The last decade witnessed an interesting development on the South African FTSE/JSE All Share Index (ALSI): in early May 2006, the ALSI was at 21 000. To the delight of many South African investors, the same index today is at 50 000.

A foreign investor, however, would be less delighted. Over the same 9½ years, the ALSI measured in US dollars has delivered much volatility but zero by way of capital growth. Aside from dividends (which have averaged 2.8% per annum over this period), an investor looking at his principal in US dollars has seen no increase.

Since almost three-quarters of South African-listed companies' revenues and 50% of their primary listings (by market capitalisation) are offshore, it is important to reflect on the ALSI in hard currencies. The reality for South African-based investors is that our principal growth has largely come from a weakening rand exchange rate. An optimist hoping for a recovery after a decade of 'lean' returns should consider that the ALSI measured in US dollars was flat for a full 23 years between 1980 and 2003 before finally taking off.

Measured in US dollars, the aggregate profits of the 170 companies that make up the ALSI have not grown over the past 9½ years and are down 45% from their peak in early 2012. It is still not obvious that profit levels are low and, as such, we remain concerned about the high levels of valuation on the market.

Our allocation to cash, short-duration bonds and hedged equities therefore remains above what we would normally hold through a market cycle.

The Fund continues to maintain a full position in offshore assets. While it is hard to argue that the rand - from the current levels of R14/US$ or R21/ GBP - is a one-way bet, our offshore assets provide diversity by being invested across a broad range of geographies and industries.

It is also worth remembering that not all of our offshore assets are US dollar, pound sterling or euro-denominated. We are finding increasingly attractive opportunities among emerging markets, which over the past five years have underperformed developed markets substantially and are currently being dismissed as being 'too risky' by many investors. While the macro environment may deteriorate further, we are finding well-managed companies that can prosper in the long run at attractive valuations.

During the quarter there have been no major changes to the asset allocation of the Fund. The Fund reduced its stake in SAB Miller and British American Tobacco somewhat, and continued to increase its exposure to the platinum and palladium listed debentures.

Commentary contributed by Simon Raubenheimer

The Fund's lower-than-average net exposure to equities reflects our continued concern over equity valuations both locally and offshore.

South African equities have outperformed cash on average by 8.4% on an annual basis since 1900. Interestingly, the difference in any given calendar year has only been within 2% above or below the 8.4% average on six occasions. Therefore, investors are very unlikely to receive the average excess return in any given year. If we focus on the last 10 years the average excess return accruing to equities above cash rises to 12% or 43% above the long-term average. This highlights the powerful effect that starting valuations, rather than economic growth, have on future returns.

One measure of value we can use for the overall market is its price-to-earnings (PE) ratio. We expect high excess returns to show up in high valuations and indeed this is the case: the FTSE/JSE All Share Index is trading on 18x its last reported 12-months earnings - an almost 50% premium to its 55- year average of 12x.

Whilst the PE ratio is normally thought of as a fundamental indicator of value, we can also look at it as a sentiment indicator. When investors are positively disposed towards equities they are willing to pay higher prices for the future; the opposite is true when they are uncertain or even fearful. The PE ratio has fluctuated between a low of 4x and a peak of 25x since 1960, with the market typically bottoming near 8x and peaking near 17x. Of course, the market could potentially return to its late 1960s valuation of 25x - but most of the data suggests it is currently overvalued. One thing we do know for certain is that the market is not depressed. Twelve years ago the market was trading on 9x low earnings, which explains the great decade that followed - despite being interrupted by the great financial crisis in 2008.

The Fund therefore continues to hold a greater-than-average allocation to relatively short duration fixed income and hedged equities. While central banks continue to punish prudent investors who retreat to cash by keeping interest rates low or close to zero, we like the option that cash provides to buy assets at lower prices when or if sentiment turns.

We continue to search for assets that are trading below fair value and which increase diversification in a world where we see little obvious value. The Fund continues to have a maximum exposure to offshore assets, which can be invested across a broad range of industries and geographies, and a position in commodity-linked instruments, such as the platinum and palladium listed debentures discussed in our previous commentary. Another example would be the Zambezi Platinum Preference shares, which have funded Northam Platinum's recent BEE deal. The preference shares accrue a dividend of prime plus 3.5% over a 10-year period, substantially secured by Northam ordinary shares. We believe they offer an attractive return at lower-than-average risk.

