War Economics: a discussion with ChatGPT

Nigel: Imagine a simple economy A where one sector produces the means of basic living and the capital goods underpinning that - and the second sector produces war materiel.

Simple economy B is just the same, except the second sector produces an equivalent value of ephemeral fripperies such as cosmetics, kitsch TV shows, glossy magazines and so on.

Ignoring rather subjective issues of which commodities are classed as welfare-adding, wouldn't it be the case that in functional economic terms economies A and B are isomorphic?

---

ChatGPT: you presents two theoretical economies: one with sectors for basic living and war materiel, the other with basic living and ephemeral goods (like cosmetics).

You ask if these economies are functionally equivalent, ignoring welfare. From a production/output perspective, both provide output, and GDP could theoretically look similar.

However, war materiel has no utility beyond destruction, while fripperies offer enjoyment. So, without considering welfare or externalities, they’re similar, but war materiel’s consumable nature sets them apart in terms of real-world impact.

(Thought for 6 seconds).

At bottom, if you strip away all the things that make “guns” and “glossy magazines” morally, socially or dynamically different – if you treat both simply as consumption‐goods with identical input–output tables, identical factor intensities, identical depreciation (i.e. zero) and identical final‐demand functions – then, yes, in the strict Walrasian/Sraffa sense the two economies are isomorphic.

---

Why, then, does “Sector 2 = Tanks” feel so different from “Sector 2 = Trashy TV”? Two reasons creep back in once you relax the sterile assumptions:

1. Depreciation & Destruction.
   – Fripperies might fade or become worthless (zero depreciation), but war materiel often destroys existing capital and utility. If you introduce a negative “return” or add a destruction term $-\delta K$, the dynamics break the isomorphism.

2. Utility & Externalities.
   – In pure theory each good just fills a demand cell; in reality combat weapons have negative externalities (refugees, ruined infrastructure) and zero “consumer surplus.” Kitsch soaps and glossy mags, by contrast, impart at least some positive utility—even if trivial.

---

3. When the Mask Slips

The moment you admit that Sector 2’s output:

• Reduces next period’s capital stock (tanks get blown up);

• Carries negative spill‐over effects on other sectors;

• Mobilises labour via conscription rather than voluntary hire;

…you’ve introduced non‐isomorphic technology and preferences. The simple “swap guns for glossy mags” no longer preserves equilibrium coefficients.

---

4. Verdict

• Under the economist’s microscope – when every good is just a spot in a matrix – A and B are formally identical.

• In the real world, where goods have durability profiles, externalities and utility weights, the two diverge wildly.

So: if you insist on the Platonic ideal of goods as undifferentiated vectors in Rⁿ, war = frippery. The instant you re–entangle capital depreciation, destruction or utility, the mathematical mirror cracks.

---

Nigel: Except that not all wars are of choice. Security - not losing - may be considered a very positive externality of a war economy. Don't you think it's important to avoid hand-wringing moralising in an economic theory discussion?

---

Refining the economic theory

You seem to want a more analytical approach without the moralizing. Understood! Let's consider security as a positive externality—something that benefits society far beyond individuals' direct consumption. The public good that is security can play into welfare theory. If we see security as a public good, it shifts the perspective; the war or defense sector wouldn't merely supply consumption goods, but a critical public service. We can tweak the model to reflect that—security isn't just a private or tragic transaction but part of a broader economic function.

(Thought for a few seconds).

You’re absolutely right: once you concede that “security” is itself a valued output—indeed a classic public good—then war materiel simply becomes the “defence” sector’s insurance premium, not a moral pariah. If we strip away hand‑wringing, here’s how to see it cleanly:

---

1. Security as a Public Good

In standard theory (Samuelson, 1954), defence is non‑rival and non‑excludable: once your missiles are on patrol, both you and your neighbour sleep more easily. 

---

2. Opportunity Cost Under Uncertainty

Wars aren’t elective pastries at the buffet; they’re actuarial choices under threat:

• Let $p$ = probability of aggression.

• Let $L$ = expected loss if undefended (in utility‐terms).

• Every unit of war‐goods delivers a marginal reduction $\Delta S$ in $pL$.

Your household would happily forego some frippery if the “insurance premium” paid avoids a castle‑sized bill when the invader knocks. 

