The HOME Plan • The full math

The Math Behind the HOME Plan

Rep. Mitchell Horner's plan to eliminate school property taxes works on paper. This page starts with his own spreadsheet, unchanged, and lets you put back the things it leaves out.

Quick definitions. A mill is $1 of tax for every $1,000 of taxable value. Georgia taxes homes on 40% of what they're worth, so on a $300,000 house one mill is about $120 a year. The “digest” is the total value of everything in the county that can be taxed.

Put his own numbers together honestly and the plan is short about $12.9 million a year — roughly the cost of 118 teaching positions.

School millage required in 2031
His model: 18.57 18.57
Corrected, on active switches
1.43 mills of room left under the 20-mill legal cap
Status
Fits, on his assumptions
Nothing has been changed yet. These are his numbers.

What would this actually cost me?

These use his own figures for his recommended rollout, Option B: an 18.5672-mill school rate once the exemption is fully phased in by 2031, scaled to a few home values so you can see roughly where you land. Not a lookup of your actual parcel: a full calculator, keyed to your own home value and move-in date, is coming. Until then, school tax only, at 18.5672 mills:

Assessed value (Georgia taxes 40% of market value, so a $350,000 home is assessed at $140,000)School tax todayHis plan, school tax, 5+ yearsHis plan, school tax, under 5 years
$90,000$1,417.50$0$1,671.05 (+$253.55)
$130,802 (county average, today)$2,060.13$0$2,428.63 (+$368.50)
$220,000$3,465.00$0$4,084.78 (+$619.78)
Assessed values are illustrative; the 18.5672-mill rate is his own, Option B, his recommended path, fully phased in by 2031. The five-year test is his own model, hardcoded in all three options. Recent buyers are typically assessed about 10% above the roll average, so a real bill for someone under five years is usually a bit higher than this table shows.

Twenty mills is a legal ceiling, not a dial.

Georgia caps a school system's maintenance and operations millage at 20 mills. Catoosa's school board sits at 15.75 today, which leaves about 4.25 mills of room to raise if something goes wrong. Under the HOME Plan that room is gone. The rate is pinned near the ceiling permanently.

So when the number above goes past 20, it does not become a higher tax rate. It becomes a cut. Everything below is drawn from Rep. Horner's own spreadsheets, his public statements, and the county and school system's own audited figures. Every item is cited.

One thing to know first. The penny sales tax on the November ballot is step one. The school homestead exemption described here is a separate vote that has not happened yet. In his words: “The school system receives no direct FLOST distribution. School relief would be presented separately to the voters.”
Show your work

Four things the model leaves out

The exemption takes homestead value off the tax rolls. The schools still need the same money. Same money, smaller pile of taxable property, so the rate has to go up on everything that is left. Georgia will not let a school system go past 20 mills. So once the rate needs to be above 20, it stops being a tax increase and starts being a budget cut.

Each switch puts a real number back into his model. Watch the readout above.

1

Give every homeowner the full exemption he promised

The plan is described as a 100% exemption. 17% of homestead value stays taxable.

To get the full exemption, a homeowner would have to clear a five-year test. Everyone else keeps paying on part of their home. His spreadsheet takes 20% of all homestead value off the exemption and books the tax collected from it as revenue that makes the plan work: $6,630,217 in 2028, rising to $9,036,306 by 2037.

“One guardrail would be a five-year residency clause. So you have to live in the county for five years.”NightTalk, Aug. 5, 2026, 47:18
“You live in the county for five years or you have to own your home for five years. Either one, it's up to negotiation with the school board.”NightTalk, Aug. 5, 2026, 47:30

The rule sweeps in anyone under five years, including longtime Catoosa residents who moved across town.

Source: Rep. Horner's model. The five-year carve-out is hardcoded as 20% of full homestead value in all three options; the tax collected from that carved-out 20% is booked as revenue at the plan's millage rate.
2

Count the homes nobody has claimed yet

His model allows for about 1,278 new filings. The county's own unclaimed-homestead list counts 3,761.

Thousands of Catoosa homeowners are eligible for a homestead exemption and have never filed, because today it saves almost nothing. Under a 100% exemption it would save their entire school tax bill, so they would file. His model does account for some of this: it steps homestead value up by four points in the first full year and labels that line “Max Usage.” That step covers roughly 1,278 new filings.

