Professional Documents
Culture Documents
HANDBOOK
DORI LEVANONI
Partner
This handbook is based on First Quadrant’s internal research and is not intended to provide specific
investment advice. Except as otherwise noted, the analysis throughout this document is based on a time
period of January 1980 to December 2016 and uses the following proxies for the referenced asset classes:
Global ex-US equities is a simulated basket of developed market equities based on MSCI World Index; Global
ex-US bonds is a simulated basket of developed market bonds based on the Citi World Government Bond
Index; US bonds is simulated US bonds based on the Citi World Government Bond Index; EM equities is the
MSCI Emerging Markets Index; and EM bonds is the JPMorgan GBI-EM Global Composite Index. Please see
the end of this paper for further details on these proxies.
Having chosen the strategic hedge ratio and whether to manage pas-
sively or actively, investors are in a position to implement their curren-
cy policy. Chapter 4 covers aspects of implementation – instruments,
counterparties and information flows – to help ensure the smooth op-
eration of the currency policy. Chapter 5 is on monitoring the currency
policy. Its purpose is to indicate what kinds of ongoing reports and
flows of information will be needed to ensure that the results of their
currency policy are correctly interpreted and therefore more likely to
be adhered to throughout a currency cycle. The final chapter provides
a summary and conclusion.
5 EXECUTIVE SUMMARY
CHAPTER 1
9 Importance of having a currency policy
11 Why now?
CHAPTER 2
14 Establishing a currency policy benchmark –
the optimal strategic hedge ratio
14 Global ex-US equities
25 Global ex-US bonds
TABLE OF 26
27
Emerging markets
Other instruments
CONTENTS CHAPTER 3
28 Active/Passive Management
28 Participants in the foreign currency market
29 Types of active management
30 Performance of currency managers
34 Performance of currency factors
35 Constraints on active hedging mandates
37 Costs
CHAPTER 4
39 Implementation
39 Instruments
41 Counterparties
41 Information flows
CHAPTER 5
44 Monitoring Currency Policy
44 Specifying objectives
44 Benchmarking
45 Monitoring return and risk
45 Communication and education
47 CONCLUSION
EXECUTIVE
Currency Hedging Handbook
SUMMARY
EXECUTIVE SUMMARY
• What is the hedge ratio that best meets their investment objectives?
• Should currency exposure be managed passively or actively?
Our analysis for US dollar investors shows that the optimal hedge ratio for
global ex-US developed equities is around 75%, though in practice, ratios
from 50% to 100% can still result in a meaningful improvement in portfolio
return/risk ratio. The 50% hedge ratio has attractive properties, for example,
providing 80% of the benefits of a 75% hedge while also allowing for symme-
try in the hedging activity around the benchmark.
FIRSTQUADRANT.COM
EXECUTIVE
Currency Hedging Handbook
SUMMARY
adding almost 6% to the volatility and more return. It is not like other risk premia
than doubling the risk compared to a fully such as the equity risk premium or bond
hedged position. term premium. Empirically, the returns to
Currency risk can also have a unmanaged currency exposure over long
meaningful impact on a diversified time periods have been small, 0.5% per
portfolio. For an investor with 60% equities annum, and depend on the time period.
and 40% bonds, hedging the currency The correlation between the returns of
exposure of foreign developed equities foreign developed equities (currency hedged)
reduces portfolio risk by the same amount and currency is close to zero on average,
as a 7% shift from stocks to bonds. though it varies over time. If the correlation
The risk reduction provided by hedging were persistently and significantly negative,
foreign currency risk allows the investor currency exposure might be diversifying, but
to reallocate the risk to higher-returning this is not the case.
assets, which can aid in meeting overall Currency volatility has also fluctuated
investment objectives. over time, although we have not detected
any persistent trends in volatility since
Why now? 1980. Some have argued that globalization
Currency risk has been growing in portfolios should reduce the currency-related risk of
as allocations to foreign assets have been foreign equities, but we find no evidence
increasing, and exchange rate volatility has of this, either in equity returns or in
been rising from relatively low levels. For corporate earnings.
investors who haven’t yet put in place a
currency policy, there are strong arguments Non-US developed equities
for doing so in the near future. Our analysis shows that for a US dollar base
currency investor, the optimal strategic
Strategic hedge ratios hedge ratio for global ex-US developed
For an investor who seeks to maximize equities is right around 75%. In practice,
return per unit of risk, the strategic hedge however, hedge ratios from 50% to 100% also
ratio can be determined using the same provide substantial benefits. One important
inputs used in developing a strategic conclusion is that it’s difficult to justify, from
asset allocation: expected returns, risks an investment efficiency standpoint, having
and correlations for the assets and a currency benchmark that is unhedged.
currencies. We undertake the analysis for Nevertheless, unhedged benchmarks are
each international asset class separately not uncommon. This implies that for many
as they have very different volatilities and investors, a meaningful improvement in
correlations to foreign currencies. portfolio return/risk can be readily achieved
Unmanaged currency risk should not be by implementing a strategic hedge ratio
expected to generate a positive (or negative) between 50% and 100%.
