Anyway, you're right.  We're supposed to be talking about raw numbers -- not per GDP.  I'm getting myself a bit mixed up.  Laffer Curve proponents essentially argue that revenues (in raw dollars) will be higher at the lower tax rate, due to stronger GDP growth, than they would've been at the previous, higher rate.

The CBO did a study of a hypothetical 10% tax cut, financed with deficits, with players broken down into three categories -- No Foresight, Lifetime Foresight and Unlimited Foresight.  Bonddad has summarized it, here.

Long story short: We'll see some change in behavior, largely among top earners (as expected), but the change is not large enough to make up for the lost revenue.  The government still ends up losing about 75%, even when the results are very favorable to the Supply-Siders' views.

According to models that account for both supply-side and demand-side effects, those effects might offset somewhat less than 15 percent of the revenue loss over the first five years.

So the CBO model is predicting that the change in behavior will offset a bit less than 15% of the loss.  That's probably a fair estimate, depend, of course, on how high the rates were to begin with.

We can all judge for ourselves whether we believe that is worth the cost, or what tax rates we would all support.  (I think I'm probably a little more in favor of lower taxes than most people on here, since I don't think income tax rates should ever exceed the 40-45% range.)  But I think the data suggests, in all or most cases (the CBO study is, obviously, just one example), that the government would've been better off with the higher rates -- again, depending on how high they were to begin with.

Be nice to America. Or we'll bring democracy to your country.