Emailed in by MBK, an article by Tim Worstall in Forbes:
However, we also know that different taxes have different deadweight costs. Sometimes the deadweight cost is actually the reason to impose the tax. This is partially true of Pigou Taxes (like a carbon tax, the aim and point is to reduce carbon emissions) and others like cigarette taxes can sometimes be so classified.
We even know a rough order of ranking of those deadweight costs.
Transactions taxes have vast deadweight costs: at least one official report on the financial transactions tax (the report from the EU itself) tells us that that tax would make the whole economy so much smaller that it would actually lose tax revenue by being levied. Lower than that are wealth taxes (a once off wealth tax that no one knows is coming does not, but a regular one does), then capital and corporation taxes, then income taxes, then with still lower deadweight costs consumption taxes (like a VAT or sales tax)* and then finally, at the bottom, we find that repeated taxes on non-movable property can even have positive deadweights. And that last can also be called a land value tax.
Please do note again that this is nothing to do with how much we tax but only with how we tax. Further, all of these assumptions have been tested emiprically and make up the heart of optimal taxation theory. We really do know that if we raise $100 in a transactions tax then we’ll lose more economic activity than if we raise $100 in a land value tax.
* He consistently ranks these incorrectly. VAT and sales taxes are transaction taxes, and hence go to the very top of the naughty list. But hey.
Debate Magazine
