This study evaluates whether modelling the existing commont trends in tourism arrivals from all visitor markets to a specific destination can improve tourism predictions. While most tourism forecasting research focuses on univariate methods, we compare the performance of three different Artificial Neural Networks in a multivariate setting that takes into account the correlations in the evolution of inbound international tourism demand to Catalonia (Spain). We find that the multivariate multiple-output approach does not outperform the forecasting performance of the networks when applied country by country, but it significantly outperforms the forecasting performance for total tourist arrivals.