Integration of automated ECG analysis techniques with the home monitoring devices can incorporate the necessary 'smartness' which can help in earlier diagnosis of Myocardial Infarction (MI), better known as heart attack, thus reducing the mortality rate. Most of the reported techniques suffer from the disadvantages of large feature dimension, computational complexity of the features and complex classifiers which make implementation difficult and time consuming. In this paper we explore the utility of Fourier harmonic phase of the ECG data for MI identification. The phase response properties of the Fourier harmonics of the time and amplitude normalized lead II ECG beats from healthy and infarction records were analyzed to identify two clearly distinguishable features related to the changes in the phase distribution pattern. Use of heuristic threshold based classification with these extracted features could classify the healthy and the inferior infarction data from the PTB diagnostic ECG database with a high detection accuracy of 97.4% and sensitivity and specificity of 98.2% and 96.3% respectively. The robustness of the proposed technique was also validated with real ECG data collected using BIOPAC MP-45 data acquisition system. The results are clearly indicative of the potential of this novel feature space for MI identification. Largely reduced feature dimension, computational simplicity of the features and the use of simple classifier contributes to the implementation simplicity of the proposed technique justifying its use in portable devices. © 2017 IEEE.