San Diego home prices have had quite a run since we last checked in on local housing valuations.
Let’s take another look at home valuations since the end of the year, what this may portend for the future and (bonus nerd topic) whether housing is in a bubble.
The graph below compares local home prices to their two most important economic underpinnings: incomes (which tell you how much San Diegans have available to spend on housing) and rents (which tell you how much it costs to put a non-owned roof over one’s head, and also how much landlords can earn from owning a house). The graph simply charts San Diego home prices divided by an equally weighted average of San Diego rents and incomes. When the ratio — the red line — is high, that means homes are more expensive than rents and incomes would suggest they should be; when it’s low, they are cheaper. The blue line is the median or midpoint of all historical values, and the orange line denotes the current level.
Here’s what this graph is telling us:
Despite the recent price run-up, San Diego homes are nowhere near as expensive as they got during last decade’s bubble.
• They are expensive, though: They are 12 percent above the historical median valuation (down from 13 percent in November), and very nearly at levels reached in both the 1979 and 1990 valuations peaks.
• The first two times we hit this level of expensiveness in the past, valuations soon began a multi-year decline. The third time, valuations went straight up for several years before totally collapsing. So what happens this time? I don’t think anyone can know the answer to that question, unfortunately. There are a number of reasons why:
• While valuations peaked at this level two times in the past, there were specific factors helping to subsequently push them down: a double-dip recession and really high interest rates in the early 1980s, and a pretty bad local recession in the early 1990s. I do believe that housing’s overvaluation puts it at more risk from negative economic events, but unless and until those events happen, valuations could remain higher than normal.
• We also have an example where valuations reached this level and proceeded to rocket upward (for a while, anyway). Highly aggressive monetary policy played a huge role in fueling that bubble, and monetary policy is currently far more aggressive then it was back then — so we can’t rule out valuations getting even more distorted in the near term.
• Or, perhaps valuations aren’t all that distorted. It’s entirely possible that the relationship between home prices, rents and incomes could slowly change over time, and that the current level will turn out in hindsight to have been a normal, sustainable, middle-of-the-road valuation. I wouldn’t depend on it, but it’s a possibility.
• On a somewhat related note, we only have three prior cycles’ worth of history here, so we should generally be cautious about assuming that history will repeat in a specific way. In 2005, values were so absurdly high that it was pretty safe to predict that they’d eventually plummet. Now, with values seemingly high but much more modestly so, the outcome is less certain.
• Home prices have gotten a huge boost from low mortgage rates — but that’s potentially a two-way street.
That last point deserves some more background. Mortgage rates became comically low, providing a huge boost to housing demand and, thus, home prices. Here’s a chart comparing monthly payments, which factor in the prevailing mortgage rate, to rents and incomes. Despite the big move up last year — a result of both rising house prices and an increase in rates — monthly payments remain well below their historical midpoint when compared with rents and incomes:
For as long as this continues to be the case, it’s plausible that home valuations could remain on the high side. But given that rates are still being artificially suppressed by central bank intervention, this seems destined to change. I have no clue as to when that might be, but when it does, a very important prop to housing valuations could be abruptly removed.
All in all, there is a lot of uncertainty as to where valuations will go from here. We do, however, know this: Housing valuations are significantly higher than their typical historical level, and which have occasionally led to pretty deep valuation declines in the past.
That alone is reason enough to be cautious.
Into the Weeds: Is Housing a Bubble?
Whether something is a bubble really just depends on one’s definition of “bubble,” which can vary widely. The important thing to know is where you are on the overvalued/undervalued spectrum, labels be damned.
A bubble is all about valuation. It’s not enough for investors to be enthused, or prices to rise rapidly — to be a bubble, something has to be really overvalued compared with its economic fundamentals.
For the question of how overvalued is overvalued enough, I turn to GMO, a well-known investment management firm that has performed extensive research on historical investment bubbles. Theirs is a purely quantitative definition: An asset is in a bubble when its valuation reaches two standard deviations above its historical fair value trend. (I told you this would be nerdy).
It is very rare for assets to reach this level of overvaluation, and when they do, the bubbles pretty much always burst. (Almost all of the major bubbles identified by GMO have eventually come back down to their fair value. The only ones that haven’t are still ongoing, so there’s a good chance they will do so too. It thus seems fair to generalize: Bubbles always burst.)
Here’s the graph of San Diego valuations again, with the green line denoting the two-standard-deviation “bubble” definition:
The bubble threshold was, unsurprisingly, reached and easily surpassed in the great bubble of the past decade. Now, however, we are nowhere close. So there it is! No bubble!
But wait. One could argue that the mid-2000s bubble was so epic that it artificially increased the standard deviation of the data series, thus significantly raising the bar on what comprises a bubble. While I agree with this in principle, I think that cherry-picking which data you use kind of defeats the purpose of using an objective, quantitative criteria for what makes a bubble.
Still, let’s take a look at what would happen if we created the bubble threshold using only data prior to 2003 (which I somewhat arbitrarily picked because it’s the year in which San Diego housing started to get really nutty):
That brings us a lot closer, but even this much less stringent definition still puts us well shy of bubble category.
My conclusion, then, is that San Diego housing is not in a bubble. But it’s still overpriced!