How much offshore is enough?

In the last few years South African-based investors would often ask how much offshore exposure is needed in an overall portfolio. There are many reasons to include an offshore allocation in a portfolio, such as slow local economic growth, low SA GDP per capita, political and government instability, heightened awareness of corruption and a much bigger global opportunity set or investment universe.

This conversation has also returned to the spotlight following the recently released Budget Review document, which – buried deep inside – stated that ‘The offshore limit for all insurance, retirement and savings funds is harmonised at 45% inclusive of the 10% African allowance’. A new exchange control circular was issued by the South African Reserve Bank the next day, confirming the provisions.

Figure 1: At 45%, offshore can now be a meaningful part of a retirement portfolio

Source: South African Reserve Bank

In addition to the absolute offshore limit, investment horizon, risk profile, asset and liability matching and blending of different investment styles and asset classes also form part of this complex portfolio construction debate. There are various ways of trying to solve the puzzle for clients.

Investors can start by asking what expected returns capital markets will deliver over the long term. What mix of assets will then help to deliver a specific real-return outcome (such as inflation +5%), within an allowable risk budget, over the long term? We typically refer to this as ‘the crystal ball method’. Investors hope and believe that the investment team or consultant can predict how markets will perform over the long term and based on those assumptions have the optimal asset allocation and asset selection methods with the appropriate volatility and risk blending.

Another way is to slice and dice the available market data to show what would have been the best way to invest with perfect hindsight. You can even try to optimise the risk profile when analysing the data. For South African investors we can access data as far back as 1900, but we would need to disclaim this approach with: ‘Past performance is not an indicator of future performance.’

What makes every analysis even more complicated is that you need to take an investment view on one of the most volatile and liquid emerging market currencies in the world. At a standard deviation of c.15% per annum over the last 15 years, which is similar to most equity markets, the South African rand is sometimes called the rattler, for very good reason. With greater portions of portfolios invested in global markets, the decision around how to manage (hedge) the currency will also become a much more material component of decision making, which could have a significant impact on future returns.

Vast difference in historic asset class performance

Looking back over the past 60 years of domestic macro history, one could easily point to episodes of low growth, rising unemployment, debt downgrades, a depreciating exchange rate and a sell-off in markets. These include the mid-1980s, the early 1990s, the early 2000s and of course the global financial crisis of 2008.

We have decided to use the last three decades as data examples of how differently the South African asset class opportunities have performed relative to global market opportunities. As can be expected, there were various reasons why markets behaved so differently over these periods, but it demonstrates just how difficult it is to answer the initial question of this article.

Figure 2: Vast difference in asset class returns per decade over the last 30 years
Annualised returns in Rand per asset class for each of the last 3 decades

Source: Ninety One, Bloomberg

At a high level, the decade from 1990 to 2000 was the period during which exchange controls went from zero to 15% (retirement funds) and 20% (collective investment scheme funds). South Africa re-entered the global stage. In the period from 1999 to 2008 we experienced a bull market driven by resources, followed by the global financial crisis and resultant bear market in 2008 . From 2009 to 2015 we experienced a broad market recovery led by industrials and financials, with resources dithering, and from April 2015 to the end of 2019 our stock exchange went sideways, with Naspers and resources carrying the market to some extent. In 2020, investors were faced with the COVID-19 pandemic. It is also fair to note that the period of 2010 to 2020 was blighted by a ‘State of Corruption’, low investment and very slow economic growth in South Africa, while global economic growth was underpinned by exceptionally supportive monetary conditions.

We have also plotted rolling SA equity returns over rolling 10-year periods (longer-term cycles) relative to offshore equity returns, converted back to South African rand, to show how different periods in the past delivered different onshore versus offshore equity returns.

Figure 3: SA Equities relative to Offshore equities (in ZAR) – Rolling 10 years

Source: Ninety One, Bloomberg. Monthly rolling 10-year relative returns from December 1979 to 31 March 2022

Figure 3 shows that over medium- to longer-term cycles money managers will continuously have to make an active onshore and offshore equity decision. We believe this decision is best made on a bottom-up selection method based on a consistent investment philosophy and process, where the local equity opportunities can be evaluated relative to the global opportunities based on the fundamentals, the price paid (valuation) and market price behaviour.

Portfolio construction matters

A typical balanced manager in SA would target 65% in overall equity as a ‘neutral through the cycle’ position.
This has proven to be a good starting point to extract the equity risk premium from markets to help achieve an outcome of inflation +5%. Roughly 22% of the portfolio would be invested in offshore equities, with the bulk (43%) invested in SA equities. To balance the risk and volatility profile of the portfolio the remainder was invested in income assets and mostly into South African government bonds to achieve good real returns, offsetting equity market volatility.

Going forward the envisaged neutral allocation could be 35% in SA equity and 30% offshore equity. This would allow investors to access the higher growth opportunity set on the offshore side compared to a more limited opportunity set locally. The balance of the portfolio would still access high-income yields from SA fixed income assets as well as select global fixed income and other asset class opportunities. We propose that these assets are currency hedged (back into South African rand), providing adequate real returns relative to SA inflation, but also to enhance the risk profile of the overall portfolio.

If a manager turned more cautious the growth assets of the portfolio would potentially be closer to ~45%,
but when the manager wanted to get the sails up, the growth assets could go as high as 75%.

