Has Household Purchasing of Confectionery Products in Turkey Changed in the Last Decade

Abstract

Although consumption of confectionery and chocolate products per capita in Turkey is considerably lower than that of the developed counties, there has recently been a rising trend in sugar consumption. The purpose of this study was to identify how and to what extent the significant changes in socio-demographic and economic structures of households in Turkey during the period of 2002-2013 influence monthly real spending on sugar, jam-marmalade, and confectionery products. The data was obtained by the annual household budget surveys of the Turkish Statistical Institute (TURKSTAT) between 2002 and 2013. These surveys are annually conducted by TURKSTAT with nearly varying ten thousand randomly selected households throughout the nation for the period between January 1 and December 31. A multivariate Tobit model was used to determine factors affecting Turkish monthly household expenditure on three foods. Cross correlation coefficients among food pairs and marginal impacts of exogenous variables were also estimated. Almost 79% of households spend on confectionery foods, followed by 73% on sugar products and low 20% on jam-marmalade foods. According to results compared to the 2002 reference year, spending on sugar and jam-like food has declined over the years, while more spending on confectionery was emerging. Over the course of twelve years period, the sample households monthly spend 13.84, 3.73 and l 12.91 on sugar, jam-marmalade and confectionery foods, respectively. The correlation coefficients among the food items were all positive and statistically significant, indicating that after controlling the role of independent variables in our model, uncontrolled variables along with measurement errors and any other functional forms induce both the spending on the two food pairs. Impacts of many socio-demographic and economic factors on monthly household spending of confectionery products including sugar were identified, and also years' impacts on spending patterns were confirmed. By years, the age variable had different effects on the food spending of the three categories. Nonetheless, it has been identified that the families whose household heads have a green card spend different amounts both for sugar and confectionery products annually. It has also been identified that while only confectionery food product spending of the families who receive in-kind and financial aids differs annually, all three categories of food spending differ significantly as the numbers of working individuals changes yearly. Results permitted us to draw relevant policy implications and to shape policy formation for future effective food programs or policies.