However the tax rates are much higher areas using the 6% method versus the 100% method. For example in NYC the tax rate is 19.157% while in San Francisco is 1.1743%. If you do the math the tax rate on market value is about the same. 6% * 19.157% = 1.149% in NYC versus 1.1743% in SF. In most places it is around 1% but can be over 2% in place like New Jersey, Illinois or upstate New York which unlike NYC relies more on property taxes.

None of this says anything about the NYC system for condos (class 2) that the article talks about but is often a source of confusion when comparing areas that use different systems.

http://www1.nyc.gov/assets/finance/downloads/pdf/brochures/c...

http://www.socketsite.com/archives/2014/09/san-francisco-pro...

EDIT: Here's a good recent review of residential real estate taxes across the US (290kb pdf):

http://www.brookings.edu/~/media/research/files/papers/2013/...

This was very confusing to me when I lived in New York. The property tax rate is terrifying (19 percent annually???), but then you realize that the basis for each property is reduced by a fudge factor such that the effective rate is in the same range that it is in other places.

I suspect that the reason that they do this is to stop people from hassling the city constantly about their assessment. In places like MA, where they make an honest attempt to keep assessments close to 1:1 on value (mine changed annually in the time that I owned the house, often going DOWN -- this was 2007-2010), people are constantly appealing their valuations.

The only relevant distinction would be if tax or assessment were curved as a function of actual value.

If you used the last sale price and didn't adjust, people who bought 30 years ago would benefit unfairly. If you used the sale price and adjust annually, how do you do it? By inflation, by average country/state/city/neighborhood increase in sale prices, etc.? Each has problems. If you hire people to assess the current market value, they can be bribed or make mistakes.

I really like property taxes as a way to fund government, but determining the value of something like a house/condo is actually a bit tricky.

Allow the government to set the rate as they see fit (with all the normal controls of an iterative democratic/representative process), but the assessed value is also a bid of sorts.

Of course there are flaws: an assessment shouldn't vary based on how willing someone is to move, it's still easy to under-assess by a little (but then that drives the rate higher to make the budget), and I'm still a little disquieted by the idea that the control against under-assessing so undermines property rights.

I think that Cambridge (MA) gets this reasonably right. My assessments have tracked the overall market in my area (at least loosely). They might be 10-20% off in a rising market, but they're not 50% off.

The problem lies in the nature of politicians... Reassessments upset people whose assessments go up, so the big players lobby to defer the re-basing. Or in the case of California, you never revisit until the property changes hands.

In a metrics-based way (#BR, #Baths, square footage)? In an aggregate neighborhood-based way (all 02138 is up 3% this year)? Via drive-by appraisals? Via invasive inside the house appraisals or "broker price opinions"? (The last clause would seem to have significant Fourth Amendment concerns.)

That's exactly how mortgage underwriting works. It's not 100% accurate, but there is an appeals process to address inequities.

This is pretty much the answer. Everyone who argues that it's impossible to figure out a fair value for the property without a change in ownership has to face the fact that lenders do it on demand and (admittedly only theoretically at certain points in recent history) are incentivized to get it right because that's the final method available to them to get their money back if a loan goes south.

2. That process is labor-intensive and therefore expensive. Typically, the borrower is on the hook for the approximately $500-1000 cost of the appraisal. (If you don't see a line item charge, it's safe to assume that was buried in the discount rate.) On the high side of that range, appraisal costs would be > 10% of the tax collected if you appraised annually. Presumably, doing every property in the municipality at one time would make this process more efficient, but it's still a labor intensive and invasive process. The city's largest commercial landlord, Boston Properties, pays about 4% of the total tax revenue in Cambridge. This process could literally siphon more money out of the system than all Boston Properties buildings pay.

Allowing a mandated and periodic governmental inspection of my property to determine its value and funding that appraisal work present significant hurdles to universal application. In this case, I think that approximately wrong is better than precisely correct.

You use square footage of livable area and land. The rates for each (which are different) are the average sale price of all houses in the neighborhood per square foot.

The only hard part is determining neighborhoods. Sure some houses are extra fancy and worth more, but the difference is not large enough to worry about since fancy houses also tend to be larger.

If you choose a whole year, your assessments will lag badly in a fast moving market. If you choose a shorter time period, you risk the small number of sales making the measure a noisy one. I suspect the whole year basis is a better balance.

Then of course, there's the "is a garage livable area? is it land? is it neither?" "is a basement livable area?" "what makes an attic livable area vs storage?" "what about porches? patios? sheds? gazebos?"

Livable area is already very well defined in law, so no need to do it again. Every real estate listing has this number. There's even an ANSI standard Z765-2003 for it.