Source Review: Metzger 2023 — Economics of In-Space Industry and Competitiveness of Lunar-Derived Rocket Propellant

Point-by-point review of claims relevant to the q4-gear-ratio leaf, with verdict taxonomy:

• Consistent — agrees with our analysis
• Different conclusion — we considered, disagree (with reason)
• Novel supporting — new info we hadn't used, integrated
• Merits investigation — flag for follow-up
• Not relevant — out of scope

Summary

Claims reviewed

Claim 1 — competitiveness framework

Quote: "The 'gear ratio on cost' for capital transport, G, and the production mass ratio of the capital, φ, are identified as the most important factors determining competitiveness."

Verdict: Consistent

Our first-principles calc independently derives the same competitiveness inequality (Eq. 8 in Metzger). G and φ emerge naturally as the load-bearing variables once costs are launch-normalized. Confirmed by Codex audit of pass-02-calc.

Claim 2 — tent sublimation φ values

Quote: "Tent sublimation technology has a value of φ that is an order of magnitude better than the threshold for competitiveness even in low Earth orbit (LEO)."

Verdict: Consistent (with caveat)

Tent sublimation φ values (Kornuta 442, Sowers 534) are roughly 10× a "~50" threshold. The order-of-magnitude framing is supported. Caveat: the "threshold" itself is destination-dependent and not a single number. Our q4.c7 reflects this.

Claim 3 — physics advantage of 24×

Quote: "the best payload mass fraction for conventional rocket technology launching off the Earth to GTO is about 2%. For launching off the Moon … the payload mass fraction can be about 48%, or 24 times higher."

Verdict: Consistent

Direct from Tsiolkovsky with realistic IMF. Our derivation gives similar order-of-magnitude (G_LEO-LS ≈ 15 for round-trip reusable). The 24× figure is the upper-bound physics advantage assuming idealised propellant delivery; real architectures eat into it via Γ_X penalties.

Claim 4 — competitive vs absolute advantage

Quote: "lunar-derived propellant needs only a comparative advantage, not an absolute advantage"

Verdict: Novel supporting

This reframing is not in our calc — we treated absolute advantage (ψ_X < 1) as the success condition. Metzger's point: lunar propellant competes against the opportunity cost of using Earth launch capacity for other payloads. With Starship-scale launch glut, lunar wins comparatively even if not absolutely. Adds nuance our analysis didn't capture.

Claim 5 — Pelech φ = 3.7 for strip mining

Quote: "For the strip mining technology P roughly estimated φ ~ 3.7. CD, J, and B predicted φ an order of magnitude larger for similar technology."

Verdict: Merits investigation

Pelech's number is 5-10× below other strip-mining estimates. Metzger attributes this to overestimating M_K based on terrestrial excavator analogies. We don't have the Pelech primary source extracted; the disagreement matters because it determines whether strip mining is viable at all. Flagged for separate Source Review of Pelech.

Claim 6 — Metzger MVP φ = 36.5

Quote: "M estimated the resulting M_K to be an order of magnitude lower than other studies … yet φ = 36.5 is still high enough to gain absolute advantage at least to GTO."

Verdict: Consistent

Our q4.c6 reflects this exactly. The "≥35 threshold" is the MVP value, not a universal cutoff.

Claim 7 — insensitivity to launch cost at high G/x

Quote: "Lunar propellant can be insulated from decreasing launch costs by achieving x < G as a capital design goal."

Verdict: Novel supporting

Counter-intuitive and important. Our calc shows the sensitivity matrix (Table 3 in Metzger) but didn't extract the design implication clearly. Adds: lunar mining firms should engineer their capital so launch cost reduction doesn't undercut them. This refutes the "Starship makes lunar mining unviable" framing.

Claim 8 — discount rate dominates timing

Quote: "S [Sowers] estimates an Internal Rate of Return (IRR) of 8.84% in a fully commercial venture, which is much less than the 21.7% that CD thought necessary to attract investors."

Verdict: Consistent

Discount rate is the bigger lever than tech for breakeven timing. Our q4.c5 captures this. Metzger's Table 1 shows 5-23 year ranges, with the variance driven mostly by financing not technology.

Claim 9 — market size weak effect

Quote: "each order of magnitude reduction in the market delays absolute advantage in LEO by about 2 years."

Verdict: Novel supporting

I didn't reason about market size in calc. Metzger's finding that 100× smaller market only delays LEO by 4 years is notable — it means policy support / customer aggregation isn't the bottleneck. The tech + finance variables dominate. Worth flagging in synthesis.

Claim 10 — LEO viability requires SEP architecture

Quote: Implied across Fig 6 + Section 4.10: lunar absolute advantage at LEO uses SEP from LLO toward Earth (I_sp = 2000 s) rather than pure chemical.

Verdict: Consistent

Our q4.c9 captures this. Pure chemical reusable RT has Γ_LEO ≈ 14 making LEO structurally hard.

Claim 11 — reliability optimisation

Quote: "the main finding is that it is unnecessary to make costly improvements to the hardware to gain higher reliabilities typical of NASA exploration missions"

Verdict: Not relevant (to gear ratio leaf)

This belongs to a different leaf (perhaps q3-isru-feasibility or q5-capital-buildup). Reliability cost optimisation is an interesting modelling choice but not load-bearing on the gear-ratio question per se.

