The cobweb model and the boom-bust cycle in the NZ dairy market

In ECON100 we teach the cobweb model of supply and demand, which is rarely taught in other courses at first-year level (see this paper by Steven Lim and I in New Zealand Economic Papers for a discussion of our overall approach to ECON100; or an ungated earlier version here). Students often ask whether the cobweb model is relevant to the real world - do suppliers really get 'tricked' by prices in markets with production lags?

It turns out they do. This New Zealand Herald article about the dairy sector from last week notes:

Fonterra has fallen into line with market expectations after oversupply and extreme volatility on world dairy markets prompted the co-operative to once again lower its farmgate milk price forecast for 2014/15.

The co-operative yesterday cut its forecast farmgate milk price for the current season to $4.50 a kg of milk solids from $4.70, having shifted it to that level in December from $5.30, and lowered the advance rate of scheduled monthly payments to its farmers. The 20c downward revision alone was estimated to lower farm incomes by about $370 million.

Fonterra said it now expected to produce slightly more milk this season despite the effects of this year's drought...

ASB Bank rural economist Nathan Penny said local production had suffered a lag effect of the season starting off at $7 a kg - at the time offering a green light for cashed-up farmers to produce more and spend more on supplementary feed - which pushed up output.

The cobweb model of demand and supply applies to markets where there is a production lag - where suppliers make a decision about how much to supply today (based on expectations about the price, which might naively be the observed price today), but the actual price that they receive is not determined until sometime later. The supply of dairy products in New Zealand fits this assumption - farmers make their production decisions today, but the dairy cooperatives (Fonterra, Westland, etc.) don't make a final decision on the price farmers will receive until close to the end of the season.

So, what has happened? At the start of the milking season, farmers' expected price was $7.00 per kg of milk solids (unless they had prescience), so they based their production plans on that price. In the diagram below, say that D0 and S0 are the initial demand and supply curves, respectively, with demand for dairy products relatively high. Farmers, who expect the price P0 ($7.00), produce Q0 units of dairy products (this is the quantity supplied on the supply curve S0, with the price P0). By the end of the season though, demand has fallen to D1. When the farmer cooperative tries to sell Q0 dairy products, the price falls to P1 ($4.50; this is the price where the quantity demanded, from the demand curve D1, is exactly equal to Q0).

Now, going into the next season farmers observe the low price P1 ($4.50) and expecting that low price to persist, they produce Q1 units of dairy products (this is the quantity supplied on the supply curve S1, with the price P1). Come the end of the season, the farmer cooperative finds that they can sell the Q1 dairy products for the much higher price P2 (this is the price where the quantity demanded, from the demand curve D2, is exactly equal to Q1). Now the price is high, farmers produce more but at the end of the season, the price falls... and so on. Essentially, the market follows the red line (which makes it look like a cobweb - hence the name of the model), and eventually the market gets back to long-run equilibrium (price P*, quantity Q*).

Will this always be a problem for the dairy sector in New Zealand? It seems likely. Farmers base their production decisions on the farmgate price forecasts of Fonterra, or AgriHQ, or their forecaster of choice. None of these forecasters have perfect foresight. So if the price forecast at the start of the season is too high relative to demand then prices will fall, and will fall further than the new equilibrium price would be, because farmers will be over-producing. And if the price forecast at the start of the season is too low relative to demand then prices will rise, and will rise further than the new equilibrium price would be, because farmers will be under-producing.

The moral of the story is that the New Zealand dairy production model locks us into a boom-bust dairy commodity cycle: any time global (read: Chinese) demand for New Zealand dairy products falls, the price in New Zealand will crash, while any time global demand for New Zealand dairy products increases, the price in New Zealand will skyrocket. And it seems unlikely that producing higher value-added products would adequately shield us from the cycle either - in years with high dairy prices, the cost of producing value-added products increases. I'm with Lyn Webster when she says: "a high milk price and a big return on value added products are mutually exclusive".

Reducing the extent of the boom-bust cycle may be possible if farmers can better smooth the price fluctuations - perhaps by making more use of forward contracts or dairy futures (see useful discussion here). That way at least, when the demand for New Zealand dairy products rises, farmers won't be 'tricked' into over-supplying (or under-supplying when the demand falls).