Tourists from different European regions have reacted heterogeneously during the Global Economic Crisis. Such variability is due to different preferences and willingness to pay for tourism. This paper explores the underpinnings behind such heterogeneity. Regional variables and household socioeconomic variables are gathered to understand tourists' expenditure cutback decision. Since the cutback decision is not independent of the destination choice, a Simultaneous Semi-Ordered Bivariate Probit model is specified, which deals with the simultaneous estimation of both decisions and endogeneity. Post-estimation results are based on GIS, contours and non-parametric analysis. They prove that during an economic crisis, tourists' cutback decisions on tourism expenditure depend on climate conditions of the place of origin, GDP and GDP growth.