Background Dog ownership is emerging as an important correlate of sufficient physical activity and therefore has the potential to positively affect a portion of the population. A growing body of literature indicates that dog-walking contributes to increased physical activity. However, most of the previous studies have been conducted in Australia or the U.S. and have sampled from the general adult population. Purpose This study examined the association between dog ownership, dog-walking, and physical activity in older Japanese adults. Methods Participants were community-dwelling residents aged 65–74 years who responded to a population-based cross-sectional survey (N=1926). Physical activity, dog ownership, dog-walking, and sociodemographic attributes were self-reported (collected in 2010; analyzed in 2011). ANCOVAs and multivariate logistic regressions were used. Results Overall, 14.0% of older adults were dog owners, with 71% reporting that they walked their dog for an average of 308.5±300.7 minutes/week. Dog walkers reported more minutes/week of moderate-to-vigorous physical activity (M±SE: 241.7±27.0) and total physical activity (M±SE: 698.6±40.6) than both non–dog walkers (M±SE: 110.7±41.8; M±SE: 527.2±62.9) and non–dog owners (M±SE: 164.7±9.1; M±SE: 519.2±13.7), respectively (p<0.05). Dog walkers also walked more minutes per week (M±SE: 508.0±33.4) than non–dog owners (M±SE: 384.5±11.3; p<0.05). Dog walkers were more likely to be sufficiently active than both non–dog walkers and non–dog owners (p<0.001). Conclusions Use of dog-walking may be a potentially viable means of intervention for increasing walking and overall physical activity in older Japanese adults.