Comparing coefficients across 2 models

Hi,

I'm woking on Stata 13 S/E, and I'm having trouble comparing coefficients across two models. May someone give me a hand?

Here's what I'm doing:
1) I run a regression without correction for selection into employment:
xi: reg wage i.educ age age_sq gender
2) I run a similar regression, but this time correcting for selection into employment using Heckman's 2-step model:
xi: heckman wage educ age age_sq gender, select(nchild_0_5 nchild_6_12 nchild_13_18 i.educ age) twostep

Now, I'm trying to test whether the coefficients associated to the "gender" variable are significantly different between regression 1 and regression 2.

I thought about using a combination of the -est store- and -suest- commands to do so, but the -suest- command is not compatible with -heckman, twostep-.
Details of what I tried:
xi: reg wage i.educ age age_sq gender
est store reg1
xi: heckman wage educ age age_sq gender, select(nchild_0_5 nchild_6_12 nchild_13_18 i.educ age) twostep
est store hckm
suest reg1 hckm
test [reg1]gender=[hckm]gender
=> When doing getting to the -suest- command, I get the error message "hckm estimated with a nonstandard vce (conventional)".

Any idea what I could do instead?

Thanks in advance!

Best,

Manon