Other remedial measures for multicollinearity?

I have log transform my model already, choose per capita variables but multicollinearity is still a problem.
I plan to drop the variable which has the highest vif (over 200)... however, I read somewhere that I will be subjected to specification error/bias.
I need to address this one since almost all of my variables were insignificant.

Are there any remedial measure that I could do to address this?

p.s. I also use first difference but all my variables shows that their all insignificant...


No cake for spunky
How do you know Multicolinearity is a problem? It is almost never a good idea to get rid of a variable that makes theoretical sense because of mulitcolinearity. John Fox wrote a sage monograph called Regression Diagnostics you may want to look at. One way to deal with it if you think it is a major issue is to combine two variables into one. But generally Multicolinearity is not a huge issue.

Note that high collinearity between variables is not Multicolinearity. Multicolinearity reflects issues in multivariate analysis and you can never tell if you have it by bivariate correlations. You need to run VIF or Tolerance or the like.