True but rarely mentioned: wealth taxes and discount rates

…the value of wealth taxes depends sensitively on the interest rate…If the interest rate is 2%, then the tax rate is “only” 1/0.02 = 50%. If the interest rate is 5%, then the tax rate is 1/0.05 = 20%. I suspect these taxes were put in place in a time of higher interest ares and nobody is really thinking about the effect of lower rates.

That is from John Cochrane, with other points of interest about tax incidence at the link.

By the way, non-inflation-indexed capital gains are in part a wealth tax, so current higher interest rates are lowering the burden of that tax to some degree.


Somebody needs to explain this to Trump.

He's busy.

"President Donald Trump's daughter, Ivanka Trump, who works as a White House adviser, is one of the names being considered as a replacement for The World Bank's outgoing president, Jim Yong Kim, the Financial Times reported Friday."

Probably not real, but funny, after Tyler's "have confidence" essay. Literally anything could happen. Peter Navarro?

Explain what? The Cochrane quote is a non-sequitor. Wealth taxes tax wealth, not income. And a 2% wealth tax will be 2% regardless interest rates.

Income is a completely different concept.

John Cochrane makes what I believe is an important error.

He is right that low interest rates increase the present value of the future tax liabilities.

But he omits to note that low interest rates *also* increase the present value of future housing services generated by the house.

This is why house prices go up with low interest rates, rather that down with low interest rates.

I believe this is a very important point.

Hm? The point I see is that market price adjusts with interest rates to match the PV of future housing services, but property tax burden does not. So if property tax is intended to tax an amenity value of owning a property, it also includes an element for up-front future amenity value that is overstated when interest rates are low - or at least, it's not neutral to the interest rate, when maybe it should be. Right?

The point is true. Imagine you buy a mortgaged house. When your mortgage is due in 10-20 years, the principal you'll need to refinance will be deflated to ridiculously low levels, but only if inflation has eaten away at the original principal. If inflation has been low you will pay back a lot in real terms. If you rent during all those years, you will still have to afford market rent at the end of the period. Whether that is a good deal compared to the mortgage depends on the real value of the principal to be refinanced.

True. But also, in high-inflation world, I think a mortgage borrower is going to pay more interest to compensate the finance provider.

Not necessarily. In the '60s inflation was low, people got very good rates in the 3-4% range. Then came the '70s, and when it was time to refinance, the principal was a pittance. The same might happen today, if your mortgage is fixed-rate.

Prop 13 ties in, in California specifically. Total future property tax liability is front-loaded and diverges from future value. A property owned 30 years may only have a 0.4% tax rate.

Not that this hodgepodge is well designed or perfect, but long term ownership is rewarded.

It would be better from an efficiency standpoint if the weighted average of tax was just applied to everyone.

's inability to correct thisI don't know about "efficiency." Prop 13 was designed to prevent people from being unable to afford to live in their house. Prudent purchases were followed by frequent reassessments and higher tax rates. This happened in the 70s and (following the Legislature's inability to see this) created Prop 13.

Building on my last point, here's a thought experiment.

Imagine that one day interest rates hit a permanent level of 0% and that property tax is a mere $1 per year. On this day, does everyone start living in the streets because houses have infinite future tax liabilities? No, of course not. People will still buy homes, even if there's a fixed yearly payment forever, as long as the services provided by that home are more valuable than the fixed yearly payment. Infinite future housing services > infinite future tax liabilities.

Still, it's a thought-provoking post by John (as usual), and I'm glad he posted it and started this conversation. I just want to emphasize some nuances that I feel were underplayed.

You know, I think I misread John and was perhaps uncharitable. I believe he understands this point. He is quite right that the burden of the tax is sensitive to the ratio of the tax to the interest rate, which makes it especially sensitive to interest rates when interest rates are low. Perhaps my thought experiment is wrong, because a house will not generate housing services in excess of its depreciation in a world with zero interest rates. I am not sure and invite anyone to clarify.

Actually, I take that back. Even if risk-free interest rates are zero, that doesn't imply housing returns zero, especially if you are the owner-resident and pay no taxes on imputed housing income. This issue seems complicated.

Capital gains are a change in wealth, not wealth itself. With all that physics envy, you'd think economists would take a cue from physics and be able to distinguish between basic concepts, just like velocity is distinguished from position and is a change in position, not position itself.

