XML Feeds

What is RSS?


Top Rated


    We have often written about the benefits of diversification and how proper diversification can both improve the returns as well reduce the risk of portfolios – by risk we mean the risk of permanent capital destruction of your investment. Important characteristics of assets that will provide diversification to portfolios is that they have an uncorrelated return profile, but that they also have the ability to generate a return in their own right.

    Ideally we are looking for diversification strategies that will provide the following return profile:

    Over the past 22 years this ‘Diversifier’ has given investors the same return as the ALSI, and when equally weighted with the ALSI (rebalanced monthly) has generated significantly better returns with much lower risk (capital drawdown and volatility).

    So, what is this ‘Diversifier’?

    Seed’s research has shown that an excellent diversifier for local investors (read ALSI / local equity market) can be found in offshore fixed income linked assets. When used wisely, these assets not only reduce the risk of portfolios, but can also enhance returns. The key aspect to consider when including diversified assets / strategies in a portfolio is to completely understand their return drivers. Offshore fixed income linked strategies will have three main return drivers:
    · Exchange rate fluctuation
    · Income generated from the capital
    · Any possible capital gains / losses

    Exchange Rate:

    At Seed we use the purchasing power parity method (as discussed in previous reports) to determine whether the rand is under or over valued versus a range of currencies. To get the best risk/return payoff in this strategy the rand should be over valued (i.e. prone to weakening) relative to the currency that the diversification is taking place in. While the rand has weakened sharply over the past couple of months, it is still slightly strong when compared to a range of other currencies.

    Income Generation:

    The level of income that an asset generates will influence the foreign currency yield that should be expected, where the preference is for high yields. Interest rates are, however, at historical lows around the world and as such one should not expect to generate significant income from an offshore investment that is linked to cash / bond rates.

    Capital Variation:

    The capital variation return driver will typically have a large impact on this kind of strategy’s total return. As capital gains / losses have a large impact on returns one should be mindful of the risks of capital losses and the possibilities of capital gains in such a strategy. With interest rates at historic lows around the world, the risk of capital losses in government bonds is extremely high, and as such we have avoided this asset class.

    The Orbis Optimal Funds are extensively used within the Seed Absolute Return Fund to provide diversification to the Fund and improve its risk/return profile. Orbis Optimal’s broad return drivers are the movement in the rand versus the US dollar and euro (we expect the rand to weaken slightly), interest rates in these two regions (negligible), and the ability of Orbis to outperform global equity markets (we expect them to outperform, particularly in down markets).

    As the Seed Absolute Return Fund is focused on generating consistent returns in excess of CPI + 4% per annum, having uncorrelated strategies is particularly important for this Fund. We expect that this strategy will provide the biggest boost to performance in times when the local market is down sharply as local market weakness is typically experienced in conjunction with a weakening rand (good for this strategy) and falling global markets (good for Orbis Optimal as all market risk is hedged out and we expect Orbis to outperform in these periods).

    The chart below compares the rolling 12 month returns from Orbis Optimal with the ALSI only in periods when the ALSI was negative (i.e. times of local market stress). What is evident is that in most 12 month periods (over 75% of the time) Orbis Optimal was up when the ALSI was down.

    As Multi Managers we are constantly on the lookout for mandates that will improve the risk/return profile of our Funds. This involves research on both the types of strategies and the managers’ best suited to manage these strategies. We then need to ensure that we increase the allocation to those strategies that are expected to produce the best returns in the given market environment, and reduce the allocation to those that aren’t experiencing favourable conditions.

    Take care,

    Mike Browne

    021 914 4966

    Permalink2012-06-26, 10:58:08, by Mike Email , Leave a comment

    Picking the Current Winner

    Who hasn’t heard the warning “past performance is no indication of future performance” before? But why do we still base our investment choices solely on the past performance of a fund?

    We cling on the hope that a winning fund will continue its outperformance. By basing your decision only on the past performance of the fund you run the risk of investing into fund after its outperformance or disinvesting out of a good fund at wrong time.

    The graph below illustrates this point. It shows the investment flows of Allan Gray Stable fund. This is overlaid with the previous year’s alpha versus the ALSI, the black line, and the following year’s alpha versus the ALSI, the red line.

    Source: Allan Gray

    The graph shows how investors chase after past performance. In January 2009 R150m of investments was made into the fund due to 36% outperformance it had the previous year versus the ALSI. The following year the fund underperformed the market by 27%.

    On the flip side, if you disinvested from the fund in December 2010, you would have missed the 10% outperformance over the following year.

    From the graph you can clearly see how the investment and disinvestments lag the past performance of the fund.

    The fund outperformed the markets in 2008, because it was able to protect capital during the credit crisis. It underperformed in 2009 because it wasn’t able to fully capture the market rally due to its low equity holding.

    A fund’s investment process is much more important than its past performances. By understanding how the fund is managed you will better know what you should expect from the fund going. You will better understand when to increase your holding or to decrease it.

    From the above chart we can see that it is evident that the classic warning “past performance is no indication of future performance” is not just an empty warning.

    Kind regards,

    Gerbrandt Kruger

    Permalink2012-06-20, 10:44:32, by Mike Email , Leave a comment

    Investment returns

    Annually JP Morgan produce a report comparing the investment returns from the main asset classes in South Africa, namely shares, bonds, listed property, Krugerrands, an investment into one’s own mortgage and fixed deposits. For the main asset classes, the data goes back to 1960, therefore a 52- year record of annual returns.