Claim 12 — pessimistic TEAs (CD, J) used SLS for capital transport

Quote: "CD assumed high G = 64.9 (per current pricing) due to the use of government-built (non-commercial) rockets for capital transport. Even with EOS/SOE and learning curve, since G is high the reduction in x has diminishing returns"

Verdict: Consistent

Architectural choice (SLS vs commercial) drives a 10× swing in G. The published pessimistic conclusions of CD and Jones were a function of conservative transport assumptions that don't reflect 2026 commercial reality. Our framework predicts the same: G dominates when high.

Claim 13 — economies of scale + Wright's Law assumptions

Quote: "a = 0.66 used as empirical scaling exponent ... b = 0.75 (more conservative learning than typical b=0.80 for new industry)"

Verdict: Merits investigation

Metzger's choices are defensible (Haldi-Whitcombe industry data; Wright's Law literature) but inherently assume business-as-usual industrial scaling. They don't account for industrial-explosion-grade automation. Under TAI-grade conditions, both a and b could collapse dramatically (effectively a near 0, b approaching 0.5). This is a research-frontier consideration that Metzger does not address.

Cross-references to internal pages

• q4.c1, q4.c2 confirmed by Metzger's Γ analysis
• q4.c3 derivation matches Metzger's Eq. 8 necessary condition
• q4.c6, q4.c7, q4.c8 directly from Table 2 + Section 6
• q4.c9 from Fig 6 + Section 4.10
• q4.c10 (TAI sensitivity) is not covered by Metzger — original to our analysis

What Metzger 2023 does not address (gaps for synthesis)

1. Industrial-explosion / TAI-grade automation sensitivity. Wright's Law and EOS assumed at conservative industry norms. No treatment of what happens under sustained automated capital design improvement.
2. Sub-2023 cost-curve realism. Starship trajectory used per Musk's 2022 statements; doesn't update for actual 2024-2026 launch cadence.
3. Strategic/national-security demand. Treated as a footnote (Listner 2023); not modelled.
4. Aerobraking as alternative to SEP for LEO. Only briefly mentioned; full architecture analysis would be valuable.

Overall verdict

Metzger 2023 is the load-bearing source for the gear-ratio leaf. Our first-principles framework converges on the same competitiveness structure independently. The main differences are scope (he models lifecycle dynamics; our calc is snapshot) and modelling sophistication (his finance treatment is more nuanced). Where his published results are firm (Tables 1 + 2, Fig. 5-9), our framework gives consistent answers when parameterized appropriately.

The leaf write pass can confidently use Metzger's TEA reassessment as the substantive answer to "what φ is needed and is it attainable" — with the TAI caveat from q4.c10 layered on.

Metzger 2023 — gear ratio & production mass ratio framework

The canonical economic treatment of lunar resource competitiveness. Develops the "spherical cow" model with two dominant variables: gear ratio on cost (G) and production mass ratio (φ). Concludes lunar propellant production will be commercially viable. Refutes the Charania-DePascuale and Jones et al. pessimistic conclusions by showing they made transportation architecture and capital-mass errors.

abstract

Economic parameters are identified for an in-space industry where the capital is made on one planet, it is transported to and teleoperated on a second planet, and the product is transported off the second planet for consumption. This framework is used to model the long-run cost of lunar propellant production to help answer whether it is commercially competitive against propellant launched from Earth. The prior techno-economic analyses (TEAs) of lunar propellant production had disagreed over this. The "gear ratio on cost" for capital transport, G, and the production mass ratio of the capital, φ, are identified as the most important factors determining competitiveness. The prior TEAs are examined for how they handled these two metrics. This identifies crucial mistakes in some of the TEAs: choosing transportation architectures with high G, and neglecting to make choices for the capital that could achieve adequate φ. The tent sublimation technology has a value of φ that is an order of magnitude better than the threshold for competitiveness even in low Earth orbit (LEO). The strip mining technology is closer to the threshold, but technological improvements plus several years of operating experience will improve its competitiveness, according to the model. Objections from members of the aerospace community are discussed, especially the question whether the technology can attain adequate reliability in the lunar environment. The results suggest that lunar propellant production will be commercially viable and that it should lower the cost of doing everything else in space.

key concepts

gear ratio on cost — Eq. 6

G = (L_K × G_{K,LEO-LS}) / L_p

• L_K = cost per kg of launching capital from Earth to LEO
• G_{K,LEO-LS} = mass gear ratio from LEO to lunar surface for the capital transport vehicle (from Tsiolkovsky)
• L_p = cost per kg of launching terrestrial propellant from Earth to LEO

Reduces to ordinary "gear ratio on mass" when L_K = L_p (same launch vehicle for both capital and competing propellant).

production mass ratio — definition

φ = M_{p,LS} / M_K

• M_{p,LS} = total mass of propellant produced at lunar surface over the life of the capital
• M_K = mass of the capital

i.e. how many kg of product each kg of capital produces across its lifetime.