FIRSTQUADRANT.COM
EXECUTIVE
Currency Hedging Handbook
SUMMARY
FIRSTQUADRANT.COM
EXECUTIVE
Currency Hedging Handbook
SUMMARY
FIRSTQUADRANT.COM
CHAPTER 1 Currency Hedging Handbook
FIRSTQUADRANT.COM
Currency Hedging Handbook CHAPTER 1
with global ex-US developed equities, an into US and global ex-US components.³ The
unhedged position adds 2.6% to volatility weight of global ex-US equities was varied
compared to a fully hedged position, from 30% of the portfolio weight (i.e., half of
an 18% increase over this time period. total equities, or roughly what it would be in a
With global ex-US developed bonds, the global market capitalization weighted index)
increase in volatility is dramatic. Currency down to 10%. This is shown in Figure 01.
risk dominates, adding almost 6% to the An investor with a 30% allocation to
volatility and more than doubling the global ex-US equities would reduce total
risk compared to a fully hedged position. portfolio risk from 9.7% to 9.0% by hedging
Unhedged emerging market equities are away currency risk. To put this in context,
riskier than EM equities in local currency.² this is equivalent to a 7% shift from stocks
The volatility of EM bonds is almost entirely to bonds. In other words, if 9.7% is the right
due to currency exposure. amount of total portfolio risk, that level
Not only can currency risk be a significant could be obtained either from a portfolio
portion of risk at the asset class level, it with 30% foreign equities unhedged/30%
0.00%
Percent of Risk Reduction
-0.20%
-0.40%
-0.60%
-0.80%
10% 15% 20% 25% 30%
Weight of global ex-US equities
10
FIRSTQUADRANT.COM
CHAPTER 1 Currency Hedging Handbook
hedging allows the investor to reallocate the Much of the reason for tackling these
risk to higher-returning assets, which can issues and bundling them into something
aid in meeting overall investment objectives. we’re calling a currency policy is due to the
The data show that currency risk is potential challenges of governance. Currency
significant for most globally diversified has a significant impact on the risk profile of
investors, but what should be done about most globally invested portfolios, and merits
it? There is a compelling case for investors the same degree of attention as other major
to formulate and adopt a currency policy, investment policy questions. Some funds
which is a summary of the analysis and require major investment policy decisions
reasoning regarding a fund’s currency to be approved by a board or committee;
management. It may or may not entail a other funds are less formal. Adopting a
commitment to hedge, but should include sound and rigorously vetted currency policy
two main parts. The first is establishing the promotes consistency through changes in
strategic hedge ratio, i.e., the hedge ratio fund personnel and during periods when
that, if applied on a static basis, does the short-term performance is disappointing.
best job of meeting long-term investment If a fund lacks a currency policy, every
objectives. The strategic hedge ratio is time there is a currency performance
the fund’s internal currency benchmark. surprise – especially when it’s a negative
It may be viewed as the currency version one – the fund will be vulnerable to
of strategic asset allocation. Absent questions about its currency exposure.
a tactical view on currency, the fund There may be pressure to reverse whatever
would hold the strategic hedge ratio. The FX positions the fund currently has (i.e.,
strategic hedge ratio is typically between switching from unhedged to fully hedged,
0% (unhedged) and 100% (fully hedged). or vice versa – and often just at the wrong
It may – and probably does – differ for time), incurring transaction costs and
each international asset class, a topic we sowing confusion. Second-guessing can be
address in the next chapter. The second avoided by having an established currency
part of the currency policy is whether to policy, which imposes discipline on both
manage currency exposure actively or the philosophy and process.
passively. This is analogous to the active/ The assumptions underlying the
passive discussion for stocks and bonds, currency policy should be re-examined
and should be addressed with the same periodically, at the same frequency and
degree of rigor. Another consideration, perhaps the same time as the fund revisits
if any sort of currency hedging, active or its strategic asset allocation.
passive, is contemplated, is whether the
hedging will be undertaken by internal or Why now?
external resources. This issue may also be Why should currency policy be a priority
included in a fund’s currency policy. now? Investors have generally been
11
FIRSTQUADRANT.COM
Currency Hedging Handbook CHAPTER 1
15.0
10.0
Average FX volatility
5.0
Dec-82 Aug-88 Apr-94 Dec-99 Aug-05 Apr-11 Dec-16
FX volatility is defined as the volatility of returns for global ex-US equities unhedged
minus fully hedged.
12
FIRSTQUADRANT.COM
CHAPTER 1 Currency Hedging Handbook
Figure 03: Global Economic Policy Uncertainty Index
300
200
100
0
Dec-96 May-00 Sep-03 Jan-07 May-10 Sep-13 Jan-17
Source: www.policyuncertainty.com
things, and these are directly connected to investors should be motivated to use their
the foreign exchange markets. risk budgets as efficiently as possible. If a
We would posit one more reason for fund has more currency risk than indicated
taking a close look at currency risk at this by its strategic hedge ratio, that risk can be
time. With the expected returns of major asset redeployed into other investments that are
classes quite low by historical standards, more likely to provide incremental returns.
13
FIRSTQUADRANT.COM
Currency Hedging Handbook CHAPTER 2
14
FIRSTQUADRANT.COM
CHAPTER 2 Currency Hedging Handbook
15
FIRSTQUADRANT.COM
Currency Hedging Handbook CHAPTER 2
movements would not be automatic; it What do the data reveal about long-
would require analysis and good judgment, term currency returns? Continuing with the
which are the ingredients of skilled active example of global ex-US equity, Figure 04
management. Passive currency exposure isolates the impact of currency hedging on
on its own cannot be expected to generate returns by comparing global ex-US equity
a positive (or negative) long-term return, unhedged returns to fully hedged returns, as
but it clearly generates risk. we did when comparing risks in Table 01 at
A simple thought experiment reveals why the beginning of this chapter.6 For the sake
this is so. Return to the example given above of brevity, we will refer to this difference as
of a US investor holding non-interest bearing the ‘currency effect’. All of the examples in
Japanese yen notes in a vault. Now consider the handbook that refer to currency hedging
the example of a Japanese yen investor use forward foreign exchange contracts with
holding non-interest bearing US dollar notes a 3-month tenor, unless otherwise stated
in a vault. At any given point in time, one (see text box on page 15 for a description
of these investors will have a gain and the of forward FX contracts).7 From January
other a loss from their respective currency 1980 to December 2016, the global ex-
positions. They both cannot rationally expect US equity unhedged return was 10% per
a positive return, unconditionally, from their annum compared to 9.5% fully hedged.
holdings, no matter how long they hold their The unhedged return exceeded the fully
foreign currencies.
