Tight monetary policy affects consumption expenditures in different ways depending on the income level of households. Individuals with more savings opportunities are expected to reduce their expenditures as a result of higher returns on deposits, while individuals with credit-sensitive demand are expected to curb their expenditures due to higher credit costs. In this blog post, we examine consumption expenditures across different income groups to enable better understanding of the recent monetary transmission mechanism.
We can monitor spending patterns by income groups through the Turkish Statistical Institute’s (TURKSTAT) Household Budget Survey data. However, these data are published annually and the latest are for 2023. Therefore, we cannot directly observe expenditures by income groups in a timely fashion and at a high frequency. That is why we use an indirect method that will also shed light on 2024.
First, using the amount consumers spend on the products in the CPI basket and household income, we calculate a representative consumer income for each product.[1] Then, we match each product with the relevant sector and use retail sector turnover and sales volume data for the respective period to measure demand for the products. Using product-sector correspondence and the representative consumer income, we categorize retail subsectors into three groups as products catering to high, middle and low-income groups.[2],[3] Finally, we aggregate sectors’ turnover according to these categories to obtain income group-based turnover and sales volumes.[4]
A detailed analysis of the course of turnover in time reveals that the annual turnover growth rate fell across the board due to monetary tightening (Chart 1). This decline is more pronounced in products catering to the high-income group. In the second quarter of 2023, just before the monetary tightening began, the annual turnover growth in products catering to the high-income group was approximately one and a half times higher than that in products catering to the low-income group, but growth rates in both product groups declined to similar levels as the monetary tightening kicked in. With the monetary tightening, turnover growth of products catering to the high-income group fell below that of the products catering to the low-income group. This implies that high-income households, in particular, preferred to save.
A similar picture appears for sales volumes (Chart 2). While the annual change in the sales volume in products catering to the high-income group was about threefold the change in products catering to the low-income group before the monetary tightening, the rates of growth fell to similar levels after the tightening. In other words, sales in products catering to the high-income group were affected more by monetary tightening in real terms than products catering to the low-income group. However, unlike turnover, the growth rate of sales volume in products catering to the high-income group was slightly higher than that of other products, indicating that relative prices of product groups differentiated.
To sum up, consumption expenditures have been slowing down across all income groups with monetary tightening. The most significant slowdown is in products catering to the high-income group, which were on a much stronger course than other groups prior to the monetary tightening.
[1] We classify the products in the CPI basket based on the four-digit level of the COICOP (Classification of Individual Consumption According to Purpose) classification. We exclude gold-related items in our calculations. The representative consumer income for a product denotes the average income of households spending on that product weighted by their share in expenditures.
[2] Using the COICOP product classification and NACE Rev. 2. activity classification correspondence table, we identify the CPI basket products that correspond to the retail sub-sector at the four-digit level classification.
[3] We group sectors in such a way that each category has an equal number of sectors. We obtain similar results when we reduce the number of sectors in the products catering to the high-income group category.
[4] We deflate turnovers by three-digit sectoral price indices to calculate sales volumes. We divide three-digit level turnover indices by three -digit level sales volume indices, both announced by TURKSTAT, and obtain price indices.