Skip to the page

About

Haliburton County is deciding how many municipal governments it should have. The Eastern Ontario Regional Network, engaged by the County to run a community-wide affordability and governance structure study, put four models to County and local councillors on 12 August 2026 with financials indicating that taxpayers can no longer afford the present two-tier structure of County and townships. Choosing between the models falls to the councils elected this autumn.

This atlas exists because of the shape that argument takes. Every case for and against turns on cost and service: whether four road departments cost more than one, whether a smaller township pays more per resident for the same service, whether amalgamation elsewhere delivered what it promised. Those are questions about figures, and before this the figures sat in four Financial Information Returns, four sets of minutes and a provincial database, on lines that do not always mean the same thing in each municipality.

So the work here is unglamorous and specific. It puts 94 metrics on a footing where the four municipalities can be set beside each other. It records 81 more as not comparable, and declines to chart those at all. And it says which is which, rather than leaving a reader to assume.

The four models

One · status quo
The County and four lower-tier municipalities as they are. The study's own financials are what rule it out rather than anyone's preference.
Two · single tier
Amalgamation. One municipal government for the county, on the pattern of Prince Edward County or Quinte West.
Three · enhanced County, four municipalities
The County takes the services that gain from scale and specialisation; the four townships remain.
Four · enhanced County, two municipalities
As above, with the four townships consolidated into two: Minden Hills with Algonquin Highlands, Dysart et al with Highlands East.

The study reports that restructuring on its own does not guarantee lower taxes, and that the gains it has seen elsewhere come from capacity, coordination and less duplication rather than from the change of structure itself. That is worth holding onto while reading anything on this site: a difference in cost between two municipalities is a fact about how they file and what they do, not a forecast of what merging them would save.

What this atlas will not do

It does not take a position on amalgamation. Nothing here is ranked overall, and no municipality is called better run than another. Most metrics in the catalogue record their direction as contested, because spending less is thrift or neglect depending on what you think a municipality is for.

It does not fill in a figure nobody published. A municipality that files nothing gets a recorded gap with the reason attached, never a zero and never an estimate borrowed from a neighbour. 301 warnings and 294 notes are published against its own figures on the quality page.

It does not confuse a place with a council. How pleasant a township is to live in and how well it is administered are different questions, and figures that answer the first are kept apart from figures that answer the second.

It is also not the study. The atlas holds none of EORN’s workings and does not restate its conclusions as though it did: what is attributed to the study here is attributed to the reporting that carried it, and the report itself is public.

The coverage

The Highlander has followed this from the service delivery review onward. Reading it in order is the quickest way to see how the county arrived at four models and an election.

The study itself, the survey and the town hall schedule are on the County’s engagement site, Wade In. Read the report before this atlas, not after it: it is the document the decision will be made from, and this is a way of checking the figures underneath it.

Who built it

A one-person project.

Colin Smillie
Developer

Colin builds tools that make Canadian public information easier to find and use: AI-assisted municipal services, open data platforms, and electric vehicle tracking.

This atlas came out of reading about the County’s service delivery review and the comparisons being drawn to counties that have already amalgamated, then finding there was nowhere to check the figures underneath the argument. Four returns, four sets of minutes and a provincial database are a poor way to answer whether one township costs more than another. So they were assembled once, carefully, with everything that could not honestly be compared marked as such.

How it is built

28 sources, each recorded with its publisher, its licence and what it structurally cannot answer. Financial figures come from the Financial Information Return every Ontario municipality files; counts and caseloads come from the municipalities’ own agendas and reports; population and dwellings come from the census. Where a municipality’s own document contradicts a provincial one, the municipality wins.

Python for the pipeline, Next.js for the site, prerendered and served from a VPS with no database behind it. Exactly one thing is loaded from another host: Matomo, on the author’s own server, which counts visits and is described on the privacy page. The code and much of the writing were done with an AI assistant, Claude Code. No figure on this site comes from a model. Every one traces to a document somebody filed, and the checks that would catch an invented number run on every build.

Everything is rebuilt from source with one command, every figure carries where it came from, and the build refuses to publish while any figure contradicts another. Every source · what the checks found