Figure 03.5: FX Returns hedged return by 0.5% per annum, but the
$4,000
$3,500
Unhedged global Fully hedged
$3,000 global ex-US
ex-US equities
$2,500 equities
$2,000 Unhedged minus
fully hedged
$1,500 global ex-US
$1,000 equities
$500
$0
-$500
Jan-80 Feb-86 Apr-92 Jun-98 Aug-04 Oct-10 Dec-16
16
FIRSTQUADRANT.COM
CHAPTER 2 Currency Hedging Handbook
17
FIRSTQUADRANT.COM
Currency Hedging Handbook CHAPTER 2
18
FIRSTQUADRANT.COM
CHAPTER 2 Currency Hedging Handbook
dollars for two main reasons (assuming prospects affects the local equity market
all else equal). First, to the extent that negatively and at the same time reduces
foreign companies generate profits in US the attractiveness of the market for foreign
dollars, the local currency value of those investors. This could lead to a positive
dollars will be greater. Second, if foreign relationship between foreign currency and
companies price their products in their foreign equity. Also, as mentioned earlier,
home currency, the US dollar price of their multinational companies located outside
products will decline, potentially leading to the US conduct business in US dollars and
greater demand from US dollar customers in many other currencies as well. The US
and higher sales volumes. The same dollar may not be the dominant foreign
influences would work in reverse when currency relationship.
foreign currencies are strong and the dollar A handful of currencies have been called
weak. This relationship could give rise to ‘safe haven’ currencies because it is thought
a negative correlation between foreign that those currencies will perform better
currency and foreign equity, which would than others during bad times in markets
indeed be diversifying. We seldom observe and economies. This may be because the
it in the data, however, in part because all countries with the safe haven designation
else isn’t equal. are viewed as open to foreign capital flows,
For example, sometimes a macro having prudent macroeconomic policies and
event such as disappointing GDP growth being less vulnerable to a global economic
Figure 04: Correlation between FX returns and hedged equity
returns, rolling 36 months
FIGURE 05 - ROLLING FX CORRELATION WITH GLOBAL EX-US EQUITIES FULLY HEDGED
(DECEMBER 1982 - DECEMBER 2016)
0.8
0.6
-0.2
-0.4
-0.6
Dec-82 Aug-88 Apr-94 Dec-99 Aug-05 Apr-11 Dec-16
19
FIRSTQUADRANT.COM
Currency Hedging Handbook CHAPTER 2
downturn or global inflationary surge. The and this is borne out in the data, where there
currencies identified as safe havens shift is little evidence of consistent correlations.
over time, and safe haven status depends Figure 05 on the previous page shows the
on an investor’s base currency. The Swiss rolling 36-month correlation of FX with
franc and Japanese yen are sometimes global ex-US equity fully hedged. The
considered safe haven currencies. We correlation will be positive when foreign
mention safe haven currencies because, if currencies move in the same direction as
the US dollar acted like one, then foreign foreign equities (fully hedged), and negative
currencies would tend to depreciate against when they move in opposite directions. The
the US dollar at the same time as foreign average correlation over the full period is
equities were weak, resulting in more an insignificant 0.05, although there are
positive correlations. That was the case in periods when the correlation is materially
the recent financial crisis, but the pattern is positive (around 0.3 or more) and negative
hardly consistent over time. During the dot- (-0.3 or less). Thus we assume a correlation
com bubble of the late 1990s to early 2000s, of zero for US dollar investors.
for example, the correlation between foreign Figure 06 builds on Figure 05 by adding
currency and foreign equities was negative. a second data series that is the difference
In searching for a fundamental in risk (standard deviation) between global
explanation for correlation, the arguments ex-US equity unhedged and fully hedged.
go in both directions for US dollar investors, Table 01 showed that, on average, the risk of
20
FIRSTQUADRANT.COM
CHAPTER 2 Currency Hedging Handbook
Fig 06
FIGURE 07 - RETURN AND RISK AS A FUNCTION OF THE STRATEGIC HEDGE RATIO
(JANUARY 1980 - DECEMBER 2016)
20.0% 0.66
18.0% Risk (LHS)
16.0% 0.64
14.0% 0.62
12.0% Return (LHS)
10.0% 0.60
8.0%
6.0% Return/Risk (RHS) 0.58
4.0% 0.56
2.0%
0.0% 0.54
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
Strategic hedge ratio
Sources: First Quadrant, L.P., Datastream
unhedged equity exceeds the fully hedged TABLE 03 - INPUTS FOR STRATEGIC
risk by 2.6%. Yet there are a few periods HEDGE RATIO
when the unhedged risk of global ex-US
equity is slightly lower than fully hedged, Expected return of FX exposure 0.5%
and Figure 06 shows that these periods
Expected volatility of FX exposure 8.4%
are driven by sharply negative correlations
between currency and fully hedged equity. Expected volatility of global ex-US 14.6%
In fully 87% of 36-month periods, however, equity, fully hedged
the risk of the fully hedged position is lower Expected correlation between FX 0.0
than the unhedged position, which argues and equity
strongly for some degree of hedging. Sources: First Quadrant, L.P., Datastream
21
FIRSTQUADRANT.COM
Currency Hedging Handbook CHAPTER 2
100%. The peak improvement in return/risk depreciation. It’s a compromise that also
is indeed attained for hedge ratios between provides most of the return/risk benefits
70% and 80%. The 50% hedge ratio has 80% of the optimal strategic hedge ratio.
of the potential improvement in return/risk. One simplification we’ve employed in
A 100% hedge ratio is slightly less optimal this analysis is to assume investors use the
than a 75% hedge ratio.13 same hedge ratio for each of the currencies
Table 04 summarizes the pros and cons in global ex-US equity. An extension could
of four different hedge ratios commonly be to undertake this analysis separately for
used by investors. It’s worth mentioning each currency, so that instead of a single
the 50% hedge ratio. One of its properties strategic hedge ratio there would be as
is that, because it’s half way between the many different hedge ratios as there are
two endpoints of 0% and 100% hedged, it foreign currency exposures. In our view,
is said to minimize the ‘regret’ that some such an approach, while not necessarily
investors experience over the course of unreasonable, is closer in spirit to active
an up-and-down currency cycle. When management than the benchmarking
an investor is unhedged, and foreign exercise we are carrying out here.