The county's chief appraiser's own unclaimed-homestead list counts 3,761 eligible, unclaimed homes. Rep. Horner's own figures on the air (84% homeownership, 75–76% of them filing) imply about 4,360.

Source: Rep. Horner's model. Full homestead uptake is modeled at 46% of the gross digest starting the first full-exemption year (42% before it), labeled in his own assumptions as “(increase due to higher usage)” and “(Max Usage).” Unclaimed count from the Catoosa County chief appraiser’s own unclaimed-homestead list, Sept. 17, 2026.
3

Use what the schools actually collect

The model says property taxes are 33% of the school budget. His own file's two numbers give 35.74%.

To decide how much the schools need from property tax, his spreadsheet starts with what they collect and subtracts the year's surplus, then locks the result at a flat 33% for every year after. The formula is =(51576257-4839544)/budget. The first number is the district's budgeted property taxes. The second is its general fund surplus.

His own file has both numbers. Budgeted property taxes of $51,576,257 against a budget of $144,309,663 is 35.74%. He gets to 33% by subtracting the year's surplus first, then locking that figure in for every year after.

Source: Rep. Horner's model, his estimated local-share figure.
4

Use what school costs have actually done

The model assumes costs rise 3.5% a year, forever.

The whole plan rests on one number typed into a formula: school costs grow 3.5% a year, every year, indefinitely. The two biggest costs the district does not control have risen far faster.

Cost the district cannot setThenNowPer year
Health insurance, per classified employee (GBPI analysis, FY2027 enacted budget)$1,953
2010
$23,220
FY2027
15.7%
  the same, just the last four years$945/mo
FY2023
$1,935/mo
FY2027
19.6%
Teacher retirement, employer rate12.28%
FY2014
22.32%
FY2027
~7.8%
in dollars

Neither rate is set in Catoosa. Both are set by the General Assembly, and the district receives a bill. Had health insurance grown at 3.5% since 2010, it would cost $3,504 per employee today instead of $23,220 (GBPI analysis, FY2027 enacted budget).

Fair caveat, stated plainly: insurance and retirement are a share of the budget, not all of it. The honest test is what Catoosa's total school budget has actually done: it has grown 5.74% a year for the last five years, and not one of those years came in under 3.5%. Leave out the two pandemic years and it's still 5.74%.
Source: Rep. Horner's model, his estimated-budget line, which grows costs 3.5% a year (typed directly into the formula, though his own assumptions tab puts the baseline at 3.4547%). Insurance and retirement rates: Georgia Budget and Policy Institute; CCPS FY2026 public budget hearing presentation.
All three options

“I offered other options”

He did. Three of them. None survives the same four corrections, and they all pass his own test for the same reason: his spreadsheet grows the tax base 3.89% a year while growing the schools' costs 3.5% a year. When the thing you divide by grows faster than the thing you divide, the required rate falls every year automatically.

The three options are not three tested designs. They are three outputs of one pair of assumptions that guarantees a passing grade. Switch the selector at the top of the page and every number on this page recalculates.

Put back…Option A
2028
Option B
2031
Option C
2037

Mills required, each option measured in the year its full exemption lands. Option C is measured in 2037, so compounding has six more years to work there than in Option A.

Option C reaches the same place a worse way. It arrives at the same 2037 endpoint as Options A and B, but only by inflating its own “Current Exemptions” row to get there. Its millage never peaks and falls back the way A's and B's do — it climbs the entire ten years, from 15.75 to 18.33 mills, with no relief along the way.
The five-year rule

Optional when he describes it. Required when you add it up.

He has talked about the five-year rule in public again and again: on NightTalk, at the July 16 intergovernmental meeting, and on his Facebook page. Every single time, it is something that could be added, not something locked in:

“We are trying to look at some guardrails. One guardrail would be a five-year residency clause.”NightTalk, Aug. 5, 2026 · 47:18
“You live in the county for five years or you have to own your home for five years. Either one, it's up to negotiation with the school board.”NightTalk · 47:30
“I think it's more than fair to come in and live five years in the county and then you have a full homestead exemption… Again, all of this is negotiable.”Intergovernmental meeting, July 16, 2026 · 25:45
“For the homestead exemptions we are proposing a resident of Catoosa County for 5 years (this is negotiable, we may not put a residency requirement on there if the school system doesn't want it.)”Reply on his public Facebook page, Sept. 21, 2026

In his Facebook posts themselves, what appears instead is the word “guardrails,” used as a reassurance: “FLOST and the homestead exemptions will finish the job and guarentee no property taxes on your home through guardrails.”

