# Market Preview: Understanding Pork Demand

US - Weekly US Market Preview provided by Steve R. Meyer, Ph.D., Paragon Economics, Inc.

One of University of Missouri Professor Glenn Grimes’ innovations in explaining markets to pork producers is the demand index – a measure of year-to-year change in demand. The index is simple, elegant, theoretically correct and widely used. The beef industry even adopted the demand index as a monitoring device several years ago, with economists at Kansas Sate University providing the computations on a quarterly and annual basis.

The index measures the movement of the demand curve for a commodity. Demand is not consumption – it is the set of quantities that consumers are willing and able to purchase at alternative prices. A line like “D,” in Figure 1, represents a demand curve or function. To measure the change in demand from D to D’, Professor Grimes performs a few calculations.

- First, he measures the percentage change in per capita consumption, here represented by (Q2-Q1)/Q2. It is a positive number in this diagram, but it could be negative.
- Next, he computes the expected change in price by using the percentage change in consumption and an estimate of the elasticity of demand. An elasticity is simply an economics measurement of the responsiveness of one variable to another. The formula is:

Estimates of the elasticity of pork demand range from about -0.65 to -0.90. Professor Grimes uses three elasticities in his calculations (-0.67, -0.75 and -0.90) to demonstrate any sensitivity there may be to this key assumption. Substituting the elasticity and the percentage change in consumption into the formula above, one can solve for the expected percentage change in price – depicted by the move from P1 to P^{E} in Figure 1.

- Finally, the actual percentage change in price ((P*-P1)/P1 in Figure 1) is computed and compared to the expected change. The difference is attributable to a shift of the demand curve from D1 to D2.

Note that this calculation uses per capita consumption and deflated prices. That means it understates total pork demand since it does not account for population growth; it measures the demand change for a fixed population. We don’t get “credit” for population growth since we don’t have much to do with that. Similarly, it does not allow price inflation to be interpreted as an increase in demand.

In addition, the calculation says nothing about why demand may be changing. It is simply a descriptive metric that says “demand moved by this much.” The shift could be caused by a change in beef or chicken prices, prompting consumers to move to or away from pork. It could be caused by a change in consumers’ income levels or an expenditure decision. It could be caused by changing consumer preferences for flavors or concerns about health attributes, or even irrational fears about influenza, for example. If your goal is to answer the question, “Why has demand moved?”, this metric is not sufficient. But it is very useful for simply monitoring what is happening to pork demand.

So now that you understand the demand index, I’m going to change it.

Figure 2 shows annual consumer-level demand indexes for pork and an alternative measurement, real per capita pork expenditures or RPCE. It is computed simply as:

Note that it uses three of the same variables used in the index calculations: per capita consumption, retail pork price and the Consumer Price Index (CPI ). Those are the same data used to compute percentage changes in quantity and real price for the demand index calculations.

The only thing missing from the RPCE calculations is the elasticity of demand – but it is actually there, as well, since the negative relationship between price and quantity represented by the demand curves in Figure 1 determine the price based on how much people consume. So the only real difference between the two demand measures is that the index uses an elasticity estimate, while RPCE implicitly uses the actual elasticity by using both the price and quantity for a given time period.

The proof of their similarity is in the results. Over time, the two measures are correlated to the tune of 0.983, meaning that 98.3 per cent of the variation of one can be explained by the other.

If you see real per capita consumption or RPCE being used to describe pork demand, realize that the numbers have changed but the measurement is still comparable and consistent. It is only a tweak to Professor Grimes’ contribution to our understanding of markets.