The other day a reader asked: "What phase of the real estate market cycle are we in?" Luckily I just attended a session on analyzing market data, where I learned some new ways of looking at numbers related to real estate. I have been adjusting my spreadsheets and creating new charts to help us get a better understanding of the real estate market in Edmonton for the past few days.
Nicolai Kolding, partner at Real Trends, presented the course I attended. During the session he showed us the following chart which, he created to help determine where we are in the real estate sales cycle. Basically, sales predict what prices will do - if sales are increasing on a year over year basis, prices will follow, if they are decreasing, prices will flatline and then start to fall:
Real Estate Sales Cycle
To use the chart, you have to look at sales and prices on a month by month basis, and compare them to the year before. For example, in February sales were down almost 20%, and prices were down 1.24% in Edmonton, compared to February 2010. That basically indicates we are at the low point on the sales cycle (phase 5), and have been there for a few months.
What does that mean for the coming months? At the beginning of the year, most people who put out forecasts agreed that the year would start off slowly, and improve in the second half of the year. Likewise, the sales cycle would suggest that we should start to see an improvement in the number of sales (compared to last year), followed by an improvement in prices.
I've broken out the past two years in the chart below, and you can see we started January 2009 off at the low point in the cycle (phase 5), worked our way around to the top (albeit very briefly) by the following January, and are now back at the bottom. The first column shows the change in sales compared to the year previous, the second column shows the change in prices, the third column shows whether we were up or down, and the fourth shows what phase of the cycle we were in.
Edmonton real estate sales cycle
So, based on this analysis of the sales cycle, what I know happens seasonally, and the predictions we made at the beginning of the year, I think we will see sales start to beat last year's numbers around May or June, and prices should be ahead of last year by August or September. Have a look at the charts I published at the beginning of the month and tell us what you think.
I have lots more to share about what is happening in the Edmonton real estate market, and what to expect, in upcoming posts.
The problem with applying a linear sales cycle to predicting the market is that we don’t live in a vaccuum (though a vaccuum may be a good idea for improving the air quality in Sherwood Park)
For example last year I could speculate that the double pluses were related to na surge in demand relaitive to the change in mortgage rules…plus with 2009 happening in conjectintion with a global economic downturn I could conjecture that the economy was improving relative to the previous year,
I think a late spring this year may delay the typical spring sales surge which may be related to another cycle (namely la nina). Also it looks like a phase 5 happened between three and four last year.
If Mortgage rates rise signifigantly we could see a phase 5, if oil goes up to $200.00 we could see a phase 1 (or a phase 5 if the prices of consumer goods ramps up out of proportion to income on account of increased costs). So all external factors remaining constant you may see a linear trend, but they don’t.
It’s not a linear trend, it’s a cyclical trend. It tells you where you are in the cycle, but not how long you are going to stay there. We were in phase 1 for all of 2006 and most of 2007, we’ve been in phase 5 for quite a long time now but last year hit phase 1 for only 2 months. Also keep in mind the phases are all relative to the year before, so prices may be higher than they were a year ago, but lower than they were a month ago. Seasonality is a huge factor, which is why it is more relevant to compare year over year than month over month.
What I think Sasquatch means by linear is that this cycle follows a predetermined or sequential path. That one state will always lead to the next state which will lead to the next….and so on.
So let me understand this then. What your saying is that it is best to compare year over year (which I strongly agree with), but then you say that some phases of the cycle will only last a matter of months. This is somewhat flawed then, as you are switching time scales.
So an example:
2003: phase 3
2004: phase 4
2005: phase 5
What if annual market conditions reflected these phases:
2003: phase 2
2004: phase 5
2005: phase 1
So then between 2003 and 2004 we went through phase 3 and 4. So we’ve gone through several phases in a matter of months, weeks or even days, but these won’t show because we are looking at annual data and comparing phases on an annual timeframe. Thus we can say that phase 3 and 4 never occured, or at least we have no evidence of them occuring, because we have no evidence in the annual data of them occuring.
So if I buy this cycle, I have to assume then that all points occur from point 1 to 8 no matter how brief? How short can a phase be then one month, a day? Isn’t it better just to look at the market at a static point and make some reasonable educated deduction as to why market conditions are the way they are?
Cheers,
OB
I’m sorry but you don’t get it at all. I’m not sure how to explain it… maybe give me a call at the office.
