Historical Prices - Edmonton Housing Bust

Category: Historical Prices

Thursday, May 28, 2009

Price to Rent Ratios

Filed Under: Historical Prices, Rental Market

Obviously there are no lack of indicators and ratios out there, we’ve looked at several here over the months. Some in my opinion are fairly useful, like absorption rate… others I feel are as useless as tits on a bull, like sales to new listings.

In any case, today we’ll look at another one… price to rent.

I briefly took a look a historical rents awhile back, but have since came across a better data set (well, same set actually but a longer horizon, and in cases like these, the more data the better). To set the groundwork, here are the rents for Edmonton and Calgary going back to 1990.

The two cities have tracked fairly similar patterns. Rents were relatively stagnant in the early 90′s, then in 1997 they started about a 5 year run up then plateaued again through 2005, after which they again took off. These figures are from October of each year for those curious.

As the name implies, the other part of the price-to-rent ratio is price. Creative name, I know. Anyway, here are the residential averages and condo averages for each city taken from October of each year.

Now our rent variable is that of 2-bedroom apartments, so obviously the condo average would be the better comparison other. So why did I include the residential average, well, it’s because I only had Edmonton condo averages going back to 1999, so I had to do some estimating to fill in the ’90-98 portion.

For the three years, ’99-01, the condo average was generally around 70% of the residential average (FWIW, in recent years that ratio has been closer to 75%). So going backwards, I used that proportionally. As prices were very stagnant over that period I felt that was an acceptable method… but if you disagree, I have also done these ratio’s using the residential average.

No such issue with the Calgary number as I had the full set. It is interesting to note that the Calgary condo numbers track extremely close to the Edmonton residential number. Also of note, the Calgary condo average is usually closer to their residential average (85%) then in Edmonton (as discussed in the prior paragraph). Why is that? I’m not sure, but perhaps condos are more prevalent in that city and thus make up a larger portion of the market.

Here we see the price-to-rent ratio for condo’s to two bedroom apartments in our respective cities (or, in case you didn’t like how I derived this as explained earlier, here is the graph using the residential average to two bedroom apartments, in any case, they yield very similar curves).

How this is calculated is to take the price, then divide it by the monthly rent multiplied by 12 (in other words the amount of rent paid in a year). Interestingly we can see that Calgary generally seems to track 2-3 points higher then Edmonton. So while Calgary prices and rents are both higher then those in Edmonton, their differences are not proportional.

Looking at the trend, through 2001 in both cities the ratios remained relatively consistent… but from ’01 through ’07 they steadily increased at a rate of around 2 points per year. A reflection of the big run up in prices, but interestingly this trend was very much evident long before prices really exploded in 2006.

We can also see the effects of the drop off in price, as the ratios for both cities fell off to the tune of about 4 points in 2008, reflecting the drop off in price while rents continued to go up as evident in the earlier graphs.

Both these upwards and downward movements display how rents are generally more ‘sticky’ then prices, in both directions. This should not be surprising though as sale prices concern only one-off transactions that take place that month, while rents are determined by ongoing agreements/relationships/leases that can be made months or even years earlier.

One of our regular commenters Chris, aka CM, mentioned a good article concerning the price to rent ratios witnessed in many bubble cities in the United States at their peaks. One passage reads.

Throughout the 1970s, ’80s and ’90s, the average rent ratio
nationwide hovered between 10 and 14. In the last few years, though,
it broke through that historical range and hit almost 19 by the time
the housing market peaked, in 2006.

And while home prices — and rent ratios — have always been higher on
the coasts, they reached whole new levels recently. In the Washington
area, the ratio went above 20. In Boston, New York, Los Angeles and
south Florida, it topped 25. In Northern California, it approached 35,
higher than it had been in any city, at any point on record.

So it’s interesting to note some of those figures with our findings above. Firstly, that the nationwide ratio in the US was generally between 10-14. This would certainly describe Edmonton’s situation until the run up… and would largely also ring true for Calgary though they were closer to 15, so they were close.

In any case, Calgary would be at the high end, while Edmonton would be right in the mix. So it’s interesting to compare the peaks, Calgary at 25.4, and Edmonton at 22.9. These ratios are very close to those experienced in Boston, NY, LA and south Florida.

While all four of those US cities experienced bubbles, those experienced in LA and south Florida were far more extreme, and both have suffered very badly in the aftermath of the bubbles bursting. This could also indicate trouble for us, as the magnitude of the bubble experienced in Edmonton is much closer to those witnessed in LA and south Florida moreso then Boston and NY… the bubble in Calgary was somewhat in the middle ground, significantly lower then the former two, but much high then the latter.

Thus, our price to rent ratio is another indicator of just who severe a boom we experienced… the question is, how bad will the bust be?

Also just want to do a quick appendices on another measure I found. These are the rented accommodation indexes, and owned accommodation indexes for Edmonton and Calgary.

These are part of StatCan’s Consumer Price Index. Not entirely sure exactly what and how many factors are included in these measures, but all things being equal, I thought they were worth a quick look.

It’s interesting to see how they’ve evolved since 1971. For both cities the two figures have tracked very closely up until 2006 when the rapid run up of real estate prices caused a significant decoupling. Rented actually have largely continued their traditional course, but Owned underwent a massive spike, to a magnitude never witnessed before.

It’s also interesting to note that in both cities in the early 70′s that the Owned index was actually significantly lower then the Rented index.

15 comments

Thursday, April 23, 2009

What is a ‘bust’?

Filed Under: Historical Prices

So, just what is a ‘Bust’?

Don’t google that.

The first hits you’ll get will probably be of the Russ Meyer-esque definition… and don’t get me wrong, big fan of that variety, but looking up that sort of thing while others are around can invite trouble from HR, or even worse, the MRS.

I’m talking about market busts, and in this case, housing market busts… and for the most part, these are safe for work.

There is no one universally accepted definition… which is also true for ‘booms’ (or ‘recessions’, or ‘depressions’, as we’ve seen of late), so that’s probably why there is such spirited debate about such things.

