Currency risk deserves more than a coin toss

Share

More Australians are investing in offshore equity and bond markets, but they may underappreciate how currency fluctuations either cushion or exacerbate gains and losses in their portfolio.

For many investors, the decision to hedge the currency exposure on offshore investment portfolios is often seen as a binary choice akin to a coin toss – heads or tails, hedged or unhedged.

Investors spend much of their effort deciding which asset class to allocate to, but often devote less consideration to how decisions around currency exposure can dramatically improve the risk-adjusted returns of their portfolio.

The hedging decision and avoiding regret

When investors buy foreign assets, they obtain exposure not only to the underlying securities but also to foreign currency. Movements in exchange rates play a significant role in determining the performance of a foreign investment. A second-order decision is whether to hedge this foreign-currency exposure. Many investors implicitly weigh up their outlooks for the respective currency and their appetite for losses due to adverse currency movement (the so-called ‘regret tolerance’).

Regret theory assumes that investors are rational but base their decisions not only on expected payoffs (‘value’) but also on expected regret. Sebastien Michenaud and Bruno Solnik (2008) found that risk-averse investors dislike volatility, whereas regret-averse investors may prefer volatility to avoid regret for giving up on the large gains that may materialise.

They found that many investors, when considering both volatility and regret, tend to err towards regret avoidance, and under-hedge even when fully hedging is the optimal decision based on all available information and traditional risk-return optimisation.

It is little wonder then, according to their research, that approximately 39% of currency investors do not hedge, 34% adopt a 50% hedging strategy, 14% adopt of a 100% hedging strategy, and the remaining 13% adopt some other hedge ratio.

The impact of currency on global portfolios

Whilst many investors make a binary choice to hedge 100% or not, our own research indicates that the optimal hedging decision varies considerably over time.

For Australian investors:

  • If the AUD is appreciating, a hedged global equity portfolio outperforms an unhedged one.
  • If the AUD is depreciating, an unhedged global equity portfolio outperforms a hedged one.

This can be seen from the graphs and table below.

MSCI World Annual Return
To Sept 2017 Unhedged 100% Hedged
1 Year 15.9% 19.6%
3 Years 12.3% 10.7%
5 Years 18.1% 15.0%
10 Years 6.1% 6.7%
20 Years 5.7% 7.3%

Source: Bloomberg, MSCI

The outperformance can be substantial and persistent. In the table above, the unhedged global equity portfolio has outperformed the hedged by 3% annually for the last five years. That is, the AUD has strengthened and it was better to be hedged than exposed to the ‘weaker’ global currencies in an unhedged position.

So then, what is the ‘best’ hedging ratio that both captures good returns and prevents extremes of underperformance?

The optimal hedge ratio over the (very) long term

The AUD is often characterised as a commodity currency, meaning that its movements are correlated to commodity price cycles. This means that the AUD appreciates during economic booms when commodity prices are higher and depreciates when the cycle turns.

To figure out the optimal hedge ratio, a set of different portfolios were created using different hedging assumptions. Their returns vs. risk were then plotted over time, essentially creating an efficient frontier (series of possible return outcomes for a given level of risk) for different hedging solutions, as illustrated below.

MSCI World
  Unhedged 100% Hedged to AUD
Annual Return 7.2% 9.6%
Annual Risk 13.4% 14.1%
Return/Risk 0.5 0.7

Source: Bloomberg, MSCI, Tempo AM

The graph above illustrates that, over this long run (about 30 years), the hedged equity portfolio generally outperforms the unhedged one. Unfortunately, while this analysis is interesting, it is not very useful for investors.

The difficulty in figuring out an optimal hedge ratio is that this ratio changes over time. Certain hedge ratios are ‘optimal’ for a certain time interval while others are ‘optimal’ for other time intervals.

For currencies, ‘optimal’ is a slippery concept

It is worth noting that while the 40% – 100% hedging ratio illustrated above is optimal over the very long term, during specific periods other hedging ratios may be better to avoid cash drawdowns required by hedging counterparties.

For example, during the GFC from 2006 to 2008, where the AUD/USD dropped from 0.98 down to 0.60, the best hedging strategy was to have no hedge at all (that is, exposure is left open to foreign currencies which increase in value as the AUD falls). Maintaining a 100% hedged strategy would have cost an extra -2.4% per year over these three years compared with not hedging.

