SS Economics of Food Markets 2002/03
Data issues in MT assignment (revised)
Getting estimates of quantities supplied and demanded in Ireland and the EU.
The source for these are food balance sheets, also called supply balance sheets. You can find these on the CSO website (for Ireland), in the Agricultural Situation in the European Union (for the EU) and on the FAOSTAT website ( for both.
For Ireland, go to the CSO website ( choose Statistics, and then Agriculture. You will find Meat, Cereals and Milk and Milk Products supply balance sheets. CSO data are also available in the annual Statistical Abstract (which provides a compendium of the most useful statistics published by the CSO) and in the quarterly Statistical Bulletin.
Domestic production is variously referred to as slaughterings or usable production, domestic demand as internal uses or (somewhat confusingly in the FAO balance sheet, as domestic supply).
Note that you can derive domestic demand from domestic production if you know exports and imports (and assuming that stock changes are zero) using the identity
Demand = Production + Imports – Exports – Increase in Stocks
Note that the meat balance is in ‘000 tonnes. As prices are often quoted in liveweight (i.e. for live animals) you may need to convert live animal weight to carcase weight. The notes to the supply balance sheets give you the appropriate coefficients to use.
The milk supply balances on the CSO website are in terms of individual milk products such as whole milk, butter, milk powder and cheese (in tonnes) rather than total milk (in litres). You can either try to assemble a total milk supply balance using this information, or you can use the FAOSTAT website.
The FAOSTAT website (see below for instructions on use) gives commodity balances for Milk Skimmed, Milk Whole and Milk- Excluding Butter. Use the latter as the total milk supply balance. The figures are meant to be in tonnes but seem much higher than those returned on the CSO website. I am not sure of the reason for this discrepancy.
Alternatively, you can try to work out a total milk balance from the information on the CSO website. For this purpose you need to know how much milk is used in the manufacture of each product. It is helpful to know that the manufacture of butter and skim milk powder are joint products, thus you should use the milk equivalent of either butter or skim milk powder but not both (this is only a rough approximation as the supply balance sheet only distinguishes milk powder and not just skim milk powder. Some milk powder is whole milk powder which is simply dried whole milk and thus absorbs some of the milk production). The notes to the CSO milk supply balance sheet give you the coefficients you need for this.
The sugar supply balance is not published on the CSO website but an alternative source is the FAO Statistics web site. You will find it easiest to work in white sugar equivalent (refined sugar). Production figures are sometimes for sugar beet, and to convert to white sugar use an extraction ratio of 16% (i.e. 100 tonnes of sugar beet converts to 16 tonnes of white sugar).
For the EU, the equivalent supply balance sheets are found in The Agricultural Situation in the European Union publication or on the DG Agri website
(look for the link to Agriculture in the European Union – Statistical and Economic Information).
To use the FAOSTAT website, go to then click on Agriculture, and under Commodity Balances, on either Crops Primary or Livestock Primary equivalent. For country, choose the European Union 15 (note that EU15+ just gives you the total, while EU15> gives you the total plus each member state including Ireland, so choose the latter). Then choose your commodity, select both Production and Domestic Supply from the third list, and finally the year. (To select more than one element from a list, hold down the Shift or Control keys).
Data on internal and world market prices
Agricultural prices come in two varieties
(a)quoted prices for specific grades or qualities
(b)unit values
The former allow comparisons of price movements over time because one is sure you are comparing like with like. Unit values can be affected by differences in the composition and quality of the output produced (for example, sugarbeet can have a different sugar content from year to year, depending on the weather). On the other hand, unit values do represent on average what farmers actually received in a given year.
For Irish prices, the CSO publishes prices for typical livestock categories on its website. You have to follow the link to Agricultural Price Indices. You don’t want an index but the absolute level of prices, and some absolute prices (in €/head) are given towards the end of those reports. It may be easier to find these in the Statistical Abstract publication. Choose a heavier animal as these will represent mature animals ready for slaughter rather than lighter animals which will be younger. These prices need to be converted to €/tonne using the weight range indicated and the conversion coefficients given in the supply balance sheet for meat (see above).
Irish agricultural prices are also reported in Eurostat Agricultural Price publications but the Library does not have a good set of these.
For internal prices, it will probably be easier to use unit values. These are obtained by dividing the value of output by the volume of output. These data are available in the Output, Input and Income in Agriculture releases on the CSO website, and the same table also appears in the Statistical Abstract.
For EU prices, the same unit value approach could be used but unfortunately there is not a similar table in the Agricultural Situation in the European Union publication. However, it does contain a table for Producer Prices in the Member States (Table 3.3.13). Although there is no EU average given in this table, you could construct a simple average of the individual states’ prices as a proxy for the EU price. An alternative approach would be to take the price prevailing in the member state which is the most important producer of the commodity.
Another approach would be to assume that the EU market price is very close to the EU intervention price and to use the intervention price (given in Table 3.3.1) as a proxy for the EU market price.
For the world price, I suggest you use the same world price for both your Ireland calculations and the EU.
World prices for some commodities are given in The Agricultural Situation in the European Union publication or its equivalent website (see above).
For other (quoted) prices, I suggest you use the commodity prices available in the IMF International Financial Statistics which is available in the Library (the web site is now a pay site). As these quoted prices are sometimes in strange units (for example, wheat is in $/bushel) you need to make the appropriate conversions to €/tonne (for example, see the bushel/tonne converter on
Exchanges rates between the US dollar and the Euro can be found, among other places in Central Bank Quarterly Bulletins which is also available on the Central Bank website (check through the Statistical Appendix for Exchange Rates: Period Averages Table B6).
Data on elasticities
These are to be found in the Tyers and Anderson book World Agriculture in Disarray, Appendix 2. For supply, both short-run and long-run elasticities are available. I don’t mind which you use provided you understand the difference, and in any case part of your assignment is to test the sensitivity of your results to different elasticity assumptions.
Data on marketing margins
The reason for including marketing margins in the analysis is that demand elasticities are estimated using retail or consumer prices while supply elasticities are estimated using producer or farmgate prices. The difference between these prices is the marketing margin.
Information on the marketing margin for each commodity is not easily available. To simulate its role, I suggest that you assume that the marketing margin is exactly 100% of the producer price, i.e. Pd= 2 *Ps. Enter this formula into your spreadsheet when calibrating your demand equation.
Alan Matthews
December 2002