Aim:The purpose of the present study was to investigate the effect of cognitive decline on household spending. Methods: The panel data covered the years 2007-2009 and were taken from the Japanese Study of Aging and Retirement, with data extracted for 253 participants aged 65 years and over. Sociodemographic characteristics, functional status, depressive symptoms, and monthly household spending excluding spending for shelter (e.g. mortgage) were used in the analysis. Cognitive function was evaluated by delayed recall and serial 7s and represented by the sum of both scores. Changes in spending and cognitive function scores were calculated by subtracting 2007 scores from 2009 scores. Participants were divided into two groups: cognitive decline and no cognitive decline. The effect of cognitive decline on changes in household spending was estimated. Results: Although the cognitive decline group had more severe depressive symptoms than the no cognitive decline group in 2009, cognitive decline was the only factor related to change in household spending. Conclusion: The results imply that cognitive decline may cause a decline in household spending in older people. Focusing on changes in household spending behaviour might help detect cognitive impairment in older people.