However, extending the analysis an extra two years and including the period from 2009 to 2010 when the AUD/USD bounced back from 0.60 back to over 0.90 again, the optimal solution would have been to 100% hedge. It’s exactly the opposite of the previous analysis!

In this second scenario, maintaining a 100% hedge ratio over the full timeframe would have benefited Australian clients by nearly 6% annually over an unhedged portfolio.

Conclusions

The complexity of investing requires that portfolio management is best done by asset class or sector specialists, and currency should be no different.

Currency management requires specialist portfolio management skills, systems, analytical frameworks, and sell-side broker relationships that are different from mainstream asset classes. Specialist currency managers are also well placed to ensure trading execution is done in the most cost- and timing-efficient manner.

Furthermore, using currency hedging specialists, as opposed to embedding currency management within investment products allows for portfolio managers and products to be adjusted independently of the currency overlay avoiding unnecessary tax consequences and trading.

 

Joseph Bracken and Robert Chapman are principals of Tempo Asset Management. Tempo is a global equity alliance partner of Fidante, a sponsor of Cuffelinks. This article is in the nature of general information and does not consider the circumstances of any investor. Reference: Michenaud, S & Solnik, B., 2008. Applying Regret Theory to Investment Choices: Currency Hedging Decisions, Journal of International Money and Finance, vol. 27, no. 5

Share
Print Friendly, PDF & Email

, , , , , ,

10 Responses to Currency risk deserves more than a coin toss

  1. Adrian February 25, 2018 at 8:14 AM #

    Surely there are still transactional costs and bid/ask spreads, no matter how small? That’s a good point on the forward curve, although it may be shifting this year as the US raises and RBA remains on hold. There may be other issues too like tax effects as the hedge gain/loss may get realised each year for tax, even though the main portfolio gain/loss is unrealised. And any working capital for the hedging (margins, etc).

    Still for me the key point is investors should think about their overall currency risk, not just their global shares.

    • Warren Bird February 26, 2018 at 3:52 PM #

      Adrian, wholesale fixed income and forward currency markets don’t operate on a brokerage or transaction cost basis. Like all markets, there’s a buy-sell spread, but this mostly comes in hedging currency from the spot rate, and those costs are greater in the purchase of the foreign asset rather than the hedging transaction. As this sort of transaction has become more commodity-like over the years, for wholesale investors the costs have become very tight.

      When I used to manage a hedged global bond portfolio, we estimated this buy-sell transaction cost impact was no more than 0.2% per annum. Historically, that’s been totally swamped by the 3, 4 or 5% pick up you earned, for the reasons I outlined in the article that I linked to.

      So the total hedging process does not cost you money.

      Besides, what does earning ‘lower returns’ mean? Yes, if exchange rates don’t change even an iota, then you will earn 0.2% per year less than the local currency market return of the global portfolio. But that’s a straw man argument because exchange rates do fluctuate. No one earns the ‘local currency’ return on a global portfolio. They either hedge and earn local currency plus their hedge premium, or they aren’t hedged and they earn local currency adjusted for exchange rate fluctuations which may be in any direction.

      I’m not arguing for hedging per se. But please don’t reject it because of wrong thinking about what it ‘costs’. Reject it because you want to accept currency fluctuations in your portfolio.

      (By the way, the cost side of things in the wholesale markets is the total opposite of the ridiculous buy-sell spreads that retail investors have to incur when they load funds on their Travel Money cards or similar. That is a rip-off in my view. )

  2. Ryan February 23, 2018 at 4:13 AM #

    There is an inherent additional cost to hedge a portfolio (to purchase futures contract etc). So if we imagine there was no currency movement, with a hedged portfolio you are automatically incurring higher costs, and therefore lower returns.

    For the realistic case where there is currency movement, I ask myself whether we believe a currency specialist can consistently and accurately predict FX price direction and turning points. I’m dubious they can add value. So for me, cost and reducing 100% exposure to AUD (as per Adrian’s comment), means I have my portfolio unhedged.

