Household consumption would come under pressure from rising interest rates and borrowing costs. But even under the same rate-changing environment, the impact of the changes in the debt-to-service ratio (DSR) on households‘ consumption can differ depending on their characteristics. However, most of previous studies neglected heterogeneity across households and primarily focused on restricted characteristics such as either income or assets. Thus, we use the Korea Labor and Income Panel Study (KLIPS) and investigate possible changes in the effect of a rise in interest rates on household consumption across different subgroups of households categorized by income, debt, and homeownership.

The results of the panel estimation show that a one percentage point rise in DSR reduces consumption of the entire household sample by 0.37 percent on average. We also find significant heterogeneity in the response of consumption across different subgroups. First, the rise in DSR leads to a considerable decrease in consumption of households categorized as high debt-low income and high debt-non-homeowner groups, well above the reduction of overall households’ consumption response. Second, when we categorize households just as low and mid to high income groups, the rise in DSR causes a greater decrease in consumption of mid to high-income group than that of low-income ones. This is because low-income group has less room to cut back as spending mostly goes to daily necessities. Higher-income group has greater discretionary propensity in spending and therefore can adjust spending on discretionary products such as durable goods and high-end apparel. Lastly, the rise in debt levels under normal setting can help to boost consumption, but for highly leveraged households (above 200 percent threshold) the additional rise in borrowing causes reduction in consumption.

These findings suggest when financing burden increases for households, the typical weak groups such as highly indebted low-income can reduce expenditure mostly on essential goods while households in higher-income groups would decrease their discretionary spending. Policy design could be more effective by taking into account of these tendencies. On the mid to longer-run, controlling household debt at reasonable level is desirable not just for financial stability but also for consumption smoothing for economic stability.