Despite the voluminous research on turnover, most studies tend to focus on individual-level predictors of turnover and have not been able to offer sufficient predictive power for managers to forecast their labour turnover projections. In this study, the popular forecasting methods of Random Walk, Historical Average, Moving Average, Exponentially Weighted Moving Average (EWMA), and Autoregressive (AR) were compared for their forecast performance on labour turnover rate. Data for this study was obtained from the Job Openings and Labour Turnover Survey (JOLTS) compiled by the US Bureau of Labor. This study also evaluated the existence of monthly seasonal effects in the labour turnover rate. The results showed that the best forecasting model for the labour turnover rate is the Autoregressive (AR) model with order 3, within-sample and post-sample. The study also found that monthly seasonal effects exist in the labour turnover rate.