Option 1 -> When inputs are scaled by factor t, output scales exactly by factor t: f(tx₁, tx₂) = tf(x₁, x₂).
Option 2 -> When inputs are scaled by factor t, output increases more than proportionally: f(tx₁, tx₂) > tf(x₁, x₂).
Option 3 -> When inputs are scaled by factor t, output increases less than proportionally: f(tx₁, tx₂) < tf(x₁, x₂).
Option 4 -> This is not a standard economic term for returns to scale.
Hence, Option 3: Decreasing returns to scale -> The inequality f(tx₁, tx₂) < tf(x₁, x₂) indicates that when all inputs are multiplied by factor t, the output increases by less than factor t. For example, if we double all inputs (t=2), output less than doubles. This diminishing response characterizes decreasing returns to scale, often occurring due to coordination difficulties or resource constraints as production scale increases. -> correct