« Diversification: Do you really understand it? (Part 2 of 3) | Main | LDI Returns: The other side of the coin »
Monday
Aug082011

Diversification: Do you really understand it? (Part 1 of 3)

Some people think that there is no such thing as a “free lunch”.  They tend to be a lot thinner than me. 

Diversification is a prime example of a free lunch.  The basic concept is easy to grasp.  Apparently grandmothers all know not to keep all their eggs in one basket.  Diversification can reduce your risk without hurting your long-term expected returns.  However, there are a few issues that need to be taken into account.

We are devoting three blogs to diversification.

  • Part 1 (this blog) looks at what it is you need to diversify.
  • Part 2 explores the effect of “diworsification”. (We’re sorry; we know it’s a corny pun.)
  • Part 3 will quantify the impact of correlation.

 

Part 1: Do not diversify your assets

Diversify your risks instead. 

We may have now lost every reader bar the incurable pedants, but sometimes this distinction can be important.  Maybe an example would help.

Imagine that you owed your local mafia some money.  A lot of money.  £1,000,000 to be precise.  I’ll let your imagination conjure up the reason for the debt and the unpleasant consequences of non-payment.  Fortunately for you, the payment is not due for another ten years.  Even better, you actually have the capital that (if invested properly) would just produce the necessary £1,000,000 in time.

The safest investment for you in this situation would be a ten-year zero-coupon gilt that paid out exactly £1,000,000 on the date required.  Even though 100% of your investment would be in a single security, we don’t think you could reduce your risk by diversifying.  The reason for this is that the only risks you would be exposed to would be the UK government defaulting (OK, we admit that this is not quite the zero-risk that some commentators once thought it to be) or fraud by your custodian. 

  • Switching some of your assets into cash would expose you to the risk that interest rate earned on cash falls, and there would be a shortfall in ten years’ time.
  • Switching some of the bond into longer-dated bonds would expose you to the risk that long-term interest rates rise, and you would suffer a capital value loss on liquidating your bond in ten-years.
  • Switching  into a mixture of cash and  longer-dated bonds would expose you to a change in the shape of the yield curve.  (In other words, the gains on your cash might not offset your losses on long bonds or vice versa.)
  • Switching into an inflation-linked bond would leave you exposed to the risk that inflation turns out to be lower than had been expected.
  • Switching into overseas bonds would leave you exposed to currency fluctuations and changes in overseas interest rates. The interest rate risk would apply even if you hedged out the currency risk.
  • Switching some of the bond into a similar bond issued by a different borrower (US Government for the brave or GE Capital for example) would reduce the risk of losing the entire portfolio through diversifying the credit risk.  However, it would actually increase the risk of not being able to pay because if any of the issuers defaulted you would lose your kneecaps.
  • Similarly, spreading the gilt holding between two custodians would decrease the risk of losing everything, but increase the likelihood of some loss (which would be unacceptable in this case).

 

So much for the theory, what does this mean in practice for pension schemes?  Here are a couple of more realistic examples.

 

  • When to use overseas assets

There is no fundamental reason why developed overseas markets should systematically outperform their UK equivalents over the long-term (or vice-versa).

So the case for having overseas (developed) assets within your strategic asset allocation revolves around reducing risk.  (This is quite different from your fund manager switching into overseas assets to take advantage of their short-term views, or using emerging markets to increase long term expected returns.)

Given that UK pension schemes’ liabilities are in Sterling, investing in overseas assets introduces currency risk.  However, this additional risk might be more than offset by the benefits of diversification.  If the reduction in risk from diversification is greater than the additional currency risk, then there is a strategic case of the overseas investment. 

Within an equity portfolio, there is diversification over a greater range of stocks, industries and economies than could be achieved in the UK alone and these benefits more than offset the additional currency risk.

However, bonds are generally held by pension schemes to match liabilities (or annuity prices for DC funds) and can do so reasonably closely (much closer than equities).  There is generally less ‘risk’ to diversify away and so the diversification benefits are much reduced, whereas the currency risk is just as significant.  Introducing overseas bonds into a UK pension fund’s bond portfolio will increase, rather than decrease, overall risk.  With bonds the situation is closer to the hypothetical mafia scenario above.

So we recommend that UK pension schemes invest their equities globally, but their bonds locally.  All because we look at ‘risks’ when diversifying rather than ‘assets’.

 

  • “Diversifying” fund managers

For many small schemes, the best practical solution involves having a single fund manager run all of their assets.  (In our members’ section we go through the advantages and disadvantages.)

However, this can often make the trustees nervous.  There is a natural tendency to want to diversify; it doesn’t feel ‘right’ to have 100% of the assets with a single manager.   But let’s look at the ‘risks’ rather than the ‘assets’.

The most worrying risk is that the fund manager becomes insolvent.  However in almost all cases the assets are kept separate from the fund manager.  The failure of a manager would not directly result in a loss for the pension scheme.  It would be inconvenient and a replacement manager would be required.  However, we are not convinced that this is a risk that can or should be diversified.

The second, and more likely, risk is that the fund manager underperforms the benchmark.

This risk can be diversified.

However, bear in mind that this risk will usually be fairly trivial compared to the risk coming from the chosen asset allocation. 

More importantly, we have yet to come across a fund management organisation where the ‘bets’ on the equity funds are replicated or even similar to the ‘bets’ on the bond funds or property funds.  So in our view, having three separate managers for your equities, bonds and property gives you no material diversification benefit over having a single investment house running all three.

 

In our next blog, we’ll look at how diversification might erode your expected returns – “diworsification”.

PrintView Printer Friendly Version

References (7)

References allow you to track sources for this article, as well as articles that were written in response to this article.
  • Response
    I think you have noted some very interesting points , thankyou for the post.
  • Response
    Response: Flappy Bird Hack
    Diversification: Do you really understand it? (Part 1 of 3) - Opinion pieces - Barker Tatham
  • Response
    Response: garcinia cambogia
    Diversification: Do you really understand it? (Part 1 of 3) - Opinion pieces - Barker Tatham
  • Response
    Response: OneTwoTrade
    Diversification: Do you really understand it? (Part 1 of 3) - Opinion pieces - Barker Tatham
  • Response
    Diversification: Do you really understand it? (Part 1 of 3) - Opinion pieces - Barker Tatham
  • Response
    Diversification: Do you really understand it? (Part 1 of 3) - Opinion pieces - Barker Tatham
  • Response
    Gown Befitting The body Kind It's hard in order to translate all o michael kors outlet stores f the most recent abercrombie and fitch online developments in order to find a method to match your person physique. Each and every lady offers some thing regarding their own entire ...

Reader Comments

There are no comments for this journal entry. To create a new comment, use the form below.
Member Account Required
You must have a member account on this website in order to post comments. Log in to your account to enable posting.