Tap a bare patch to plant it. Planters get flowers. Long grass gets wildflowers.
Does building infill change what nearby homes are worth?
Infill means building new homes inside a neighbourhood that is already there, instead of on empty land at the edge of the city. It takes many forms, including duplexes, fourplexes, sixplexes, rowhouses, small apartment buildings and condos.
Zoning rules set what can be built on a piece of land. Edmonton changed those rules in 2024 to allow more homes on a single lot, so more infill is going up now than before.
Once a year the City works out what every home in Edmonton is worth. That figure is called the assessed value, and the City uses it to set the tax bill. This page measures what happened to the assessed value of the homes around new infill buildings.
How to read a number about house prices
Three things that change how any number about infill should be read. Each one is shown using the City's own records.
A neighbour tells you their house is worth 30 percent more than it was ten years ago.
That sounds like good news. But nearly every house in Edmonton gained value over those same ten years. So the number by itself does not tell you much.
The chart below does not show dollars, because a $300,000 home and a $900,000 home would be impossible to compare. Instead every home starts at 100 in 2015. A home worth $400,000 that year and $480,000 today has gained 20 percent, so its line reaches 120. The chart averages every home this page looked at into one line.
A student's mark rises from 60 to 85 over the year. That looks like real improvement. Then you find out every student in the class rose by about the same amount, because this year's test was easier. The rise says more about the test than about the student.
Run the check and this fills from the City records.
Two houses sit on the same street. A fourplex was built next door to one of them. Ten years later, both houses are worth about 40 percent more.
Here is the same chart, split into two lines. The red line is homes within 50 metres of a new infill building. The grey line is homes in the same neighbourhoods that sit more than 200 metres from any new building.
Both groups of homes sat in the same city over the same years. So anything that pushed house prices up across Edmonton pushed both lines up. Anything that held prices down held both lines down. If the two lines pull apart, house prices across the city cannot be the reason, because both groups had the same city prices. Only the space between the lines could be about infill.
A school wants to know if milk helps children grow. Watching one child will not answer it, because that child would have grown anyway. So you follow two groups. One group drinks milk every day. The other group does not. At the end of the year you compare how many centimetres each group grew.
Suppose the milk group grew more. That still might not be the milk. If those children were younger, they would grow faster anyway. If they were already growing faster in the years before the milk started, then something else was making them grow and the milk is not the answer. A fair study checks both of those things. So do the measurements further down this page.
Run the check and this fills from the City records.
One block has a house built in 1955 and a house built in 2005.
Homes built in different decades gain value at different speeds. This is true no matter what gets built near them.
Each bar below is one decade. The height of the bar is how much those homes changed in value.
Runners of different ages do not run at the same speed. Put a 12 year old and a 40 year old on the same track and they will post different times, whatever training either of them has done. The difference comes from their age.
Run the check and this fills from the City records.
If old homes and new homes already change at different speeds, then comparing the two groups measures their age as well as their distance. This page handles that by comparing homes of the same age. A 1950s bungalow beside a new building is compared against a 1950s bungalow four blocks away.
What the records show
The measurements, with those three things taken into account.
A fourplex was built beside one house in 2016. Another was built beside a different house in 2023. Both are used. Each house is compared with what homes on its own street were worth the year before its building went up.
Before anyone builds, they need permission from the City. That permission is called a building permit, and the City publishes a record of every one it gives out. Each bar below counts the permits given out in one year, for buildings of the size set at the top of this page.
Across the whole record there is about one of these buildings for every hundred homes that were already there. On a street with twenty houses, that works out to roughly one new building every five streets. The drawings on this page show more than that, so the different building types can be seen.
Two loaves of bread go into the oven at different times. To see how each loaf baked, you count the minutes from when that loaf went in. You do not check both loaves at the same time of day.
The City decides what your home is worth once a year. It uses prices from July of the year before. So a building finished this spring is not in anyone's value yet. It takes about two years before a new building can appear in these numbers.
Run the check and this fills with the permit record.
Your house is three doors from a new sixplex. Your cousin lives eight blocks away in the same neighbourhood. Nothing new was built near your cousin.
Homes within 50 metres of a new building go in the first group. Homes in the same neighbourhood that sit more than 200 metres from any new building go in the second group. Both groups are measured the same way. The City sets a value for every home each year. The measurement uses the value from the year before the nearest building was approved, and the value from a chosen number of years after that. That number can be changed at the top of the page.
You want to know how much a heater warms a room. You measure the air beside the heater. You measure the air in the far corner, where the heater reaches nothing. The air halfway between is a little warm as well. If you used that halfway air as your cold reading, the heater would look weaker than it really is.
The pale grey homes sit between 50 and 200 metres from the new building. They are too far away to count as next door. They are also too close to be a fair comparison. So they are left out of both groups. If they were counted as the comparison group, the same building might be changing both sides.
The problem with using each building's own date
Every home is measured from the year its own building was approved. So two homes on the same street can end up covering different years. A home beside a 2016 building is measured from 2015 to 2019. A home beside a 2022 building is measured from 2021 to 2025. A year when prices jumped affects one of them and not the other.
The other choice was to pick one four year stretch and drop every building outside it. That would throw away about two thirds of the record, so this page keeps all of it instead. For each neighbourhood it reports how far apart in time the two groups ended up. Those reports appear in the running list of steps at the top of the page, on the lines marked CAL.
What a unit is
A unit is one home inside a building. A duplex has 2 units. A fourplex has 4. A single house on its own lot is 1 unit. The control at the top of the page sets how many units a building needs before it counts here.
Three different things people mean by value
Market value is what a buyer would actually pay for a home. Assessed value is the City's estimate of that, worked out once a year. Your tax bill comes from comparing your assessed value with all the others, rather than from how much your own went up.
