Tuesday, December 8, 2009

Big in Japan II: Electric Boogaloo

Greetings brothers from other mothers. Today we're going to do a big of a supplement/appendix to Saturday's post on Japan... this time around we'll do some comparing and contrasting of their HPI with ours, as well as the Case-Shiller index from the States.

Japan/Canada/U.S.A. - HPI
Here they are charted out without any adjustments. Kind of a mess, we have different index points, some peaking before the index point, others after, yadda yadda yadda, I'm really tired today it's a mess. So, lets do some adjustments and see how it looks then.

Japan/Canada/U.S.A. - HPI
Alright, since we're obviously concerned with "bubbles", lets find the respective peaks and count back six years. Why six years, well, it would have been ten, but the Canadian data doesn't go back that far, nor does the U.S. Composite 20... so basically I went back as far as the shortest data set allowed.

We can see by far the biggest bubble was in the Japanese major city index, followed by the U.S. composite 10, then U.S. composite 20, the Canadian composite 6, and then finally the Japanese nationwide. Interesting to note that Japan bookends the high and low ends. Also since their asset bubble was back in '91 they have the long tail.

So, as far as the Canadian figure goes our bubble is bigger then Japan's (as a whole), but well below the U.S. bubble. It is worth noting the Canadian cities peaking has been a little more spaced out then our American counterparts, and in light of the recent national flurry we're in all likelihood going to set a new peak as the fall numbers become available... so the Canadian curve would thus need to be reset in that event.

In any case, it's unlikely the Canadian peak will get anywhere near the U.S. figures... unless we just go completely off the rails for another year or more (particularly Toronto). After the last eight months though, never say never, but it would take a whole lot to pull us up to that level.

Japan/Canada/U.S.A. - Decline From Peak
These are the respective decline from peak figures. Again, Japan's extends far beyond everyone else. Interestingly, we can see the two U.S. composites have declined at eerily similar patterns despite having varied peaks as we noted in the prior graph.

We can also see that the current ultra-low interest rate environment has not just spurred prices in Canada, but also the U.S. as their curves too have suddenly made a stark move up. Difference being that south of the border they're still down 30% from peak even after this upswing... whereas in Canada we were still early in our decline, thus much of those losses have been erased and we're now approaching a new high (which would reset this graph).

Japan/Canada/U.S.A. - HPI - Cities
National numbers only do you so much good as real estate is more a local creature and can vary widely from region to region. So lets take a look at some city numbers. As I mentioned in the prior article, I have been unable to obtain any city specific numbers for Japan, but we'll compare some major cities in North America to what happened to the Japan major city composite. At least that will give us a general idea.

For Canada I picked Calgary, Vancouver and Toronto as they are our most hotly discussed real estate markets. To make them easier to pick out, I made them the dashed lines. For the U.S. I selected Miami, Phoenix and Seattle. Miami having the biggest bubble (as per Case-Shiller), Phoenix was about 7th out of the 20 but much discussed, and Seattle was about in the middle of the pack.

We see here the Miami curve is actually quite close to the Japanese major market composite. Down a little, we see Calgary close to Phoenix, with Vancouver not far behind. Down a little more we have Seattle, then finally Toronto's curve looking downright flat compared to the others.

Unfortunately for us in Edmonton we aren't included in the Canadian HPI. But if we want to get an idea of where we might be on this scale lets start by comparing Calgary resale stats from this period with their HPI... indexing Calgary's various resale averages/medians using the same method (index point six years prior to peak) we find peaks in the 225-to-240 range. Knowing their HPI topped out at 226, seems we're right in the range, near the low end.

Now looking at Edmonton's resale averages/medians, we come out with peaks in the 260-to-280 range... the low end of which would put us right up at the top with Miami. The same relationship between resale and HPI holds for both Vancouver and Toronto as well. So, take that for what it's worth, but it would suggest that the bubble we experienced in Edmonton could very well be right up there with the worst in the United States.

Japan/Canada/U.S.A. - Decline From Peak - Cities
Finally here is the decline from peak for all the cities (the Japanese one is cut off in this and the prior graph, but the full plot is available in the earlier graphs). Pretty serious stuff in Phoenix and Miami, with 54.5% and 48.5% respective declines from peak. Even Seattle which had a much more moderate bubble has fallen over 20%, well beyond the classification for a "bust."

We can certainly see the effects of the interest rate plunge on both sides of the border (except it seems in Seattle). Even in the U.S. markets where residential real estate had been written off for dead prices have jumped significantly in recent month. It'll be interesting to see how the next couple years play out, as rates stay low for now and what will happen when rates start jumping. But only time will tell.