currencies depreciate substantially, with The focus here is on the US dollar
the benefit of hindsight, they may wish they investor, but it is instructive to see how
had a 100% hedge. On the other hand, if a strategic hedge ratio of 75% compares
an investor is fully hedged, and foreign to the strategic hedge ratio for investors
currencies appreciate, they may wish they with different base currencies. Is a 75%
had instead been unhedged. A 50% hedge hedge ratio an outlier, or are US dollar
gives investors half the gain of any foreign base investors members of a broader
currency appreciation or half the loss of any community of global investors who would
22
FIRSTQUADRANT.COM
CHAPTER 2 Currency Hedging Handbook
TABLE 05 - STRATEGIC HEDGE RATIOS FOR base currency equities. The exhibit has
INVESTORS IN GLOBAL EQUITIES, BY BASE three columns. The first is the investor’s
CURRENCY base currency. The second column is the
Optimal Optimal optimal strategic hedge ratio using the
(Historical (Zero historical correlation between foreign
correlation) correlation) currency returns and fully hedged foreign
AUS 66% 111% equity, and the third column is the same
as the second except we used a correlation
CAN 37% 102%
of zero instead of the historical one.
DEN 114% 121% Whether to use the historical correlation
or a zero correlation depends on whether
EMU 64% 69%
you believe there is a fundamental basis
HKG 67% 51% for the historical correlation, or if it may
be distorted by special factors or is simply
JPN 100% 84%
noise and should be ignored. In the case
NWY 88% 121% of the US dollar base, we believe zero is
the right answer because we don’t find
NWZ 101% 152%
a sufficiently strong fundamental case
SNG -10% 67% to use the historical correlation, and the
correlation is quite close to zero in any case.
SWD 43% 75%
As shown in the table, the US dollar
SWI 84% 68% strategic hedge ratio is not an outlier. Most
other base currencies also have hedge
UK 79% 95%
ratios between 50% and 100%. When you
USA 82% 75% work through the math, given the inputs
of volatility, correlation and return, several
Sources: First Quadrant, L.P., Datastream base currencies have theoretically optimal
strategic hedge ratios greater than 100%,
be better off hedging a substantial but in practice few investors would be
portion of the currency exposure of their inclined to hold a strategic hedge ratio
foreign equities? greater than 100%.
Table 05 uses the same analysis as For the safe haven Swiss franc and
above, but from the perspective of a Japanese yen, the optimal hedge ratios
selection of different developed market using their historically positive correlations
base currencies. In each case, investors are 16% higher than if the correlation is
have the same objective of maximizing assumed to be zero, a modest difference.
expected risk-adjusted returns of the Several currencies exhibit the opposite
currency exposure of their global ex- behavior, where the correlation between
23
FIRSTQUADRANT.COM
Currency Hedging Handbook CHAPTER 2
foreign currency and foreign equities tends three commodity exporters, the optimal
to be negative, so the optimal hedge ratio hedge ratio ranges from 66% to 101%.
using the historical correlation is lower than Returning to the earlier discussion on
if the correlation were assumed to be zero. how time horizon affects risk, we repeated
For most of these base currencies, some the analysis using multi-year inputs. This
foreign currency exposure is diversifying. assumes that investors are indifferent to
For some of these countries – notably short- and medium-term currency risk and
Australia, Canada, New Zealand and Norway only care about returns at the end of the
- commodity exports are an important part holding period. When the results are assessed
of their economy. During global economic for rolling periods varying from 2 to 20 years,
downturns, commodity prices are often the average optimal hedge ratio was about 50%.
weak, putting downward pressure on
the currencies of commodity exporters. Drawdowns
Investors in commodity-exporting countries The optimal strategic hedge ratio of 75%
experience base currency weakness was derived using standard deviation as
when global equities are weak, reducing the measure of risk, but for some investors
– though not eliminating – the benefits drawdown is a more pertinent measure of
of currency hedging. For Canadian dollar risk. Would switching the risk measure to
base investors, the optimal hedge ratio is drawdown change the optimal strategic
a relatively low 37%, although for the other hedgeFigure
ratio? If07
the concern is mitigating the
: Optimal Hedge Ratios For Different Objectives
120%
100%
80%
60%
40%
20%
0%
Return/Risk One-month Max Drawdown Mean
Drawdown Drawdown
24
FIRSTQUADRANT.COM
CHAPTER 2 Currency Hedging Handbook
worst peak-to-trough drawdown, then the bonds. The conceptual framework and
most effective hedge ratio would have been assumptions we set out above also largely
100% (fully hedged), based on the historical apply to bonds. 14 The inputs, however,
data. If the concern is mitigating the mean are quite different. Table 01 showed that
peak-to-trough drawdown, then 70% would currency volatility dominates the risk
have been optimal – quite close to the 75% profile of global ex-US bonds. Currency
figure. A 70% hedge ratio would also be volatility, on its own, is almost double the
optimal if the objective is to minimize the volatility of global ex-US bonds fully hedged.
worst one-month drawdown. Thus, based on An investment in global ex-US bonds
the analysis, the 75% strategic hedge ratio is unhedged is effectively an investment in a
not only the best answer for maximizing the basket of foreign currencies, which means
Sharpe ratio, but also quite close to optimal a large increase in risk with no incremental
for mitigating mean or worst one-month expected return.