This page uses assessed value, because it is the only number the City puts out for every home every year. Alberta keeps real sale prices private, and only a handful of homes sell near any one new building anyway.
If a new building changed what nearby homes are worth, the change should be biggest for the house right beside it. It should get smaller four blocks away.
Every home with a value in both of its years is sorted into a group by how far it is from the nearest new building. This chart uses all of them, not only the two groups from the last section. If distance changes what happens to a home, the bars will step up or down in order across the chart.
Stand beside a loud speaker. The sound is loudest there. Walk away and it gets quieter. If the sound is just as loud in every room, then you are not hearing that speaker.
Run the check and this fills from the assessment records.
A duplex with 2 homes in it goes up beside one house. A building with 16 homes in it goes up beside another house. If the building is what changes what neighbours are worth, the bigger one should change it more.
The chart above sorted homes by how far they sit from a new building. This one sorts them by how big that building is. Every home in it is within 50 metres of a new building, and every one is measured against the homes more than 200 metres away in the same neighbourhood.
A doctor wants to know whether a medicine works. If it does, two pills should do more than one pill. If one pill and five pills give exactly the same result, then something other than the pills is producing it.
Run the check and this fills from the permit and assessment records.
The other measurements on this page compare homes near a building against homes further away. This one compares near homes against each other. If the size of the building makes no difference at all, then the gap in the other charts is coming from something the buildings did not cause.
On some blocks the old houses were already selling fast and getting fixed up. This was happening years before anyone applied for a permit.
This chart follows the same two groups backwards in time. Each home starts at 100 in the year before the building next to it was approved. The shaded part of the chart covers the years before that approval.
Builders choose where to buy. They look for lots where the land is worth more than the house standing on it. That usually means an older house on a wider lot. So the blocks that get infill are not an ordinary slice of the neighbourhood. They were already different from their surroundings before anything was built, and some of those differences could change home values on their own. This chart shows what those blocks were doing beforehand.
A group of students starts getting help with math in March. Their marks go up. But their marks had already been climbing for two years before March. Something was already raising their marks before the help arrived, so the help does not explain the whole rise.
Run the check and this fills from the assessment history.
The homes beside new buildings are mostly from the 1950s. The homes further away are mostly from the 1980s.
Homes near new infill are older than homes further away, so comparing the two groups straight across includes their age difference. Instead the 1950s homes near a new building are compared against the 1950s homes further away. Then the 1960s against the 1960s, and so on for every decade. Those results are then combined. Each row below is one neighbourhood, showing the figure before and after that is done.
Two running clubs want to know whose training is better. Club A has mostly older runners and Club B has mostly younger ones. A straight race between the clubs would mostly show which club has the ages that suit the distance. So instead you race the 12 year olds from A against the 12 year olds from B. Then the 40 year olds against the 40 year olds, and so on. Then you put those results together.
Run the check and this fills once homes are matched by decade built.
Two houses sit on the same block. One is 49 metres from a new building and one is 51 metres. The first counts as near and the second does not, even though they are almost the same distance away. Move the line to 60 metres and they swap sides.
To measure anything at all, four numbers had to be fixed. Near means within 50 metres. Away means beyond 200 metres. Each home is followed for 3 years. Homes are matched by the decade they were built.
Those four are definitions, not findings. Whatever a new building does to the homes around it, it does not stop at a line. It fades. So there is no distance where near becomes away, and the line has to be drawn somewhere anyway. The same goes for the other three. Nothing in the City records points to 50 rather than 40 or 60.
So the analysis runs again with every combination of them. If the results come out close together, then the answer holds whichever numbers are used. If they come out far apart, then the answer changes with the numbers, and no single figure from this page can be quoted on its own.
You step on five bathroom scales. Four say 70 kilograms and one says 90. You cannot trust any single reading, because you do not know which scale is the odd one out. If all five said 70, you would trust it.
Each row is one setting. The dots on that row are every re-run that used it, and the black dot is their average. If the rows for a setting sit on top of each other, that choice makes no difference. If they sit apart, that choice is moving the answer. The caption names the setting that moves it most.
Run the check and this fills with one dot per re-run.
In one neighbourhood, homes near new buildings did better than their neighbours. In another neighbourhood, they did worse.
One figure for the whole city can hide neighbourhoods that point in opposite directions. Each bar below is one neighbourhood, measured from the zero line. The thin line through the bar shows the range the true figure could fall in.
A class has an average mark of 70. Every student might have got 70. Or half the students got 90 and half got 50. The average does not tell you which.
Stand at a window and count the students in a schoolyard. You count about 40. You might be off by a few, so the real number is somewhere near 40. If you only glanced for a second, you could be off by 20. If you counted carefully for a minute, you might be off by 2. The thin lines on the chart work the same way.
A short line means that neighbourhood had many homes to measure, so the figure is close to right. A long line means it had few homes, so the figure could be well off. If the line crosses the zero mark, the two groups in that neighbourhood may not differ at all.
Red means homes near new buildings rose less than the comparison group. Green means more. Grey means the range crosses zero. Select a circle for the numbers behind it.
Run the check and this fills neighbourhood by neighbourhood.
The sections above report measurements. This section reads them. Every number used here also appears on a chart above. The settings behind those numbers can be changed using the controls at the top of the page.
Run the check and this fills in from the actual result.
The two city context
Calgary publishes the same kinds of records as Edmonton. The chart this button fetches shows each city's average residential assessed value by year, with both lines set to 100 at their first shared year, so the two paths can be seen side by side.
This one is optional and asks Calgary's open data portal as well as Edmonton's. Press the button to fetch both.