Correlation is not Causation

We hear "correlation does not imply causation" quite a bit in the investment world. It means we can't really know for sure there is a definitely cause-and-effect relationship between two variables based on the observed correlation between them.

It's really easy to make this mistake unintentionally in analyzing economic and market data. We are also wired to look for patterns in data, and we find them even when they might not really exist.

I was thinking about this when looking at the chart below. The relationship seems intuitive enough based on simple supply and demand dynamics. In 2006, existing home sales began to decline and home prices soon fell after as demand crumbled.

An equilibrium wasn't found until 2010 or so a home decline of 30%. But WHY did existing home sales fall? Consumers tapped out on leverage? Over-supply of houses? All of the above? The cause was realistically a combination of things. The question is, does a sharp fall in existing-home sales CAUSE home prices to fall?

Home prices have stayed stubbornly high in the face of rising mortgage rates and a resilient consumer. But the sales decline is very real. This is likely where you insert the term "LAG". It's always different this time as circumstances are always changing. This time around we don't have enough houses and existing home buyers don't want to trade for a higher mortgage. And this is all "pre-recession".

So while the cause for the decline in existing home sales may be different, I think the effect will be similar over time.