    A critical decision that each investor must make is to how to allocate between various investments. Looking back as far as possible and making comparisons of returns over various time periods helps in this decision making process.

    Some interesting facts that come from the analysis include:

    The arithmetic average of 52 calendar years of annual returns from the JSE is 19,1%. Over this same period, bonds would have achieved 9,7%, the same as fixed deposits, while inflation ran at 8,5%. Looking at these annual returns from a more accurate wealth effect angle, i.e. on investing R1000 in 1960 and reinvesting the income each year, one arrives at the following: A basket of goods costing R1 000 in 1960 would now cost R64 680. An investment into fixed deposit would be worth R117 000, while R1000 invested in 1960 into the JSE, and with all dividends reinvested would now be worth some R3 154 000. All calculations assume no tax or transaction costs.

    While equities have outperformed over the long period, because there is much higher volatility compared to other asset classes, it is instructive to look at rolling periods of returns. By using five-year rolling intervals from 1960, a clearer picture emerges for the average investor looking at his returns over a five-year period. Since 1960, there have been 48 five-year periods and so far in not one of these rolling five-year periods have investors lost money in nominal terms.

    However when adjusted for inflation, there have been 4 rolling five-year periods when equity investor returns were below inflation. The worst period was 1973-1977 when investors lost 5,2% p.a. in real terms. Across all 48 rolling five-year periods, the best five-year investment period was from 2003-2007 when local shares gained 24,4% in real terms. More recently however from 2007-2011, investors have had a tough time, gaining only 1,2% p.a. in real terms. This more recent five-year period is the worst since the 1994-1998 period and merely demonstrates the volatility over various shorter time horizons

    The chart below clearly demonstrates the more recent poor performance of equities, bonds and cash when measured after inflation. The starting and ending period over any five-year period has a major influence on the results. Therefore an investor into equities at the start of 2008 will have a poor five-year experience, whereas a 2009 start (after shares had fallen some 40% from their peak) is highly likely to produce a superior result, with the total return up some 17% per annum nominal.

    Chart 1: Real returns – five-year rolling periods

    Source: Morgan Stanley

    The chart below is indicative of the typical returns across the various asset classes that an investor can expect. Over this period on an after tax basis, equities have outperformed all others. On a pre-tax basis however property unit trusts have been the best performer.

    Chart 2: 20-year average annual total returns- before and after tax – 1992-2011

    Source: JP Morgan

    Over the more recent five-year period however, when measured from the beginning of 2007, shares have not performed as strongly as in past years. Over this period, as the world underwent the global financial crisis, Krugerrands on both a pre and after tax basis, outperformed all others. Listed property unit trusts also performed very strongly.

    Chart 3: Five-year average annual total returns- before and after tax – 1997-2011

    Source: JP Morgan

    If the 52-year history is anything to go by, over time so-called risk assets such as equities and property have a high probability of outperforming fixed income and inflation. However five-year rolling returns do vary considerably and therefore investors need to have a process to remain underweight when assets are expensive and overweight when they are cheap.

    Kind regards

    Ian de Lange

    021 914 4966

    Permalink2012-06-18, 08:59:36, by Mike Email , Leave a comment

    “Life Cover Is A Waste Of Money!”

    There are many people that have a mental block when it comes to life cover. The very idea that someone else is going to “benefit” from their death and that they have to “waste” their hard earned money on something that will give them no benefit, causes them to lose every trace of rational behaviour.

    There are 3 basic reasons why one would need life cover, namely;

  1. covering one’s future income stream for the benefit of dependants,
  2. covering one’s debts

  3. for the provision of liquidity in one’s estate (rather than a fire sale of prime assets).

    Once these basic obligations have been fulfilled, then one typically has sufficient life cover.

    Below is a copy of a letter I wrote to a client who, apart from his aversion to life cover, is an extremely intelligent and rational person.

    “As discussed with you previously, your present situation is that R20 000 pm is needed by your dependents as a future income. You presently have a shortfall of approximately one million rand to cover that income. One million rand life cover will cost you R261.55 pm. If we divide R261.55 into one million rand you’ll see that you have to pay this R261.55 pm for 318 years to make up the one million rand (if we assume a no interest).

    You have stated that you will rather invest R1 000 pm into SATRIX 40. If we assume a growth rate of 10% pa compounded monthly; it will take you 22½ years to build up one million rand capital. If you invest R738.45 pm into SATRIX 40 and R261.55 into a life policy it will take you just over 25 years to build up one million rand capital, but you would not have the down side risk of insufficient capital should you prematurely pass away.”

    Despite the obvious risks that the client’s dependants are exposed to while he is building up capital, he decided not to take the cover. To date, they haven’t needed it, but they still have 22 years before the SATRIX 40 investment will probably reach one million rand!

    Another argument I often hear against the use of life cover, especially when it is for individuals with high value but illiquid estates (like buildings or farms), is that their heirs should sell a building or take out a bond. However if we look at a 50 year old male who has a sizeable estate and needs to pay R5million in estate duty, a R5 million building yielding 10% would produce an income of R500 000 pa, while a bond of R5 million over 10 years would cost approximately R66 075 p.m. It generally does not make sense that an otherwise rational businessman would want to choose these options rather than pay a life cover premium of R2 600pm.

    Unfortunately life cover, like rugby and politics, seems to be one of those areas where reason often takes a back seat and emotions take control. It is important to make an accurate assessment of future obligations and then to assess the most efficient way to meet this obligation.

    Kind regards,

    Barry Hugo

    021 914 4966

  4. Permalink2012-06-05, 11:58:44, by Mike Email , Leave a comment