You would expect Anon to tell us what "physics envy" means here in a human system where humans create wealth, rather than just parroting memes the academic equivalent of unemployed woke Twitter.

'You would expect Anon to tell us what "physics envy"'

Well, a classic example predates twitter - there actually is a Nobel Prize in Physics, but it took the Swedish Riksbank to create something that is called the Sveriges Riksbank Prize in Economic Sciences in Memory of Alfred Nobel.

And notice the 'Sciences' part for an academic field that is actually a part of the humanities, and whose origin is in moral and political philosophy.

Anybody who has actually worked with various academics in an American university setting in the last generation is very aware of how economists see themselves, and yes, envy of a real science that is actually capable of describing the world we live in is generally not that difficult to discern (call that an American attempt at understatement).

Which just might explain why an economist who actually was quite successful describing the world we currently live in, without using mathematics, is so easily ignored. But then, putting Veblen aside just might be the equivalent of putting Kropotkin to the side when talking about Darwinism,

Testable hypothesis frameworks are really good and useful to non-believers. This is what heterodox economists / anthropologists / sociologists tend not to get when they construct their ideological theories.

Saying that the the money you would need to pay the tax from interest is, in the example give, £1million is true. And saying that the amount put aside depends on interest paid on that amount is also true. Saying that the impact of the tax depends on how much value of house in increasing is also true.

Saying that this means a £1million house actually costs £2million (1% tax and 1% interest rate) seems quite the stretch, no?

Who pays their tax by setting aside a lump sum and using the interest to pay the tax? No one? A few very rich people? And why is this any different from any other tax? If I pay £20,000 a year in income tax, I could pay it with 1% interest on £2million. Do we say that's like a £2million tax? I suppose we could, but it just highlights that a £10,000 a year property tax costs the same as any £10,000 a year and quoting it in lump sum needed to pay off of interest is bizarre.

2. If you did put money aside to pay the tax you would still have the £1million, so the cost to you is simply the interest paid which you've used to pay the tax on your house. (Plus interest you would have earned, but that applies to any money you pay in tax).

"Who pays their tax by setting aside a lump sum"

Try being outside PAYE. See how your behaviour might change.

"and using the interest to pay the tax?"

If interest rates are low...

Legislatures that raise taxes on wealth are like individuals who inherit wealth and then spend it all on depreciating assets such as cars, yachts and also lots of holidays. Once it has gone it has gone.
Maybe a solution (assuming the politics of envy remain supreme) is to somehow earmark tax income from wealth for infrastructure projects such as roads and buildings that in themselves will generate income for the common good.

"Maybe a solution (assuming the politics of envy remain supreme) is to somehow earmark tax income from wealth for infrastructure projects such as roads and buildings that in themselves will generate income for the common good".

I.e. More money for the rich. No! We paid for the scientific research thad made the Brazilain Midwest the breadbasket of Asia (soybeans, wheat and meat). We paid for the infrastructure. Now, rich "terratenientes" are saying the money are all theirs, they did everything alone... and we need to give more money anyway. No! It is my money, my taxes. I want it back. I paid for it.

"cars, yachts"

This is MR, where we believe the guy buying a base model Honda Civic, so he can get to work, should the same sales tax rate as a guy buying a Lamborghini or a hundred foot motor yacht.

Hear hear!

Luxury sales taxes are very efficient. They are almost impossible to dodge given that they are highly transportable goods and the demand is inelastic since there are few to zero substitutes on luxury spending in general. It's not like you can buy a Yacht in another country and then move it. They're not like vehicles or anything crazy.

This 'anonymous' guy econs.

You pay sales tax on most boats at the DMV, when you register them, like cars. You might have to write off a few percent super rich who manage to maintain foreign registration, but it gets complicated.

This Guy was being sarcastic.

Part 2 covers "Geographic or Time Limitations on Use" of foreign vessels.

By the way, the metaphysical argument is about whether consumption is always good, the more the better?

Are we the world better off with more 10 miles per gallon supercars? Because jobs?GDP? Fun?

Obviously at one end most consumption has an environmental cost, but at the other it feeds and clothes people.

That does not seem to be the metaphysical argument at all.

The percentage of people on this board who would support a consumption tax in lieu of an income tax is probably anywhere from 3-5x what you would find in the wild.

Environmental quality is a consumption good like anything else and has proven to steadily increase as a society becomes wealthier.

