“Tax burden” is a standard term in public finance that refers to the loss of consumption imposed on someone that results from the COLLECTION of a tax. It is, as a matter of definition, a question of “sources” of government income, and is completely unrelated to the effects of the “uses” of that income. For the reasons I laid out in my original comment, there is simply no tax burden in the unusual state of affairs you posited.

YOUR point is simply that if the government spends the tax revenues it collects from Kendrick in a pattern that’s different from the way they’re being spent now (which is whatever way that Kendrick’s capital is being invested), the pattern of consumption will change. That’s a perfectly fine observation to make, but it ignores the very important fact that the entire POINT of taxing and spending by the government is to alter the pattern of consumption. In the standard case, the goal would be to shift production away from “private” consumption goods toward “public” goods. The welfare effect of such a shift depends on the extent to which the government spends its tax revenues efficiently, i.e., on goods for which the private sector faces higher costs of provision (including organizational costs) than does the public sector.

Given your past writings on the taxation of capital gains, I think the point you intended to make in your example was that the burden of a general tax on capital is shifted onto other factors of production when capital is in perfectly elastic long-run supply. That is a completely valid logical proposition. But a tax on a single individual–which is the way you set up the Kendrick example–will absolutely be borne by that individual, because changes in his supply of resources to the market will not be large enough to affect any general input prices at all. The weird thing about the case you posited is that targeting Kendrick doesn’t even hurt him, because his marginal utility of income is zero, and so there’s no tax burden at all.

There is a long gray line between “real” and “financial” assets.

No there’s not. Real assets are inputs to the production function.

Here’s an example you may be able to understand, which exactly replicates my previous discussion in simple terms: Suppose Kendrick had kept his wealth as a gigantic stash of canned food in a warehouse. He never consumed any of it, and he never planned to consume any of it. Now the government taxes him by taking away some or all of his canned food and giving it to people who will eat it. Do you really think that there’s a “burden” to this tax?

If you start a syllogism with a faulty premise, you’ll reach a faulty conclusion. The faulty premise here is that there is “always” a burden of taxation, when the assumptions of the case at hand make it clear that there is none.

All sorts of other premises may be stipulated for this example, such as that Kendrick actually lent his wealth to other people and kept re-lending the interest he received on that wealth, which would be like letting people take his canned food in exchange for IOUs. Then a tax on Kendrick would cause a redistribution of consumption if the tax proceeds were spent in some way different from the status quo. This seems to be what Landsburg had in mind. There will be a reallocation of consumption due to the government’s spending pattern, but that is simply not what economists generally mean when they discuss the burden of a tax.

I think the idea is that if such resources exist, the government can simply “employ” them. The only reason to take Kendrick’s money first – and thereby call it “spending” — is if there is a resource owner who requires compensation (and some underlying moral or political calculus suggests this is a more palatable way to consummate the acquisition.) But that’s just saying there is in fact a cost, and we’re back to determining who really pays it at the end of the day. And that’s where the parallel to printing money is instructive.

Yes. Exactly.

Why should the source of the wealth matter? Clearly in this case someone prior to Kendrick continued to earn well beyond the point where their “budget constraint was binding” because they left a lot of it to him. And why assume all of the wealth generated by a company over time is calibrated to someone’s initial budget constraint? Perhaps the venture turned out to be more successful than they had expected. And sometimes people just have a passion for creating value.

“none of this implies that people other than Kendrick are bearing the “burden” of the tax”

I don’t think there’s a “paradox” here, just much confusion over what the point of the initial post was (and wasn’t). Landsburg stated as clearly as humanly possible that taxes generally impose a burden on someone (unless perhaps there are truly idle resources). So if Kendrick gets zero marginal utility from his Pareto-optimal Lagrange thingy, that implies that the “someone” is other people (as Landsburg described, via an increase in prices and/or interest rates if consumption increases with no increase in the supply of real productive resources (as was assumed)).

“it is like the site value of land in midtown Manhattan. It can be taxed with no resulting change in the allocation of resources”

I’m not sure what this post has to do with the price of real estate in Manhattan. It is true that no matter how we tax it, a plot of land is unable to exhibit changes in its consumption pattern and it will continue to exist where it presently is. But surely taxing it more affects its value to potential owners (e.g. its “best and highest use” could change, which could be a form of reallocation). To the extent this can be translated to human beings, the fact that I can’t avoid a tax doesn’t mean I’m not impacted by it. I pay my taxes because I don’t want to go to jail (among other reasons), but surely my consumption suffers.

I think the response would be that, as with steel or medicines etc., Kendrick is not hoarding any productive jobs either. If we can find some idle real resources we might be able to give some people something “real” to do. Otherwise we can fund make-work projects or extend unemployment benefits. Any of these things could be similarly accomplished by printing new money rather than taking Kendrick’s (without risk to inflation in the Keynesian scenario in either case).

vic – “idle resources that can be employed at no social cost will in fact be so employed only if the likes of Mr. Kendrick are taxed”

I think the idea is that if such resources exist, the government can simply “employ” them. The only reason to take Kendrick’s money first – and thereby call it “spending” — is if there is a resource owner who requires compensation (and some underlying moral or political calculus suggests this is a more palatable way to consummate the acquisition.) But that’s just saying there is in fact a cost, and we’re back to determining who really pays it at the end of the day. And that’s where the parallel to printing money is instructive.