Inelastic supply is a powerful cause of instability and the mechanism is clear. When quantity cannot respond to price, any change in demand must be absorbed almost entirely by price, so a shift that would barely move the price of a manufactured good produces a large swing in the price of a commodity. The same applies to supply-side shocks: a poor harvest cannot be made up from anywhere, so the price rises sharply. The goods with the most volatile prices in any economy — agricultural products, minerals, oil, housing — are precisely those with the most inelastic supply, which is strong evidence that the relationship is real.
But it is not the only cause, and treating it as the main one hides several others. Inelastic demand does the same work from the other side: where consumers cannot reduce their purchases, a shift in supply must be absorbed by price rather than quantity. Instability needs one side of the market to be unresponsive, not specifically the supply side, and for food and fuel both are inelastic at once — which is why their prices are the most volatile of all.
The frequency and size of the shocks matter at least as much as the elasticities. A market with perfectly stable demand and supply would have a stable price however inelastic it was, because nothing would be disturbing it. Agricultural prices swing partly because supply is inelastic and partly because the weather keeps moving the supply curve; a market with equally inelastic supply but no shocks would be quite calm. Elasticity determines how violently a market responds to a disturbance, not whether disturbances occur.
Speculation and expectations add a further source. When traders buy because they expect prices to rise, their buying can create the rise, and when they sell in a falling market they deepen the fall. That instability comes from behaviour rather than from the shape of the supply curve.
There is also the lag in the supply response, which is related to inelasticity but distinct from it. Because production decisions are taken long before output arrives, high prices attract investment that comes on stream years later, often just as the shortage has ended — so the market overshoots in both directions and cycles rather than settling. It is the delay, not merely the low elasticity, that turns a single shock into a cycle.
My judgement is that inelastic supply is the most important amplifier of instability rather than its cause, and the distinction matters. Shocks, expectations and lags are what disturb a market; elasticity determines how large the resulting price movement is. A market can be inelastic and stable if nothing disturbs it, and it can be elastic and unstable if it is disturbed violently enough.
That also points to what can be done about it. Because elasticity is the amplifier, the most effective way to reduce volatility is usually to raise it — building storage, improving transport, diversifying sources of supply — rather than trying to prevent the shocks, which is generally impossible. Buffer stock schemes for commodities work on exactly this principle: they do not stop harvests failing, they make the effective supply more responsive when one does.