I did some googling of my own, and found all sorts of numbers thrown around, but it seemed most were in the 10-15% range peak-to-trough.I ended up stumbling upon a couple pretty good long term studies that I think I’ll go with.

One was from the IMF and published in 2003. It studied 20 busts from 1970-2002 in 14 industrialized nations. It’s findings were that to qualify as a ‘bust’ the peak-to-trough contraction must be at least 14% in real dollars (inflation adjusted).

Also perhaps of interest, on average they lasted about 4 years, and the average correction was 30% (again, in real dollars). There is actually a lot of really interesting findings in there as it compares equity to asset bubbles, and it’s quite prophetic when you compare it to what’s happened in the U.S. of late.

The other study was by the FDIC in the US, and was published in 2005. It studied US markets from 1978-2003 and they concluded that to be a ‘bust’ the peak-to-trough contraction must be at least 15% in nominal dollars.

This may not sound like a big difference from the IMF definition, but the distinction between real and nominal dollars is actually a significant one. Which I’ll display later when I compare our current situation with the bust in the 80′s. Personally I’d lean towards using real dollars, especially over the long term, but there is an argument to be made for the use of nominal dollars, particularly in the short term.

First we’re going to take a look at the IMF definition. Here we see the various price measures here in Edmonton and their percentage declines since peaking respectively when adjusted for inflation.

From this graph we can see that for all measures have satisfied the IMF definition of a ‘bust’ already… so, if you didn’t think the name of my blog was correct, how are them apples!

Now the FDIC definition. This graph yields an interesting observation… both condos and single-family homes have surpassed the threshold and are therefore busts… but somehow the overall residential average (which those two make up), has not…. how does that work?

Basically the answer is proportion. When prices were peaking, the ratio of condos/SFH’s selling was much higher for whatever reason *coughspeculationcough*… since then the ratio has dropped, thus, more SFH’s and fewer condos are counting towards the average… and when you have more higher priced units selling and fewer lower priced ones, it’s brings up the overall average.

So, even though the residential average hasn’t passed the threshold just yet, everything that makes up the average actually has. So, by the FDIC measure too, Edmonton qualifies as a bust.

Clears as mud? Good. And we wonder why so many people hate statistics!

Now, as promised, a comparison with the 80′s bust. Since we’re were just talking about nominal prices, we’ll stay there and do them first. To compare apples-to-apples, we’ll stick with talking the residential average as I don’t have the other price breakdowns going back that far.

Here we see the 80′s bust took about four years to go from peak-to-trough, and the total contraction was 29%. We can also see that our current bust is tracking very closely thus far.

Also interesting to note the volatility in the prices. The overall trend is down, but there are a ton of spikes along the way. Evidently there is a lot of bouncing left in a dead cat. So, for those eager to call bottom don’t get too excited… even a couple or three consecutive strong months are not unusual, and can be quickly erased. Bottoms are only revealed long after the point.

Now the comparison using real prices. As we can see, by this measure the 80′s bust was much longer and much deeper.

Took 100 months before finally hitting bottom, over 8 years (though I suppose one could argue that it effectively hit about two years earlier, the absolute bottom wasn’t struck until month 100), and the total contraction came in at a whopping 48%.

We can also see that our current bust appears to be tracking faster, passing the 14% threshold two years earlier.

An interesting note is that for the 80′s bust, the real peak actually occurred two and a half years before the nominal peak. Real prices had already dropped about 13% before nominal prices peaked. This could happen because at the time there was very high inflation, and also something of an extended plateau in price.

This is in stark contrast to our current peak, where inflation was relatively low and there was no plateau to speak of… thus real and nominal peaks were simultaneous.

In conclusion, it appears that by most measures we are not just heading towards a bust, but have already arrived. And it’s not one of those things that bouncing back above the threshold undoes, once you cross the line, it’s done.

Like virginity in many ways, you can argue about the criteria, like it should be nominal dollars not real… or I was drunk, and it was Mexico… or it should be 14% not 15%…. or it was just the tip, just for a sec, just to see how it feels… but wherever you draw that line, once it’s crossed, it cannot be uncrossed.

34 comments

Tuesday, April 21, 2009

Population and Migration

Filed Under: Historical Prices, Macroeconomics

I’ve been meaning to take a look at this for a while and finally got off my fat ass and did it… or I guess technically, sat on my fat ass and did it… um… I digress.

So, today I’m going do another reader request and to take a quick look at population/migration and it’s effect on real estate prices. Many speculate that it was high migration and growth that not only triggered the run up in prices, but also argue that it should sustain that level of prices. We’ll see if that theory stands up and earns my unpatented stamp of approval, or not.

Lets start with population. Here is a look at Edmonton’s reported population from 1962 to present, charted against the inflation adjusted residential average price in Edmonton.

As we can see there, the population curve is quite gentle and consistent. Contrast that to the price curve, which the late 70′s bubble is quite pronounced, as is the spike in prices witnessed of late.

It could have likely been deduced from the first graph, but here is some further derivation to back up that overall population growth doesn’t appear to have much impact on prices. Prices are very volatile, and largely appear to move independent of population.

Included in this graph is the percent change of both Edmonton and Alberta populations, I didn’t include Alberta’s population in the first for scaling purposes. Alberta’s, like Edmonton’s population, generally fluctuates between 0%-5% growth year-over-year (FYI, the spike in Edmonton’s population in 1964 was due to their annexation of Jasper Place).

Thus we will conclude that overall population growth has no direct relationship with residential real estate prices. Now, onto migration.

That is a graph of net migration into Alberta since 1962, and it’s two primary components, interprovincial and international migration. As we can see, interprovincial appears to be the primary driver of net migration. From here on out we’ll just deal with net migration, I just wanted to give you all an idea of what how international and interprovincial figures stacked up.

Now here is a graph of net migration and year-over-year price change (bear in mind those migration figures are for the entire province, not just Edmonton). If there was a relationship between the two, one would expect that migration would be the leader, but as can be seen in 70′s through the 80′s it was actually price that lead through the first bubble. We can also see that through the 90′s, even with fairly good migration, prices really didn’t seem to respond.