The US, Japan, and Finland consume more than India and Bangladesh.

Which set do you think has a cleaner environment?

This shit isn't rocket science.

I believe the proper economic view is that you just price all externalities, and then your consumption is good.

But pricing isn't really sufficient is it?

You can put any environmental tariff you want on a bluefin tuna, but when a guy will still pay 3 million dollars for a fish .. at what point does that bring the bluefin back?

We need two things, both far from popular acceptance now. You need the pricing, and you need to use the price to pay for remediation.

A "simple" but unavailable solution.

This is gibberish.

What CSU abortion of a school did you attend?

Pfft, in which he doesn't recognize negative externalities.

Fullerton? Long Beach?

I will say I’m impressed with your ability to rememeber the word externality. You don’t seem to understand what it is, but that’s okay. For a CSU grad it’s a miracle you can spell with spellcheck.


This is MR, so most of us, even if we could afford a lambo, would be driving a honda. :)

The 2014 purchase price of the house was $550K. I live in Cypress, California. I pay about $7300 in taxes, not $5,500 which would be 1% of the house purchase price. Why? Cypress is a nice city, but living in a nice city is expensive. The city of Cypress issues bonds to cover its expenses: water; PD; FD; rather good school district. That is, I pay $1800 on top of the property tax to live in a nice city.

Correct. Prop 13 taxes are 1% of purchase price + bonded indebtedness

In an era of high interest rates, tax deferral is the same as tax avoidance. I know because I was a tax lawyer in the era. Most of our work at the time was devising strategies for tax deferral, which effectively made tax practice one devoted to the time value of money. Boring yes, but the risk of a long vacation at government expense was much lower. Today, deferral has a much lower value, so the goal has become tax avoidance (evasion being the unlawful version of avoidance). The distinction between avoidance and evasion has become less clear, which creates incentives for techniques that are on the evasion side of the ledger. Couple those incentives with a general hostility toward both government and taxes, and what results are schemes with little or no substance to deflect income to tax havens, shifting the tax burden to those who cannot take advantage of the schemes and generating enormous government debt for Cowen's preferred future generations over the present generation.

Yesterday I viewed a podcast of current tax strategies. I thought our strategies for tax deferral were complex, but today's strategies for tax avoidance are beyond complex. It takes more complexity to avoid taxes than to defer them. It also takes a higher level of risk tolerance.

Get the rates low enough and they'll just pay the taxes and forgo spending the money on accountants and lawyers. Talk about eliminating the rent seekers!

What? And put all the young rayward's out of work? The horror of it all.

He says:

Let's set up an account that will pay your property taxes. If you get 1% interest on that account, you need to put $1,000,000 in the account!

But isn't it actually much worse than that? The money you set aside to pay the tax doesn't just have to pay the 1% tax, it also has to grow enough to match inflation (the 1% tax payment will be higher next year) and to cover the annual income tax on the interest.

The post is set in the context of California property taxes. Those do not rise with inflation; why would the account that pays the tax need to grow enough to match inflation?

Under Prop 13, the assessed value on which you are taxed can have an "inflationary increase" of up to 2% every year.

1%(.01) of 2%(.02) = .0004(.4%) pretty negligible

If your tax is $10K this year, it will be $10,200 next year assuming 2% inflation. Not a huge increase, but the money you invested to pay the tax will also have to grow by 2% annually to cover it (and that's over and above the money taken out to pay the property tax and the income taxes on interest).

The house has an annual convenience yield of $Y(t) and every year you pay $X(t) in taxes. Then the purchase price of the house could just be $Y(t) - $X(t) as a discounted cash value. The imposition and/or increase of taxes can cause that value to go down. A 100% tax on the stated value of the house (eg a 3% annual tax in a world of 3% interest rates) does not mean it is a 100% tax on wealth because of the yield effect; the yield of the house pre-tax does not need to be the neutral rate. In essence you might have a bond yielding 4% but having to pay a 1% extra tax in a world where neutral rates are 3%. That buyer does not suffer the incidence of tax at the time they buy if taxes follow the expected path and the market is efficient.

Changes in tax do affect the wealth of the buyer, but some buyers may get something for their money (taxes collectively increase convenience yield) and some changes in tax can be priced in (effectively, X(t) is not necessarily a martingale).