In the spreadsheet it is not optional. It is written into all three options, in every year, and it is worth more than the entire margin the plan has under the cap. He has also described what it is doing:

“And that 5% could be picked up with either a five-year residency because you're bringing in additional revenue, so you could lower your mill rate to not being at that 20 mill rate.”Intergovernmental meeting · 40:50
Both ways lose. Drop the five-year rule and every version of the plan is at or over the legal cap. Keep it and roughly a fifth of homestead value keeps paying school property taxes, in a plan sold as eliminating them. The negotiation he describes is one his own numbers cannot survive.

He has never told the public that his numbers already assume the rule is in effect. He has never told the public what happens to those numbers if the school board negotiates it away — which is the thing he keeps saying publicly they might do.

His second workbook models a full homestead exemption with no five-year carve-out at all. It finds it needs exactly 20.000 mills — the legal maximum, zero headroom — every year from 2028 on. This is his own file, distributed by him, needing no correction, showing what happens without the rule he keeps describing as optional.

A smaller inconsistency worth noting: two of his workbooks disagree with each other on the homestead-uptake assumption. His first workbook uses 46%. His second workbook uses 42%. Same author, same month, same plan.

No residency or tenure condition appears in HB 581, the county resolution, or the IGA. The five-year rule exists in his spreadsheet and his public remarks. It does not exist in the law that would put any of this into effect.

One row, ten years

He said it washes out. His spreadsheet says it grows.

Asked on the air whether people who have not lived here five years would lose out, he said the problem solves itself:

“By the time it's fully implemented… you will be able to take full advantage, because we're talking four years away.”NightTalk · 47:50

This is the line in his spreadsheet that collects money from those same homeowners. It does not wash out. It rises every year for a decade.

Source: Rep. Horner's model, his additional-revenue line under Option A. The same figures appear identically under Option B.

If costs behave normally

The year Catoosa runs out of room

The first three corrections hit all at once. Cost growth is different: it compounds. A small error in the growth rate is invisible in year one and decisive by year ten. This shows the millage the schools would need each year, against the ceiling they cannot go past.

His assumptions Current settings 20-mill legal cap

Recalculates with the switches above. Corrections are applied from the year the full exemption lands. Earlier years show his own figures, because until then the exemption is only partly in effect.

What it adds up to

Put it all back and the plan is short $12.9 million a year

There is no millage left to raise once the plan is in place. A shortfall that size leaves the district choosing between cuts or asking voters for another sales tax.

In staffing terms, that is roughly the cost of 118 teaching positions.

The other name on your ballot

You’ve seen the math. Want to get to know the other candidate for House District 3?

Meet Margaret Spear at spearforga.com →
Take this with you

Five questions worth asking him

  1. Is the five-year rule part of the plan or not? If it is dropped in negotiation, what is the millage?
  2. The county's chief appraiser's own unclaimed-homestead list counts 3,761 eligible homes that have never filed. Your model allows for about 1,278. Which number is right?
  3. Your model puts the schools' share of local funding at 33%. Your own file's two numbers give 35.74%. Why the difference?
  4. School costs are assumed to grow 3.5% a year forever. Health insurance has grown 15.7% a year since 2010. What happens if 3.5% is wrong?
  5. You have said this should be stress-tested before adoption. Who is doing that, and when will the public see it?
Where this could be wrong. Rep. Horner has not commissioned an independent fiscal impact study, and neither has anyone else. These are his numbers, corrected with the county and school system's own figures, and every correction is listed with its source so anyone can check it. If a number here is wrong, tell us and it gets fixed.

Method. Every figure comes from Rep. Horner's own model, obtained through an open records request, his public statements on the record, the county's property records, audited Catoosa County and Catoosa County Schools budget documents, or published state budget analysis.