Voltage fluctuations caused by rapidly changing loads, such as arc furnaces, can propagate to different parts of a power system. Although the flicker level at its origin can be high, levels that are measured at other sites are subject to attenuation, a process that is influenced by fault levels, transformer impedances, line impedances, and composition of the connected loads. This paper presents the methodology, measurement results, and data analysis in relation to synchronized flicker measurements carried out in a high-voltage (HV)/medium-voltage (MV) power system which contains an arc furnace supplied by a dedicated feeder connected to the HV busbar. The flicker transfer coefficients derived from measurement results clearly indicate that flicker transfer from the arc furnace site to the upstream HV busbar is governed by the fault levels at the two locations. However, the transfer of flicker from the upstream HV busbar to other downstream busbars is dependent on the downstream load composition. These flicker transfer coefficients are vital in the application of methodologies described in many reports and standards in relation to establishing planning levels at various voltage levels and in the allocation of flicker emission to customers.