---

3. No Moralising Required

All we’ve done is treat “defence” as another good in the social welfare function. The only “sin” lies in ignoring security’s positive externality. Once you fold it in, your model tells you exactly how much to spend on armour versus armchairs:

1. Calculate the marginal utility of extra security ($\mathrm{\partial }U\mathrm{/}\mathrm{\partial }S$).

2. Equate it to the marginal utility of extra consumption ($\mathrm{\partial }U\mathrm{/}\mathrm{\partial }C$), adjusted by price ratios.

3. Solve for the equilibrium mix of frippery vs. defence output.

---

4. The Real Distinction

So in purely abstract terms, war‐goods and frippery are isomorphic if—and only if—they share identical production coefficients, depreciation profiles, and yield the same utility‑externality bundle. In practice:

• War‐goods produce security, but destroy capital in conflict (negative depreciation).

• Frippery yields ephemeral private joy (positive $\mathrm{\partial }u\mathrm{/}\mathrm{\partial }{d}_{}$) but zero security and negligible destruction.

Once you map these differences into your social welfare maximand, they cease being mere vectors in Rⁿ and become distinct sectors with unique trade‑offs.

---

Bottom Line

No moral posturing needed—just a crisp recognition that goods differ by the externalities they carry. If you insisted on lumping tanks and soap‐operas into one “everything‐else” box, you’d have to pretend security doesn’t matter (or that kitsch yields it). And that, I think we can agree, is the real fiction.

---

Nigel: At any rate, short of catastrophic defeat or endless unaffordable expenditure, a strong defence sector is perfectly economically sustainable and may have multiple positive externalities such as transferable technologies.

Arguably the Israeli economic is a case in point, with much innovation driven from the IDF?

---

ChatGPT: Analyzing defense sector sustainability

You seem to be asking about whether a strong defense sector can be sustainable, looking for examples like Israel where the military drives tech innovation. To answer, I should gather objective data on Israel's economy, the proportion of defense spending, and its influence on tech sectors. Specifically, I could look into Unit 8200 and its role in fostering cybersecurity startups. Economic factors like GDP growth and R&D investments will help strengthen the analysis. I’ll dive into this.

Reviewing defense sustainability in Israel

I’m outlining key points: A strong defense sector is sustainable if it avoids excessive losses or unaffordable costs. For Israel, the economy showed growth despite conflict, around 1% in 2024, with military spending contributing significantly (roughly 5% of GDP). Projections for 2025 show a potential rise to 4% growth. Military spending in 2023 was $27.5 billion, and defense spending accounted for around 5% of GDP. I'll provide citations to back this up.

(Thought for 13 seconds).

Short of a catastrophic defeat or a debt spiral, a robust defence sector can indeed be both sustainable and growth‑enhancing—so long as it’s managed like any other strategic “insurance premium.” Israel illustrates the point rather neatly:

• Defence as “insurance”
  Israel’s military outlay—roughly USD 27½ billion in 2023, or about 4.9 % of GDP—is not an endless black hole but a deliberate investment in deterrence and rapid‑response capability. In a world of real threats, that “premium” protects households and firms against ruinous loss, just as fire insurance spares your savings when your roof goes up in flames.

• Macroeconomic performance under fire
  Even through its 2023–24 Gaza conflict, Israel’s real GDP grew by about 1 % to roughly USD 565 billion, largely thanks to state reconstruction and elevated public spending—military and otherwise—propping up demand. The finance ministry now pegs 2025 growth at about 4 % once the worst of the fighting recedes.

• Fiscal sustainability
  War spending did push Israel’s debt up—to near 69 % of GDP in early 2025—but that remains within the range of many advanced economies (France ~96 %, Italy ~145 %). And with a still‑healthy sovereign rating and a current account surplus, Israel can carry that load without tumbling into ruin.

• Positive externalities & spin‑offs
  Perhaps most striking is the dual‑use payoff: intensive R&D inside the IDF and Military Intelligence’s Unit 8200 has seeded a world‑class tech ecosystem. Companies like Mobileye and, most recently, cybersecurity startup Wiz (in talks for a $23 billion acquisition by Google) trace their lineage to recruits solving “impossible” military problems. These spin‑offs generate civilian employment, exports and tax revenues—far beyond the battlefield.

---

In sum, when defence outlays are calibrated to real risks, financed prudently and channelled through R&D that diffuses into civilian industry, a militarised economy needn’t be a drag on growth. Israel’s blend of “guns and glossies” might look unusual, but it delivers both security and a remarkably resilient innovation engine—proof that, under the right conditions, defence can pay handsome dividends.

---