What a Road District Would Cost
Key Facts
The Bill a District Has to Cover
The number a district needs is not a matter of opinion. Boulder County surveyed Niwot’s pavement, and a treasurer’s workbook built from that survey carries the cost of bringing every street up to standard and holding it there. Amortized over the life of the work, it comes to about $3.05 million a year in 2025 dollars, with the capital side priced on the condition the pavement will be in by 2027.
That figure is the numerator, and it does not change no matter how the district is drawn. What changes is the divisor.
Every levy on this page is given in whole mills, rounded to the nearest. Two decimal places would claim a precision this chain does not have — the cost rests on a pavement survey priced at a generic $11 a square foot. Where we are being careful is in which figure we publish, not in how we round it: the bottom of the range is the scenario most favourable to a district, and it is the one every comparison below is priced at.
Who Is in the Base
A mill levy is a ratio: the money needed, divided by the taxable value standing behind it. Niwot’s taxable value is unusually concentrated. Fifteen industrial parcels on Dry Creek Parkway and Monarch Park — the Boulder Tech Center — carry $40.4 million of assessed value plus $13.1 million of business equipment — about a quarter of everything a townwide levy could fall on. Downtown’s 29 parcels are a second question, because the 2025 citizen proposal deliberately left Second Avenue out: it already pays a 1% improvement-district tax of its own.
Two questions, four answers. All four raise the same $3.05 million.
| Who is in the district | Parcels | Taxable base | Mills | Median home |
|---|---|---|---|---|
| Everyone — downtown in, Tech Center in the bottom of the published range | 1,837 | $207.9M | 15 | $1,045 |
| Downtown out, Tech Center in | 1,808 | $196.6M | 16 | $1,115 |
| Downtown in, Tech Center out | 1,822 | $154.4M | 20 | $1,393 |
| Downtown out, Tech Center out the top of the published range | 1,793 | $143.1M | 21 | $1,463 |
Losing downtown costs everyone else about a mill. Losing the Tech Center costs everyone else five. That is not a finding about anyone’s intentions; it is what happens arithmetically when 15 parcels out of 1,837 carry a quarter of the value.
Why we publish all four, and not the friendliest one
The tempting figure is 15. It is the lowest, so it is the most generous to a district, and it is the only row levied on the same base as our own 4 mills — which makes it the right anchor for a comparison. We use it as that throughout this page.
But it is the wrong thing to publish as the levy, because it is not a measurement — it is a measurement plus an assumption: that both downtown and the Boulder Tech Center are inside the district. No boundary anyone has drawn puts both of them inside — the 2025 proposal deliberately left downtown out — and nothing anywhere commits 15 industrial parcels to a district that does not yet exist. A single number would carry that assumption invisibly, and a reader would have no way to see what it was resting on. So the range is the honest figure and the table above is the reason for it.
The range varies exactly one thing — who is in the district. That is what separates it from a range that hides how unsure we are about the cost. It is about $3.05 million a year in all four rows.
Against Incorporation, at the Same Home
All three rows below are priced at Niwot’s median dwelling, $1,094,500, under the same assessment arithmetic: 6.8% of actual value after subtracting the lesser of 10% or $70,000.
| How the roads get paid for | Per year | Per month |
|---|---|---|
| Incorporation 4 mills plus 2.5% sales tax — roads, snow, land use, wage authority, a permanent vote | $752 | $63 |
| Townwide district 15–21 mills — pavement only | $1,045–$1,463 | $87–$122 |
| Neighborhood district 14–47 mills — your own streets only | $975–$3,274 | $81–$273 |
Incorporation costs less than the cheapest version of a district, in 32 of Niwot’s 34 neighborhoods. Against the most expensive version it costs less in all 34. We test the claim at the bottom of the range, because an argument that only works on the unfavourable end of someone else’s assumption is not an argument.
The gap is not the interesting part. What the money buys is. A district is a single-purpose taxing body: it can finance capital improvements and maintain them, and that is the end of its powers. It cannot plow snow. It cannot write a land use code. It has no standing at a county hearing, no seat on a regional transportation board, and no authority over a minimum wage. The roads-only option is priced like a roads-only option because that is what it is.
It could not plow it.
Where the Price Comparison Is Close
At the bottom of the range — 15 mills, a district with everyone in it — two of Niwot’s 34 neighborhoods price below incorporation, and a third is a $3 tie. We would rather say so than round it our way. These are the three with the least valuable homes, which is the whole reason: a district charges only on property, and incorporation charges partly on spending, so the lower the home value the closer the two run.
