Cheap housing is on cheaper land

Click for larger, interactive map

UPDATE Aug 29 2021:  If anyone familiar with Chicago doubts that removing improvements from the tax base will ease the burden on low-income homeowners, this map will be instructive.   The original, mapless post from Aug 25 follows.

We sometimes are told that a land value tax (LVT) would punish the poor person who has a small rundown house on a high-value lot, while benefiting the person next door who has a large fancy house on an identical lot. And that’s not wrong, it’s just atypical.  In practice, we believe, poor people mostly live in neighborhoods where housing is cheap and land is cheaper, thus they would tend to benefit from a shift to LVT.

As the quality of Cook County assessments has been improving, we expect to be able to show this by analysis of that data. In the meantime, we have some estimates of land and improvement value from William Larson and colleagues at the Federal Housing Finance Agency.  Using appraisals produced for mortgage underwriting, they estimate land and improvement values for homes in most zip codes (and census tracts) nationwide.  Their source data includes only single family properties which were appraised for mortgage purposes.  They consider only parcels where the improvement is less than 15 years old, and exclude vacant land as well as land where the appraised value is very close to the assessed value (in case appraisers might have relied on low-quality or obsolete assessments).  Also excluded are zip codes with an insufficient number of single family home transactions.

The chart below shows, for Chicago zip codes, the ratio of land value to total value (vertical axis) and total value of the property (horizontal axis. What stands out is that the ratio tends to be lower where the properties are cheaper.  That is, a revenue-neutral shift of property taxation to land values only, ignoring value of improvements, would tend to reduce the taxes on low-value zip codes, while increasing it in higher-value areas.

 

The table below shows the data for each zip code, sorted from high proportion of value in land to low.   Clearly the more affluent areas have lower proportion of improvement value, and the areas with low income population have a higher proportion of improvement value.

Estimated land value proportion and related data for single family properties in Chicago zip codes
ZIP Code Lot Size building sq ft floor area ratio Property Value (As-is) Land Share of Property Value
60640 4300 2320 0.540 $809,500 50.70%
60614 2920 2950 1.010 $1,516,300 47.10%
60613 3740 2590 0.693 $1,128,200 46.90%
60625 4220 1860 0.441 $528,000 46.30%
60618 3580 1890 0.528 $615,800 45.90%
60660 3820 1940 0.508 $553,500 44.50%
60657 3210 2600 0.810 $1,133,300 44.40%
60622 3120 2330 0.747 $911,800 41.80%
60631 5620 1690 0.301 $390,700 40.90%
60610 2250 2900 1.289 $1,443,300 40.20%
60646 5370 1820 0.339 $445,900 39.80%
60647 3250 1880 0.578 $600,800 39.20%
60654 1990 3580 1.799 $1,661,200 37.40%
60630 4310 1550 0.360 $315,500 37.20%
60642 2230 2140 0.960 $667,200 37.20%
60641 4230 1640 0.388 $316,400 35.70%
60605 1840 2200 1.196 $774,700 35.30%
60659 4310 1800 0.418 $367,500 35.10%
60626 5410 2050 0.379 $407,400 34.20%
60656 5100 1500 0.294 $334,600 34.20%
60612 2910 1920 0.660 $424,300 33.90%
60645 4420 1870 0.423 $373,600 33.70%
60634 4240 1470 0.347 $264,100 32.40%
60608 2960 1530 0.517 $302,200 31.70%
60615 4460 2630 0.590 $615,700 31.50%
60616 2570 1950 0.759 $447,900 31.40%
60607 1970 2140 1.086 $641,400 28.90%
60655 5140 1490 0.290 $258,000 28.80%
60707 4940 1600 0.324 $258,700 28.70%
60637 3920 1970 0.503 $350,000 26.30%
60609 3320 1450 0.437 $209,700 25.20%
60639 3900 1420 0.364 $202,900 25.20%
60638 4280 1340 0.313 $225,100 24.70%
60632 3860 1300 0.337 $173,600 20.60%
60643 5960 1600 0.268 $200,800 20.60%
60652 4730 1330 0.281 $176,400 18.30%
60629 4040 1340 0.332 $155,300 17.40%
60651 3920 1440 0.367 $158,300 16.90%
60653 3290 2360 0.717 $390,100 16.60%
60644 4580 1650 0.360 $137,400 16.00%
60633 4690 1320 0.281 $120,200 15.20%
60649 4960 1870 0.377 $170,800 13.80%
60623 3510 1340 0.382 $118,500 13.60%
60624 3300 1540 0.467 $122,200 12.80%
60621 3880 1490 0.384 $69,300 11.50%
60617 4200 1360 0.324 $116,300 11.20%
60619 4320 1430 0.331 $127,700 10.50%
60636 3540 1270 0.359 $65,200 10.30%
60620 4180 1410 0.337 $117,800 10.20%
60628 4320 1340 0.310 $99,800 9.20%
source: https://www.fhfa.gov/PolicyProgramsResearch/Research/Pages/wp1901.aspx