There are some periods that may appear to support a relationship, like in the late 80′s and in ’05 and ’06, but on a whole it doesn’t appear there is any direct relationship. It’s also interesting to note that the recent spike, which reached record level in 2006, still doesn’t appear to be anything exceptional, and migration retreated quickly in 2007 and 2008 back to historical norms.

This leads me to conclude while there is likely some short term effect of migration on price, but in our recent boom its effect was probably over blown. I would hypothesize that the shortage of inventory and rental units experienced was more rooted in speculative buying then actual increased demand.

As we’ve seen an explosion of listings since the peak, particularly of condos and smaller homes, largely of which have been sitting empty. This appears to support the hypothesis that we saw a great increase in properties bought, and then held to be flipped at a later date.

This sort of behaviour was also evident in the rental market, where the number of total units dropped from 75,267 in 2005 to 67,907 in 2008. These were largely due to condo conversions. That kind of contraction will of course increase demand, even if the population increases. The number of occupied units actually only increased about 775 from ’03 to ’06, then dropped by over 4,000 in ’07 and another 3,000+ further in ’08.

Beyond that, our current situation where we find the city seriously overbuilt and with another 10,000 units in construction. We can’t get to that point without demand being perceived as much higher then it is in actuality.

So, while a rapid increase in migration could likely pinch supply short term, the escalation of prices we experienced here was more rooted in speculation and good old irrational exuberance.

And as far as long term price support goes, not a chance. As I’ve said before, Edmonton isn’t Manhattan, we can build for as far as the eye can see in every direction and generally it’s not any problem getting materials or labour… any short term situations where demand outweighs supply, will quickly be corrected… or as we’ve witnessed, over-corrected.

At least that’s my opinion.

EDIT: Had a request for the graphs of change in net migration vs change in price, so here they are. I had already prepared them, but they got cut from the original post to keep the length down.

17 comments

Wednesday, April 15, 2009

The ‘Burbs

Filed Under: Historical Prices

I was doing a quick search of Google News local real estate stories, and stumbled upon this story from Sherwood Park talking about how prices had dropped $120,000 since their peak in July 2007.

They appear to be privy to some figures that are not publicly available, so I can’t speak about that claim. I’ve been trying to piece together as much data as I could on the prices in the suburbs, and while I’ve gotten a fairly complete collection, it’s still a bit patchy at this point. So, won’t be any graphs in this post, but hopefully by this summer I’ll be able to at least do some good ones for decline from peak.

While I don’t have the complete data, I do have enough to offer up some interesting examples.. and it gives me an excuse to reference an flick from Tom Hanks bad comedy career phase. So, in some cases the peaks I note may not be the absolute peaks reached by the market, but they are close as I have all the second half ’07 figures, and the Edmonton market peaked in mid-07.

So I’m going to take a quick look at the respective bubbles, first the prices five years ago (March 2004), the highest value I have for them in 2007, then their values as of March 2009.

St. Albert
March 2004 Average – 226,074
March 2004 Median – 209,900

Sept 2007 Average – 497,380
Sept 2007 Median – 477,500

March 2009 Average – 376,944
March 2009 Median – 354,500

St. Albert saw prices more then double during the bubble, and since then have seen prices both average and median prices drop over $120,000 respectively, roughly 25%.

Sherwood Park
March 2004 Average – 235,700
March 2004 Median – 227,725

August 2007 Average – 468,400
March 2007 Median – 439,000

March 2009 Average – 400,074
March 2009 Median – 390,000

Sherwood Park has actually had a surprisingly good March, as in January the average and median prices were 376,545 and 353,500 respectively. Even with that they have seen average price drop almost 15%, and median 11% since peaking.

Spruce Grove
March 2004 Average – 185,417
March 2004 Median – 173,899

Sept 2007 Average – 420,990
Sept 2007 Median – 416,000

March 2009 Average – 328,509
March 2009 Median – 314,500

Spruce Grove is another entrant in the six-figure club, seeing their median price already fall $101,500 since the peak. Their average had also passed that mark in December, but has since rebounded a bit. Spruce Grove was also the bubbliest of the larger ‘burbs (populations over 20,000), seeing their median price go up 139% and average 127% during the bubble.

I also have notes on some of the smaller suburbs, so instead of typing all that out, I’ll just give you some of the overall highlights.

Bubbliest Suburb (Percentage) – Morinville – Average +166%, Median +183%

Bubbliest Suburb (Dollars) – St. Albert – Average +271,306, Median +267,600

Biggest Percentage Decline – Stony Plain – Average -25%, Median -35%

Biggest Dollar Decline – (Average) St. Albert -120,436 (Median) Stony Plain -149,000

So while the city/region as a whole hasn’t eclipsed the 100K mark in decline from peak, at least four of it’s suburbs have the dubious distinction, St. Albert, Spruce Grove, Stony Plain, and Morinville. There may already be more in fact, but as I said, my dataset is not complete just yet.

In any case, I figured this information might be interesting to some out there. It’s also very relevant since the suburbs are so intertwined with the city and make up such a large part of the CMA population.

7 comments

Monday, April 13, 2009

Seasonality

Filed Under: Historical Prices, Inventory, Sales

Hope you all enjoyed your long weekend. If it was anything like mine, it was likely full of sun and way too much refined sugar.

If you’ve been reading the releases coming from the various real estate associations, you may have noticed they are now talking up the month-over-month sales and price figures more so then they have in the past.

Traditionally they usually focused on year-over-year figures… but as those are coming in as declines as of late, they are referencing those less and less. Something of a marketing move, as they’re in the business of stimulating sales, and hearing about prices and sales going down doesn’t make their message very receptive to buyers or sellers.

As a result of this, it’s also increasingly being pointed out that month-over-month gains are actually normal in the spring, as real estate for whatever reason has a lot of seasonality. So today I’m going to take a quick look at the traditional seasonal movements of prices, sales and inventory.