All three reverse at the top of the range. If the Boulder Tech Center is not in the district, a district costs more than incorporation everywhere in Niwot without exception. We are not resting the argument on that, which is why the table below is priced at the figure least helpful to us.
| Neighborhood | Incorporation | District at 15 mills | Difference |
|---|---|---|---|
| Countryside median home $367,100 — a tie, and the sign turns on the rounding: at the unrounded 14.68 mills it is +$4 | $334 | $337 | −$3 |
| Cottonwood Park Condos median home $339,500 | $327 | $312 | +$15 |
| Cottonwood Park West median home $649,000 | $632 | $596 | +$36 |
A difference of $3 a year is a tie, and $36 is close enough that a household there should decide on something other than price. So here is the argument that does hold, and it does not depend on a dollar figure: a district would levy on these households and still not touch their streets. Countryside and Cottonwood Park Condos are served entirely by private roads, and a public road district has no authority to spend public money on a private road. Those households would pay the district levy and go on funding their own pavement through their association — on the order of $400 per home for an overlay in Countryside and $1,015 in Cottonwood Park Condos, and two and a half times those figures if the pavement is left long enough to need rebuilding. Cottonwood Park West is mostly county road with one private cul-de-sac, so it is the weaker case of the three, and we say so. No district of any kind can plow snow.
Whether a District Is Actually Available
Everything above prices a district as though one could be formed. The record of the last thirteen years is worth knowing before anyone treats that as settled.
- 2013. Boulder County put a county-wide subdivision paving district on the ballot at 7.15 mills. It failed.
- 2025. A citizen group proposed a Niwot district at 12 mills — a figure that assumed a county capital contribution and interest-free county financing. Both have since been withdrawn. The proposal was pulled before reaching the ballot.
- Today. The same work costs 15 to 21 mills, with no county contribution in the estimate. There is no proposal on the table and no organization assembling one.
Two attempts, thirteen years — one defeated at the ballot, one pulled before it got there — and a price that has roughly doubled between them. That is the history; readers can weigh it themselves.
One structural asymmetry, stated as structure
A district’s boundary can lose parcels; an incorporation’s cannot. Incorporation is on the ballot on November 3, so the boundary being voted on is fixed and every property inside it is in. A district could not reach a ballot before 2027, and in that window a boundary can change — the 2025 proposal is the example, having lost neighborhoods before it was withdrawn.
That matters here because of how concentrated the base is. The Tech Center shares a boundary with the City of Boulder, and the four-row table above says what its departure would cost the households left behind: about five mills. We are not predicting what anyone would do, and we do not think it is our place to. What we will not do is assume the favourable answer and print it as a single number. That is the whole reason the figure on this page is a range. A household weighing a district is entitled to see the span it would be signing up for, not the best corner of it.
Common Questions
Why does a district cost almost four times the property tax and still not cover as much?
Because a district has one revenue source and incorporation has several. A district can only levy property tax, so the whole road bill lands on the mill levy. A town levies 4 mills and collects a 2.5% sales tax, and roughly half of that sales tax is paid by people who are not Niwot households — visitors, and businesses buying equipment and services delivered to an address inside the boundary. Incorporation is cheaper for a Niwot household because the bill is shared with people who use Niwot without living in it.
It is worth saying plainly that this cuts the other way too: the sales tax is a real cost, it is included in the $752 above, and it is the reason the comparison narrows in the neighborhoods with the least valuable homes.
Does the range mean you are unsure what it costs?
No. The cost is a single figure — about $3.05 million a year — and every row in the table above raises exactly that. The range is not doubt about the cost; it is the unanswered question of who pays it. Take 15 parcels out of a 1,837-parcel district and the same bill lands on fewer people, so the rate goes up. Nothing about the roads changed.
We could have published the low end alone. It is the figure that flatters our argument least — the one most favourable to a district — so it would have been defensible as a comparison. But as the levy it would have been a conditional with the condition hidden: it assumes both downtown and the Boulder Tech Center are inside a district nobody has drawn. That is not something anyone should assume on a reader’s behalf.
The levy also rises over time rather than falling. Property values grow about 3.8% a year in the pro forma; construction costs grow about 3% and pavement deterioration adds about 2.4% more of scope as streets drop below the thresholds where an overlay stops being enough. Costs outrun values by roughly 1.6 points a year.
What about a district for just my neighborhood?
It is more expensive, not less, because a small district loses every economy of scale and has a much smaller base to spread the work across. Across the 24 neighborhoods with a costed road program, the levies run from 14 mills on Monte Vista Avenue to 47 mills in Centrebridge. The average Niwot household’s own neighborhood district would be about 33 mills.
Eight of Niwot’s subdivisions have private roads, and for those a district changes nothing about who is responsible for the pavement.
Where do these figures come from?
The road costs and the district boundaries come from the treasurer’s workbook built on Boulder County’s own pavement survey. The tax base is measured parcel by parcel from the Boulder County Assessor’s roll, using the county’s own assessed values, with exempt property — churches, schools, government land — excluded, because exempt property pays no levy whatever it is worth.
Every figure on this page is reproduced by working/neighborhood-pid/derive.py in the campaign’s published repository. The full derivation is in PID Cost Methodology, and the history of the two attempts is in The PID Proposal.
A note on what this page is not. Nobody currently proposes a townwide district, and this page is not an argument against the people who once did — they were trying to solve the same problem, and the 2025 group’s work is part of why the cost is now well understood. It exists because “why not just do a PID?” is a fair question that deserves an arithmetic answer rather than a rhetorical one.
The roads get paid for by the people who drive on them either way. The question on November 3 is what else you get for it, and who decides. See what it costs a household, run your own number, or read the roads case in full.