I see (www.cotlook.com) cotton prices have hit a 2½ year high on fears that the Pakistan floods will cut supply – the country produces nearly 7% of world supply. The dynamics of this are pretty easy to lay out, just like many other cyclical industries:

• The total of industry current capacity [a resource] determines maximum output
• … which balanced against current demand determines price through normal elasticity mechanisms.
• If prices rise, so do supplier-profits – and crucially, so do expected profits [an intangible]
• … which encourages producers to invest in new capacity.
• The flow of ‘new capacity per month’ enters an implicit resource of ‘capacity in development’.
• During the time this takes to come on-stream, prices remain high, encouraging still more suppliers to add to the flow of new capacity.
• Eventually the resource of ‘capacity in development’ is completed and flows into the resource of ‘current capacity’ – and since so many suppliers invested, the capacity resource quickly grows to exceed demand
• … so prices collapse, and suppliers make losses (which can be huge)
• … so the flow of new capacity additions fall to zero, and high-cost capacity gets shut down [an out-flow from current capacity].
• No-one likes to give up, so this may take some time, until the least efficient producers run out of cash [another resource] – during which period prices remain depressed.
• Lower prices stimulate rising demand, which eventually grows to overtake the falling capacity
• … so prices rocket upwards again.

… and so the whole cycle starts over – as we see repeatedly with oil prices, shipping, real-estate, computer-memory etc. etc. The whole cycle can be exaggerated if another resource [investor confidence] responds slowly to current profits and delays both new investment and dis-investment. In many cases it may be possible to see the switch-points approaching – when the sum of new capacity additions means there is more potential capacity than demand, and on the down-cycle, when rising demand is approaching current capacity. What to do about this? The smart operators do not invest in new capacity when prices and profits are high [the new capacity itself may be costly, because the constructors themselves are enjoying a boom in demand !]. Instead, they bank the cash from high prices, wait for the collapse, then buy up bankrupt capacity on the cheap when product prices fall.   How can industry leaders [if they want to!] moderate these cycles? Make public announcements about the impending switches in supply/demand to signal to weaker players they are about to get killed.