During the quarter there have been no material changes to the composition of the Fund, but we have continued to increase the position in the platinum and palladium listed debentures.

Commentary contributed by Duncan Artus

The majority of the Fund's 4.7% exposure to commodity-linked instruments is to platinum and palladium metal bars through listed debentures. These metal bars provide no yield, leaving investors entirely dependent on the change in their price for a return. Why does the Fund have a bigger exposure to these zero-yield metal bars than to the property sector, which currently yields 5.1% on average?

The FTSE/JSE SA Listed Property Index has returned a stellar 25.2% p.a. over the last three years. However, a big chunk of this return has come from investors paying an ever higher price for each rand of net rental income. Distributions per share have grown at a steadier 7.7% p.a., but even this flatters the true underlying growth in net rental income. Distribution growth has been augmented by acquisitions of property at higher yields (funded by new debt and share issuance) and by companies borrowing more against ever increasing property valuations.

In short, the true underlying performance of South African property companies has not been as strong as their share prices suggest. And while a platinum bar in a vault may not provide a yield today, it will be just as shiny in 20 years' time. The same can't be said of a shiny A-grade office block - in 20 years' time it will probably need renovation, and may even be on the wrong side of town. Because property companies distribute virtually all of their net rental income, they are forced to borrow to fund substantial renovations. This works well when property valuations are rising, but not when they stagnate or fall. Just as leverage has augmented returns during the up cycle, it will detract from returns in the down cycle.

We have more conviction on the potential for rand capital gains on platinum and palladium from today's valuations. At current prices, less than half of South Africa's platinum mines generate enough revenue to cover cash operating costs and capital expenditures. South Africa has the vast bulk of the world's known platinum resources, so it is hard to see a significant new source of mine supply. Global mine supply of these metals has been falling for years already and it will probably accelerate if prices remain at these levels. Moreover, the cost to mine an ounce of platinum has grown by roughly 15% p.a. over the last decade, and it is likely to continue to grow.

Commentary contributed by Ian Liddle

The Fund's total return of 9.0% constituted a good first nine months of the year and a more difficult final quarter. The FTSE/JSE All Share Index (ALSI) returned 1.4% for the fourth quarter, but Sasol underperformed and the Orbis portfolio, which accounts for 25% of the Fund's assets, also had a weaker period.

The rapid decline in the oil price was the root cause of both Sasol and Orbis' tough quarter. Orbis owns some very attractively priced Russian businesses, as well as a few oil-related companies, that sold off with the oil price. In early September oil was trading at US$100 per barrel (/bbl), a long way from end Decembers US$57/bbl. A slowdown in emerging market oil demand coincided with strong supply growth from the US. As the price fell, many commentators became very negative on the price outlook, forecasting surpluses through the second half of 2015.
Rather than trying to forecast the near-term price fluctuations, we focus on assessing what we believe is a long-term sustainable price using the information available. Our long-term assumption, using currently available information, is US$85/bbl. The high price over the past few years has enabled oil companies to invest heavily in new projects using both operating cash flow and debt. Many of these projects are not feasible at lower prices and companies do not have the cash required to fund them. Despite the oil price only falling below US$80/bbl in November, at least half the major companies have already announced spending cuts for 2015. Lower capital spending should lead to a tighter market and price recovery. At current levels, Sasol is discounting a long-term oil price well below US$85/bbl. We think the company offers excellent value, especially in the context of a fully valued JSE.

The Fund's relatively low net share exposure of 55.9% is a function of the South African market offering few compelling value opportunities. Similarly, global markets have had an excellent five years and do not offer the value of a few years ago. In this environment we prefer low-risk fixed income investments and the Orbis Optimal SA funds to many JSE-listed equities. These liquid, low-risk assets put the Fund in a position to take advantage of opportunities as they present themselves.

Commentary contributed by Andrew Lapping