Complete data for per capita consumption is available only two months in arrears since, as readers of this letter know, consumption is a function of production and exports/imports and the export/import data run two months behind. About the best we can do without using export/import forecasts or predictions is to measure the status of demand two months ago.

But even with those data, the question of measuring demand is not completely clear and is muddied by misuse of the term demand. Some claim “demand is good” when large quantities of product are moving without regard for the prevailing price level. Or conversely, they say something like “high prices have reduced demand” — a statement that is equally incorrect. Demand is not quantity. It is the set of quantities of a product that consumers are willing and able to purchase at alternative prices — an entire group of price-quantity combinations that define the demand curve in a classic supply-demand diagram. Our goal has always been to use a measure that includes both price and quantity and thus tells us something about the position of the demand curve in the standard quantity-price space of the supply-demand diagram we learned in Econ 101 or its equivalent course at your favorite university.

Our tool of choice for that comparison has long been the demand index first used by Professor Glenn Grimes at the University of Missouri. This index puts the position of the demand curve in Q-P space in terms of a base year and thus measures its movement by the difference between the index levels each year. An increase in the index means the demand curve has moved up and to the right. A decrease means just the opposite. The index does not explain why demand changes. It only states how much it changes. A complete explanation of index computations, written by Dr Ron Plain, a long-time colleague of Professor Grimes at Mizzou, can be found at agebb.missouri.edu/mkt/demand/index.htm.

The data used in the demand index calculations are per capita consumption, nominal retail price, a measure of inflation (the CPI for all goods, in this case) and an assumed elasticity of demand. The first three of these can be used to simply compute real per capita expenditures (per capita consumption x deflated retail price) or RPCE. Embedded in the year-on-year change in RPCE is the actual negative relationship between price and quantity — or the elasticity of demand.

Are the two measures similar? As can be seen below, they are, historically, practically identical for pork. The same is true for both beef and chicken. The only source of difference would be variation between the assumed elasticity used in the index calculations (here it is –0.75) and the actual elasticity exhibited by consumer purchase decisions.

So how can this information be used? Charts of monthly RPCE for beef and pork appear below. The chicken chart is also below.

The impact of the “pink slime” smear campaign is very clear in the beef chart as RPCE dipped sharply in April. The good news is that it recovered in May to a level 3.8 per cent higher than one year ago. RPCE for beef is up 4 per cent year-to-date and, for the past 12 months, is 3.4 per cent higher than the same period (June-May) one year earlier. We believe the latter comparison is the most valid since it includes an entire year of seasonal variation where year-to-date comparisons do not do so until we get to December.

As we expected, pork demand appears to have been hurt by the LFTB situation as well. It, too rebounded in May to move slightly above the 2011 level. Pork RPCE (demand) is down 0.4 per cent from one year earlier over the past 12 months.

Finally, chicken RPCE (demand) continued to recover from its abysmal February level but remains 2.7 per cent lower than one year earlier over the past 12 months.