5 Aug 2009

Optimal Taxes -- Property Tax

In this post [originally published 15 Sep 2006], I suggest that property taxes replace all other taxes. Property taxes can be more efficient and more equitable than the alternatives.

How can government raise taxes most efficiently? Economists have written libraries on optimal taxation. They conclude that the best tax does not distort behavior or fall heavily on a particular group. Taxing cigarettes would distort behavior -- increasing cigarette smuggling or driving people to other habits; taxing people with blue eyes would be unfair -- they would protest.


Income taxes distort behavior: The more you earn, the larger the tax-bite on your last dollar earned. More importantly, people need to report their income, which is a pain for the law-abiding and an opportunity for evasion or avoidance for those unwilling to be "suckers".

The higher your potential tax bill, the larger the reward for evasion, which means that corporations and rich people pay accountants a lot of money to avoid taxes. Self-employed and informally employed people (eg, drug dealers) evade taxes altogether. (Some Swedish econmists estimate that households with one self-employed member underreport their taxes by around 30 percent.)

Consumption taxes are better on incentives ("consumption -- vs. saving -- is bad, so tax it"), but the reporting burden falls more heavily on fewer people. It's easy for businesses to collect a retail sales tax from customers, but hard to make sure the businesses send in the proper total. The Value Added Tax (VAT) increases honesty by comparing the cost of inputs that retailers report to the sales of outputs wholesalers report. Unfortunately, the VAT increases complexity.

Property taxes are relatively easy to administer and hard to avoid. The property register of owners and assessed values identifies who pays and how much to pay. Better yet, property taxes do not distort behavior: they do not vary with one's actions, job, habits, purchases, etc.

Property tax discriminates against recent buyers, since it's based on the property's assessed value (AV, the purchase price plus some inflation adjustments), which rises more slowly than the market value. (I am assuming that market values are rising.) Imagine two identical houses, side-by-side. The owner who bought her house in 1986 will have a lower AV (and thus property tax bill) than a new owner who bought the house next door in 2006, even though both houses have the same market value. This "distortion" is the result of an argument of fairness -- that people should not have their tax bills jump because of events beyond their control. (Changing the tax rate is an easy way to increase tax revenue with fixed AVs. That happens for income and sales taxes, but is less common for property. The claim is that fixed-income pensioners would "suffer" from increases, but all long-term homeowners benefit. New owners and those who pay other taxes are harmed, of course.)

In 2006, according to the Tax Foundation, Americans work for 116 days/year to pay local, state and federal taxes. 70% of the total comes from income tax, 20% from consumption taxes and 10% from property tax. What about social security and corporate taxes? Social security taxes are another tax on income. Corporate taxes are actually paid by people -- not corporations -- either through reduced income (as workers or shareholders) or higher prices (as consumers). This brings up the important notion of incidence, i.e., who pays a tax. First point: only people pay taxes, not corporations. That means that the one-half of social security that your employer pays for you is actually coming out of your pocket (through lower wages) or customers' pockets (through higher prices). The Congressional Budget Office estimates that 70 percent of corporate taxes are "paid" by workers and 30 percent by shareholders. Someone always pays -- either directly (e.g., sales tax) or indirectly (e.g., corporate tax).

So who pays property taxes? Property owners, of course. Does that mean renters pay no property taxes? No. They pay property taxes indirectly through their rental payments to investor/landlords. The bottom line is that only individuals pay taxes, and some pay them indirectly -- through higher prices or lower wages.


Relative to property taxes, income and consumption taxes create more distortions, more paperwork and single out those foolish enough to honestly report income and or expenditures. This latter point -- reporting -- is a small one in the Developed world but imagine trying to set up a tax system in a Developing country. How would it work? First, you need a widespread, honest and competent bureaucracy; second, you need citizens willing to declare income or expenditures (i.e., merchants reporting consumer expenditures as sales); third, they must have the ability to do the accounting. These features are hard to find in the Developing World.

Property taxes, on the other hand, are easy to understand, need only be collected once or twice a year, and -- as I mentioned above -- do not distort behavior.


