{"id":299202,"date":"2025-06-17T18:21:26","date_gmt":"2025-06-17T18:21:26","guid":{"rendered":"https:\/\/peraltafinancing.com\/business\/real-estate\/is-the-gta-average-home-price-being-propped-up-by-a-change-in-market-dynamics\/"},"modified":"2025-06-17T18:21:26","modified_gmt":"2025-06-17T18:21:26","slug":"is-the-gta-average-home-price-being-propped-up-by-a-change-in-market-dynamics","status":"publish","type":"post","link":"https:\/\/fivemor.com\/?p=299202","title":{"rendered":"Is The GTA Average Home Price Being “Propped Up” By A Change In Market Dynamics?"},"content":{"rendered":"


\n<\/p>\n

\n

Put on your tinfoil hat, folks!<\/p>\n

And get ready to explain how and why the moon landing was faked.\u00a0 I mean, it\u2019s not like you could launch a rocket from a\u00a0round\u00a0<\/em>earth, right?<\/p>\n

But seriously, I\u2019ve heard multiple times this year, and from several different people that the GTA average home price is somehow\u00a0misleading<\/em>.\u00a0 More specifically, I\u2019ve been told that the average home price is being \u201cpropped up\u201d by a change in market dynamics.<\/p>\n

What do we mean by this?<\/p>\n

Well, it\u2019s essentially the \u201claw of averages.\u201d<\/p>\n

Let\u2019s explain by using a very rudimentary example.<\/p>\n

In fair\u00a0Verona<\/em> where we live, every single month, we see exactly ten properties sold as follows:<\/p>\n

5 Condo Sales
5 Freehold Sales<\/p>\n

Let\u2019s say that condos are worth $500,000 and houses are worth $1,000,000.<\/p>\n

This means that the average sale price in Verona is $750,000<\/strong>.<\/p>\n

Now, let\u2019s say that the market dynamics in Verona change, and we start to see the following every month:<\/p>\n

6 Condo Sales
4 Freehold Sales<\/p>\n

Suddenly, the average declines to $700,000<\/strong>.<\/p>\n

We can play this game all day, right?\u00a0 Do you see where this is going?<\/p>\n

Ignore the small sample size, just for a moment, but recognize that this\u00a0is\u00a0<\/em>something we should talk about just as one of the readers suggested in response to last Monday\u2019s blog.\u00a0 We\u2019re only using a small sample size here to prove the point about how a change in market dynamics can be affected by the law of average, and vice versa.<\/p>\n

Now let\u2019s look at the inverse<\/em> of the situation described above.<\/p>\n

What if the market changed and we saw\u00a0fewer<\/em> condos being sold in Verona and\u00a0more<\/em> houses, proportionally?<\/p>\n

Picture this:<\/p>\n

3 Condo Sales
7 Freehold Sales<\/p>\n

Now all of a sudden, the average home price in Verona\u00a0increases<\/em> to $850,000<\/strong>.<\/p>\n

But are all the houses in Verona really \u201cworth\u201d 13.3%<\/strong> more today than they were when our split between houses\/condos was 50\/50?<\/p>\n

Of course not.<\/p>\n

But\u00a0this<\/em> is how the law of averages can affect a particular data set, and there\u2019s a prevailing theory in our Toronto real estate market the moment that the average GTA home price does not \u201caccurately\u201d reflect the market conditions.<\/p>\n

So let\u2019s start from here:<\/p>\n

\n

Last month, the average GTA home price increased<\/em> by 1.2% from $1,107,463 in April to $1,120,879 in May.<\/p>\n<\/blockquote>\n

Some people, including the odd Realtor,\u00a0 suggest that this is \u201cmisleading\u201d because they have seen, witnessed, or experienced a decline in home prices.<\/p>\n

But others are quick to note that they\u2019ve lost in bidding wars, or seen houses sell in mere days, and have seen an overall increase in activity.<\/p>\n

After all,\u00a0sales increased<\/em> by 11.5% from April to May, moving from 5,601 to 6,244.<\/p>\n

The question I would ask is this:<\/p>\n

\n

Is the data set large enough to account for variables that could manipulate the data unreasonably?<\/p>\n<\/blockquote>\n

In fair Verona, where only\u00a0ten<\/em> properties sell each month, the market average is far more susceptible to variability.\u00a0 But with 6,000 sales every month in Toronto, isn\u2019t our data set large enough?<\/p>\n

To that point, let me explain why and when a smaller sample size can mislead you.<\/p>\n

I was doing a listing presentation for a builder two weeks ago and I was demonstrating the real change in market value for a \u201ccomparable home\u201d by looking at a new-build that sold in August of 2020 and turned over yet again this past April.<\/p>\n

