[w4]

> 1. Allow unrealized losses to be deducted.

This seemed really reasonable to me until I started thinking about how it might work in practice. The sequence of returns can make this proposal ineffective in practice, even if it makes sense on first blush.

By way of explanation: Let's say you're the founder of Pets.com in an alternate universe where unrealized gains have always been taxed (and correspondingly unrealized losses can be deducted). It's 2000, and you've just had an incredible run. You have also paid incredible taxes along the way.

Then your company blows up and goes to zero.

Now you've payed an incredible amount of taxes on your paper gains, and have realized no gains whatsoever. So the entire enterprise only resulted in an enormous real loss to you. Sure you can now carry forward those losses, but so what? You're never going to make up the difference, unless we're also letting your heirs carry forward those losses into the next century or two.

Given the exposure to massive tax bills without any actual profits, who in their right mind would start or invest in a business under that tax code? Who would dare invest a large portion of their personal worth in public equities given the risk that they plummet, as they did in 2022, 2020, 2008, 2001, 2000, 1987, 1962, 1929, 1907, etc.? Who would take a gamble on a big real estate development? And so on.

It seems to me that a tax on unrealized gains massively disincentivizes investment and the creation of anything new, and therefore the only way to tax capital gains that makes sense is if we calculate the tax due based on when chips are taken off the table. Issues like Buy, Borrow, Die are better addressed with other changes to the tax code that undo the weird incentivizes presently in place (e.g. eliminating the step-up basis, possibly at some threshold if the goal is to make the tax code more progressive). Unless, that is, your goal is to actively disincentivize entrepreneurship and investment. Which if it is, I guess fair enough, but then none of us should be surprised to find ourselves with a lower standard of living in a decade as a result.

[Rury]

I think what you do is simply tax stock ownership. Say you own 100 shares of stock. A 2% tax would mean the government would confiscate 2 of your shares, so you then own 98 shares. The government then proceeds to sell their confiscated shares on the open market (not at once, but spread out over the next year) and use the proceeds as tax revenue. You as an investor can maintain your 100 shares of stock by simply buying back the shares on the open market (or not and so pay a smaller capital gains tax than you otherwise would when you sell. This is your cost basis being adjusted). This also doesn't necessarily disincentive investing, as the tax proceeds can be used for funding jobs (ie investment), and owning stocks can still be worthwhile.

That said, I don't see why there's a need for a deduction here. There isn't one for property taxes. Sure there is one when you sell your property at a loss, and that's also already the case when selling stock. Additionally such a tax like this won't ever cause you to lose your entire stock ownership as it's always based on a fraction of your ownership. And, last but not least, you could also impose caps or progressions.

[jandrewrogers]

This tacitly assumes a pretty naive model of how these markets work. The dynamics of poor liquidity, dead equity, stock restrictions, intangible asset loss, etc materially change the outcomes you can expect. In many cases it may cost the government more than the revenue generated, and the counter-party as well. This doesn’t work like your retirement account. Similar types of scenarios historically created by civil litigation have had many adverse consequences to business in practice.

And property taxes are deductible in the US. I’m not sure where you got the impression they aren’t.

[Rury]

>This tacitly assumes a pretty naive model of how these markets work. The dynamics of poor liquidity, dead equity, stock restrictions, intangible asset loss, etc materially change the outcomes you can expect. In many cases it may cost the government more than the revenue generated, and the counter-party as well

In all seriousness, that is extremely unlikely to be the case, especially in broad terms. I mean, to purchase public stock at all you already must do so from a licensed broker, who mind you, is already required by law to report the cost basis of shares purchased by investors. To require them to regularly move 2% of shares owned by investors, to the government's ownership, would be rather trivial in cost to do. Hell, it could be completely automated.

And the cost of it doing that would likely be much less than property taxes, which, is a far less liquid asset, and much more costly to assess than equities, but is nonetheless profitable to tax. I mean the SP500 alone has a market cap of ~$47 trillion, which is, surprising almost the same exact value as the entire US real estate market, but much more liquid.

Additionally, whether it is even profitable at all could be besides the intent of the tax here. It doesn't necessarily have to be profitable, from which perspective, poor liquidity and changes in outcomes isn't a problematic at all. It could be even the intent.

> And property taxes are deductible in the US. I’m not sure where you got the impression they aren’t.

Okay fair. I suppose you could do something similar, but also don't see why it's necessary, just because we do so for other taxes.

[kortilla]

The vast majority of wealth is not traded on a liquid market like the S&P. Your scheme only works in nice divisible liquid assets.

Otherwise, when do you expect the government to actually realize its gains? If I have a house I intend to live in until I die and the govt take 2% of my stake away each year, when would it be allowed to force a sale to realize it?

Do I just lose my house once it has a majority stake? Does it just wait until I die? If it does, how is this better than an inheritance tax that is significantly easier to implement?