drawdown risk. Figure 08 on the previous The correlation of monthly returns
page summarizes the optimal hedge ratios between foreign currency and global ex-US
for these objectives. bonds fully hedged is 0.6. Foreign currency
exposure, far from diversifying global
Global ex-US bonds ex-US bonds, adds additional risk. The
We repeat the exercise for US dollar significant positive correlation may indicate
investors holding global ex-US sovereign that currency investors and global bond
Volatility difference:
10.0 global ex-US bonds, 0.6
9.0 unhedged minus
8.0 fully hedged (LHS) 0.4
7.0 0.2
6.0
5.0 0.0
4.0
3.0 -0.2
2.0 -0.4
FX correlation (RHS)
1.0
0.0 -0.6
Dec-82 Aug-88 Apr-94 Dec-99 Aug-05 Apr-11 Dec-16
25
FIRSTQUADRANT.COM
Currency Hedging Handbook CHAPTER 2
TABLE 06 - STRATEGIC HEDGE RATIOS FOR correlation between foreign currency and
INVESTORS IN GLOBAL BONDS, EXCLUDING fully hedged bonds is zero, or 110% if one
BASE CURRENCY believes the 0.6 correlation is meaningful.
Investor base Optimal Optimal Either way, a reasonable strategic hedge
currency (Historical (Zero ratio would be 100%.
correlation) correlation)
Table 06 compares the foreign bond
Australia 109% 104% strategic hedge ratio for investors with
Canada 113% 103% different base currencies. The hedge ratios
are all fairly close to 100%. The US dollar
Denmark 99% 109% investor is not a special case with respect
Euro area 89% 95% to hedging the currency exposure of bonds.
26
FIRSTQUADRANT.COM
CHAPTER 2 Currency Hedging Handbook
from emerging market investment. Our of expected return, risk and correlation
research has shown that hedging away to foreign currency. Because holdings in
EM currency risk from EM equity and bond alternative asset classes vary substantially
indices essentially provides (expensive) from one investor to another, it is more
developed equity and bond exposures, difficult to generalize about their expected
respectively, while retaining very little EM- return, risk and correlation than for the
specific exposure.15 traditional classes analyzed above, and
Table 07 summarizes the strategic therefore more difficult to supply an
hedge ratios derived in Chapter 03. optimal currency hedge ratio.
For example, within hedge fund
TABLE 07 - STRATEGIC HEDGE RATIO MATRIX allocations, the weights of specific
FOR US DOLLAR BASE CURRENCY INVESTORS types of hedge funds vary widely across
Equities Bonds investors. Some investors will have a
higher weighting of equity long/short
Developed markets 75% 100% funds while others may emphasize fixed-
income arbitrage. The optimal currency
Emerging markets 0% 0%
hedging treatment of these two types of
Source: First Quadrant, L.P. hedge funds is quite different.
Without knowing, at a minimum, the
Hedging other types of investments weights of the different types of funds
A currency policy should cover all foreign within an investor’s hedge fund allocation,
assets, including illiquid and alternative it would be pure speculation to supply an
assets, and the framework presented here optimal currency hedging ratio for a hedge
can be extended to these investments. The fund allocation.16 For alternative assets, one
required inputs are the same: estimates size rarely fits all.
27
FIRSTQUADRANT.COM
Currency Hedging Handbook CHAPTER 3
28
FIRSTQUADRANT.COM
CHAPTER 3 Currency Hedging Handbook
PORTFOLIO
M&A, FDI FLOWS
4%
8% For one thing, the critical importance of
TOURISM
3% CURRENCY currency markets to the real economy
CENTRAL MANAGERS
BANKS means that currency is among the most
5%
2% susceptible of the financial markets to
politically motivated intervention. This adds
Sources: BIS, IMF, US Federal Reserve,
World Tourism Organization, Datastream, to both the challenges and opportunities
First Quadrant, L.P. for active managers in this field.
29
FIRSTQUADRANT.COM
Currency Hedging Handbook CHAPTER 3
income, have for many years analyzed the as value, carry and trend. Commoditization
relationships between the fundamental occurs when a factor receives large inflows
characteristics, or factors, of a currency of investor money, driving down returns
and its subsequent performance. Rigorous and reducing diversification benefits. A
factor-based investors require that disadvantage of less well-known factors
each factor they use has a fundamental is that there is less academic research on
economic rationale for why it might work.17 them, so the underlying rationales may not
In currency markets, three factors have be as well understood.
achieved elevated status: they have been
known for many years and are supported Dynamic hedging
by a large body of academic work; there Dynamic hedging is an approach to
are plausible reasons for why they work; managing currency that has as its main
they have worked in many time periods objective mitigating downside risk, though
and in many currencies; and they are it may also seek to add returns. It is a
investible even for large investors. The kind of portfolio insurance for currency
three factors are value, carry and trend. exposure. Some versions use put options
Precise definitions for the factors differ while others use the forward FX market
among investors. The text box on the next dynamically to create option-like payoffs.
page summarizes these factors. There are A benefit of this approach is downside
also many other potential factors beyond protection, but this comes at the cost
these three (see discussion of systematic of an explicit or implicit premium. The
fundamental managers below). approach is implemented by reacting
Factor investors implement their to currency moves and does not require
chosen factors systematically. The factors currency forecasts. Whether the approach
are transparent and straightforward to can add value depends on the nuances of
implement and there are many variations the particular strategy and the behavior of
on how managers combine factors. They are currency markets. If the markets exhibit
sometimes called ‘smart beta’ strategies. significant trend-like behavior, dynamic
hedging may add value. A potential pitfall
Systematic fundamental with versions of dynamic hedging that seek
Systematic fundamental management is to replicate option-like behavior without
an extension of the factor-based approach. actually using options is gap risk – the
This approach goes beyond value, carry and risk that markets move too rapidly for the
trend, employing a range of less well-known strategy to respond effectively.
factors based on fundamental aspects of
currency market and investor behavior. An Performance of currency managers
advantage of using less well-known factors is Earlier in this chapter we showed that
that they are not as prone to commoditization a sizeable fraction of currency market
30
FIRSTQUADRANT.COM
CHAPTER 3 Currency Hedging Handbook
participants are not primarily in the to maximize the value of the currency
currency market in order to maximize the exposures they manage. If, as a group,
value of their currency exposure — other they are able to add value consistently
motives matter more. It’s conceivable that over time, it suggests that there are
this creates inefficiencies in the currency exploitable features of the currency market.
market, although it is difficult to know with If collectively they cannot add value, it could
certainty. One piece of evidence, albeit mean that the currency market is highly
indirect, is the track record of currency efficient, or that most managers are not
managers. This is one group of currency sufficiently skilled to capitalize on the
market participants who are motivated opportunities that are present.
