Lagged Dependent Variables and HAC Standard Errors

[Mila]

Hello,
I wonder if anyone can shed some light on the following:
I seem to get conflicting thoughts on whether Newey West HAC Standard Errors can be used in the presence of autocorrelation in a specification that contains lagged values of the dependent variable on the RHS.
Some say the correction is not valid in the presence of lagged dependent variables, i.e. one either uses Newey West in a model without lagged dependent variables or constructs a specification with lagged dependent variables, and finds other means of correcting autocorrelation if it's still present. On the other hand, I see papers that use the HAC SEs even when there are lagged dependent variables.
Any clarification on the above is much appreciated!