The Compendium holds these ~flat (0%).
Back to 2:1 competition
2029
Queue at 2048
0
Holds ≤2:1 to 2048?
Yes
Prevocational queue, 2025–2048 (FTE)
Do nothing reproduces the Compendium's trajectory (15,900 → 38,222). Prioritise drains the queue below the 2:1 threshold.
Do nothing Prioritise (your settings) 2:1 competition threshold (2K)
Formula (§4.3): St+1 = max(0, St + Gt+1 + γt+1 + θ·At+1 − K − x). Start S₀ ≈ 15,900 FTE (Compendium Table 5). K = 0.15 × 38,222 ≈ 5,733 places/yr. Domestic inflow G grows 3,274 → 3,812 over six years; grandfathered tail γ = 3,000 / 1,500 / 500 in 2026–28. The return to 2:1 is insensitive to the auxiliary assumptions (§5.1) - it lands in 2028–29 across the plausible range.
§3 & §8 · Priority-group ratio & PGY (Paper Figure 2)

What happens to competition - and how senior you'll be

The competition ratio is simply applicants per place (R = 1/p). The average postgraduate year at entry follows the queue-wait formula PGY ≈ G₀ + (R − 1). As the priority queue from tab ① drains, the ratio faced by domestic and grandfathered doctors falls below the healthy 2:1 level. New entrants get in at the minimum eligible year: PGY 2–3, not PGY 8. These curves are driven by the same levers as tab ①.

Try a scenario
The central case: existing cohort grandfathered, future IMGs earn priority after five years' service (θ = 0.3). The backlog clears and stays cleared.

Levers

The same queue parameters as tab ① - kept in sync.
Share of arriving IMGs re-admitted to the priority pool from 2031, after 5 years' service.
Held constant, anchored to AHPRA first-time international registrants.
Prevocational doctors leaving the queue each year.
Annual growth in accredited training places.
Ratio 2048 · do nothing
6.7×
PGY 2048 · do nothing
PGY 7.7
PGY 2048 · prioritise
PGY 2
Competition ratio
Applicants per training place, 2025–2048.
PGY-at-intake
Postgraduate year on entry to training.
Do nothing Prioritise / priority group (your settings) 2:1 target
Who fills the training places?
Share of each year's accredited places, by group - this is what θ (tenured re-entry) and IMG arrivals move.
Domestic + grandfathered Tenured re-entry (5-yr service) Newly-arrived IMGs (non-priority)
Equal footing after tenure. Once IMGs earn priority through five years' service they join the same priority pool as domestic graduates, so the competition ratio they all face is the whole priority queue ÷ places - Rp = S ÷ K, with PGY = G₀ + max(0, Rp − 1), G₀ = 2. That is the same queue S as tab ①, so all four levers move these curves: raising θ or IMG arrivals adds tenured re-entrants to the pool and, once their inflow exceeds spare capacity, pushes competition and PGY back up (try θ = 0.6). The design holds a healthy ≤2:1 for θ up to about 0.5 because the backlog clears before re-entry begins in 2031; beyond that the pool re-grows. The composition chart above shows who fills each year's places; the curves here show the pressure the pooled priority group feels.
§6 · Sizing the place expansion

Prioritisation fixes the queue. Only more places fix the shortage.

Who trains is set by prioritisation. How many specialists the system produces is set by the number of accredited registrar places. The Compendium projects a specialist shortfall of ~12,800 FTE by 2048. Each extra place adds specialist supply only after a ~6-year training lag. And only ~0.77 FTE of it survives attrition and part-time practice. So how many extra places a year does it take to close the gap?

Levers

Phased in from a start year over four years.
Sustained annual increase, on top of the current ~5,733 intake. ~1,150 (+20%) closes the 2048 gap.
First Fellows emerge ~6 years later. Start by 2031–32 at the latest to matter by 2048.
Specialist gap 2048
12,812 FTE
Gap remaining 2048
0 FTE
Intake needed 2048
6,883
Specialist supply vs demand (FTE)
Demand = do-nothing supply + Compendium shortfall. Your expansion adds supply after the training lag.
Demand Supply · do nothing Supply · with your expansion
Formula (§6.3): extra FTE(t) = c·φ·Σ ΔN(τ) for τ ≤ t − L, with training lag L ≈ 6 yr, completion c ≈ 0.85, FTE factor φ = 0.9 (so ≈ 0.77 FTE per place). Domestic graduates (~3,812/yr) can't fill even current places, so the expanded balance is met by IMGs directed to genuine area-of-need, or by later, calibrated CSP growth.