The 100 value is the relative value in January… and from there it behaves like percentages. So 110 = 110%… so if the price was 200,000 in January, that value would be equal to 100… so 110 would be 110% of the January price, so 220,00.

Here we see how prices typically move through the year. The blue line is the historical trend, and as you can see it usually shoots up about 7% in the first half of the year before leveling off, then drops off a couple points in the summer, and ends the year up about 5%.

We can also see that over the last year we’ve severely deviated from the norm. In 2008 it tracked normally through March, then fell off slightly through spring and then really dove throughout the second half… and 2009 thus far has been an extension of that.

Here’s a look at how prices behaved during the boom relative to historical norms. In this case normal seasonality barely registers as a speed bump, and while the boom was on through mid 2007, normal seasonality was no predictor what-so-ever. Which has also been true thus far in the bust.

Basically, seasonality is no indicator of prices during a bubble. That said, it is noticable that during the downturn, spring still does appear to be the strongest period, and falls off from there. Which seems to indicate the second half of 2009 could witness a very big decline.

Here we look at sales. Similar pattern, but in much larger proportions. Sales continually improve through May (typically more then doubling the January tallies), then going back down the rest of the year.

In 2008 we can see that it tracked a closer pattern in sales then prices. This also holds true though the boom, so while the scales may change, the general seasonal pattern remains.

It’s also notable that 2009 appears to be gaining very strongly, but this may be more of a result of an overly poor January. Again though, it is tracking quite closely to the historical seasonality.

Finally we’ll take a quick look at inventory levels. Again we’re seeing levels track fairly normal curves, 2008 went a bit higher, and 2009 is shaping up to be lower.

On the other hand, during the the boom, and the onset of the bust it looked quite different. In 2006 inventory actually tracked the opposite pattern, and declined through to the summer. This obviously caused a inbalance of supply and demand, and really fed the skyrocketing of prices.

Then in 2007 it was a very different story when inventory tracked normally, then picked up, then did some skyrocketing of its own. Which then caused a glut of inventory, and halted then reversed prices. Obviously a very volatile period there in 2006 and 2007.

So, as we can see, we should not be surprised by sales increasing in the spring, and the real estate boards out there should enjoy trumpting month-over-month gains while they can, cause come summer they’re going to have to find something else as both month-over-month and year-over-year figures will be poor.

As for some April predictions, I’ve been working on some different models but they’re hardly perfect… but I’m a glutton for punishment, so lets throw out some numbers and see how I do.

As mentioned before during a bubble prices are kind of anyones guess, but I imagine they’ll hold their prices in April. Active Inventory I’m going to say will end April around 7,900. Sales I’m going to peg at about 1,700 for April.

Though that could be a ways off, on one of the local agent blogs they are citing the EREB claiming 574 sales already this month, which by my calculations would put us on pace for almost 1,800 on the month (there were 1,830 last April)… though in the paragraph prior they talk about how much things have slowed sales wise and gave quite a low tally for the week.

So, in conclusion… who knows?! Time will tell.

13 comments

Sunday, April 5, 2009

Oilberta Redux

Filed Under: Alberta, Canada, Commodity Prices, Historical Prices

Earlier this week I did an entry on the relationship between oil prices and housing prices in Edmonton, and there were some interesting findings which led to new hypothesis’. There was also some good ideas that came up in the resulting discussion.

So tonight I’m going to take a bit of a deeper look at possible relationships with housing prices… or maybe I’m just doing it because that last one got lots of comments and I’m an attention whore… it’s really hard to say.

In any case, beyond just oil we’re going to look at natural gas, stock markets, and several other possible indicators. Be forewarned, there is going to be a ton of charts and graphs in this one, so if that isn’t your thing, well, you’re probably on the wrong blog. Anyway, this one will be a big’uns.

Like the last entry, I’m going to use Toronto as my control sample. Partially because Toronto isn’t exactly known for oil and gas (well, other than hot air… I kid, I kid), and it’s also the city I have the next best data set on.

This time we’re going to be using yearly averages rather than monthly, so you’ll notice the graphs look a bit different. As real estate tends to be a bit of a lagging indicator and prone to a fair bit of seasonality, yearly may actually be the better measure. I also have the yearly averages for Toronto going much further back, so that should improve the findings.

And here we have a plot of oil (West Texas Intermediate) and natural gas (Henry Hub) over the same period, also yearly. You’ll notice they chart fairly similar paths. You may be thinking to yourself that this graph doesn’t show prices getting as high as you recall, or diving last fall, but remember, these are yearly averages and/or spot indexes. These won’t look nearly as volatile as daily, weekly or monthly figures… for example here is the the monthly figures semi-transparent over the yearly ones.

As discussed in the prior entry, oil and home monthly prices in Edmonton had a high correlation… but they also had a very high correlation with Toronto prices. Which led me to conclude it was a spurious relationship.

These correlations remain using the yearly figures over a longer term… and are also present with natural gas. Though that shouldn’t be surprising, as noted earlier, oil and gas followed quite similar paths and actually have a correlation of 0.89 from ’72-’08.

Over the full term, ’72-’08, Edmonton had a correlation of 0.90 with oil, and 0.86 with gas. Toronto came in at 0.71 with oil, and 0.82 with gas. While the gas values came in quite close, there was a fair difference in oil.

This appears to be a timing issue though, as over the last 10 years both cities came in at 0.89, and over the last 20 years Toronto came in at 0.91 and Edmonton at 0.92.

On the natural gas front Toronto actually came in higher over the shorter terms, 0.84 to 0.79 over the last 20, and 0.80 to 0.63 over the last 10. Quite a large difference there on the latter.

As a result, I would again hypothesize that the correlations are due to a spurious relationship, due to a common lurking variable.

Here are a couple scatter points to illustrate the what we’re looking at.

You’ll notice the R² value is lower than the correlation. For those unfamiliar, R² is the coefficient of determination, and is the correlation squared (so, obviously its symbol would just be R).

In those graphs you can see the general trend formed over time. The higher the correlation, the tighter the plot points correspond with the trendline.

