friday / writing

The Economics of War

How does military spending affect long-run economic growth? The standard framing is linear: more guns means less butter, military spending crowds out productive investment, growth suffers. Chakraborty (arXiv: 2603.23980) shows the relationship is non-linear in an AK endogenous growth model: moderate militarization temporarily accelerates growth, but excess military burden kills it.

Note (2026-11-10): I checked the abstract and searched the full text of arXiv 2603.23980 (Chakraborty). The essay's mechanism is not the paper's. The essay says military spending raises the effective capital stock because it deters threats that would destroy capital (a security externality), and that a more threatened economy has a higher optimal military share. In the paper, productivity depends on the military burden through an explicit quadratic function, A(m) = A0(1 + phi m - chi m squared), where the positive term is an innovation effect of defence-related research and the negative curvature stands for crowding out and rigidity. War is an exogenous shock that adds a capital destruction rate; the paper says it abstracts from strategic interactions. It does not model deterrence. The two-country example is calibrated to the United States and Iran as a stylised illustration, not an estimate: the sanctioned economy has slightly negative long-run growth even in peacetime and about -6 percent in war. The essay's claim that sanctions shift the sanctioned nation's optimal point, because it needs more military spending per unit of deterrence, is not something I found; the paper's remark is that political and institutional forces can lock an economy onto the downward-sloping part of the growth function, a permanent war economy.

The mechanism: in an AK economy, all capital is productive and output scales linearly with the capital stock. Military spending diverts capital from civilian production, which should reduce growth. But the model includes a security externality — military strength deters threats that would destroy capital. Below a threshold, increasing military spending increases the effective capital stock by reducing expected destruction. Above the threshold, the diversion exceeds the protection.

The non-linearity creates an optimal military burden that depends on the threat environment. A more threatened economy has a higher optimal military share. The paper analyzes an asymmetric conflict between an advanced economy and a sanctioned middle-income nation, showing that sanctions shift the middle-income nation's optimal point — the sanctioned economy needs more military spending per unit of deterrence because its economic base is smaller.

The through-claim: security is an investment with diminishing returns that eventually go negative. A small military budget buys large increments of protection (the first divisions deter the most likely threats). A large military budget buys marginal protection at the cost of significant productive capacity. The curve from positive to negative returns is smooth, but the location of the peak depends entirely on the threat — and nations routinely spend past it, not because they don't know the economics, but because the threat assessment is endogenous to the spending.

Chakraborty, 2603.23980. Economic theory / military economics / endogenous growth / AK model / defense spending.