This paper demonstrates for the first time how heart rate (HR) can be extracted from novel neck photoplethysmography (PPG). A novel algorithm is presented, which when tested in neck PPG signals recorded from 9 subjects at different respiratory rates, obtained good precision with respect to gold standard ECG signals. Mean absolute error (MAE), standard deviation error (SDAE) and root-mean-square error (RMSE) resulted in 1.22, 1.54 and 1.98 beats per minute (BPM), respectively. HRneck estimation showed strong correlation (R=0.94) with reference HRECG. Good agreement between both techniques was also demonstrated by Bland-Altman analysis. The bias between mean HR paired differences was -0.16 BPM and 95% limits of agreement (LoA) were (-4.7, 4.4). Comparatively, for widely used finger PPG, errors were slightly smaller (MAE=0.38 BPM, SDAE=0.48 BPM, RMSE=0.62BPM) and the correlation with reference ECG was also very close to 1 (R=0.99). Bias of -0.04 BPM and 95% LoA (-1.5, 1.4), also showed high degree of agreement. However, these findings show the potential the neck could have as an alternative body location for wearable monitors, aiming to reduce the number of sensing sites whilst still providing access to a wide variety of physiological parameters.