This research aims to model the Pakistan long-run per capita growth under Markov switching regimes to explain regime changes in the economic growth for the period 1970 to 2015. The objective is attained with the help of Markov Switching Models, namely, the Hamilton (1989) Markov Switching Model and Diebold et al. (1999) Time Varying Markov Switching Model. Wefound the evidence ofnonlinearity in the per capita economic growth, and recognized two different levels in the data related with stagnation and stable growth regimes. In particular, the probability of remaining in the stagnation regime is high as compare to the stable growth regime. Result also indicate that the probability of remaining in the stable growth regime increases with a fall in inflation and with the increment in the terms of trade and foreign direct investment. [f the economy is in stagnation regime, an increase in foreign direct investment and terms of trade decreases the probability of remaining in this state, while a rise in inflation increases this probability. From the AIC value, it is found that the Markov switching model with Time Varying Transition Probabilities (TVTP) is the best fitted model then the Hamilton Markov Switching Model. Supervisor:- Dr. Hafsa Hina