In my optimal system there is good news and bad news. The good news is that you need not file a 1040 ever again, and you can forget about 7.25% or 8.5% surcharges on your shopping trip to the mall. The bad news is that property taxes would go up and fluctuate.

In California (where I live), Proposition 13 has limited property taxes to about 1.3% of a property's basis, its assessed value or purchase price, which is allowed to rise by 2% each year -- regardless of the property's actual, market value. Limits like Prop 13 would have to go if the government was going to collect all of its revenue through property tax. If government expenses vary, taxes will as well. If basis does not change, the tax rate will have to.


If we set income and sales tax to zero and wanted to raise the same amount of tax revenue for the government, property taxes would go to 10-times their current level. (Average per capita taxes in the US: $10,200, of which $1,100 from property tax. For California, the numbers are $11,200 and $963. Tax Foundation figures for 2004.)

Before panicking remember that the switch should be revenue neutral, ie, total government taxes will not change. Also remember that removing distortions and broadening the base (so everyone pays property tax, directly or indirectly) can reduce the actual burden for current tax payers.

For example: Mr. A, B, and C are citizens. A and B pay $12/each in income taxes; C evades his taxes. All three pay $1 each in property tax ($27 total). If we used property tax only, they pay $9/each; A and B are better off ($9 is less than $13) because C has to pay his "fair share".


More interesting is the calculation of basis. The current system allows basis to rise by a small percentage each year. If all taxes were based on property, there is an incentive to minimize your basis. In the US, property owners often declare the real value (ie, sales price) because they must get loans to buy the property; a lower declared value would reduce the loan and require the buyer to bring more cash to the sale. In a Developing country, however, cash sales make it easier to reduce your declared basis.

I suggest therefore that the basis for property taxes be the minimum of either declared value or 90 percent of the average of declared values of 20-100 local properties (controlling for size, improvements, etc.) An owner declaring a property value of $250k when average value is $200k has a tax basis of $180k. Since there's no reason to under-report one's own value, everyone will tend to be honest (to get bigger loans, among other things) -- pushing the "declared" average closer to the true average.

Let's explore this further: A, B and C have property.
Basis...$250k..$350k..$360k.<-- 90="" average="" br="" of="">
A sells for $850k to D. The new figures are:


Total basis moves from $960k (250+350+360) to $1,430k (540+350+540) so tax revenues go up. D is happy because she pays less than her value. B is happy because his basis stays the same. C is less happy because his basis has gone up, but it's still less than his value. With a higher basis and the same tax rate, total taxes will rise; if the total is held constant, then the tax rate will drop. (The outcome depends on political budget factors I do not cover here.)

Some people will worry about the impact on people with fixed incomes, etc. Although their basis may not change, their tax burden may, if rates are free to fluctuate. If total basis is rising faster than government spending, however, the rate on a property with the same basis will fall.



1) This system would work in Afghanistan as well as NYC. It's also very progressive: Rich people in expensive/multiple houses would pay more than apartment dwellers. To reduce distortions, no property would be exempt. That means government, non-profit, church, etc. (Big lobbying problem there.)

2) If a property has a basis that's "too low", either because of fraud or long ownership, the government might audit properties with values less than, say, 40 percent of local value. The owner could be allowed to declare a higher value or the property could be put up for auction, with the winning bidder buying it from the owner (and readjusting the basis). The owner could buy it for herself, of course, but the price is likely to be far higher than if she declared a value that was, say, 90 percent of market price.

3) There is some good reason to tax activities that produce externalities (eg, a carbon tax or sin tax) to reflect the "true costs" of that activity. That's anther discussion.

Sept 15 Addendum: Today, I read a paper by Fred Foldvary that proposes a "Geo-Rent" on land only. My proposal differs by taxing land and improvements. I prefer my version because it is more progressive (valuable houses are build on valuable land -- magnifying the difference between rich and poor areas, easier to understand (assessed value is easier than the "land value", which may be arguable), and passes rent onto renters more easily (since they occupy the improvement. In Foldvary's scheme without taxes on these, the land-owner without an improvement pays the same tax as one with land.) I do envy the sprawl-reducing effect of his tax, but it does not make up for these other deficiencies.