Sara, a linear pattern is one where one stage would lead to a defined next stage and so on, so even though it is indeed a cycle we are saying phase I goes to phase II and so on.
I think that’s overly simplistic. For example if last year we had an early spring and this year we have a late spring we will see the impact of that in year over year data as we know better weather brings out house buyers.
So that circumstance may put us in what looks like a Phase 5 (prices and sales down from last year) because of the weather versus a defined market cycle.
Or we now have an election, if one of the political parties promised incentives for first time home buyers as part of thier platform we may see market activity decrease while buyers are waiting to see if that promise is implemented and then a surge of activity once it is.
All things being equal we could see defined market cycles, but I think external factors (ie financial meltdowns, change in lending policies, weather) are stroinger impacts.
Theoretically the price of Oil/Commodities are also cyclical but start bombing oil/resource producing countries and the theory unravels quite quickly.
I wish the market did function exactly cyclically as I’d make a fortune.
It may be weather, or interest rates or an election that causes the cycle to shift to the next phase. Whatever causes it is irrelevant to this analysis, what the cycle shows is that is is extremely unlikely (if not impossible) that we would go from say phase 5, to phase 1 in one month… in order to get from decreasing sales and decreasing prices, to increasing sales and prices, we would see flat sales and decreasing prices first, then increasing sales and flat prices, etc. So when you start to see the cycle shift from one phase to the next, you may want to take an action before the next phase comes. The cycle can also be used as proof that one of the external influences you mentioned is or is not actually having an affect on the market.
So for example, we have been stuck in phase 5 for months, but we just had some pretty big changes to the mortgage rules. We didn’t think the rule change would have much of an affect on the market, but if we are still stuck in phase 5 at the end of this month, the cycle helps suggest that the mortgage changes did not, in fact, affect the market.
Good post. The concept seems pretty clear to me.
It would take more work, but if you were to chart price and volume, rather than using a table, and then overlay the chart with cycle identifiers, most people would probably find it more intuitive. That way everything, including the duration of each phase, would be represented visually to scale.
Is there any data on the historical average duration of each phase of the cycle?
I think the idea behind the theory is that it is always sequential like discribed on the diagram. But the length of each of the cycle is undetermined. So you cannot say:
2003: phase 2
2004: phase 5
2005: phase 1
In reality it must have been like:
2003 Jan-Oct: phase 2
2003 Nov: phase 3
2003 Dec: phase 4
2004 Jan-Sep: phase 5
2004 Oct: phase 6
2004 Nov: phase 7
2004 Dec: phase 8
2005: phase 1
Or something to that matter.
I think this theory makes lots of sense. And probably looking into historical data a few years back would prove it.
But essentially this theory does not tell you when the prices are going to rise/fall. It just tells you that they are going to rise/fall eventually.
FYI to AB
Not attacking you, I just REALLY have a hate on for this thing. POPPYCOCK! I say good sir!
*Adjusts monocle*
Cheers,
OB
FYI to OB,
Not attacking you, I just REALLY have a hate on for this thing. STATEMENTS WITHOUT ANY EVIDENCE! AND STATEMENTS BEGINNING WITH “Not attacking you”. Usually those statements come just before the attack.
I say good sir!
*Adjusts monocle*
Cheers,
AB
“Usually those statements come just before the attack.”
AH HA! You’ve seen right through my clever ruse! I was about to strike a mortal blow! I was going to be all over you like a fat, hungry man on a sandwich! Alas, I will have to wait for another day!
Cheers,
OB
It’s fun to see the new ways real estate agents have learned to show housing data in a positive way.
Will we get a different “tool” (different way to look at data) in your new toolkit each
week from the course you recently took?
Sales have declined YoY since May 2010 with prices declining since Sept of 2010.
If I use your circular chart, we have been in phase 5 since Sept and were in Phase 3 from May 2010 – August 2010.
Personally I like prefer to work with 15 phase charts. 8 is too little.
72 phases, MINIMUM! 128 phases if you really want to explain how things are going…
It’s fun to see the new ways real estate agents have learned to show housing data in a positive way.
You know it’s currently in phase 5. Meaning sales down and prices down. So how is that positive?
Guess you must want a “CRASH” or “See I told you so” phase. Man… you just can’t make some people happy.
They forgot to provide the users manual, but plenty of rope.
The “long term” market, that this graph is relevant to, is a 20 year cycle.