You may be wondering if there is a relationship between year-over-year gains… and to answer that in a word… no. Regardless of the city and the commodity (or any of the multitude of other indicators discussed later for that matter) any relationships are negligible at best.

To give you an idea of that, lets contrast the above graph of Edmonton vs natural gas prices to Edmonton’s year-over-year price change plotted against those of natural gas. You’ll note in the earlier graph that the points are often grouped and you can see an overall trend.

Now note in this year-over-year graph that the points are all over the place with no apparent rhyme or reason. The R² is also almost 0, which just reinforces that there is little or no relationship.

As I have said before, I think the relationship between oil/gas and real estate is a spurious relationship due to a common lurking variable… and I’ve theorized that variable could be the greater economy and/or financial markets as a whole.

To further explore that line of thinking, I compiled a couple spot indexes for the NYSE Composite and S&P/TSX Composite. Here is a look at how those have performed over time.

These being composite indices they give you a good idea of the overall stock markets. With the NYSE being the largest stock exchange (by dollar value) in the world, and TSX being the dominant one in Canada they should be pretty good indicators for our situation.

The findings also may support my earlier musing, with both having significant correlations with real estate prices in both Edmonton (NYSE-0.87 TSX-0.92) and Toronto (NYSE-0.88 TSX-0.89).

To go a little deeper even, I consulted a Statcan report of the leading business indicators. These include ten different measures (including the S&P/TSX index) from various sectors of the economy.

Here is a list of the others, average work week-manufacturing, housing index, United States composite leading index, money supply, new orders-durable goods, retail trade-furniture and appliances, durable goods sales excluding furniture and appliances, shipment to inventory ratio, finished products, business and personal services employment.

Here is a look at just a few of those and how they’ve charted out.

I’m not going to look any deeper into these specifically, I just wanted to give you guys an idea of what they graphed out like. All the measures listed above appear to have a decent correlation on one level or another with real estate prices, except the average work week one which didn’t appear to have any correlation.

What I found most interesting was the composite of those ten different figures. As it covers many different sectors of the economy, it gives you a measure of the economy as a whole. As such, I’m going to focus primarily on that single measure, and here is how it graphs out.

And here is a scatter plot of the composite vs Edmonton real estate prices.

You’ll note that the points all plot fairly close to the trendline. Also interesting to see the snakelike pattern, which is present in the the natural gas plots earlier, but are not as clear as in this one.

Here is the time series for the leading indicators and Edmonton’s real estate prices. We can see there is far less variability in the composite than in housing, but they both are clearly trending up. This also reveals itself in a high correlation between the two, 0.92. The relationship also remains very strong regardless of the term, something that cannot be said for oil or gas which vary from very strong to moderate depending on the period.

Here we see the same composite graphed with Toronto prices. Again they look quite similar, and real estate remains the more volatile. The correlation for these two come in at 0.90.

FINALLY, we’ve arrived at the last graph. Here is the Edmonton prices plotted out with the composite, oil and natural gas prices for comparison. Yes it’s a bit of a mess, but I tried to make it as understandable as possible considering the multiple axis’ values. But I think it’s a good visual anyway.

I’m sure some will take exception with the scaling, but I tried to be as fair as possible and match up the various lines with housing prices long-term as best as possible. So just as a disclaimer, scaling can be misleading, so take it for what it’s worth and remember these can be made to appear to back up any conclusion.

Alright, here we can see the relative volatility of the four indicators. Oil and gas being the most so, then real estate, and finally the composite which is very smooth and relatively straight. It’s this smoothness is probably what yields the consistently high correlations, as all the prices increase over time.

I suppose one can see any number of things from the graphs, but I feel that these findings largely back up my earlier hypothesis that housing prices are probably more closely tied to the overall health of the economy, than to individual commodities.

While positive correlations exist between home prices and oil/gas in Edmonton, the same correlations exist in Toronto, a market without such resources close at hand… therefore implying a spurious relationship between the two factors.

While oil and gas certainly effect financial markets and the economy as a whole, throughout Alberta and Canada, it is still but one factor, granted one of the bigger.

That said, I’m not sure a direct causal relationship between be proved between the economy and real estate either. The economy is too complex for even an army of economists with lifetimes to study it to truly understand… much less one half-assed blogger.

So again, I conclude that while oil and gas certainly has a big effect on the local and national economy, I feel they do not have a direct causal relationship with real estate prices. But I don’t think any single indicator can claim a direct causal relationship. Real estate prices are effected by hundreds, if not thousands of variables… earnings, supply, demand, seasonality, land, labour, materials, just to name a few.

Oil and gas prices would certainly effect several of these, but there is a limitless interplay amongst these variables and countless others, and the effects of all these just cannot be quantified in an acceptable fashion.

That the same correlations exist between oil and gas prices and cities in both producing and non-producing regions indicate that real estate prices drill a lot deeper. Pardon the pun.

6 comments

Tuesday, March 31, 2009

Oilberta?!

Filed Under: Canada, Commodity Prices, Historical Prices

One of our frequent commenters, Two-Thirds, offered up a couple local real estate folklore that he’d like to see tackled… his wish being my command, here is the first, the relationship between oil prices and real estate prices in Edmonton.

Those that have been reading this blog for awhile have already seen graphs of Edmonton’s historical prices several times, so we’ll start with a look at historical oil prices instead. For this post I’ll be using a spot index of West Texas Intermediate Crude… that seems to be the most commonly cited oil price, so should be a good standard.

To give you a better idea of the prices, here is the inflation adjusted price, and in Canadian dollars. As you can see from the graph, oil prices can be pretty volatile and undergo some very big swings, quickly.

It only goes back to 1971, this is because that is as far back as I could obtain exchange rates for… but that’s okay, since as you can see from the prior graph, prices were pretty much stagnant before the ’73 Oil Crisis, and Edmonton house prices were also pretty stable up to that point anyway.

It’s also good because creating these graphs over such long periods makes my year old iMac behave like a 486 trying to run Quake.