27 Feb 2018 (!) Addendum: I link to this post alot, but I have not updated it. First, the addendum above is referring to Henry George's idea of taxing land value alone. Such an idea would give owners an incentive to improve property (to get income off of it in excess of the land tax), but I prefer the "tax wealth" goal to George's "encourage use" goal. Second, I think that property tax is morally justified as a return on citizenship, i.e., "it's our country, so we should benefit from the value of its property." Such a claim does weaken property rights by assuming that ownership does not free one from the property tax, but I think it's reasonable to assume that property value is based on a nation's institutions rather than terra nullis. Finally, I'd return these taxes to all citizens as a "basic income right" appurtenant to citizenship.


Anonymous said...

Big problem: property value does not imply ability to pay. Income and transaction taxes do. I couldn't pay your proposed property tax, nor could many who own property but have little income. You would reverse the long trend of increasing home-ownership in America.

Texas Ann said...

Property tax and sales tax are the two methods the state of Texas uses to raise revenue. Appraisals are based on market prices to some extent, since appraisers compare sales figures for similar properties in similar neighborhoods. Ag land is appraised based on a five-year production average, which can get sticky if a couple of bad years follow a boom year. Mineral and oil property values are based on market prices.

Although some always complain about their appraisals, I believe property tax is probably a good system. The state also offers generous exemptions (homestead exemption, elderly and disabled, disabled veterans) which can be stacked.

Texas may someday have an income tax, but it won't be our lifetimes.

R. said...

Albionwood is right, but beyond the academic abstraction, the reality this idea could impose upon my life is quite harsh. I learned in my economics class that our state (New Mexico) is income-poor and property-rich. Even at my present rate of taxation I'm very concerned about of being able to hold onto my home because although I own my home, I have always lived on a fragile shoestring.

I don't' know how other New Mexicans got into this predicament, but I arrived at this difficulty because I would never have been able to afford my own home unless I built it myself—buying it one brick at a time.

Rural New Mexico simply does not offer the same income opportunities as the big city and sweat equity can create modest wealth in places where economic opportunities are lean.

However after investing all my resources into my home, even with the homestead exemption, my “creation” is taxed at a rate that is quite difficult to meet. I’m not nearly as hypersensitive to new tax ideas as other Americans, but please do be careful. I would hate for the fruit of my labor, the culmination of my dreams and my retirement security to be lost because I couldn't afford the taxes on it.

MS said...

Great to know you are a fellow quasi-Georgist. I sometimes describe myself as a "Georgist libertarian," basically due to arguments like yours. I'm not sure between you and Foldvary (and other Georgists who share Foldvary's position), but if we are down to debating these issues, we are pretty close to a decent solution - in theory.

In terms of strategies for implementation, I am increasingly interested in the quasi-anarcho free zone approach as a means to make this highly sensible Georgist libertarianism real. Hong Kong and Singapore are, of course, highly Georgist and highly libertarian.

For more, see the "Women's Empowerment Free Zone" paper,... Read More [PDF]

as well as "Free Zones as an Additional Option for the Cambrian Explosion of Government," at "Let a Thousand Nations Bloom" for more

jwetmore said...

As MS said Henry George advocated taxes based on the value of the undeveloped land as the best form of taxation.

I have wanted to understand his arguments for a long time but have never taken the time to read his work in the original.

That said, the post is thought provoking and could open an interesting line of discussion. There are numerous points made that I agree with as well as those I disagree with. Responding to the many points would take an extended comment. Today I will adress a few of the points in the second paragraph.

Efficient collection of taxess is good.

In the case of negative externalities we want to provide a mechanism to influence behavior, but as Johnn Stuart Mill argued we should not attempt to practice social engineering through tax policy.

We also know that when you tax things you generally get less of the thing you tax. While the supply of land may be more or less fixed (think of the Netherlands) the amount of property that is owned is not fixed. I understand that in some western states, Wyoming for example, the federal government owns approximately half of the land. Taxing land could have the effect of privately owned land reverting to the local, state or federal governments. Undesireable in my view because of the Tragedy of the Commons.

