The bullwhip effect

The price of lumber hit a record high of over $800 per thousand board feet last week, an increase of over 300% since early April 2020. As coronavirus spread across North America mills cut production in expectation of plunging demand. The mills didn’t count on a surge in home renovation projects and an exodus from the cities as urban dwellers pined for a country retreat.

The lumber market is just one of many examples across commodity markets and other supply chains caught out by the bullwhip effect.

The bullwhip effect is a phenomenon that can occur in supply chains in which irregular or unexpected orders by consumers reverberate up the supply chain. As participants in the chain react to the new information the impact is amplified, resulting in wild swings in inventories.

The bullwhip (or whiplash) effect was first coined by Proctor Gamble in the early 1990’s when they noticed that erratic orders for nappies were amplified through their supply chain. The effect has been recognised in markets as diverse as computer memory, shipping containers and my favourite, beer. Each of them have seen orders in which the variability cannot be explained by fluctuating consumer demand alone.

There are four sources of the bullwhip effect – demand signal processing, rationing game, order batching and price variations. Demand signal processing is where demand fluctuates but players in the market use past demand information to update forecasts. The rationing game refers to the strategic behaviour of buyers when a supply shortage is anticipated. Order batching occurs when there are economies of scale from buying in bulk. Finally, price variations are exactly that, non-stable prices.

The bullwhip effect was first gamified into the “Beer Game” – an experimental game in which players make independent inventory decisions relying only on orders from the neighbouring player as the sole source of information. The task is to produce and deliver units of beer: the factory produces and the other three stages deliver the beer units until it reaches the customer at the downstream end of the chain. Try it out for yourself

The bullwhip meets the cobweb

The Cobweb Theorem was developed in the 1930s by two economists at the Bureau of Agricultural Economics, G.C. Haas and Mordecai Ezekiel. The two researchers were trying to make sense of the volatility they saw in the price of hogs, and help develop a model to help farmers adjust supply more rapidly to demand.

The theory shows how supply and demand responds in a market where the amount produced must be chosen before the price is observed. Agriculture is a great example of where the theory might apply, since there is an interval between planting and harvesting.

For example, because of unexpectedly bad weather, farmers go to market with an unusually small herd of hogs, resulting in higher prices. If the farmers expect these high price conditions to continue, then they will fatten up more hogs relative to other crops in the following season. When the farmers then go to market with the second year’s supply of livestock, supply will be high, resulting in a drop in the price of hogs.

And so it goes on. If farmers then expect low prices to continue, they will reduce the size of their herd for the subsequent season, resulting in a return to high hog prices yet again.

While farmers should become more efficient in their production choices over time – something called adaptive expectations – there is no guarantee that will happen. Indeed, Haas and Ezekiel noted that you only need to make small changes to the underlying assumptions for the price to become increasingly more volatile, moving further and further away from equilibrium.

Traders in lumber futures and investors in the companies that supply timber should be aware of how quickly the bullwhip can snake in the opposite direction. The lumber market is one of the most thinly traded of all commodity futures markets. Lumber’s price history suggests gravity reasserts itself quickly and violently.

Related article: Why are commodity prices volatile?

(Visited 96 times, 16 visits today)