Oil Prices vs. Edmonton Residential Average

Now here is a look at how oil prices chart against Edmonton’s residential average price going back to 1971. We can see they have somewhat similar patterns, but it doesn’t appear that Edmonton’s real estate prices are nearly as reactive to oil prices as some may think. It is hard to say though, as real estate is something of a lagging market, not nearly as reactive as the oil market.

Realistically it takes time for the benefits of higher oil prices to makes its way through the economy. It takes months, if not years, for new projects to get off the ground, and the money from that to circulate.

So, when looking at the first boom in the 70′s, it could be argued that the rise in prices was, at least in part, due to the big spike in oil prices in January 1974. That delayed reaction could also explain why there was no apparent effect on real estate prices after the further spike in ’80 when prices briefly eclipsed $120/barrel (2009 dollars) then started shooting back down.

Also as oil prices started to move up in the late 90′s, real estate prices again started to creep up by the early 00′s… but real estate also started to decline while oil prices were still rocketing up too.

To counter that though, we can also see that real estate prices were declining during a period when oil prices, while dropping, were still well above what they were in the mid 70′s when they may have triggered the boom. Of course there were external factors at play at that time, like the NEP, which effect would be extremely difficult to quantify.

Then there is the big drop around 1986 when the price of oil plunged over 60%, and stayed there… while real estate prices seemed to have no effect, delayed or otherwise.

So, what’s it all mean? Hard to say, I guess one can see in those graphs what they wish.

To take a more statistical approach, we can take a look at the correlation between the two… this actually yields a seemingly remarkable result… a positive correlation of 0.68 since 1971. Anyone familiar with the measure knows that actually indicates a significant relationship.

But there could be many different factors at play, so to get an idea of what kind of correlation is normal I decided to also run the numbers against a control city that isn’t generally associated with oil and therefore one would not expect to find such a correlation… in this case, Toronto.

I only have the full numbers for Toronto going back to 1995, so to compare apples-to-apples as best as possible, I re-ran the number for Edmonton over the same period. Here are the graphs of those.

Oil Prices vs. Toronto Residential Average

If you were shocked by the high correlation between Edmonton prices and oil prices earlier, you haven’t seen anything yet. Edmonton from 1995 to now is an astounding 0.87. Seems like that would make the relationship a slam dunk!

Not quite it seems. Sit down for this one. Toronto during the same period had an even higher correlation with oil prices… 0.89.

The two are very close, and this stays true (though in lower correlations) for the periods of the last 10 years and the last 5 years, inflation adjusted and nominal. Even figuring in moving averages with terms as long at 3 years to account for lagging reactions, there just doesn’t appear to be substantial differences between the two cities.

So, while a high positive correlation remains, I think that this finding of a non-oil and gas market having as high or higher correlations would pretty much refutes an actual relationship between oil prices and housing prices in Edmonton. Home prices here appear no more linked to oil prices then other cities in Canada.

Edmonton and Toronto Residential Averages

To look at it from another angle, there is an equal correlation between real estate prices in Edmonton and Toronto, as their with between Toronto and oil prices. Though this should not be surprising since both also had similar correlations with oil.

In any case, I would have to conclude any kind of relationship between real estate prices and oil prices is anecdotal at best. It looks to be a spurious relationship caused by some lurking variable(s), and likely present in most if not all Canadian markets.

As is often said in statistics, correlation does not imply causation.

It seems that the real driver of real estate prices has probably more to do with the overall financial markets of which oil is a part of, or perhaps the economy as a whole… which would at least in part explain Toronto having just as high a correlation.

And just for shits and giggles, here is a little measure I derived… basically it’s how many barrels of oil it would take to buy an “average residence” in Edmonton at the market rates.

As we can see, this can be very volatile, with values anywhere from 2,500 all the way to 6,500 not being unusual over the last two decades. The median since 1971 has been 3,835 barrels, with a standard deviation of 1,270 barrels. Such a large range again would make me question any kind of hard relationship between oil price and house price, even figuring in real estate being a lagging indicator.

So in conclusion, while I’m sure oil prices have an overall economic impact on our fair city, of which housing prices would be a part of, from the data I’ve ran I see no tangible evidence of a direct causal relationship between oil and home prices. Any implication of such a relationship appears to be spurious. So, in the absence of any otherwise compelling evidence, this one is…

32 comments

Wednesday, March 25, 2009

Incomes and Interest Rates

Filed Under: Historical Prices, Personal Finances

I had been wanting to do this for while, but could never find a solid set of data regarding historical incomes in Edmonton… but late last night I finally stumbled upon a Statcan report that was my dream come true.

So today I’m going to take a look at incomes, interest rates and prices over the last 30 or so years here in Edmonton. This may get a bit long and be a bit graph heavy, but their are a lot of different angles to look at, but I’m going to try to narrow things down as best as possible.

To get things rolling lets take a look at this graph (like all the graphs, you can click on the image to get a view of a larger version).

Edmonton - Median and Average Economic Family Incomes

Here we see the median and average economic family incomes for Edmonton from 1976 through 2006 in nominal dollars. A fairly steady, gentle curve up as time goes on. We’ll also note the average always is proportionately higher then the median, but that they track very similar patterns. As I’ve mentioned before, I think median is the more reflective stat, so we’ll be using it rather then the average.

Now the same graph adjusted for inflation (2009 dollars).

Edmonton - Median and Average Economic Family Incomes - Inflation Adjusted

It’s striking how relatively flat earnings have been over the last 30 years. It’s also interesting to note this chart picks up in 1976, during the run up of the last big housing boom in Edmonton (as discussed in this entry) and ends in 2006, effectively the run up of the current boom.

Unlike home values which has seen a small growth above inflation, earnings have experienced negligible growth above inflation. You may be asking yourself how we can afford to be paying more for homes when earnings are stagnant… that ties into this next graph.

This graph shows the values of average interest rates on mortgages (5 year terms) in Canada since 1962. As you can see, they have been very volatile and have a huge range. We can also see that historically we’re currently at very low levels… even lower still when you consider todays rates are not represented in the graph (though it is worth noting that in the early-60′s, rates then were in around 7%).