31
FIRSTQUADRANT.COM
Currency Hedging Handbook CHAPTER 3
What do the data show? Using a 0.2. This suggests that if investors had only a
database of currency manager track random ability to select currency managers,
records, we plotted aggregate manager no better than throwing darts at a board, the
performance for rolling 3-year periods from average investor would outperform gross of
2002 to 2016.18 The plots show the gross fees in each 5-year period. Even if investors
performance of the median manager in selected the lower quartile manager, they
each 3-year period and the upper and lower would have outperformed (gross) more often
quartile managers. Figure 11 shows this than not in each five-year period.
data for excess return over each manager’s To put performance of the universe
respective benchmark, and Figure 12 (next of currency managers in context, we
page) for the information ratio. compared it to a universe of global
Figure 11 shows that the median developed equity managers in Table 09
manager in every 3-year period generated (page 34). The managers in the currency
a positive excess return gross of fees. Table universe performed, on average, better
08 (next page) divides the same underlying than the managers in the global equity
data into 3 non-overlapping 5-year periods universe. The information ratios of the
(2002-2006, 2007-2011 and 2012-2016). upper quartile, median and lower quartile
The median information ratio averaged currency managers were consistently
0.5 (gross) in those periods, and the lower higher than the information ratios for
quartile managers had an average IR of the global equity managers at the same
Upper
Ex Return (%pa)
Median
Lower
Median
Source: Mercer
32
FIRSTQUADRANT.COM
CHAPTER 3 Currency Hedging Handbook
Upper
Information Ratio
Median
Lower
Source: Mercer
performance quartiles in each of the Another way of interpreting the data is that
5-year periods. if an investor is more skillful in choosing
The track records suggest that if an global equity managers than currency
investor is able to select currency managers managers, the reward from active currency
as well as they can select global equity management may nevertheless be similar to
managers, the currency universe offers that achieved from global equity managers.
better opportunities for outperformance. Over the 15-year period, for example, the
# Managers 40 48 32 40
Source: Mercer
33
FIRSTQUADRANT.COM
Currency Hedging Handbook CHAPTER 3
information ratio of the upper quartile global equal-weighted combination of them. Table
equity manager was 0.6, which is slightly 10 (next page) is a numerical summary of
higher than the 0.5 information ratio of the the same data.
median currency manager. The three factors all generated positive
The number of currency managers returns with information ratios ranging
in this universe averaged about 40 over from 0.2 to 0.4 (excluding costs). Carry
the time period, and 32 in the recent was the riskiest of the factors, especially in
5-year period. While the number is small terms of downside risk. The 3-factor equally
compared to other types of investments weighted combination performed well, with
– the same database has over 160 global an information ratio of 0.6 and less downside
equity strategies – capacity should not risk than any individual factor. This study
be an issue.19 This is because developed shows that naïve strategies in the public
market currency is one of the most liquid domain added value, another indication that
financial markets in the world, with over US there is scope to add value in currencies.
$5 trillion in daily volume. The live track records of managers and
the performance of naïve strategies are
Performance of currency factors not entirely independent, however. Some
Another piece of evidence on opportunities managers use these factors as one of
to add value in currency is the performance several inputs to their strategies. These
of factor-based strategies. The results two analyses taken together provide some
can be found in Figure 13 (next page).20 In evidence of the opportunities available for
addition to the three factors, we included an active currency managers.
34
FIRSTQUADRANT.COM
- bonds
CHAPTER 3 Currency Hedging Handbook
Figure
FIGURE 13 09: Value,
- TIMESERIES carry,
CHART trendVALUE
OF CARRY, and combination,
AND TREND cumulative returns
(SEPTEMBER 2000 - DECEMBER 2016)
180 Carry
170 Value
160
150
140
Combination
130
Trend
120
110
100
90
Aug-00 May-03 Feb-06 Oct-08 Jul-11 Apr-14 Dec-16
35
FIRSTQUADRANT.COM
Currency Hedging Handbook CHAPTER 3
36
FIRSTQUADRANT.COM
CHAPTER 3 Currency Hedging Handbook
exposures. The first and third sample the size and liquidity of the market. The
constraints at the beginning of this section first kind of trade, which we’ll call ‘spot’
have this effect. Mandates for currency trades for convenience, can be more
hedging are often constrained in this way, expensive than the second because it
but they don’t have to be. Constraints can be entails a transfer of exchange rate risk, and
much looser and untethered to underlying exchange rates are not static. Roll trades
currency exposures. are less expensive because their risk is
For a skilled manager, loosening determined by interest rate differentials,
constraints tends to improve risk- which are usually fairly stable.
adjusted performance. With respect to other types of
instruments, FX futures contracts also
Costs incur generally low transactions costs,
Currency hedging, passive or active, incurs though without the benefit of custom
several kinds of costs: settlement dates. Options, on the other
• Transaction costs hand, can be expensive and, depending on
• Management fees (active) or the structure used, can require an explicit
implementation costs (passive) outlay of premium.