Whether X() is determined based on net benefit of services to the buyer, or something more like what the market will bear, might make this in the end more like an income or a wealth tax (eg high property tax in a city with many undercapitalized people with high earned income is probably more like an income tax; high property tax in a resort area with many overcapitalized people with low earned income probably more like a wealth tax; neither is obvious to me).

I did not realize property tax rates in California were so low at only 1%. Here in flyover country, I pay almost 3%.

I am assuming that this is because housing prices here are much lower. But this basically suggests that, on a nationwide basis, property taxes are highly regressive as all areas raise a similar amount of money so areas with cheap houses have to charge much higher rates.

Perhaps if property taxes were somehow equalized, there would be less resentment at coastal elites who enjoy effortless growth in their wealth through faster housing appreciation while paying 1/3rd the tax rate, and the massive gap in home values would also be less than it is currently.

"I am assuming that this is because housing prices here are much lower. "

Yes, and by the same token, the 1% property tax rate limit in California (which was established by Prop 13) is part of the reason California housing prices have risen so much in the decades since the initiative was passed. A 3% mortgage on $1M in CA works out to $40,000 in taxes and interest. In a region with 3% property taxes, a mortgage on a $650K house costs about the same per year as that $1M house in CA.

The initial tax may be higher, as Catalan notes above, but the big Prop 13 difference is that it looks lower over time. Presumably in 20 years Catalan will think his taxes "cheap."

New buyers and old owners are (often?encouraged to be?) the same people in different stages of life.

Obviously this reduces costs for the elderly, but reduces economic efficiency and population mobility.

"The initial tax may be higher, as Catalan notes above, but the big Prop 13 difference is that it looks lower over time. "

Yes. The property tax is 1% of taxable value, but that taxable value cannot rise faster than inflation, so there are folks living in $2M houses who are paying just a few thousand in property taxes (much less than 1%), for example here is a house with a $2M asking price where the current property taxes are just $3700 a year.

"Obviously this reduces costs for the elderly, but reduces economic efficiency and population mobility."

Yes, and it's even worse than you might think given that the low property valuations and taxes don't reset when the house is passed to the next generation (and not even if the property is used as a rental rather than a primary residence!) Seems quite insane (but I guess we have to remember that we are talking about California).

I would be in favor of some law or court ruling that required uniform taxation. No special deals for Tesla factories or Amazon headquarters either.

But pretty much everyone wants to cut a deal, from Elon and Jeff, down to your neighbors.

Your brilliant plan is to force proposition 13 on the rest of the country through judicial fiat?

Moderate Republican indeed good sir.

Carry on then.

I'm pretty sure I just said the opposite.

Seriously? You're defending this (because, hey man, everybody has an angle)? California shouldn't change these inequitable laws unless and until there's a national law requiring uniform taxation (which, of course, would never happen and is probably unconstitutional anyway)? Sheesh.

I am not defending, I am just uniformly cynical.

(uniform-progressive that is)

In San Francisco county, the tax rate is about 1.2% of your assessed value, which is the price you paid for the property, subject to up to 2% increases every year.

In a different, non-coastal state in which I own property, the tax rate is about 3.7% of assessed value. However, the assessed value is about 1/3 of land and improvement value, which is less than what I paid for the property. So in the end, my effective rate based on market value of both properties has been about the same.

(I am talking broad strokes here, since SF caps the rate of increase of assessed value, while the other county does not; the rate of increase in market values are different, with my non-SF property actually appreciating faster than my SF property in recent years; and SF is much more prone to random special assessments from time to time.)

The real return on investments is about 5% and Pickety says it has been for Centuries. For GDP calculation the rental value of owner occupied homes is estimated to be 5%. Using the interest rate instead of return on capital does not seem to be right.

Could it be possible that Piketty is wrong? Is this based on records for two countries again, or Jane Eyre?

"The real return on investments is about 5% and Pickety says it has been for Centuries. "

This is kind of a dumb comment. I'm sure Pickety is referring to average real returns. There are plenty of people with significant investments in Treasures. They aren't making a 5% real return.

"non-inflation-indexed capital gains are in part a wealth tax, so current higher interest rates are lowering the burden of that tax to some degree."

I can see why non-inflation-indexed capital gains are a wealth tax:

1. Have $100,000.

2. Experience 10% inflation, so you now have $110,000.

3. Pay 10% capital gains tax on your unadjusted $10,000 of capital gains, leaving you with $109,000. After accounting for inflation, your $100,000 has dwindled to $99,091. You had zero actual capital gains, so the tax is purely a wealth tax.