To get back to the earlier question of why we can afford to pay more for homes when earning effectively earning the same income… it’s because interest rates are historically low. As you can see from the graph we’re well below the long term trendline, and have been to varying degrees for most of the last 15 years.

It’s also interesting to note the contrast of interest rates during the last housing bubble to the current one. The last one saw prices increase then hold from about ’73 to ’81… a period when interest rates were climbing. Significantly. Rates were already trending up leading into it, then went from about 10% in ’73 to as high as 22% in ’81.

Numbers like that are unheard of today, credit cards aren’t even into the 20′s! Prices were rocketing up, even when the cost of borrowing was doing the same… compare that to our current bubble when the cost of borrowing was a relatively paltry 6-7%.

Now that we have the interest rates and the prices, we can figure out what income it would take to qualify for a conventional mortgage (25 year amortization) using the 32% Gross Debt Service Ratio (but ignoring taxes and heating costs, so effectively people would actually have to earn more, but for the purposes of this example they’re ignored).

Edmonton - Median Family Income vs Income Required for Financing

This is unadjusted for inflation. We can clearly see the two bubbles, when the income required for financing blows WAY past the median family income of the city. This was also for Single-Family Homes, the same graph for residential average can be viewed here.

It’s interesting that both bubbles appear roughly similar in that graph, but when we adjust for inflation…

Suddenly that boom in the late 70′s-early 80′s looks massive…. and that big spike it almost entirely due to the staggering increase in interest rates, as prices relative to inflation had largely plateaued from ’76 to ’81. Rising interest kept putting more and more downward pressure on prices, and eventually prices gave as affordability disappeared.

In any case, as we can see, eventually prices got back in line with earnings (interest rates coming down also contributing to that) and stayed relatively close for about 20 years or so… until the current bubble took off. It should be noted, the income data is not as current as the real estate prices, hence they cut out two years early.

Another factor to take a quite look at is interest rates and inflation rates… here is a little graph of that (these are yearly averages unlike the earlier graph of interest rates, which were monthly).

Not surprisingly, they track together. Like interest rates were much higher during the last boom, so was inflation. This revelation also reveals that lowering interest rates would have cushioned the fall during the last bust… so even if you overpaid at least you’re going to see your payments decrease in the future and inflation would be bringing up incomes thus improving affordability at higher prices. This is in stark contrast to where we find ourselves now when the prime rate is already effectively at rock bottom and inflation is minimal if not negative.

Since the 90′s economic policies have been trying to minimize inflation, trying to keep it 1-2% a year when possible… contrast this to the last bubble when inflation was anywhere from 6-12% annually. So, while inflation could go back up to those levels, it would also cause interest rates to go up and these effects would largely offset each other.

Which begs the question, what happens this time?

It’s a good question… because prices are still out of line, interest rates are effectively as low as they can go and inflation currently is negligible (and some are even speculating we may see some deflation)… in this situation, prices are the only thing that can give.

Even in the event of inflation because of all the printing of money, interest rates will inevitably also increase… which will still leave prices as needing to make up most of the difference and financially destroy most who bought during the boom when their mortgage comes up for renewal or are on variable rate. It seems there is going to be no cushion for falling home prices as they get back in line with earnings.

Ultimately it’s really hard to say what’s exactly going to happen in the long run, other then eventually prices and incomes will get back in line… how, is anyones guess, because the global financial crisis largely being uncharted waters.

In the short term though, real estate prices only appear to have one way to go. Down.

27 comments

Monday, March 9, 2009

A Tale of Two Cities

Filed Under: Alberta, Historical Prices

Calgary and Edmonton, two cities of similar sizes, linked together by geography, politics, and a healthy dislike of each other. In many ways it’s inevitable they’d be compared.

One’s a little bit more country, the other a little bit more rock ‘n’ roll… one’s a little more white collar, the other a little more blue collar…one’s sports teams win a lot of championships, the others, um, well, they wear a lot of red… but it’s not all bad down South, after all, they have chinooks, so they do have something going for them!

I kid, I kid. Calgary’s a pretty good town.

Anyway, with two cities quite similar in demographics, geography and governance, they should provide an interesting comparison for home prices. So lets get on with it.

As you can see here, the actual price of homes in our cities have followed very similar patterns over the years. Though it should be noted Calgary has had higher prices over the last 25 years, we’ll be touching on that later. Overall they track about the same, booming at the same time, declining at the same times. That really shouldn’t surprise any anyone, their economies are very closely tied.

Here’s a look at the same graph adjusted for inflation:

Obviously just reaffirms the above. Also as we discussed in the prior entry, those that bought during at the peak of the boom during the late 70′s-early 80′s took about 25 years for the value of their homes to return to the price they paid for them when inflation is accounted for.

One interesting observation is that during that prior boom it was also Edmonton that started seeing prices decline before Calgary did. Coincidentally, the same was also true during the current bubble.

In the comments discussion in the last entry, someone brought up a similar inflation adjusted graph done on a Calgary specific blog about a year ago. They conversation concerned the best-fit line in that graph, finding annual 1.2% growth above inflation in Calgary vs. my using a line showing 1.8% growth above inflation in Edmonton.

Just to clarify, those two plots were not of the same thing, and should not be compared as such. His was a purely statistical measure, whereas mine was looking more to mimic a long-term trend and was based more on intuition. Both have their strengths, though a statistician would probably argue that my number might as well have been pulled from my rear-end.

In any case, just so we can compare apples-to-apples I’ll do best-fit trendlines for both cities based on current numbers and 2009 dollars.

This is Calgary, 1973-current. In red is the inflation adjusted price, and in black a best-fit exponential trendline. According to this, over that period Calgary experienced a 1.33% annual yield beyond inflation (also perhaps of interest, it tells us the residential average has about $100,00 to drop before it reaches the long term trend).