• Monitoring / oversight costs
• FX prime broker fees (optional) Transaction costs
Transaction costs are incurred when Passive hedging programs require trades
trading instruments used for currency when they are initiated, plus additional
hedging, and also if it’s necessary to trades when the amount of underlying
buy or sell portfolio assets to meet the assets changes due to movements in
liquidity demands of the hedging program. the underlying markets and inflows or
Transaction costs depend on market outflows. Passive programs also utilize
conditions, the amounts traded, the method lower-cost roll trades, typically once every
used for trading, as well as the types of three months. We estimate the transaction
instruments (discussed in Chapter 4) and costs of a passive 100% hedge may be as
which currencies are being traded. There low as 5 basis points per year.23
are no explicit per-trade costs or fees, Transactions costs can be higher
however, with the possible exception of for active management than for passive
prime broker fees in some cases. hedging because of the potential for a
Two kinds of trades are common with greater volume of transactions, although
forward FX contracts: 1) trades that change this varies according to the type of
the amounts being hedged, and 2) trades management. An additional potential
that roll existing hedges forward to a future modest cost is an FX prime broker (please
date. In developed market currencies, both see text box on prime brokers on page 40
kinds of trades are quite inexpensive given for additional details).
37
FIRSTQUADRANT.COM
Currency Hedging Handbook CHAPTER 3
38
FIRSTQUADRANT.COM
CHAPTER 4 Currency Hedging Handbook
39
FIRSTQUADRANT.COM
Currency Hedging Handbook CHAPTER 4
any amount of currency (see text box on currency exposure, but an expensive
forwards on page 15). one. See also the discussion in Chapter
3 on dynamic hedging, a technique using
Options forwards that seeks to replicate some
Someone who buys an option has purchased aspects of option behavior. Most currency
the right to make a currency exchange at options trading is OTC, though there are
the agreed strike price, but is under no a limited number of exchange-traded FX
obligation to do so. An option buyer will get options. OTC options are subject to similar
the better outcome, regardless of whether credit and counterparty considerations
the exchange rate moves up or down, but as forward FX contracts, and exchange-
at the cost of the option premium. Options traded options have similar credit and
are a powerful instrument for managing counterparty considerations as FX futures.
40
FIRSTQUADRANT.COM
CHAPTER 4 Currency Hedging Handbook
41
FIRSTQUADRANT.COM
Currency Hedging Handbook CHAPTER 4
42
FIRSTQUADRANT.COM
CHAPTER 4 Currency Hedging Handbook
43
FIRSTQUADRANT.COM
Currency Hedging Handbook CHAPTER 5
44
FIRSTQUADRANT.COM
CHAPTER 5 Currency Hedging Handbook
There are two benefits to having this CURRENCY RISK HAS BEEN
type of currency-only benchmark. First, it
GROWING IN PORTFOLIOS AS
is the most transparent and accurate way
ALLOCATIONS TO FOREIGN ASSETS
to benchmark a currency manager. Second,
HAVE BEEN INCREASING, AND
it is an effective way to monitor the impact
EXCHANGE RATE VOLATILITY HAS
of currency on the portfolio independently
from the underlying assets. BEEN RISING FROM RELATIVELY
The fund needs a suitable benchmark LOW LEVELS. FOR INVESTORS
for each mandate, and it also needs a way WHO HAVEN’T YET PUT IN PLACE
to aggregate the performance of individual A CURRENCY POLICY, THERE ARE
mandates in order to track performance for STRONG ARGUMENTS FOR DOING
each asset class and for the entire fund. In SO IN THE NEAR FUTURE.
this example, the fund’s benchmark is global
ex-US developed equities, 50% hedged.
The aggregate portfolio performance for limited to a single measure. Some investors
this segment is the performance of all monitor standard deviation of returns as
of the relevant equity managers plus the well as drawdowns.
performance of the currency manager’s Assessing the difference in performance
portfolio of forward FX contracts. between portfolio and benchmark is crucial
for evaluating how well the currency
Monitoring return and risk policy is being implemented. It is also
The return of the currency hedging portfolio useful to track the absolute performance
is calculated using an approach similar of the currency benchmark and portfolio
to the one described in the section on because these contribute to overall fund
benchmarking. The underlying foreign performance and investor objectives.
currency exposure is exactly the same as in
the benchmark. The return of the forward Communication and education
contracts (or other instruments, if applicable) Communication on currency policy is
is added to the return of the foreign currency especially important because opinions
exposure.29 The difference in returns between differ across the investment industry
the portfolio and benchmark therefore comes on how to deal with currency exposure.
down to the difference in returns between the Once they have implemented their
actual forward contracts and the hypothetical currency policy, investors should regularly
ones (if any) in the benchmark. communicate to their management
Portfolio and benchmark should be its objectives, the principles on which
compared on the basis of return and risk. it is based and performance. In such
Risk should be evaluated in the terms most communications, it will be useful to
relevant to the investor, and need not be make a clear distinction between: (1) the
45
FIRSTQUADRANT.COM
Currency Hedging Handbook CHAPTER 5
performance of the strategic hedge ratio; returns and/or reducing risk relative to
and (2) its implementation. The strategic the strategic hedge ratio.
hedge ratio is designed to maximize Communications regarding a currency
return/risk and facilitate efficient use of the hedging program should, ideally, be
risk budget. Regarding implementation, supplemented with a regular program of
if a passive approach is employed, the education regarding the currency market
objective is to execute the strategic hedge environment and how it relates to the
ratio at a reasonable cost. An active investor’s currency policy. Investment
implementation, on the other hand, will managers, consultants and counterparties
have its own objectives of augmenting can often assist with this.
46
FIRSTQUADRANT.COM
CONCLUSION Currency Hedging Handbook
47
FIRSTQUADRANT.COM
Currency Hedging Handbook CONCLUSION
48
FIRSTQUADRANT.COM
CONCLUSION Currency Hedging Handbook
20
Source: Deutsche Bank. The Carry Index (“Carry”) 23
Costs could be somewhat higher, depending upon
systematically buys the top three-yielding currencies the trading method employed and counterparties,
out of the G-10 and sells the bottom three-yielding among other factors.
ones based on three-month yields, with the position 24
Futures are transacted by broker-dealers.
re-appraised every three months. As with any carry 25
The credit risk of exchange-traded instruments is
trade the aim is to extract the risk premium for limited to a one-day price movement. One-day price
taking currency risk. movements, however, can be quite large.