Those are my comment on just the second paragraph, so there is ample content to generate much interesting discussion. Thanks David for (re)opening such a interesting thread.

David Zetland said...

@abiowood and R -- First, property calues would change if the tax base changed (more taxes make them *less* valuable). Second, taxes have to come from somewhere (god knows where they are SPENT), and the relationship between lifetime income/expenditure and property values is strong. Thus, I am merely proposing to switch taxes to the most-efficient base (measurement, collection, etc.)

Of course nobody likes to pay more taxes, but also note that base doesn't move as long as you do not...

@jwetmore -- I'd agree that the states should tax federal land (as well as church land, etc.

@all -- I forgot to mention (or emphasize) that this system is BEST for developing countries...

Spencer said...

I studied municipal budgeting and taxation issues a good bit in college, and while there’s an argument for property taxation, there are definite problems as well. Relying solely on it (or any other one source) for revenue would magnify its particular distorting properties and weaknesses. In revenue budgeting, as investing, governments (and arguably citizens) are most secure through diversification.

High property taxation would distort a very fundamental consumer behavior (i.e. where we live) with wide-ranging impacts, both major and minor:
-It would drive higher apartment living as well as ruralization as people seek out cheap property. This would strongly affect gasoline consumption, business location and myriad aspects of family and civic life.
-Quality of housing will decline because of a strong disincentive to improve or maintain one’s property and thus increase its taxable value.
-Existing property values would decline, causing current owners to sustain major losses and making tax assessments extremely volatile on the short term.
-The effects of job loss, being a student, retirement or any other impact on income would be magnified and force people to move or become delinquent because their tax burden wouldn’t change with ability to pay. The only way to avoid the tax is to move to a different property. That’s a lot harder than sacrificing discretionary spending.
-Abandoned property would rise, with rapidly growing tax liens further inhibiting its sale.
-Currently, land use and taxation is primarily locally controlled. Federal and state governments would inevitably become very involved and interested in setting assessed property values (a very subjective determination, unlike income or sales), tax rates and redistributing revenue among governments. Plus, imagine the political fights over trying to constantly adjust rates or the number of assessments that would be challenged.

The "tax revolt" that resulted in Prop 13 and many other tax & expenditure limitations across the country is a valuable study in behavioral economics and a cautionary tale, because people get very upset about being forced to move. Place is practically and emotionally important, and it’s extremely difficult, slow and disruptive to change it. Under your plan, the US would look a lot more like Japan, where people (esp. families) crowd together in small living units, and housing doesn’t appreciate and is frequently rebuilt.

What’s better: a lot of small distortions than can be adjusted here and there and help balance each other, or a few very huge distortions with the potential to destabilize society when taxes don’t readily correspond to ability to pay?

Anonymous said...

DZ, your response to my comment and R's seems to have missed our points. In fact it resembles the kind of abstract thinking that characterizes some of the worst products of economists... For example, "the relationship between lifetime income/expenditure and property values is strong" - does not answer my objection. Unless you propose to collect the taxes only at the end of one's life, when the lifetime income would be a meaningful statistic... Otherwise, year-on-year income fluctuations would make property ownership tenuous. How would you deal with farmers, for example, whose income fluctuates drastically year-on-year, while property values remain relatively stable?

"Efficiency" is a shibboleth.

David Zetland said...

@albionwood -- I understand exactly your point, which is "I don't want to pay taxes and/or I can't afford them now." My point is that we *plan* for taxes and a switch to a different base would not cause trouble if it's gradual, so that LIFETIME income and property values align.

David Zetland said...

@Spencer -- I agree about the problem of distortions being magnified. I disagree on most of your "marginal" effects -- mostly b/c you need a cost-benefit to put them into context against the current system.

In addition, your "distortions" depend on narrow-neoclassical behavior that we just don't see, e.g., "Quality of housing will decline because of a strong disincentive to improve or maintain one’s property and thus increase its taxable value." may be true at a very tiny margin, but it's not like people will stop caring about paint, their gardens, etc. In fact, I'd say that the incentive to stop working after 40 hrs is much stronger than the incentive to stop painting your siding...