Here is the same graph for Edmonton, over the same period. According to that, Edmonton only experienced 0.6% annual growth beyond inflation… less then half what Calgary did, and quite low in general. Though, as mentioned in the Calgary blog, not unprecedented, as mentioned there over 345 years the actual growth above inflation there was a mere 0.2% annually.

One issue with this kind of measure is that it can change significantly just by adjusting the period. To give you an example of that, and since I have data for Edmonton going back to ’62, here is a graph of prices and both the ’73-current trendline, and a ’62-current trendline.

Here we can easily see the contrast. The ’73-current line is almost flat, whereas the ’62-current line shows appears much steeper (relatively speaking), coming in with 1.45% annual growth above inflation. It’s really quite a significant difference.

This of course begs the question, which one is right?

The answer is… who knows, they’re just numbers.

As any good statistician could tell you, you can probably make the numbers tell you whatever you want to hear. The numbers are a conversation piece and you can certainly make an argument with them, but going forward they aren’t proof of anything.

Thus, I still kind of like my old arbitrary graph here…

Pretty, ain’t it?! And as different as the measures are, they all seem to be telling us that residential average should be heading toward the $200,000 mark.

Anywho, so another topic bandied around in the prior entries discussion was the ratio of prices between Edmonton and Calgary. As noted earlier, while the two cities follow very similar patterns, Calgary has experienced consistently higher prices for a good while.

Here we see the Calgary price relative to the Edmonton price since 1973 (Edmonton’s prices =100%, Calgary’s price plotted).

Until the early 80′s the two were pretty close, then as the oil boom ended, as we noted earlier, Edmonton declined faster and we saw the first extended separation, but eventually the bust hit Calgary prices as well and they were again about even between the two for a few years.

After 1985 though Calgary pulled ahead, and has stayed there ever since. Trended about 20% above through the mid-to-late ’90′s when it shot up to about +40% (getting as high as +51% at times)… but then cooled back down into the 20′s (but did briefly spike back to +51% )… and has since cooled further, and is not back at levels not seen since the mid-90′s.

Lately we’ve been seeing Calgary at around +15-to-20% above Edmonton in the last few months. Depending on the period, the overall average is a bit different. If you include everything right back to ’73, Calgary is 21% higher (median 20%)… but over the last 20 years, they have been 27% higher (median 26%).

So, to make a long story short… I’d say Calgary prices we should generally expect be about 20-25% higher on any given month.

On a side note, this is actually a bit out of line with affordability, while Calgary does have a higher median household income, it’s only about a 10% difference currently. So we may see the margin narrow a bit in the years to come.

Oh yeah, and some more good news for our friends from the South. I heard they’re planning on remodelling the Saddledome… it’ll be perfect in every way… except for one thing… the seats will still face the ice!

Enjoy your Monday!

1 comment

Wednesday, March 4, 2009

Historical Prices and Inflation

Filed Under: Historical Prices

I had been meaning to do this one for awhile, but never got around to doing all the number crunching… hopefully you guys will find this entry interesting, cause it took a fair bit of work.

As you have likely deduced, that is a graph of the historical average residential price from January 1962 right through February ’09. Looks rather dull until 2006 when suddenly things went crazy. Prior peaks and troughs look quite gradual, and the big boom/bust in the late ’70′s early 80′s doesn’t seem so big.

But when we adjust for inflation…

That tells a bit of a different story, eh?

For those curious, I used the Bank of Canada’s inflation calculator, and did all the calculations in 2009 dollars, since that’s what we’re dealing with today.

As you can see from that graph, when adjusting for inflation, those that bought at the peak of the prior bubble, didn’t recover their initial investment for 25-30 years… coincidentally just after the current boom really kicked in.

That’s some very unsettling news for those first time buyers that bought in in the last two years. Also a reality check for all the young people who thought real estate never goes down… it can, has, and has done it right here in Edmonton.

The late ’70′s bubble was a bit of a different beast then the current one. The run up took about 5 years… things levelled off for about 5 years…. then declined for about 5 years. Quite gradual compared to the relatively explosive run up from January ’06 through mid ’07… then the decline started almost immediately.

I plan on doing a comparison of the two bubbles later, so I’ll move on to something else. But there is a lot interesting stuff in there.

Here we see the inflation adjusted prices, against a curve of anticipated prices assuming 1.8% growth above inflation. The 1.8% figure is somewhat arbitrary, I just took the prices from 1962 and 2005 and calculated the annual growth, since it seemed the market was reasonably balanced in both years. The curve also nicely dissects the other one, so I’d say it’s a plausible figure to go with.

In any case, we can see that the peaks and troughs always return to the anticipated curve. The bubbles far exceed the line, while the troughs tend to follow it a bit closer. It’s interesting to note that the largest trough (late 90′s through early 00′s) was not so much caused by dropping prices, but instead by stagnant ones.

This is a behaviour that I wouldn’t be surprised to see repeat itself once our current bubble has returned to Earth.

What should be a very startling observation from that graph for first time buyers is that even as far as prices have already fallen, they’re still ~$100,000 higher then the anticipated curve. That’s a rather scary prospective fall, and surprised even me when I first noticed it.

Here is one more graph just for a little bit more food-for-thought.

This one is just something of a melding of the aforementioned. This is the actual unadjusted average residential price from graph one, plotted against the anticipated curve from graph three with prices adjusted for their times.

You’ll notice the anticipated curve in this case isn’t nearly as smooth as in the prior graph, that’s because inflation varies from year to year, and that curve is of inflation + 1.8%.

Again, because the numbers are unadjusted for inflation, everything appears muted compared to the adjusted figures. So, while the current bubble looks quite extreme currently, should we have another boom in 30 or 40 years our current one will look much less extreme, and the one from the 70′s probably won’t even register.

Anywho, hope you found this somewhat interesting and could follow along. I was just hoping to convey that looking at raw numbers can be misleading over a long time span, and thus how important it is to account for inflation.

It’s quite sobering to realize that those who bought during the peak of the last bubble took at least 25 years to see their property values recover to what they paid for them when accounting for inflation. It should also serve as a warning to those looking to buy into the current market, as prices are still well above historical measures.

32 comments