I want to estimate a fixed effects model (with both country- and year fixed effects) and ADDITIONALLY apply HAC standard errors. Although I didn’t get any error message after running the following code in r, I’m not really sure whether I’m actually getting HAC standard errors. Can someone please confirm? In particular, I’m getting exactly the same standard errors if I change HAC3 to HC3 or even remove the option vcovHAC (or vcovHC) completely, etc.

fe1 <- plm( log(DEB) ~ log(IMP),
                   model = "within",
                   data = data,
                   index = c("country","year"), 
                   effect = "twoways")

fe2 <- plm( log(DEB) ~ log(IMP),
                   model = "within",
                   data = data,
                   index = c("country","year"), 
                   effect = "twoways",
                   vcovHAC = "HAC3")

fe3 <- plm( log(DEB) ~ log(IMP),
                   model = "within",
                   data = data,
                   index = c("country","year"), 
                   effect = "twoways",
                   vcovHC = "HC3")