1. Trade and Markets Division Colombo, Sri Lanka, 30 January – 1 February 2012 INTERGOVERNMENTAL GROUP ON TEA A Demand Analysis for the Tea Market

2. Trade and Markets Division OUTLINE Introduction Definition and Methodology Empirical results Conclusion and issues to consider 2

3. Trade and Markets Division Introduction At its last session, the IGG on tea recommended a review of demand elasticities to gain some insights into factors affecting consumption and update the world tea model. This presentation presents the major findings of this effort and discusses follow up actions needed to improve the results in light of the limited data availability. This analysis should be considered as a work in progress. 3

4. Trade and Markets Division Definition and methodology Several factors influence the demand for tea, including the traditional price and income variables, demographics such as age, education, occupation, and cultural background. In an earlier analysis of the United States tea market the Secretariat found that both income and prices had relatively small impact on consumption, but demographic factors did. The study also revealed that health had a greater influence on tea consumption. 4

5. Trade and Markets Division Definition and methodology During the last two decades, consumer demand analysis has moved toward a system(s)-wide approach. Numerous algebraic specifications of demand systems now exist, including the linear and quadratic expenditure systems, the Working model, the Rotterdam model, translog models, and the Almost Ideal Demand System (AIDS). For this study we used the AIDS approach. The analysis covered five tea markets: China, India, Japan, the UK, the United States. 5

6. Trade and Markets Division Definition and methodology The Secretariat had initially hoped to build a comprehensive demand system, which would cover both complements and substitutes for tea, demographic characteristics for each of the major market covered, retail prices, and other preference shifters. Unfortunately, it was impossible to gather all this information. The analysis was therefore restricted to three products of the beverages group, namely, black tea, green tea, and coffee. Information was gathered from different sources: FAO world tea database, Comtrade, IMF, ICO, and IGG Members. 6

7. Trade and Markets Division Applying the LA/AIDS 7 and where, in observation t; wit is the budget (expenditure) share of the ith good; pjt is the nominal price of the jth good; lnXt is total expenditure; uit is the random or error term; and lnPt is the translog price index substituted by the stone index. We computed two types of elasticities: Marshallian elasticites and Hicksian elasticities.

8. Trade and Markets Division Key results….. India Most of the estimated parameters of the LA/AIDS model were significant, suggesting a significant influence of prices on budget share. Also the sign of price and expenditure variables were consistent with the theory, and their magnitudes were within the expected range. 8

9. Trade and Markets Division Key results….. India 9 Coffee and tea were found to be complementary. Marshallian elasticities:

10. Trade and Markets Division Key results….. Japan 10 Coffee and black tea were found to be substitutes. Hicksian elasticities:

11. Trade and Markets Division Key results….. UK 11 Coffee and black tea were found to be substitutes. Marshallian and expenditure elasticities:

12. Trade and Markets Division Key results….. United States 12 Coffee and black tea were found to be substitutes. Hicksian elasticities:

13. Trade and Markets Division Key results….. China 13 Black tea and green tea were found to be substitutes. Hicksian elasticities:

14. Trade and Markets Division A comparison with some previous estimates of price elasticity for tea 14 *Not significant

15. Trade and Markets Division Summary and follow up Price elasticities (Marshallian) for black tea varied between -0.32 and -0.80. Estimates for prices elasticities (Marshallian) for green tea ranged between -0.69 and -0.98 In general, price elasticities for tea were found larger than previous studies. This study also suggests evidence of substitutability between tea and coffee, and in some cases with green tea. Consumers tend to spend proportionally less on tea and more on coffee, as household expenditure on beverages increases. 15

16. Trade and Markets Division Summary and follow up Prices and incomes have relatively smaller impact on the consumption of tea. Marketing efforts should focus on: –demographic differences amongst consumers, –increased accessibility through improved visibility, and retailing. –support innovative technologies, especially at the consumer level, as demonstrated by the coffee industry. More analysis is required utilizing comprehensive information of all competing beverages in order to establish the nature and magnitude of the changes in consumer choices, if any. If such analysis should continue, the Group is requested to provide the information required to undertake the analysis. 16

17. Trade and Markets Division Thank You