Most of seismic codes regarding seismic isolation prescribe upper and lower bound analyses to account for the variability of the mechanical properties of seismic isolators (ASCE 41, ASCE 7, EN 15129, EN 1998). This procedure aims to increase the reliability of base-isolated structures and should be carried out based on experimental data provided by the specific manufacturer during the qualification tests of the devices. In the case that such data are not available, codes provide the so-called property modification factors or l-factors for different kind of devices, giving upper and lower bound coefficients to be applied to nominal properties of bearings. Different sources of variability are taken into account from the variability due to the bearing production to environmental or behavioural effects (such as temperature, contamination, ageing or repeated cycles). However, the origin of these values is not straightforward: some of them are strictly related to the limit imposed by the codes for the qualification tests, whereas others are based on indications given by technical literature. More in details, in the European context, both the EN 1998 (part 2 – bridges) and the EN 15129 (anti-seismic devices), provide values derived from old American codes for seismic isolation of bridges, in particular the AASHTO 1999, which in turns proposes values derived from the MCEER-99-0012 report “Property Modification Factors for Seismic Isolation Bearings” (Constantinou at al, 1999). The aim of this paper is to illustrate in detail such factors and to make the reader aware that most of factors derive from limited and/or old data, especially for HDRBs (High Damping Rubber Bearings), for which updated experimental campaign have not been carried out. Finally, for this last kind of bearings, linear equivalent parameters are calculated in the varied conditions, according to l-factors proposed by European codes, in order to show their influence in the design of base-isolated structures.