-0.060, and -0.032 obtained by Kinnucan et al. (2001). All of the own-price
parameters are statistically significant, confirming the strong influence of prices on
the allocation of consumer spending. The intercepts terms in both models reveal a
positive consumption trend for soft drinks and a negative consumption trend for
milk, and coffee and tea. The main difference lies in that Kinnucan et al. (2001)
found that advertising enhanced demand for juices, while model I does not report
any statistically significant own advertising effect or price-advertising interaction
effect.
Results of the best-performing AIDS model indicate that advertising might
have the ability to make the demand curve steeper for milk, and coffee and tea, as
well as the ability to make the demand curve flatter for soft drinks. For milk and
coffee-and-tea advertising, this is the case depicted in figures (2d) - (2f), where
advertising flattens the probability distribution of WTP. For soft-drink advertising,
it’s the reverse. The implications are, although this might not be the true intention of
producers who advertise their products, advertising of milk, and coffee and tea
appeals better to consumers who have high WTP for them, while soft-drink
advertising appeals better to consumers who have low WTP for it. Our policy
suggestion based on the AIDS model, therefore, is that milk and soft-drink firms
might enhance profits by timing advertising to coincide with high- and low-price
periods, respectively.
18