Sustainability Standards and Decertification: A Bayesian Analysis of SA8000

Abstract

Social Accountability 8000 (SA8000) is widely recognized as the leading certification-based standard for social sustainability. The common assumption is that the number of certified companies increases steadily over time. However, this prevailing narrative lacks empirical support owing to the absence of comprehensive studies on SA8000 decertification. Using a beta-binomial model estimated within a Bayesian approach to inference, we analyze a 25-year dataset of SA8000-certified companies. Our findings indicate that growth in certifications may not reflect a stable, long-term commitment to the standard. Instead, a significant proportion of companies discontinue certification, often after the initial three-year period. This study reveals that decertification rates are higher among companies with fewer certification renewals and vary considerably by industry. These insights contribute to a refined understanding of the certification and decertification patterns and highlight the need to further explore the long-term maintenance of social accountability standards across different industries and contexts.