Figure 4: Active asset allocation is a key lever to enhance returns and manage risk
Asset allocation for Ninety One Balanced strategy over time

Source: Ninety One. Monthly asset allocation of the Balanced strategy from 31 December 2005 to 31 March 2022

Figure 5: Exposure to equities is actively managed to mitigate potential risk, but ensure growth opportunities are captured

Source: Ninety One. Range of exposure to equities within the Balanced strategy from 31 December 2005 to 31 March 2022

We believe the management of this asset allocation framework on a strategic view (medium to longer term) and especially over the shorter term as opportunities and risks unfold, can be done more effectively and be better optimised if part of one integrated portfolio. The monitoring and management of this holistic portfolio can be managed daily by an investment team and risk management team. Transparency of what is held in the portfolio from a bottom-up perspective and the level of granularity is only possible when a holistic fund management approach is taken. By segmenting the underlying investments (specialist component approach each with a unique benchmark), the consolidation and management of all the underlying components will be removed from the investment team’s decision-making framework. This shifts the responsibility of daily consolidation and risk management squarely onto the shoulders of the client, consultant or aggregator.

In our view, what you hold within each of the asset classes is even more important than how much you own of each of the asset classes. For instance, you need to consider the local equities you own in the context of your offshore equity holdings (and vice versa) to ensure you have an optimal selection of equities.

To illustrate, investors are familiar with the global rand hedge Industrial equities in our market (Naspers, Richemont, British American Tobacco, Bidcorp), the resources stocks (Anglo American, BHP Group, Glencore, Sasol, Gold, Platinum and Paper) and the SA Inc. basket (banks, insurers, retailers, property). How much you own in each of these broad local sectors should influence your sectoral decision making (technology, healthcare, utilities, food & beverage, financials, materials, energy and semi-conductors) and your geographical decision-making (developed markets such as US, Europe, Japan or more emerging markets such as China, India, Brazil). You would not want to add risk by buying more of the same offshore and thus end up with a highly risky portfolio.

The rattler and forex management

Currency management to help manage and reduce return volatility will now become a bigger debate as up to 45% offshore will bring more currency volatility for a South African-based investor. A stronger than expected rand could negatively impact offshore returns for investors as we have experienced more recently. In addition to the onshore and offshore conversation, currency or forex management as a return calculator should also be viewed as an asset class or return enhancement opportunity.

Using the returns (as shown in Figure 2 above) and volatility of the different asset classes as well as the correlations between the different asset classes, the graph below shows the ‘optimal’ multi-asset balanced portfolio for each of the last three decades. It is clear that due to ongoing changes in market conditions, the ‘optimal portfolio’ has shifted dramatically from one decade to the next.

Figure 6: Optimised asset allocations per decade over the last 30 years

Source: Ninety One, Bloomberg. Portfolios for each period are optimized to maximise risk-adjusted returns for a CPI+5% return profile (Offshore capped at 45% and Equities capped at 75%). Hedge asset class returns reflect the US$ return plus the relative interest rate spread between SA and the US

In the 1990s, a portfolio with a primary allocation to global equities (but with the offshore exposure hedged back into rand) and domestic bonds would have resulted in the best risk-adjusted outcome to meet CPI +5% objectives. In the next decade, the best portfolio would have almost been exactly the opposite when compared to the previous decade. The ‘optimal’ portfolio for the 2000s would have been a large allocation to domestic equities and offshore bonds (also fully hedged) with almost no exposure to offshore equities or domestic bonds. Considering the significant currency depreciation we have experienced over the last decade, it is not surprising to note that optimising a portfolio based on market returns in the 2010s would result in basically no currency hedge, with a large allocation to global equities and domestic bonds.

Given these changing market conditions, coupled with significant currency volatility, investors should be extremely careful not to hang their hat on a single long-term ‘optimal solution’ and then allocate assets on a fixed basis. When looking just at the volatility for each of the asset classes over the different periods, it is interesting to note that offshore cash in rands has been more volatile than SA Equities – purely due to exchange rate volatility. Volatility and the management of currencies as well as the onshore versus offshore allocation on both a longer-term strategic and over the shorter term will be critical for fund managers to manage on behalf of their clients.

Global equities (unhedged) offer the opportunity to access greater growth ideas on a global scale, but it will introduce more volatility to performance. Actively adjusting allocations and currency exposures, driven by a repeatable process, should help to deliver a better risk/return outcome.

It depends

Managing the opportunity of greater global access has become the greatest talking point for trustees, consultants, pension fund members, retail investors and fund managers. We embrace the fact that more flexibility should and will help to deliver better return outcomes for our clients. Greater GDP growth in other geographies relative to the sub-2% expected GDP growth for South Africa should deliver better return opportunities for growth assets over time. But greater flexibility also means more responsibility and potentially bigger return volatility.

We have tried to give some context to our thinking and an indication of the tramlines we use when we construct a globally integrated portfolio. Based on our analysis and modelling we believe that the answer to the question, ‘How much should be invested outside of South Africa if there were no regulatory limits,’ is that it depends on the bottom-up opportunities that markets provide together with a rigorous and dynamic strategic asset allocation framework. Our objective is to construct the optimal global multi-asset portfolio on behalf of our clients using a globally integrated approach and a consistent asset allocation process (based on fundamentals, valuation and market price behaviour) across various asset classes and geographies, combined with an integrated bottom-up selection philosophy and process.

Figure 8: Ninety One Balanced with International outperforming all the primary asset classes
Annualised returns: 15 years as at 31 March 2022

Source: Ninety One, Bloomberg. Indices: SA Inflation (SA Headline CPI), SA Cash (SteFi Composite), SA Bonds (ALBI), SA Equities (Capped SWIX | SWIX pre 010415), Global Equities (MSCI ACWI in ZAR), AF LMW (Global LMW median)

How you use the historic data, your decisions around currency management, as well as your investment and expected return outlook will have a significant impact on your strategic portfolio construction and desired investment outcome. We have successfully managed integrated onshore and offshore portfolios within responsible risk and regulatory allowances over the last three decades. Wherever regulations continue to evolve, we will evolve with it and remain committed to delivering good investment outcomes for our clients.

Download PDF