the competitiveness condition — Eq. 8

[(x + G) φ^(-1) + ω + ξ] · Γ_X < 1

Where:

• x = launch-normalized equipment cost (ζ/L_p)
• ω = launch-normalized operations cost (λ/L_p)
• ξ = launch-normalized finance cost (f/L_p)
• Γ_X = G_{LS-X}/G_{LEO-X} = propellant use ratio for delivery to location X

When this inequality holds at destination X, lunar propellant has absolute advantage over Earth-launched propellant at X.

launch-normalized capital cost — Eq. 9

χ = (x + G) φ^(-1)

pre-delivery cost ratio — Eq. 10

ψ_X = (χ + ω + ξ) · Γ_X

Lunar wins at X when ψ_X < 1, i.e. ψ_0 < 1/Γ_X.

numerical values — model baseline

Table 2 — φ values per extant TEA studies

The "≳35× threshold" commonly cited as "the gear ratio threshold" is approximately Metzger's MVP value (36.5) — the φ at which his Minimum Viable Product technology achieves absolute advantage at GTO. It's not a universal threshold; the actual threshold for absolute advantage at location X depends on Γ_X (delta-v to X) and the other cost terms.

Table 1 — years until lunar absolute advantage (baseline)

Note: market size has surprisingly weak effect on competitiveness — even a 100× smaller market only delays LEO crossover by ~4 years. The crossover times for everywhere except LEO are essentially market-insensitive.

Table 3 — parameter elasticities (cost ratio ψ_0) as function of G/x

Crucial insight: when G/x is large (capital transport dominates equipment cost), the lunar propellant cost becomes insensitive to launch cost L_0. This means Starship dropping launch prices doesn't undercut a properly-designed lunar mining operation — it actually helps lunar (because both terrestrial competitor AND lunar capital costs drop together, but lunar wins on physics).

Design implication: lunar mining firms should engineer to achieve x < G so their cost structure is dominated by capital transport, insulating them from terrestrial price wars.

structural conclusions per TEA reassessment (§6)

headline economic claims (verbatim where possible)

• Section 3.1: "the best payload mass fraction for conventional rocket technology launching off the Earth to GTO is about 2%. For launching off the Moon … the payload mass fraction can be about 48%, or 24 times higher. If this were the only difference between the Earth and the Moon, lunar propellant would be 24 times cheaper than Earth-launched propellant in GTO."
• Section 3.1: The competitive question reduces to: "can we (1) transport capital to the Moon, (2) tele-support its operation on the Moon, and (3) work with difficult raw materials in a harsh environment, with a total economic penalty that is less than a factor of 24 so it does not eat up the entire positive margin afforded by the physics?"
• Section 3.2: "lunar-derived propellant needs only a comparative advantage, not an absolute advantage" — because Earth-launched propellant competes against the opportunity cost of launching more lucrative payloads instead.
• Section 5.3: "Maximizing φ appears to be the dominant strategy for lowering the cost of lunar propellant"
• Section 5.3: "Lunar propellant can be insulated from decreasing launch costs by achieving x < G as a capital design goal."
• Section 6.7: "All models except J and B indicate that an absolute advantage is gained in GTO no later than year 12. Both tent sublimation studies predict economic viability in LEO from year 5."
• Section 7: "With 12% PPP discount rate, all studies except J and B predict absolute advantage in GTO by no later than year 10."

structural sensitivities

From the parameter sensitivity analysis (Fig. 10):

• Tripling φ has roughly the same effect as cutting M_K by 3× — both compress the cost curve down
• ζ (fabrication cost rate) is dominant in year 1 but its importance fades over time as EOS/learning curve operate
• G remains dominant in long run because physics doesn't compress

relevant for the gear-ratio leaf question

The leaf asks: what gear ratio threshold must lunar capital achieve, and is it attainable?

Metzger answers:

1. There's no single threshold; the threshold for absolute advantage depends on destination orbit (Γ_X) and other cost terms.
2. Tent sublimation (φ = 442-534) achieves the threshold at LEO from year 5 with conservative assumptions.
3. Strip mining (φ = 22-44) achieves GTO threshold but struggles for LEO.
4. Metzger's own MVP (φ = 36.5) gets to GTO by year 5 with optimistic market.
5. The threshold is ATTAINABLE with current technology choices, especially tent sublimation. The bottleneck is reliability and ground-truth on lunar ice deposits, not the gear ratio itself.

what's missing / what to chase in subsequent passes

• Independent first-principles derivation (the calc sub-pass) — verify the φ ≥ 35-ish threshold from physics + capital cost first principles, without consulting Metzger's numbers
• Pelech, Charania-DePascuale, Sowers, Kornuta primary sources — to cross-check Metzger's reassessment quoted above
• Bennett et al. reassessment of Jones — primary source needed
• Post-2023 updates: anything from 2024-2026 that updates these numbers given Starship's actual progress (vs. assumed)
• Sensitivity to "modulo TAI" — Metzger explicitly assumes business-as-usual industrial scaling (Wright's Law b=0.75, EOS a=0.66). What happens if AI/automation collapses MK or compresses the buildup period?