Surprisingly predictable?

The FT Lex Column had a piece last week on serial correlation in the housing market. The piece discusses the apparent predictability of price changes in the housing market.

Over the past 20 years, UK housing has shown clear momentum; any quarter’s move reveals much about what the next quarter will bring. You should not see this in a proper liquid market; if you could, the opportunity would quickly be traded away.

This is something that’s always niggled me about the commercial real estate market too. Basically, when prices start to rise in a period, they tend also to rise in the next one and the next one etc.  Likewise when they start to fall in one period, they tend to continue to fall in the next period.  Obviously there have to be turning points and you do see the odd exception to this – a kind of glitch.  Sometimes you get an isolated quarterly fall in a period of increasing prices and vice versa.   But they’re few and far between.

As Lex notes, this doesn’t happen in more efficient markets where patterns of prices changes from one period to the next are uncorrelated.  That is, the pattern of price changes from one period to another tends to be more like the toss of a coin.

As politicians seem to be always saying these days, let’s be clear what this means because it is very odd. If you see real estate values starting to rise, then you should buy because it is almost certain that they will continue to rise for at least several years.  It is, therefore, hard to lose money.  In turn, if you observe values starting to fall, then it is almost certain that they will keep falling for 1-2 years.  As I said, this isn’t guaranteed, sometimes you get short-term glitches.  For instance, values fell during the LCTM crisis in 1998 for one month in the middle of a long run of monthly value increases. There’s also a frequency issue.   The higher the frequency of measurement period e.g. monthly, quarterly, annual etc., the higher tends to be the level of correlation between the previous value changes and current value changes.

It seems that anticipating the turning points that tends to be critical. When prices start to fall, it can be hard to sell.  Transactions take at least three months and often more.  When prices are falling, liquidity tends to disappear, buyers tend to become fewer and trading volumes can collapse.

Also, I just had to go through what I’d written changing prices to values. Commercial real estate indices are based on appraisals – not actual prices.  These are essentially guesses of selling prices rather than actual prices.  Many analysts considered them to underestimate market movement and to anchor on previous appraisals.  It’s been a hotly contested issue (er, by academics) but there’s almost certainly an underestimation of price movement going on.  However, the fact that house PRICES display some similar momentum and predictability suggests that price change predictability in the commercial real estate market is not a data artefact and that the direction is capital returns is very predictable in real estate markets.

I’m not sure why I’ve never done anything about it in terms of investing. Risk aversion and lack of money, I suspect.  Hindsight can make you look misleadingly smart.  As Lex indicated, Brexit is creating a lot of market uncertainty.  It could create a glitch or it could mark a turning point.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s