We do know that people are very bad at recollecting past experience and at forecasting future experience. Rather than a tax on unhappiness, however, I would favor a subsidy on hedonimeters (see Kahneman and Rees, p. 292). People could in principle overcome these affective biases if they could precisely measure and record how happy they are.

I like where you are going with this, but these data are a bad place to start for trying to figure out what makes us happy and what doesn’t. Work, for example, provided it is matched to our ability, can supposedly make us very happy, and fulfilled. Long commutes and big houses make us slightly LESS happy than short commutes and small houses – hence in this instance the consumption tax would be efficient for maximum happiness.

This might be a good place to start:

http://www.happinesshypothesis.com/

Millian, in the signaling model, the $5M of wedding services is entirely wasted, in that as far as we know, its only purpose is signaling the couple’s vanity. In the post-tax scenario, the couple enjoys the same vanity benefit (after all, it’s still signaling $5M in disposable income), but part of the resources are transfered to the government rather than being wasted.

In other words, the purpose of taxing conspicuous consumption is not just to discourage activities we dislike, but to capture a socially-created rent that would otherwise be dissipated.

Where exactly is the reduction of positional harm? It is a consumption tax. Nothing more.

Yvian, yes there’d be a lot more work to do to establish than an activity contributed to happiness overall, all things considered. Contrary to you, I don’t at all have the impression that Frank or Miller are up to that task.

First, about unhappy activities: we usually do them to gain something else further down the line. I hate homework as much as any student, but I do it because I think I’ll be happier later if I learn more. If I’m right and you tax homework in a way that successfully discourages me from doing it, you’re hurting my future happiness. Likewise, when we don’t do something that makes us happy, we often have a good reason for it. Most people don’t spend their whole lives partying (my next-door neighbors excepted) because they know it would impact their health. You’d need a very different list from this one if you wanted to fully analyze the eudaimonic consequences of an activity. Some people might say that in such a list, we would always find ourselves doing activities in proportion to the utility they gave us (we’d never deliberately choose a low utility action) but I don’t think that’s quite true (see paragraph 4).

Second problem – assuming our goal is a government policy to increase happiness and that we think subsidization works – we should subsidize activities that make people happy if and only if subsidizing them would make people do more of them, and doing more of them would make people happier. We should tax unhappy activities if and only if taxing them would make people do less of them, and doing less of them would make people happier. There may be a law of diminishing returns: I enjoy the parties I go to, but going to more parties wouldn’t make me even happier. By your logic, you could say “Yvain likes pizza, so we should make him eat pizza every night to maximize his happiness.”

Now I know you’re holding up this position as a reductio ad absurdum and don’t really believe it, but the problem is that it does make sense in a few specific cases. If there’s some activity that people are doing too much of for irrational reasons, even though they don’t like it and it brings them no benefit, and taxing it would decrease it and make people happier, then taxing it is a good policy. Likewise, if there’s some activity people are doing too little of for some irrational reason, even though they like it and doing more would bring them great benefit, then subsidizing it would make a good policy.

We have a lot of evidence that things in this class do exist; we know from happiness research that people are terrible at predicting what would make them happy and often avoid things that statistically are very likely to make them happier. This suggests that, if you have this philosophy that one should use government as a tool to increase happiness, taxing and subsidizing activities in this class is a good idea. But you can’t simply assume that every possible activity that increases or decreases happiness is in this class, just cause you found it on a Big List Of Activities With Happiness Numbers Next To Them.

And even in that case, you’d need to demonstrate or at least reasonably believe that the benefit you got from shifting people to better activities was greater than the cost in penalizing people with strange tastes (as nazgul pointed out).

Straw man argument, needs more subtle thinking. From the few quotes I’ve seen from Frank and Miller, I get the impression they had that subtle thinking.

Is not the point of this consumption to demonstrate your wealth? It seems by making such displays even more costly we’re right back where we started with only the super-rich having access to cars and barbecues and an even larger number of people envying their neighbor’s wealth.

Wouldn’t the logical conclusion be subsidizing these displays of consumption to increase access instead of taxing to limit it?

It’s obvious that doing anything efficiently should buy more happiness than doing that thing inefficiently.