In my first post "What is a Supply-Side Liberal?" I wrote
I believe the harm to the productive performance of the economy caused by taxes and regulations is serious (though seldom serious enough that a reduction in taxes would raise revenue).
Mark Thoma’s post (title above is a link) shows that the weight of informed opinion is behind my parenthetical remark. I can verify that since I started econ grad school in 1982, having attended many economics seminars and having had many informal discussions with economists, I have never in person heard an academic economist argue that tax cuts raise revenue, with the possible exception of Larry Lindsey (Greg Mankiw’s and my boss when we were both section-leaders in Harvard’s Ec 10). Larry Lindsey argued that cuts in capital gains tax rates would cause investors to change the timing of capital gains realizations enough that cutting capital gains tax rates would raise revenue now—implicitly at the expense of revenue later, though he didn’t emphasize that.
One reason that tax cuts don’t raise revenues is that the effect of taxes on GDP is itself complex, and can go either way. See my posts "Can Taxes Raise GDP?" and "Why Taxes are Bad." If marginal tax rates can be cut both now and in the future it raises efficiency (a good thing) but it will typically make people feel richer as well, so that work hours won’t go up much, if at all (also a good thing).
Update: Scott Sumner pointed out to me that the disagreement of economists with the statement
A cut in federal income tax rates in the US right now would raise taxable income enough so that the annual total tax revenue would be higher within five years than without the tax cut.
indicates that the overwhelming majority of economists don’t believe that tax cuts can raise revenue for Keynesian reasons either. (And actually, disagreeing with this statement means not believing that the combination of supply-side and demand-side effects of tax cuts is enough to lead to an increase in tax revenue.) On the question of whether tax cuts can raise revenue for Keynesian reasons, see Valerie Ramey’s Powerpoint discussion of a recent paper by Brad Delong and Larry Summers.