    • Warren Bird February 23, 2018 at 6:02 PM #

      Ryan, such transaction costs are minimal in the scheme of things and don’t apply when a professionally managed fund hedges. The idea that ‘hedging costs you money’ is one of the great misconceptions. Hedging earns Australians a premium as I explained in my article on the subject a couple of years back: .
      http://cuffelinks.com.au/managing-foreign-exchange-risk/

  3. Andre Lavoipierre February 22, 2018 at 3:44 PM #

    Does anyone know who provides Currency Overlays for individual clients as most only provide this service for institutional investors? Can use AUDS etf but that is expensive as you need to buy 40% to 50% of the value you wish to hedge and these funds earn nil so there is a cost of carry here.

  4. Adrian February 22, 2018 at 12:01 PM #

    Some investors might like to have exposure to overseas currency. We are naturally hedged as we travel overseas, and even many goods & services consumed domestically have an underlying cost derived from overseas currencies. If you hold all your investments in AUD (residential property, ASX shares and global hedged shares) you might seem quite exposed to a falling AUD. Personally I don’t mind leaving most of my global shares unhedged for this reason. Maybe best to think of risk management holistically, not just based on the return of your global shares in isolation?

    • Graham Hand February 22, 2018 at 12:14 PM #

      Great comment, Adrian. I feel the same way. I expect to spend a lot of time travelling overseas and spending USD and EUR, so happy to be unhedged in global investments.

  5. Warren Bird February 22, 2018 at 11:00 AM #

    OK chaps, let’s get a fact straight. At the start of 1997 the Australian dollar was trading at about the same levels as it as at the moment. During January 1997 against the USD it was between around 76.5 and 79 cents. It’s 78 cents right now. It didn’t reach the lows that Ashley refers to until 2000-2001.

    So the criticism that this chart biases the hedged returns because the AUD has appreciated over the period is wrong. The starting point matters, but in this particular case it’s not an issue.

    The reason for the outperformance in the chart is because of the hedging premium that’s consistently been picked up by investors with a hedged portfolio.

    The bigger questions to ask about this article, in my view, are:

    – on what basis can we have any confidence that a specialist currency manager can add any value compared with adopting a simple strategy? I remember 2000-01 well because we had a currency specialist in place doing an overlay on some funds at the time, and just when they should have been lifting all our hedges because at 50 cents it was way undervalued, they changed their view and believed that the low exchange rate was here to stay! They were terminated fairly soon after, not just for getting the market wrong but for demonstrating that their process was a bit too flexible.

    – more importantly, on what basis do you think that the hedge premium will continue to deliver positive returns above local currency outcomes? It’s disappeared between Australia and the US recently. Is that likely to be permanent or temporary? Rather than speculating about the volatility of the currency, the key question for me is whether to reduce my strategic hedge ratio.

  6. Frank D February 21, 2018 at 10:31 PM #

    Agree with Ashley, the table highlights that starting point is a very important consideration rather than some historical analysis.

    Not sure about paragraph under the first table, it refers to “substantial and persistent” but then quotes a 5 year number. Yet in the same table there is a longer dated number… which could represent greater persistency… which is also substantial… but is in the opposite direction! Shows how difficult defining ‘optimal’ is.

  7. Ashley February 21, 2018 at 10:25 PM #

    In any of these studies, the starting point is arbitrary. This happens to select a 1997 starting point – which is when the AUD was at its dot com low – therefore hedged returns will always look better with that particular starting point.

    · There is no such thing as an ‘optimal’ hedge ratio because correlations are not static and stable

    · But they are right in pointing out that hedge ratio decision is usually a casual afterthought – but it should be as important as the weighting decision on global shares. Personally I have added far more value to portfolio returns over the years from the hedge ratio decisions than active managers have added in alpha. Eg active manager alpha can add 2-3% or so (if they are lucky) but hedge ratio decisions in the big swings adds 10+% to returns – eg unhedged in 2008 sell-off then hedged in 09 rebound, unhedged in 2011 sell-off then hedged in 2012-14 QE boom, un-hedged in 2015 sell-off then hedged in 2016-7 rebound.

Leave a Comment:

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Register for our free weekly newsletter

New registrations receive free copies of our special investment ebooks.