Overdispersion can result from many reasons. Hilbe (2007) provides an excellent discussion of this issue, differentiating between apparent and real overdispersion. Apparent overdispersion can occur as a result of outliers, the exclusion of relevant risk factors or interaction terms. In this respect, Denuit et al. (2007) highlight that overdispersion arises because differences in driving behaviour among individuals cannot be observed by the insurer, such as swiftness of reflexes, aggressiveness behind the wheel, consumption of drugs, etc. When the resolution of these issues does not have a conclusive response, the overdispersion is assumed to be real and it may be due to unobserved heterogeneity related to the equidispersion hypothesis.