4 ms·
> unimproved land to zero Yep! > dynamic supply and demand effects Yep! > ...then not if housing units aren't the most profitable use of land in the local a
by ItsMonkk 5y ago
> unimproved land to zero
Yep!
> dynamic supply and demand effects
Yep!
> ...then not if housing units aren't the most profitable use of land in the local area it won't.
Agree. But they almost always are.
> Confiscatory
Housing can either be an investment or it can be affordable long-term, it can not be both. Therefore the Land Value Tax is the only correct rate. And a UBI for at least a decent portion of it nullifies any potential regressive nature of it.
> don't reduce tax rates for the average person
It does, over time, as housing supply improves and the entire economy gets out of the mud. The wealthy own the most expensive land.
> catastrophically drive down land value
The tax is based on the rental value of the land, that is, the flow. You could see it as the neutralization of the capture of the positive externalities of your neighbors.
As in your first example where an empty lot will settle at $0 market price, that would not mean it has a $0 tax. It would still have whatever tax it would have. You determine that tax based on market rates, and it is a set cost based on that particular property - it's not some percentage of the market rates as we do today.
I agree that in the short-term, with most people having most of their networth in their home, this would be a bad thing. We can transition to this process slowly, and only increase as the incentives improve the economy and allow us to improve. The best way to start is to have a net zero change where we remove taxes on improvements and instead to tax the land.
Making better use of the city land would in fact drive the overall land flow value up hugely in value. SF has much more room to improve than NYC, NYC makes pretty good use of it's land. But so can CT and NJ, and the rise in supply of non-NYC land outside of NYC will drive more value into NYC.
- jjav 5y ago> Housing can either be an investment or it can be affordable long-term, it can not be both. While this is one of the most often cut&pasted comments on HN, I haven't seen a thorough discussion of what it actually means in practice. Any references with substantial analysis of proposals to achieve this and what the unintended consequences could be? You seem to advocating for an immense increase in property taxes. This line from the previous comment: > The money available to collect from NYC and SF land taxes alone could fund a nation-wide UBI. suggests that the increase in taxes would be quite enourmous indeed. That would certainly make housing a bad investment. Nobody would want to own any if the recurring property taxes are that high. But of course it doesn't make property affordable either, if the tax rates are sky high.
- ItsMonkk 5y agoPennsylvania has a few cities that have implemented a partial LVT. Here's[0] a Strong Towns article on the topic. They show exactly what I mentioned earlier - overall taxes are up, most peoples taxes are down. Business is booming. It's not Silicon Valley, but they are doing better than the cities around them that do not have the LVT. The best introductions are Lars excellent book and blog articles[1] over on Scott Alexander's blog. > Nobody would want to own any if the recurring property taxes are that high. But of course it doesn't make property affordable either, if the tax rates are sky high. The tax rates are set by the market, so they are by definition affordable. You are taxed on the land alone, what capital you put on top of it is entirely yours to earn. [0]: https://www.strongtowns.org/journal/2019/3/6/non-glamorous-gains-the-pennsylvania-land-tax-experiment https://www.strongtowns.org/journal/2019/3/6/non-glamorous-g... [1]: https://astralcodexten.substack.com/p/does-georgism-work-is-land-really https://astralcodexten.substack.com/p/does-georgism-work-is-...