A Chicago zip code map is here.

Also of interest, even tho the low-value areas have a high ratio of improvement to land value, this isn’t because of large houses on small lots.  The floor area ratio is generally lower in the areas with lower land value proportion.

Overall, the above data is consistent with Georgists’ assertion that low-income residents usually benefit from a switch to LVT.  I might be taking a further look at this dataset.

Why trust corrupt governments to honestly administer a land value tax?

bar chart of what folks say they're afraid of
source: Chapman University Survey of American Fears

I don’t know that governments are always and inevitably corrupt, but there sure seems to be a lot of corruption going on.  It isn’t a new development; maybe it’s worse nowadays or maybe just more visible.

So how can we single taxers say that we want the government to collect all, or nearly all, of the economic rent? Don’t we know that it will be stolen or, at best, wasted?

Not necessarily.  Consider the following:

In the U S at least, real estate tax is administered and collected at the local — that is, substate– level. This is where the records and expertise needed to operate a land value tax exist.

Unlike income tax or sales tax, nearly all the data involved in real estate taxation is public information.   Most of this data is accessible to everyone with internet access, generally without fee. I can see how much real estate tax my neighbor paid.  I cannot see how much income tax they paid. The same goes for sales taxes and most other kinds of taxes. So cheating in real estate tax can be seen.  That doesn’t mean it will always be impossible for people to cheat, but it provides a much greater possibility that cheating will be observed and rectified.

Government corruption seems to be a function of government size.  A survey earlier this year found that “87% of voters nationwide believe corruption is widespread in the federal government. Solid majorities believe there is also corruption in state (70%) and local (57%) government.”  Looked at the other way round, only 13% of us believe the federal government is possibly honest, compared to 30% for states and 43% for localities.  I actually believe that one of the local governments to whom I pay taxes is pretty honest and efficient.

State and federal governments might logically collect some of the economic rent.  Examples currently include severance taxes and could reasonably include rents for electromagnetic spectrum should our rulers become persuaded to levy and collect them. Existing federal agencies are able to review and evaluate collection efforts.

 

What LVT is and why it is such a good thing

A “not so concise explanation of Land Value Tax and some brief responses to some of its most common objections” at Jock’s Place

A good balance between failing to adequately describe it and going beyond the readers’ attention span.  British, but comprehensible.

Progress in Pennsylvania, maybe Connecticut

The March ’08 issue (pdf) of the Center for the Study of Economics’ Incentive Taxation newsletter has a couple of positive notes.

Washington, PA, a split-rate city for decades, solved a budget deficit by raising the tax on land to 82.63 mills, while taxing buildings at a rate of only 3.5 mills. Thus the added tax burden goes mainly to people who are leaving land idle, and doesn’t discourage productive construction or investment.

And in New London, CT., the Re-New London Council recommends a land tax because of its benefits to, among other things, housing affordability. Had such a policy been in place over past decades, perhaps the Kelo case would never have happened.

NAHB report endorses Land Value Tax

A new report prepared by a consultant for the National Association of Homebuilders reviews dozens of strategies which have been proposed or used to promote affordable housing.  It points out that an increased tax rate on land values, balanced with decreased taxation of improvements, reduces real estate taxes for most homeowners, while encouraging owners of vacant or underused land to get their land developed, often increasing the supply of housing.

The report also notes that it costs virtually nothing to tax land at a higher rate than improvements.  Examples cited include Harrisburg and Allentown, PA.  Information is from Josh Vincent of the Center for the Study of Economics.

In  Illinois, the Cook County Board could pursue a similar strategy using existing authority to tax land and improvements as two different classes of real estate.  As previously discussed here, the Assessor could take a big step in this direction by just valuing vacant land as prescribed by existing laws and ordinances.