Diversification - but not as we know it
Currency exposure in global portfolios presents a set of risks not present in other investment decisions. International investing means the currency decision cannot be avoided. Whether that means passive hedging, active management, or deliberate or accidental unhedged currency exposure, there is no universally accepted safe position to take.
Around the world, funds are placing ever increasing amounts offshore in the search for extra returns against a current world environment of diminishing opportunities at every turn. The increasing importance of currency is a global phenomenon associated with this expansion. The rationale for different policy decisions on currency are diverse and sometimes compelling.
The selection of the right policy for a strategic benchmark is beyond the scope of this article. However one of the omnipresent reasons for taking on currency exposure (that is, unhedged currency positions) is diversification. The mistaken belief that currency diversification is the same as asset diversification is a key starting point in deciding what style of currency management may be right for a given fund.
It is common for currencies to be treated in the same way as asset investments when it comes to asset allocation and other policy decisions. This is inappropriate and leads to poor investment choices. Currency exposure should be considered in the same way as any overlay in a fund. An overlay with one very important difference – the level of currency exposure is typically much higher than the allocation to any overlay manager.
Currency is an exposure, not an investment. A fund invests capital in foreign currency denominated assets (such as equities, bonds or private equity). Currency exposure does not require the fund to invest further capital. Currency comes together with the foreign currency asset. Currency exposure is a long/short position: long the foreign currency, and short the base currency. So there are two components of international return: the return on the capital invested in the international asset and the currency exposure translation swings.
As a concept, diversification is well understood. So the textbook theory goes – additional assets in a portfolio will lower the total risk so long as the correlation of the new assets with existing assets is less than one. This is absolutely correct but, unfortunately like many issues in investments, it is oversimplified and misleading. The true picture of diversification is shown in Figure 1.
Each line represents the introduction of a new asset to an existing portfolio and shows the risk reduction received depending on the riskiness of the new asset and its correlation to the existing assets. The incoming asset has a weight of 25%. This weighting was chosen for comparison with a currency overlay representing 25% of the portfolio below .
The legend indicates the level of risk of the incoming asset. For example, the 200% line shows the effect of introducing a new asset with double the risk of the current portfolio. The red arrow shows the correlation at which the 150% risk line begins providing diversification – it crosses the 0% reduction level for correlations below 50%.
The 100% (yellow) line is what we normally associate with diversification – adding an asset of equal risk to the current portfolio delivers a risk reduction for all correlations except +100%. In most cases it is unlikely that a new asset addition will have the same level of risk as the existing portfolio. Adding an asset of less risk, the bottom line on the chart at 20% of the risk, always produces a reduction in risk (of between 20% and 40%). Conversely adding a much riskier asset, say the 200% line with double the risk, only delivers a risk reduction if the correlation is less than +10%. This chart is a reflection of the property that new assets added to a portfolio reduce the weights of existing holdings. This is what we call asset diversification.
Currency, on the other hand, has no effect on existing asset weights. The new risk that accompanies currency is additive, and means that currency diversification is quite different, as shown in Figure 2.
Although the two charts look similar, note that the risk lines for the overlay are all shifted up and to the right. The other striking feature is the lack of chart lines completely below the 0% level, compared with the previous chart. When you add an overlay of equivalent risk and 25% weight to an existing portfolio, diversification occurs only when the correlation between the overlay and the exiting portfolio is less than -10%. Even more important, the potential for risk to be increased is dramatic. It is likely that the risk of an overlay compared with a diversified portfolio will be higher than 100%, thus the probability of adding total risk at the fund level from an overlay of this size is quite high. For correlations greater than 0%, increasing total fund risk by 20% to 40% is a strong possibility.
Our starting proposition therefore is that large currency exposures require robust risk control because currency behaves differently and interacts differently than other components of a global fund. Currency is a risk first and return second activity. Moreover, it is important to understand the difference between risk control and alpha. Alpha strategies spend part of the fund’s risk budget. Risk controlling strategies can increase the amount of risk available for return generating activities. They contribute to the diversification of the total portfolio – in essence increasing the unused portion of the risk budget.
Large sophisticated funds in the US and Europe are benefiting from this: They employ a barbell strategy of a low-risk overlay combined with an aggressive build toward a 10% allocation in high-risk alternatives. This is only possible because of the extra risk budget made available by the risk control currency overlay, as shown in Figure 3.
Currency risk and return does not fit into the normal alpha/beta classification. Incorrectly forcing it to reside there leads to poor investment choices. Currency risk is specific risk with no expected return, but which uses up a significant portion of the total risk budget. Currency exposure of the average European fund would use up 13.7% of the total risk budget if left unmanaged .
Active currency risk control takes existing risks and intervenes in such a way as to reduce downside risk and add value. An alpha strategy, on the other hand, creates risk that didn’t previously exist in an attempt to add value. The risk budget effect of these clearly divergent strategies is striking. Risk control currency management is the only active process that actually reduces the deployed risk in the fund. This makes additional risk available for extra investment elsewhere.
Accepting the necessary risk focus of currency exposure, it may seem that passive hedging is an effective technique for managing this risk. Indeed passive hedging through forwards is a widely used and accepted low-risk response to currency risk management. However, large negative cashflows when the base currency is weak, expose the Achilles heel of this seemingly simple solution. Moreover, locking in a passive 50% hedged position (the principle of “least regret”) actually guarantees regret going forward, as performance will always be second best to one of the extreme alternatives.
A properly implemented risk controlled active currency overlay has the ability to simultaneously minimise the regret of passive hedging and add significant value in the long run without additional funding being required. Unlike a pure alpha strategy, overlay has the advantage that the value added is delivered in such a way that the purchasing power of members’ funds is protected. Cashflows are positive when the translation effect on the underlying assets is negative, and negative cashflows are minimised when there are unrealised gains in the underlying. An alpha strategy may add value, but not necessarily when it is needed most.
The management of currency exposure needs to be considered differently from almost every other component of a global portfolio. Offshore investment is an ever-increasing portion of global funds with a larger array of components than ever before – hedge funds, long-short and private equity to name a few. Unlike the majority of these alternatives, currency often represents a significant fund level risk in its own right.
It is vital for the risk budget to work as hard as possible in delivering the best outcome to a fund. Unmanaged currency uses up a large portion of this risk budget for no expected return, and has dubious diversification benefits at best. Due to its unique risk/return characteristics, currency necessarily needs to be separated from the alpha/beta debate and considered separately in risk/return terms.
Ironically, the search for extra return in a world of falling equity risk premiums, falling returns to hedge funds, rising bond yields and general pricing uncertainty in many other investments, is leading to increased international allocations and thus increased currency risk. Currency therefore represents the barrier to greater international exposure and the possible releasing of funding pressure for many.
One of the few viable solutions to this dilemma is to reduce the amount of unrewarded risk budget that is squandered by unmanaged currency exposure. Employing active currency risk management can both add value and free up valuable risk budget. This can then be deployed more effectively in separately specified alpha strategies. This recycling of unrewarded risk offers the opportunity to enhance total fund return for the same or lower level of total risk.
Jim Coleman is senior vice president, research and Michael Shilling is managing director portfolio services at Pareto Partners in New York