If the open reading frames of your retrotransposon follow the order of a Gypsy/BEL family you should classify it as a Gypsy, nevertheless if you suspect that is a ERV that has recently lost the env protein:
1) you could check with censor or repeatmasker if your element presents higher homology with some ERV
2) you should find in the genome of the species you are studying the sequence of the ancestral element with a env ORF
3) if a sequenced genome of your species is not available, you should find the hypotetical ancestral form anywhere in the genome through a PCR with a couple of primers among the INT domain and the LTR 3' of your element. If an ancestral ERV exists you will find a shorter band corresponding to your element and a longer one corresponding to the complete ERV emboding the ENV domain.
If you don't find a clue with these analysis, in my opinion it is unlikely that all ERV present in the species you are studing have undergone a deletion without leaving any residues of the env domain.