The Momentum Index (“Trend”) systematically 26
Of course this also works in reverse: when the
buys the three currencies with the highest return counterparty bank has a large unrealized gain on a
over the previous year and sells the three currencies forward, the investor is a debtor.
with the lowest returns. The idea here is to profit 27
A counterparty with a higher credit rating might
from the fact that currencies trend over time as have a larger position limit than a counterparty with
prices adjust slowly to new information. a lower rating, for example.
The Valuation (PPP) Index (“Value”) buys the 28
Some index vendors publish returns for currency
three most undervalued currencies and sells those benchmarks where the currency weights match
three that are most overvalued based on the concept the weights of the foreign assets in that vendor’s
of purchasing power parity – on the basis that corresponding equity or fixed-income indices. Some
research suggests this concept holds up in the long vendors also publish returns of fully hedged currency
term, which means currencies tend to move towards exposures (again, excluding underlying asset
a measure of fair value. returns). So if the strategic hedge ratio is zero or
We show the factor returns starting with the 100%, there may be an off-the-shelf index available.
first month for which all three are available, which is For ratios other than zero or 100%, however, it may
September 2000. For additional information, please be necessary to create (or have an index vendor
see https://gm.db.com/about/disclaimer.html create) a customized currency benchmark, which
21
Net currency exposure is defined as the exposure is a weighted combination of unhedged and fully
from the underlying asset combined with the hedged currency indices.
manager’s hedging. 29
Ideally, the same spot exchange rates should be used
22
For example, a 100% limit would mean that to value both the live portfolio and the benchmark.
hedging activity is limited to the amount of
underlying foreign assets.
49
FIRSTQUADRANT.COM
Currency Hedging Handbook CONCLUSION
Model, backtested or simulated performance presented in this paper and further described below are for illustrative
purposes and are not a representation of strategies currently managed by First Quadrant, L.P. and are provided solely
for conceptual discussion only; therefore no fees are discussed or illustrated within this paper. Results from any live
implementation of strategies discussed would in actuality be reduced by fees charged against an account deploying
the strategy. Model, backtested, or simulated performance is no guarantee of the future results in a live portfolio
using the strategy. Potential for profit is accompanied by possibility of loss. Hypothetical or simulated performance
results have certain inherent limitations. Unlike an actual performance record, simulated results do not represent
actual trading. Simulated trading programs in general are also subject to the fact that they are designed with the
benefit of hindsight. Further, backtesting allows the security selection methodology to be adjusted until past returns
are maximized. No representation is being made that any account will or is likely to achieve profits or losses similar
to those shown. Unless otherwise noted, performance returns for one year or longer are annualized. Performance
returns for periods of less than one year are for the period reported. Reinvestment of earnings was assumed. The
simulated performance used in this presentation may differ from live performance experienced using the same
strategies for a variety of reasons such as: The simulations assume that the strategy guidelines are constant
through the life of the portfolio, whereas, the guidelines for live portfolios may change over the life of each portfolio;
the simulations assume zero transaction cost whereas live portfolio transaction costs will be variable; dependent
on market conditions such as liquidity and portfolio size, the types of instruments traded within the simulations
may not have been permissible in a live portfolio; etc. Details specific to the simulations: Global ex-US equities
includes simulated returns for the following development market countries: Australia, Austria, Belgium, Canada,
Denmark, Finland, France, Germany, Ireland, Italy, Japan, the Netherlands, New Zealand, Norway, Spain, Sweden,
Switzerland, and the United Kingdom. The unhedged returns for each country is a simulation of the MSCI World
ex-base country measured (not hedged) in the base country’s currency, where the country weights are rebalanced
on a monthly basis. The hedged returns for each country is a simulation of MSCI World ex-base country hedged
back to the base country’s currency, where the country weights are rebalanced on a monthly basis. The hedges are
established with currency forward contracts on each respective currency and adjusted monthly on a per currency
basis. Due to available index and hedging data, start dates of data for Finland and New Zealand is January 1988, for
Norway is March 1984 and for Switzerland is February 1980. Before 1999, we used the average of all the “legacy”
EMU countries as a synthetic EMU country in the dataset. Global ex-US bonds: includes simulated returns for the
following development market countries: Australia, Austria, Belgium, Canada, Denmark, Finland, France, Germany,
Ireland, Italy, Japan, the Netherlands, Norway, Spain, Sweden, Switzerland, and the United Kingdom. The unhedged
returns for each country is a simulation of the Citi World Government Bond Index ex-base country measured (not
hedged) in the base country’s currency, where the country weights are rebalanced on a monthly basis. The hedged
returns for each country is a simulation of Citi World Government Bond Index ex-base country hedged back to the
base country’s currency, where the country weights are rebalanced on a monthly basis. The hedges are established
with currency forward contracts on each respective currency and adjusted monthly on a per currency basis. Due to
available index and hedging data, start dates of data for Finland is May 1988, for Norway is February 1985 and for
Switzerland is February 1980. US Bonds: Includes simulated returns for the United States.
For index definitions and trademark language used in this handbook, please visit https://www.firstquadrant.com/
index-definitions for further information.
50
FIRSTQUADRANT.COM
This material is for your private information. The views expressed are the views of First Quadrant,
L.P. only are subject to change based on market and other conditions. All material has been
obtained from sources believed to be reliable, but its accuracy is not guaranteed.
FIRST QUADRANT, L.P. | 800 E. COLORADO BLVD. SUITE 900, PASADENA, CALIFORNIA 91101
MARKETING SERVICES INFO@FIRSTQUADRANT.COM | OFFICE 626 683 4223 | WEB FIRSTQUADRANT.COM
Copyright © by First Quadrant, LP, 2017, all rights reserved.