But I don't see how high interest rates lower the burden of the tax. Suppose inflation was 30% instead of 10%. Then the example would look like this:

1. Have $100,000.

2. After 30% inflation, you have $130,000.

3. After paying 10% "capital gains" tax on your unadjusted gain of $30,000, you have $127,000. Adjusted for inflation, you have $97,692 -- which is much much worse than the counterfactual where interest rates were lower.

Michael Watts: your examples suggest that interest/return on capital is exactly equal to inflation. Under that condition you are correct that inflation and capital gains taxation is a pure wealth tax that is increasing in the rate of inflation. Add in a real earnings component so that interest/return % is higher than inflation, and the analysis is more complicated.

I made that assumption because those two quantities are tightly coupled; each exerts a causal influence pulling the other one towards itself. Thus, if the Fed raises interest rates as a policy action, I assume that that just results in long-term inflation.

Higher real returns on investment would lower the burden of a wealth tax, just as they lower every other kind of burden, but you can't just declare higher real returns by fiat.

Michael, you are misreading it. Tyler didn't say higher inflation he said higher interest rates. Furthermore, your examples conflated inflation and interest rate return.

Better example:

Assume 10% inflation & 10% capital gains in both cases:
a) Interest rates = 10% & inflation =10%, then per your math, real returns are negative;
Net = $99,091

b) Interest rates = 30% & inflation =10%, then
Nominal result = $130K nominal gross
Minus 10% taxes = $127K nominal net

Real Net = $115,455

Cochrane is usually more insightful. A wealth tax is a wealth tax, not an income tax on the income generated by the wealth. It doesn't matter if the rate of the wealth tax represents 2%, 10%, 50%, or 500% of the income generated by the wealth (or -150% for that matter, if bad investments have made the income of the wealth be negative for a year).

I'll keep reading, but right our of the gate, I'm pretty sure a property tax isn't part of the 70% (or whatever) marginal rate John mentions. Guh.

Isn't Treasury income exempt from state (and local?) income taxes?

Cochrane writes:

If you buy a house that costs 5 times your income -- say someone earning $200,000 per year buying a $1 million house -- then that is equivalent to 5 percentage points additional income tax. On top of 42% federal, 13.2% state, 9% sales, and other taxes, it's part of my view that we're past 70% top marginal rate now.

This is a clear error. He is taking the property tax as a percentage of income without including imputed rent in income. Of course he gets a high rate.

In expensive markets annual rents can be as much as 3% or more of home rental. So a 1% property tax equates to a 33% marginal rate on the imputed rent.

Of course there is an issue of whether the NPV approach is a sensible way to look at this matter. Suppose you are in the 25% marginal tax bracket and get a $10,000 raise. Interest rates are at 4%. Do you complain bitterly because you've been hit with an unexpected bill for $62,500?

"This is a clear error. He is taking the property tax as a percentage of income without including imputed rent in income."

It's hard to imagine that matters. In real life, out of my $200K income, I have to pay $10K in taxes or 5% of my income. If you assume the cost of the house raises linearly with income, then it's effectively raising my marginal tax rate by 5%.

House prices should be related to rent based on something like

P ( 1 + rf ) - R = P ( 1 + d ) - t P


P ( rf - d + t) = R

where P is price, rf is risk free rate, d is the appreciation on houses under RNM, t is property tax, and R is rent less upkeep costs.

In most cities, R - tP ~= 0.08 P (look up the cap ratio for your city.) That means that under the risk neutral measure, house prices go DOWN in expectation. Given rf ~= 3%, we need d ~= -5%.

The right way to think about property tax is a tax on phantom rent (i.e. the rent you would pay if you were renting the house). The tax is t P and rent is R. So the tax as % of rent is

t P / R = t / ( rf - d + t) = 1 / ( 1 + ( rf - d) / t )

If rf = 1% and t = 1%, we get a tax as 14% of rent.

If rf = 4% and t = 1%, we get a tax as 11% of rent.

If rf = 0% and t = 1%, we get a tax as 17% of rent.

(In the Bay Area, the cap rates are much lower, say 4%, reflecting higher expected growth, which gives d = -1%. That gives us 50%, 20%, and 100%, respectively.)

Comments for this post are closed