Comments on: Diversification and Correlation

By: Ethan

Ethan — Wed, 17 Feb 2010 05:29:13 +0000

I always like to point out that if your careful asset allocation fails to produce a diversification benefit – fails completely and utterly over your entire investing horizon, producing a correlation coefficient of 1 – what have you lost? Nothing. For the careful investor, fees probably aren’t even any higher. And since they performed identically you don’t have any rebalancing costs or time investment.

When failing carries no repercussions, and success has positive repercussions, why would you not try?

By: Mike

Mike — Wed, 25 Nov 2009 21:44:11 +0000

Paul: I’ve actually been full-time writing since last November–pretty much since before anybody was reading this blog.

By: RJ Weiss

RJ Weiss — Wed, 25 Nov 2009 16:46:37 +0000

I was talking to a financial adviser the other day who really believes that buy and hold is dead.

Mike, like you said, it’s exactly the opposite. For whatever reason, the public perceives that buy and hold means you hold the S&P 500 and that’s it.

We seem to forget that buy and hold means indexing not just in stocks but other asset classes as well.

By: Paul Williams @ Provident Planning

Paul Williams @ Provident Planning — Wed, 25 Nov 2009 16:37:17 +0000

Mike, great article once again!

When did you go full time on this? I must be out of the loop!!! Congratulations!

Dylan, you’re right about the ultra-short term, high-credit debt having a slight negative correlation to U.S. equities. The info I have shows that 0ne-month CDs and the Vanguard Short-term Bond Index both have a -0.04 to -0.05 correlation to the S&P 500 and about a -0.12 correlation to US Small Caps (based on CRSP data and actual returns). The Lehman Brothers Intermediate Gov’t/Credit Bond index has a slightly positive correlation (~0.10 to 0.15).

By: Rick Francis

Rick Francis — Wed, 25 Nov 2009 15:56:11 +0000

>It seems to me that diversification didn’t fail at all. It worked perfectly >according to plan–practically a banner year for the “here’s why you should >diversify” message. So what failed? The general public’s expectations and >understanding of diversification.

Absolutely! Sadly, the general public is likely to abandon diversification, as they expected it to prevent losses- rather than mitigating losses. Happy Thanksgiving!

-Rick Francis

By: Mike

Mike — Wed, 25 Nov 2009 14:41:02 +0000

Dylan, you’re right.

When I said “predicting,” I didn’t mean predicting future returns. I really meant something more like this: “if you were told the return of asset class A for a given year, and were asked to guess the return of asset class B for that year, if they have zero correlation, the information about A will be of no use.”

By: Dylan

Dylan — Wed, 25 Nov 2009 14:25:44 +0000

Mike, you are exactly right. Diversification didn’t fail us (neither did indexing or buy and hold investing). Capital markets decided stuff is worth less money. That’s true whether one is diversified or not, and the undiversified investors as a whole cannot do better then the diversified investors. So we cant really say diversification failed and non-diversification did not.

I think historically, ultra-short-term, high-credit (insured muni bonds and 90-day T-Bills) debt has had a slight negative correlation to US equity markets (and much lower return).

“The movement of one has no value for predicting the movement of the other.”

I think this applies to any correlation coefficient, not just zero. It’s never predictive because variations are concurrent. I think a better way to explain zero correlation is that the frequency and degree of positive correlation matches the frequency and degree of negative correlation, resulting in a wash.