This study addresses relative permeability prediction from well test data for low permeability, rich gas-condensate systems. Characteristic of these systems are high velocities and large pressure gradients within the near wellbore region. Within this region the relative permeabilties are rate sensitive and non-Darcy effects can be important. This study combines both the non-linear (in velocity) terms into a single "effective relative permeability" term. Effective relative permeabilities are estimated through non-linear regression with both synthetic and field data. Results show that a two-parameter simplified correlation is adequate for representing effective relative permeability. These parameters can be obtained by matching well test data. Mechanical skin was needed to match field data considered in this study. Non-Darcy effects can decrease the flowing bottom-hole pressures by about 480 kPa in high rate gas-condensate well tests. A well test design is proposed from which gas and condensate relative permeabilities can be estimated.