I showed him the market statistics that compared August of 2020 to April of 2025, specifically:<\/p>\n

1) GTA average home price
2) Toronto-416 average home price
3) GTA HPI
4) Toronto-416 HPI<\/p>\n

He asked me, \u201cCan you run the same data for\u00a0only\u00a0<\/em>detached homes and\u00a0only\u00a0<\/em>in C11?\u201d<\/p>\n

I told him, \u201cYou have four averages right here; two that exist to eliminate seasonality and variability (the HPI\u2019s) and two that are based on several thousand sales.\u00a0 To look at only detached homes in C11 would lead to a sample size error.\u201d<\/p>\n

He furrowed his brow and said, \u201cI don\u2019t understand.\u00a0 Just look at the average for detached in C11.\u00a0 That will tell us what we need to know.\u201d<\/p>\n

The problem is, there were 10,775 sales in August of 2020, which is a large enough sample size to compare to the 5,601 sales in April of 2025.<\/p>\n

However\u2026<\/p>\n

\n

\u2026Looking only at detached sales in C11, there were a mere\u00a09<\/em> recorded in August of 2020 and only\u00a013<\/em> that changed hands in April of 2025!<\/p>\n<\/blockquote>\n

Regardless of how this data is used, the sample size is simply too small to produce a result that can be effectively and accurately relied upon.<\/p>\n

You could have one<\/em> mansion selling in either month that completely skews the numbers!<\/p>\n

So with that in mind, let me return to the theory that\u2019s been expressed over the last couple of months.<\/p>\n

Here is what I wrote in last Monday\u2019s blog post:<\/p>\n

\n

A colleague of mine suggested that the only reason why the GTA average home price wasn\u2019t lower is because the condominium market is so weak, condo sales are scarce, and thus the average home price is being \u201cpropped up\u201d by the fact that there are proportionally fewer condo sales than usual.<\/p>\n<\/blockquote>\n

I have two hours at the laptop right now, free of children, so I want to dig into the data and see if we can confirm or deny this hypothesis.<\/p>\n

Alright, let\u2019s start with a simple data point:<\/p>\n

\n

In May of 2025, there were 6,244 total sales and 1,482 were \u201ccondominium apartments.\u201d<\/p>\n

That\u2019s 23.7%.<\/p>\n<\/blockquote>\n

Keep in mind, there\u00a0is<\/em> a data point for \u201cTownhouse\u201d which considers both freehold townhouses\u00a0and<\/em> condominium townhouses, but there are so few condominium townhouses sold, and they are typically at a higher price than \u201ccondominium apartments,\u201d that I don\u2019t think it\u2019s worth adding this data point.\u00a0 Not only that, but consistency<\/em> is what keeps data clean.\u00a0 So long as we\u2019re only using \u201ccondominium apartments\u201d across\u00a0every<\/em> comparison, then the data will be completely accurate and can be relied upon for our experiment.<\/p>\n

So what do we make of that figure above?\u00a0 23.7%.<\/p>\n

Is that a lot?\u00a0 Is that a little?\u00a0 Why is it important?<\/p>\n

Well, the average condominium apartment in May of 2025 was $683,413.<\/p>\n

Meanwhile, the average semi-detached home was $1,098,447 and the average detached home was $1,425,264.<\/p>\n

\n

This infers that, as our example in Verona above showed us, the\u00a0fewer<\/em> condo sales we see, the\u00a0higher<\/em> the average home price will show.<\/p>\n<\/blockquote>\n

So what can we compare this data point to?<\/p>\n

Let\u2019s compare this to\u00a0April<\/em>, for starters.<\/p>\n

In April, we saw 1,430 condominium apartment sales amid 5,011 total sales.<\/p>\n

That\u2019s 25.5%.<\/p>\n

So the proportion of condominiums sold\u00a0decreased\u00a0<\/em>from 25.5% in April to 23.7% in May.<\/p>\n

Go back to our example from fair\u00a0Verona<\/em> for a moment.<\/p>\n

The average home price was $750,000 when there were 5 condo sales and 5 freehold sales.\u00a0 But what happened when condo sales declined<\/em>?<\/p>\n

The average went up to $850,000.<\/p>\n

So what happened in Toronto from April to May?<\/p>\n

-Condo share declined from 25.5% to 23.7%
-Average home price increased by 1.2%<\/p>\n

Here\u2019s where a person could argue:<\/p>\n

\n

The\u00a0real<\/em> average home value, all things being equal, remained the same.\u00a0 The decline in proportional condominium sales provided a lower number of cheaper<\/em> sales, and thus resulted in a higher average home price.<\/p>\n<\/blockquote>\n

