Almost all fisher models assume time-invariant parameter values of the underlying biological growth function except for an i.i.d. error term. We examine the economic implications of cyclical growth parameters in both single and multi-species models, which are frequently observed in many real-world fisheries. Neither optimal harvest rates nor optimal escapement (remaining fish stock after fishing) remain constant as current models would predict. The amplitude of the optimal escapement is incrasing in the amplitude of the biological growth function. Moreover, the optimal harvest rate lags the cycle of the biological growth function, i.e., the highest harvest rate is observed after biological conditionos have started to decline and the optimum escapement level has alrady decreased. This is in sharp contrast to current policies which are in phase with biological conditions and hence imply and increasea/decrease in harvest quotas when the biological system is improving/deteriorating. In our model, harvest closures are only optimal during time periods when growth parameters are improving most rapidly. We show that once the periodicity of the biological growth function is incorporated, many of the traditional policy prescriptions reverse.