Objectives Favourable long-term prognosis in proliferative lupus nephritis (LN) is associated with the achievement of complete renal response (CR), which is defined as a urine protein/creatinine ratio (UPCR) of < 0.5. However, it is unclear whether a more stringent cut-off for proteinuria (normal value of proteinuria; UPCR < 0.15) is better than CR. We aimed to evaluate the effect of stringent CR, defined as a UPCR of <0.15, on long-term renal outcomes in proliferative LN. Methods We included 87 patients with class III or IV LN who achieved CR at one year after induction therapy. Clinical and laboratory data were compared between the stringent and non-stringent CR groups. Logistic regression analysis was performed to identify factors associated with achievement of stringent CR. Cox analysis was performed to analyse the risk factors for renal flare and development of chronic kidney disease (CKD). Results The stringent and non-stringent CR groups included 58 and 29 patients, respectively. The two groups showed no significant baseline differences in terms of the clinical, laboratory and pathological classification. The sustained CR rates during five years were 91.3% and 50.0% ( p = 0.014) in the stringent and non-stringent CR groups, respectively. In Cox analyses, the achievement of stringent CR was associated with a lower risk of five-year renal flare rate (hazard ratio (HR) = 0.161, 95% confidence interval (CI) 0.063–0.411, p < 0.01) and development of CKD (HR = 0.189, 95% CI 0.047–0.752, p = 0.018). Mycophenolate mofetil induction therapy was associated with achievement of stringent CR at a borderline level of significance (HR = 7.268, 95% CI 0.894–59.089, p = 0.064). Conclusion Achievement of stringent CR predicted lower risk of renal flare and development of CKD in proliferative LN. These findings suggest that stringent CR is a valuable treatment target in proliferative LN.