Passive hedging of currency exposure of non-domestic equity portfolios can be problematic because of fluctuations in the net asset value of the portfolio itself. The purpose of electing a passive hedging strategy is to eliminate currency risk. However, the portfolio volatility will continuously create a mismatch with the forward contracts that the manager has put in place to hedge portfolio currency risk. In practice, the impact from this can be significant. For example a sterling-based client investing into MSCI Europe and adjusting currency hedges quarterly could have lost a total of 8% in performance since 1988.
The process by which managers adjust the currency hedges is here referred to as the rebalancing strategy. In theory this mismatch can be eliminated by making an infinite number of adjustments to the currency hedges, thereby creating an exact match to the portfolio value at all times. This, however, would lead to an infinitely cumbersome and costly process that is neither a viable nor practical solution. It naturally raises the question as to what is the best way of eliminating as much currency risk as possible without inccurring excessive costs. This article demonstrates that it is possible to find a strategy that provides the optimal trade-off between performance impact and transaction cost and will also suggest a measure of hedge efficiency for the various strategies available to the manager.
Rebalancing Strategy Has No Impact
Occurring On Cash Flows!
One of the characteristics of currency hedging is that it creates a temporary cash deficit/surplus generated by the requirement to settle the marked-to-market values of the forward contract every time that they are rolled (which is only matched by an unrealised gain/loss from the currency exposure on the underlying assets). This can have an impact on portfolio performance as assets might have to be sold to cover a potential cash deficit although the manager can also gain some market exposure via futures in an attempt to alleviate this.
Note from this explanation that it is the choice of roll (or settlement) date that determines when the cashflow effect impacts the portfolio, not the date that the hedge is initiated. Since the purpose of the currency hedge-rebalancing strategy is to adjust the notional of the forward contract (buy/sell the currency out to the existing settlement date) it follows that for a given settlement frequency the impact from hedge adjustments is insignificant. It is therefore appropriate to exclude cashflow considerations when selecting an optimal rebalancing strategy.
Establishing The Trade-off
It can be shown that if we assume that the initial portfolio value is always fully hedged, then the portfolio impact for any given period is equal to the return of the asset multiplied by the return of the currency for any given period. Transaction costs have here been measured in their narrowest possible definition as the bid-offer spread. This corresponds to the cost associated with buying back part of a previous currency hedge and has been set at 0.05%. No cost is added when increasing a currency hedge. Using these assumptions and formulas, we can calculate the effect of applying various adjustment strategies on a series of actual daily return data for equity and currency exposure. The rebalancing strategies tested are various fixed frequency (periodic), asset trigger (percentage change in asset) as well as a combination of the two (quarterly rebalancing and asset trigger). The portfolio analysed in this example is that of a sterling-based investor into MSCI Europe since 1988. It is important, however, to stress that this concept is generic and can be applied across all developed currencies and indices with identical conclusions.
Figure 1 provides a graphic expression of the results. (Percentage numbers on chart indicate the various trigger levels and the numbers are days-in-between hedge adjustments so that "1"=daily, "5"=weekly etc). First, it shows the trade-off between expected annualised transaction costs and tracking error for each strategy. As expected, the maximum expected cost of around 4.5bps a year comes with daily rebalancing ("1"). Decreasing frequency and/or increasing trigger levels will rapidly reduce costs toward zero. This will however eventually result in a significant increase in expected tracking error. An information ratio can be calculated to represent this relationship but it is also visually apparent that the hedge frequency can be reduced to weekly without adding significant tracking error. Hereafter the benefit of reducing costs quickly disappears. Second, the chart shows that it is always possible to find a periodic strategy with a better trade-off than the other options.
Hedge Efficiency
As a measure of perfect hedge efficiency try and imagine a portfolio with an effective hedge ratio of 100% and assume that hedging costs are zero. By reducing the hedge ratio in small increments a series of portfolios, increasingly sensitive to currency fluctuations (and thereby less efficient), can be generated. These portfolios can now be used to benchmark the hedge efficiency of the various strategies analysed above in a simple percentage term. The table below illustrates the hedge efficiencies available to the sterling-based investor.
We see that (because of the high volatility of MSCI Euro) that it is not possible to achieve more than 98.5% effective hedge ratio if the portfolio hedges are adjusted on a daily basis. The weekly rebalancing frequency provides the manager with a 97% effective hedge at approximately half the transaction costs. The expected annual tracking error is 12bps and 23bps respectively. Also note the poor hedge efficiency (90%) of the basic quarterly (65-day) rebalancing strategy.
Summary
Using the technology presented above it becomes relatively straightforward to determine the optimal rebalancing strategy for a given portfolio as long as the underlying volatility is known. For equities this implies that hedges are rebalanced on a weekly basis. Note that this does not imply that the entire forward position should be rolled in weekly increments. The rebalancing strategy is only concerned with adjusting forward hedges to the value of the underlying asset in order to reduce unwanted currency impact. This means that any hedge adjustment is executed out to the existing roll (or forward settlement) day and does not have any practical cash flow impact.
The hedge efficiency measurement proposed allows the manager to compare and identify the strategy that best suits their risk and cost profile. Interestingly, the trade-off with tracking error and cost can be shown to be independent of asset and/or currency volatility. The more volatile the asset, however, the more significant the rebalancing strategy becomes. Given typical manager objectives this would indicate that rebalancing processes using monthly or quarterly intervals might be sub-optimal for equity portfolios
This week's Learning Curve was written by Henrik Pedersen, v.p. in the FX risk advisory group at Citigroup in London.