Well, hot dog!<\/p>\n

Now, let\u2019s look at the trend<\/em> through 2025:<\/p>\n

\"\"<\/p>\n

The share of condominium sales is\u00a0declining<\/em>.<\/p>\n

This makes sense for several reasons:<\/p>\n

1) The condo market is weak.
2) The condo market has become weaker.
3) The freehold market is more cyclical.
4) The freehold market typically peaks in April and May.<\/p>\n

But do points #3 and #4 discount points #1 and #2, or do they work in tandem?<\/p>\n

Let\u2019s try something.<\/p>\n

Let\u2019s look at the proportional condominium sales share in a chart along with the average GTA sale price:<\/p>\n

\"\"<\/p>\n

Are there any conclusions to draw here?<\/p>\n

The proportional condominium sale share is\u00a0declining<\/em> as the average home price is\u00a0increasing<\/em>.<\/p>\n

Again, one could argue that \u201cfewer condo sales will lead to higher average home prices,\u201d and theoretically, they would be correct.<\/p>\n

Once more, consider the average prices in May:<\/p>\n

Condominium: $683,413
Semi-Detached: $1,098,447
Detached: $1,425,264<\/p>\n

Fewer condominium sales = higher average home price.<\/p>\n

The point has been made.<\/p>\n

However, we\u2019re only looking month-to-month.\u00a0 Shouldn\u2019t we at least consider year-over-year data and see what evidence it provides?<\/p>\n

Let\u2019s look at January through May and compare 2024 to 2025.<\/p>\n

\"\"<\/p>\n

First and foremost, we note that the proportional share is lower in four out of five months this year, compared to last year.<\/p>\n

Secondly, we note that the share is declining at an increasing rate as we move through spring.<\/p>\n

But to that end, didn\u2019t we just say that the freehold \u201cpeak\u201d is typically in April and May?<\/p>\n

I would argue that we might come to\u00a0expect<\/em> a declining proportional share of condominium sales, based on the freehold peak.\u00a0\u00a0However<\/em>, we didn\u2019t see this in 2024.\u00a0 In fact, the proportional share was essentially the same in February, March, and April, and only declined by 0.6% from April to May, whereas the decline in the same period in 2025 was 4.0%.<\/p>\n

But that\u2019s only 2024 vs. 2025.<\/p>\n

Let\u2019s look at a longer history:<\/p>\n

\"\"<\/p>\n

Alright, so many of you are just looking at a maze of %percentage% signs and are moving on.<\/p>\n

Read through at your heart\u2019s content, but here\u2019s what I would propose: we need to understand what the 2025 figures mean against\u00a0all\u00a0<\/em>the previous years, not just 2024 vs. 2025.<\/p>\n

Let\u2019s take an\u00a0average\u00a0<\/em>of 2021 through 2024 and compare that average to 2025.\u00a0 That is the best way to eliminate a sample size error and answer this question:<\/p>\n

Are there proportionally fewer condos selling in 2025 than usual?<\/p>\n

Here\u2019s the improved data set:<\/p>\n

\"\"<\/p>\n

The answer is\u00a0yes<\/em>.<\/p>\n

Save for February, when a proportionally higher than average\u00a0<\/em>number of condo sales appeared, the sales have been lower in three of the other four months, with March showing us a draw.<\/p>\n

This would, in fact, lead me to conclude that there are fewer condos selling than normal, as a percentage of\u00a0overall<\/em> sales, and thus this could have an effect on the overall TRREB average home price.<\/p>\n

How\u00a0much<\/em> of an effect?<\/p>\n

We can only speculate.<\/p>\n

Last but not least, let me demonstrate the share in a graph since a<\/em> picture paints a thousand words<\/em>:<\/p>\n

\"\"<\/p>\n

As we can see, the month of May did\u00a0not<\/em> provide us with the lowest proportional share of condo sales in the last five years, as the low point was in 2021, but the trend in 2025 has been downward.<\/p>\n

Back to the speculation!<\/p>\n

I think the following would be fair:<\/p>\n

If we saw the proportional share dropping from 35% to 15%, that\u2019s going to have an effect.\u00a0 If we saw the proportional share dropping from 35% even to 25%, that\u2019s going to have an effect.\u00a0 But if we\u2019re talking between 1% \u2013 3%, I don\u2019t know if we should be getting out the\u00a0tinfoil hats<\/em> just yet.<\/p>\n

In an unrelated story, the\u00a0new\u00a0<\/em>TRREB average, which takes proportional sales share into account, will be called the \u201cOswald Average,\u201d and I expect long-time blog reader and infrequent commenter\u00a0Writer<\/em> to calculate it for us every month.\u00a0 \ud83d\ude09<\/p>\n

Happy Monday, folks!<\/p>\n