Uncorelated Assets

doubleJackD

I have seen mention of uncorelated assets (UA) on this forum. I just wondered what people were investing in to get these "uncorelated assets"?

Most of my personal wealth is in shares. What assets could I buy to get some UA? As I see it most financial assets largely move in harmony......

westy22

Traditionally bonds and gilts - the rule of thumb is that they should rise as share prices fall and vice versa. Also commercial property, as the theory is that long lease commercial rents remain steady despite the vagaries of the stock market.

gadgetmind

Westy has given you the short answer, but that's not enough on which to make investment decisions.

I suggest you read.
1) Smarter Investing:Simpler Decisions for Better Results - Tim Hale
2) The Intelligent Asset Allocator: How to Build Your Portfolio to Maximize Returns and Minimize Risk - William Bernstein

If you only read one, read the first as it's UK-centric and has lots of practical advice. However, both come at things from a different direction, and (2) is slightly older. While this could be bad, it's good to see the historical perspective.

The first book has a STRONG leaning towards passive investing, but there is no reason why you can use the same principles with an active (or even hybrid in my case) portfolio.

kidmugsy

The tricky bit is when you have to guess whether assets will be uncorrelated in future.

grey_gym_sock

isn't a mix of bonds/gilts and shares likely to have less volatility than just shares, but a lower total return? the uncertainty of the final result - viz. the real value of investments when one finally wants to spend it - matters, but the volatility along the way is something one can learn to live with.

you might (or might not) choose to hold fewer equities to reduce the uncertainty about the final result, even if that means a less favourable expected* final result. if you do, aren't the expected returns from fixed-rate and index-linked bonds/gilts the same as far as one can tell? and the uncertainty much higher for fixed-rate than for index-linked (because the former have inflation risk)? so shouldn't one's bonds/gilts all be index-linked?

(i've ignored commercial property - that's a different discussion.)

this may all be answered in tim hale's book. feel free to summarize it so i don't have to read it!

*"expected" in the mathematical sense: the average predicted outcome, weighted by probabilities.

westy22

Grey Gym Sock - I don't disagree with what you say and I wasn't advocating any particular strategy; merely illustrating to the OP what are generally considered to be some uncorrelated asset classes to hold alongside equities.

grey_gym_sock

fair enough - i wasn't arguing with anybody (though i guess it looked like i was)

gadgetmind

grey_gym_sock wrote: »

this may all be answered in tim hale's book. feel free to summarize it so i don't have to read it!

Yes, it is covered, and there are loads of graphs and tables, some of which come from back-testing and some from monte carlo analysis.

You can add a whole load of bonds (40%?) before you start to see any significant reduction in return. This is because the (usually!) lower return from bonds is offset by the advantages of rebalancing uncorrelated assets.

TCA

I intend to read the Tim Hale book, but just want to ask now, what is the advantage or rationale to holding bonds or gilts over cash if holding equities too? Assuming you fixed your cash at around 4%, why would you want a lower interest bearing asset?

Perelandra

grey_gym_sock wrote: »

isn't a mix of bonds/gilts and shares likely to have less volatility than just shares, but a lower total return? the uncertainty of the final result - viz. the real value of investments when one finally wants to spend it - matters, but the volatility along the way is something one can learn to live with..

I believe the answer to this (and this post is more a 'test my own knowledge' than 'listen to me, only I know the truth'!) is- surprisngly- it depends on how you manage your portfolio.

If you re-balance your portfolio at regular intervals (either through buy/sell, or by dripfeeding) between the uncorrelated assets, then over time you can expect a *higher* return than if you'd invested in equities alone. The rebalancing will enforce a level of "buy low, sell high" that you wouldn't necessarily have followed if you'd simply bought into equities alone.

So a portfolio comprising of balanced, uncorrelated assets managed with regular (but not too frequent) rebalancing could well result in a higher overall return provided that all asset classes are cyclical, with underlying growth, and that you ultimately don't "cash in" on an asset during a trough.


EDIT:

As I was writing this, I was thinking to myself "this is right up gadget's street"... only to realise he'd said something similar at 8:04! Oh well... lol

gadgetmind

kidmugsy wrote: »

The tricky bit is when you have to guess whether assets will be uncorrelated in future.

There are tables showing the correlation between different asset classes based on long-term historical data. Using these, you can construct a portfolio that will weather pretty much anything you throw at it, and that will only need occasional rebalancing (and ideally regular feeding!) to keep growing nicely, albeit not smoothly.

Of course, all of this assumes that you subscribe to modern portfolio theory.

http://en.wikipedia.org/wiki/Modern_portfolio_theory

Here we start getting into technical areas where the experts can argue for decades!