Even the short term economic cycle in RE is at least a year and up to 5. We are clearly in a phase 3 transitioning to 4. how soon will we move to 5? How long will it last? Can external forces cause 5 to be mitigated? Who knows.
the cycle is so long because RE inventory takes a long time to build up, a long time for a person to pay off a unit, a long time for new entrants to enter the market, etc.
Those of you trying to apply this down to the month / week / day / sale are out to lunch.
The cycle is just a basic supply and demand cycle. That as price goes up supply goes up and demand goes down and as prices go down supply goes down and demand goes up. Buyers are sellers are slow to react when it comes to house ownership which creates the cycle.
The cycle can move slightly faster or slower but it’s based on the reaction time of buyers and sellers. The cycle is always there and always affecting the market.
However the cycle is only a part of home prices and I think more important is external factors such as the economy, mortgage rules and interest rates.
“..However the cycle is only a part of home prices and I think more important is external factors such as the economy, mortgage rules and interest rates…”
Not attacking you at all here Steve, but what you’re saying is that this “cycle” exists, but external factors can override it? Fair enough, but every moment of everyday the market is exposed to external factors, thus completely totally negating the cycle at anytime. If the real estate market existed in a vacuum then I would agree with you that it could follow a predictable pattern. Even then to convince me you would have too:
1. Clearly define what a decrease/increase in price actually is (some set level, say at least +/- 5 percent) and how you derived this threshold.
2. Clearly define some timeframes for a phase. Example: have monthly prices show a decline for a minimum of 3 months but no greater than 6 months.
3. Establish what timescale you are examining data on and stick to it. Annual, monthly? As I said before, you can’t say: “2003 looked like phase 2 and 2004 looks like phase 5, therefore phase 3 and 4 had to have occurred during those 12 months…”. You are examining data on an annual timeframe, therefore any monthly changes would be “washed out” in the year end numbers. I have about as much evidence for phases 3 and 4 occurring as I do for the existence of the Loch Ness monster or aliens.
Simple things like cycles are easy to understand, this is why we all have a tendency to use them to explain complicated things like housing prices. I strongly agree with you economy, mortgage rules and interest rates play a way greater role in explaining why housing costs as much as it does, as opposed to taking a poorly constructed, poorly defined model and fitting it into a dataset.
One more thing, please don’t turn this into a real estate agent bashing thread. I commend Sara and Sheldon for posting information like this up to provide some debate!
Cheers,
OB
Seasonality is a huge factor… interest rates are below baseboard… oil is up…
Then WHY?
Price Reduced! $308,000 4829 32 AV
2 Bedrooms, Status: Active. Residential
Price Reduced! $389,900 3558 MCLEAN CR
4 Bedrooms, Status: Active. Residential
Price Reduced! $399,900 1157 RUTHERFORD CL SW
4 Bedrooms, Status: Active. Residential
All a price reduction means is that the seller was, and may still be, asking for too much money.
At the end of the day a house is only worth what someone is willing to pay for it.
I’d say the price changes are because the house is not selling in a timeframe that meets the sellers expectations. If I am motivated to sell I may lower my price while someone may wait longer and get a better price. A few price reductions isn’t a macro trend.
Sometimes people willing to wait get higher prices, in fact Realtors when listing thier own properties typically wait longer and get a better price according to the UChicago study referenced below. Fact is price changes are often related to the level of motivation sellers, and not market conditions.
http://home.uchicago.edu/~syverson/realestate.pdf
It’s going to an interesting spring.
So, given Nicolai Kolding’s real estate cycle hypothesis (as stated), lets use it to make some predictions for the next year. Then we can re-visit the predictions that the hypothesis made and see how it did. A fundamental aspect of a scientific hypothesis is that it makes predictions. If those predictions come true, it may someday graduate and get to be called a theory.
I can’t remember where I first read about this, but some economist gave advice to another (regarding public predictions): if you give a number, don’t give a date. If you give a date, don’t give a number.
Let’s be specific here.
Sara: you gave dates and relative values in your earlier post. Could you save us the time of digging up the historical stats and declare a prediction (the sales volume and $/sqft or median SFH / condo prices, however you want to do it) for May, June, August, September this year?
I did, in the original post, last paragraph:
So, based on this analysis of the sales cycle, what I know happens seasonally, and the predictions we made at the beginning of the year, I think we will see sales start to beat last year’s numbers around May or June, and prices should be ahead of last year by August or September.
