There is a sort of interesting argument offered about Mankiw's column: to wit, that the inheritance tax does not, on net, reduce economic output, because parents who work hard now in order to pass wealth onto their children, simply enable their children to be idle.
This sort of depends on things like the level of their taxation, the size of the inheritance, and whether or not the heirs want to pass wealth onto their children bequest. To the extent that the worrywarts are right, and the upper-middle class is cementing the elite status of its children less by passing on money than by passing on social capital like a peer network that values workaholism, this may only be true at a very narrow margin. Of course, in that case, the same is probably true of Mankiw's tax problem.
But say it is true--I've certainly seen a lot of lives ruined by trust funds. Are the rest of us necessarily worse off?
Stipulate, for the nonce, that in some metaphysical sense, the heirs themselves may be worse off--but that this is none of our business. If Greg Mankiw wants to ruin his kids' lives with a trust fund, and his kids want to accept it, we, as a nation, have better things to worry about.
Let's just focus on the public aspect: should we skew the tax code so that rich people work less hard, and their kids work harder?
Not necessarily. Assume that, on average, the parents are extremely talented at whatever they do to make money. I will entertain arguments to the contrary, but they had better be good arguments, not some variant on an extended whine that management are just a bunch of butt-kissers and no-talent cutthroat hacks. If there's one thing business school gave me, it's a healthy respect for all the different talents I don't have; I'm quite sure that I could not run IBM, or even a smaller company. In my experience, the kind of people who spend a lot of time complaining that management are all idiots tend to make very poor managers; they may (or may not) have excellent technical skills, but the reason they think managers are usually idiots, seems to be precisely because they aren't very good at identifying management challenges. (This effect is extremely pronounced among skilled professionals who tend to work autonomously--professors, journalists, some sorts of programmers and engineers.)
Now, of course, some people get wealthy merely by being awful human beings. But this percentage is not really that high. Being an awful human being is, all by itself, not very valuable, which is why serial killers are not usually found lounging in the pool of their Beverly Hills mansion. Indeed, in most professions, being a really awful human being is an actual hindrance to advancement. We notice the terrible people who succeed precisely because they stand out against a background of non-terrible people.
The people who make money are also lucky in all sorts of ways--they were in the right place at the right time, they usually started with quite a bit of social capital, and so forth. But a lot of that luck consists of endowments that are actually valuable, like skills and contacts. Whether or not they deserved these early endowments, deploying them now creates value for the rest of us.
Statistically, the children of these talented parents are likely to be less talented than they are. The children of outliers tend to display regression to the mean: unusually tall parents produce kids who are still tall, but shorter than their parents; unusually smart parents produce kids that are still smart, but not as smart as their parents; musical prodigies produce children who aren't quite as gifted, and so forth.
That means that for all sorts of wealthy people, the marginal value of their children's output is likely to be lower than their own. Society as a whole could very well be better off getting more output from today's entrepreneurs, even if it means that their children are idle tomorrow. Would you rather have another ten years of output Steve Jobs, or thirty years worth of the output of his offspring?
This does not mean that we shouldn't tax inheritances--that question involves a lot of different values, and I myself vacillate between opposing the estate tax, and believing that it should be 100%. But I don't think that this particular argument is very convincing--at least when we're talking about parents who are actually earning money, rather than simply conserving inherited capital for the next generation.