(I disagree with your other points, but I don't want to type all day :)

jwetmore said...

@David -- Yes each level of government should be able to tax the land owned/controlled by the other levels within its jurisdiction. Some people think this is like taking money from one pocket and putting it in another, but I believe it reduces distortions and makes cost more transparent. My background is in transportation and since transportation is generally funded by user fees and property taxes I have always thought that transit buses and government vehicles should pay fuel taxes just like private users. Failure to do so hides costs and provides incentives for the government sector to perform services that might be more efficiently performed by the private sector.

Bottom Line: The former Soviet Union collapsed because their system could not properly recognize and account for costs, and because they could not differentiate between cost and value.

Back to the main thread later.

Anonymous said...

DZ - No, that is/was NOT my point! Argh. I'll try again. I have no more objection to paying taxes than any other taxpayer (I'm no teabagger). I do object to shifting the entire burden onto a single class, in this case landowners. And my objection is rooted in the disconnect between property value and free cash flow.

I do not dispute your assertion of a strong correlation between lifetime earnings and owned property value - I assert that this statistic is not only meaningless but dangerously misleading. (For one thing, you're looking at averages, but the data are not normally distributed...) Assuming property taxes must be paid on some recurring basis (mine are due twice per year), the owner must have some source of cash with which to pay them. If the taxes are not a large burden, income fluctuations can be accommodated with savings. If the tax becomes large in proportion to income, and income temporarily drops, a crisis is created with large risk of property loss. Since almost everyone experiences income fluctuations, the effects of shifting all taxes onto the property-owning class would include increased property turnover. The ramifications of that are, I think, largely negative.

David Zetland said...

@jaywetmore -- great point. That's why church and gov't land would be taxed :)

@albionn -- sorry for misunderstanding you. 1) I agree on the problem of "one base," and perhaps we only agree that property can be taxed more. 2) Given the fixed costs of a collection system, I'd want to dump income tax first and then expenditure tax. 3) Property renters would pay tax as part of their rent 4) Fluctuations in income would cause problems, yes, but that's what financial instruments are for. A cautious person (like me) would avoid such a hazard by living somewhere cheaper :)

jwetmore said...

My knowledge of taxation theory is limited but perhaps other readers of this blog can fill in some of the missing knowledge.

I know of two approaches to apportioning taxes.
1. According to benefits received.
2. According to ability to pay.

These are not nessesarily at odds as individuals with more income, property, or other wealth may derive greater benefits from government actions such as the national defense.

I think several other itmes should be considered, but I have never seen a coherent discussion of these points. Thye include:
1. The efficiency of collections (Fuel taxes are incredibly efficient to collect. In Minnesota, for example, they are collected at the wholesale level and I have read there are two wholesalers in the state. Vehicle Milage Taxes that are being promoted do not share this attribute of conventional fuel taxes.)
2. Fairness and transparency considerations. (The more transparent the better.)
3. The behavior distorting impact of the tax. (As David argues some taxes distort eeconomic behavior less than others, and they should be favored.)
4. Opportunities for Rent Seeking. (I think this is an under appreciated aspect. I also think that property taxes would compare poorly with other taxes, though I can think of few taxes that would rank high under this test. One could imagine that significant lobbying would occur to identify different classes of land, some supposedly more beneficail to society than others, whose owners would lobby for relief. Conservation land is an example. Conservation land requires fewer governmental services than commercial land and provides sustainability benefits to society, the owners would argue and therefor deserves advantageous rates.)
5. Incentives for tax evasion. (The higher the rates, generally the higher the incentives and stakes. Low tax rates reduce incentives for evasion. Any single source of tax revenue would have higher tax rates than if there are multiple sources. A clear tradeoff between low cost of collection and low incentives for evasion.)
6. Provide meaningful participation and sense of belonging for all citizens. (Much of our current tax policy seems based on taxing other people for our own benefit.)

@David -- Church land yes (not taxing it would seem to be a violation of the Establishment Clause), and don't forget about educational land. Think how much tax revenue Cambridge, Massachusetts looses on just its two best known universities.

Post a Comment

Note: only a member of this blog may post a comment.