I saw those predictions.
What I was looking for was for you to dig up the historical numbers for those respective months and re-state the prediction with numbers rather than just referencing the months.
Sara, your predictions are based on what?
In theory if we compare last year vs. this year:
- 2010 – 35 years mortgage vs. 2011 – 30 years mortgage – so for the same income you can afford less house…5 years less mortgage
- Last year the lowest interest rates in history.. this year the interest will start to go up – again you can afford less house for the same income.
I can continue but for the first time it is enough. If we need we can add later.
So only basic math in these affirmations.
What are your arguments other than speculation and also affirmations to scare people.. buy now or the prices will go up..
As I said in my blog post: “So, based on this analysis of the sales cycle, what I know happens seasonally, and the predictions we made at the beginning of the year…”
Look at the chart for sales I posted at the beginning of the month. Sales peaked in May last year and had a long slow death after that. Is it really that revolutionary to predict we will see more sales this year in June than last year, when we know sales normally peak in June, we know the job market is improving, and we are almost certain Alberta is going to do way better than the other provinces?
Prices are going up, look at the weekly charts. They’ve gone up $20k since January. I’m not making that up, and nowhere have I ever said buy now or the prices will go up. Whether you buy now or later will have NO affect on prices, they are going to go up and down and up and down for as long as there is a real estate market.
The only opinion I have ever had about the right time to buy, is that it is when it is right for you.
Wow the mark was really missed with this real estate market cycle model. I would recommend hiring a statistician to describe the basics of trends and models. Throw in an economist while you are at it. Why do you think a one day course will give you the insight needed to professionally speak to this, Even a Realtor needs more than a brief intro course to do their job (which is a lot easier than predicting cycles in the real estate market).(which is a lot easier than predicting et).
No, the mark was not missed. I did not invent this cycle after one day in a market analysis course. An analyst created this way of looking at the data after studying all the data on record and determining how the cycle works. All I did was plug in our data for Edmonton. Perhaps what you need is to take a course to understand it, since I couldn’t explain it in a simple blog post. If you understood it, it would make sense. It is a way of taking the temperature of our market, and giving a sense of what is coming. It does not tell you when changes are coming, or how big they will be, or how long they will last. But when you use this tool, in conjunction with all the other information you can gather about a market, it can help you make predictions.
Sara, prices went up from January for 2 reasons.. one was the 35 to 30 y reduction for mortgages and second because the homes which were more expensive lowered the prices.. So is it correct to say the price went up if a house was last year on the market with 600k and it is sold for $550k which is higher then the average? And I might agree with higher prices but only in a situation when the high end real estate will lower the prices by at least 10%. We’ll see.. Interesting is the fact that in Calgary we see higher prices in January and February and now they already started to get lower again… It might be the 35 vs. 30 y theory.
Second, improved economy doesn’t mean higher salaries.. Improved economy means, the balance sheet of the company looks better.. do you know how? Better balance sheet means a company does better with less people.. it is more efficient and also if you need a new position, you don’t do the mistakes from 2005-2008 to offer high salaries just to get somebody.. that time is gone for 25 years… You have now tones of applicants so there are no issues to choose.
On top of these, over 70% ownership rate, highest debt per household in history will also be factors to contradict your theory.
When you do statistics you consider a longer period of time. We saw the boom but not the bust yet… Check the Real Estate or what market you want stats and you will see both boom and bust are part of the cycles.
Look for the elections also.. I hope we don’t see again a National Energy program as in 80s.
Did you consider the fact that prices went up from January to now because that is what prices typically do in the first quarter? Everything else you said makes almost no sense.
I forgot also to comment your opinion: “The only opinion I have ever had about the right time to buy, is that it is when it is right for you”…
The right time to buy something is when it is correct priced. It doesn’t mean if I have money I have to buy a 97 Ford for $50k just because I want a car…
Seriously? That’s how you interpret my comment? No wonder you don’t get the concepts in this article.
If we learned anything from last year, lending changes, we are just starting phase 5. Most of the consumers were most likely rushed into a house purchase. Prices were slightly raised (due to demand) and the rest of the year will be missing the newbies (home buyers). I actually would like to suggest phase 5 will be long (due to the fact buyers set the price not sellers). Economic fundementals my house rich friends.