Estimating tobacco price elasticity in Kosovo: using the micro data from Household Budget Survey (2007 – 2017) and Deaton demand model
More details
Hide details
Centre for Political Courage, Kosovo
Publication date: 2019-03-26
Tob. Prev. Cessation 2019;5(Supplement):A96
Download abstract book (PDF)

Literature shows that in low- and middle-income countries (LMICs) tobacco expenditure represents a significant portion of the household budget especially in poor household, it exacerbates poverty by increasing healthcare costs, reducing incomes, and decreasing productivity. This analysis aims at estimating price elasticity of demand for cigarettes in Kosovo.

Using data from Kosovo Household Budget Survey for the years from 2007-2017, in total 10,217 households with positive expenditure on cigarettes enter the sample for the estimation. Deaton (1988) demand model is used and it uses unit values as a proxy for price, spatial variation, and the structure imposed by a weak separability assumption account for the effect of the quality of the good. Basic idea of the model is that all households within a cluster (typically a small territory unit, such as municipality or village) face the same market price and that within-cluster variations in purchases depend only on total household expenditure and household characteristics, while cross-cluster variations in purchase are due to genuine price variations, among other factors. The estimation of the model consists of three stages. In the first stage, the effects of total household expenditure and other household characteristics are by the means of regression analysis purged from the budget share of the consumption and unit value. In the second stage, cluster average values of budget share and unit values are used to estimate unit value elasticity of consumption. In the final, third stage, we use separability assumption to separate the effect of price elasticity from the quality effects contained in unit value elasticity.

Results of the estimation indicate a negative price elasticity of cigarettes which amounts to −0.288. Standard error of the elasticity calculated via bootstrapping procedure (1000 replications) indicates that the elasticity is significantly lower than zero (ξ= -0.288; SEξ = 0.097, t = -2.969).

In Deaton’s model unit value of cigarettes as approximation of the prices are used and find a negative price elasticity of demand for cigarettes of -0.288. Although the previous estimates for Kosovo are non-existing and therefore it is impossible to make any comparison, this result is in line and validated with the theoretically expected and previous estimates of price elasticity for cigarettes for low and middle income countries.

The research I conducted is funded by the University of Illinois at Chicago Institute for Health Research and Policy through its partnership with Bloomberg Philanthropies.
Journals System - logo
Scroll to top