Finance

European funds often fall short on currency hedging

The theory of optimal currency hedging has been well established for decades. This column uses new regulatory data to analyse how 2,806 European investment funds use currency forwards and swaps and how the use compares to theoretical benchmarks. It finds that only a minority follow hedging policies close to the portfolio-theory benchmark, and observed derivative trading does not significantly contribute to lower fund risk. Yet a feasible optimal strategy could lower volatility by about 10%, at very modest additional transaction cost.

International investment is usually about two consecutive decisions. A fund first chooses which foreign assets to own and then decides how much of the resulting currency exposure to retain. For example, a euro area fund that owns a dollar bond earns the bond return in US currency but is also exposed to movements in the dollar. Selling dollars forward can remove some, or all, of that exchange rate risk.

The theory of optimal currency hedging (conditional on previous asset choices) has been established for more than three decades (Anderson and Danthine 1981, Glen and Jorion 1993, Jorion 1994). We use European regulatory data to analyse how European investment funds use currency forwards and swaps and compare the results to the theoretical benchmarks (Bräuer and Hau 2026).

Currency risk is a portfolio problem

Hedging every foreign-currency position in full does not necessarily minimise a fund’s overall risk. A full one-for-one hedge is optimal only in a special case. In general, the best forward position depends on the entire asset portfolio: how the currencies move with one another, how they covary with bond and equity returns, expected currency returns, and trading costs. A dollar exposure can therefore be hedged partly with a dollar forward and partly through positions in other currencies if those contracts offset portfolio risk more effectively.

This distinction gives us two benchmarks. The first is a mean-variance optimal benchmark. Its risk-minimising component uses all available forward contracts and accounts for the full covariance structure across currencies and assets; it may additionally include a return-seeking component based on forward premia, exchange-rate expectations, and transaction costs. The second is a unitary hedge, which matches each foreign-currency asset exposure with a corresponding short position in the same currency of similar size. This second benchmark is intuitive and easy to implement, but not optimal from a portfolio perspective.

A new regulatory window on fund hedging

Our analysis links derivatives reported under the European Market Infrastructure Regulation (EMIR) to monthly fund holdings from Refinitiv Lipper. The sample covers 2019-23 and starts from 7,029 euro-based funds with foreign investments and no strict currency-hedging mandate. Of these, 40% use foreign exchange forwards or swaps at least once. We focus on these 2,806 funds, for which derivative use is discretionary: 806 bond funds, 1,109 equity funds, and 891 mixed funds.

For each fund and month, we reconstruct net positions against the euro in six major currencies – the US dollar, pound sterling, yen, Swiss franc, Swedish krona, and Australian dollar. Forwards and swaps account for 94% of the sample’s foreign exchange derivative holdings. We combine these positions with the currency composition of each fund’s assets and estimate time-varying covariance matrices for representative bond and equity returns. This produces an optimal benchmark tailored to each fund’s actual asset portfolio and allows us to compare observed with theory-implied optimal hedging. 

Bond funds come closest to the optimal benchmark

The strongest alignment with theory appears among 728 bond funds that use derivatives only to sell foreign currencies and buy euros. The model calculates how much each fund should hedge accounting for covariance between assets and exchange rates. If actual positions matched the model exactly, the estimated regression coefficient would be −1. For these bond funds, the coefficient is −0.94, suggesting close alignment on average. In contrast, equity and mixed funds’ derivative positions are only loosely connected to the position implied by their assets and covariances, with a coefficient of only −0.17 and −0.27, respectively.

Figure 1 illustrates this comparison for US dollar derivative and asset positions of European funds. It compares each fund’s observed US dollar hedge with the hedge implied by the risk-minimising model. Both positions are measured as shares of assets under management and US dollar hedge positions indicate euro long and US dollar short derivative positions. Each marker represents a group of funds, with individual observations aggregated for confidentiality reasons. The dashed 45-degree line indicates exact alignment: funds below the line hedge less than the model suggests, while funds above it hedge more. Bond funds show the clearest positive relationship and tend to lie closest to the benchmark. Equity funds’ derivative positions frequently remain close to zero even when the model implies a substantial hedge, while mixed funds generally lie between the bond- and equity-fund patterns. The figure therefore shows that bond funds come on average closer to the mean-variance optimal benchmark, but also that individual hedging practices vary considerably within every fund type.

Figure 1 Observed versus model-implied USD/EUR hedge positions

Notes: The horizontal axis shows the risk-minimising EUR long positions implied by each fund’s asset holdings and the covariance structure of returns. The vertical axis shows the fund’s observed (time-averaged) EUR forward position. Markers represent fund-level positions, and the dashed 45-degree line denotes exact alignment with the model. 
Source: Bräuer and Hau (2026), Figure 5.

Most funds hedge one currency at a time

The apparent proximity of bond funds to the mean-variance optimal benchmark in Figure 1 is somewhat misleading. Because bond and currency returns are only weakly correlated, the optimal hedge for bond funds is relatively close to a simple unitary hedge that treats each currency exposure separately. Thus, bond funds can appear close to the optimal benchmark without adopting the comprehensive portfolio approach implied by mean-variance optimisation. Their positions instead are dominated by the unitary component: the fund sees a dollar asset and sells dollars forward, then treats another currency separately. We find that 44% of bond funds hedge in line with the unitary benchmark. The corresponding shares are 28% for mixed funds and only 7% for equity funds.  

Observed derivative trading does not lower average risk

We next compare observed fund returns with four counterfactual strategies. Removing derivatives altogether would slightly reduce annualised volatility for bond, equity, and mixed funds by 0.07, 0.10, and 0.26 percentage points, respectively. None of these differences is statistically significant. The striking conclusion is therefore that the observed currency derivative positions of European funds do not significantly reduce the average fund’s return risk.

A full unitary hedge works better for bond funds, lowering volatility by 0.28 percentage points, or about 4.6%. But it raises volatility for equity funds by 0.39 percentage points and for mixed funds by 0.24 percentage points. Thus, simply matching each currency exposure one-for-one can be counterproductive when asset returns and exchange rates provide natural diversification.

Figure 2 Optimal portfolio hedging would reduce volatility for every fund type

Notes: Change in annualised return volatility relative to observed derivative trading. Positive values denote lower risk; negative values denote higher risk. The no-derivative changes are not statistically significant. The optimal-hedge reductions are significant for all three fund types.
Source: Authors’ calculations based on Bräuer and Hau (2026), Table 6.

What optimal hedging could achieve

The optimal mean-variance portfolio benchmark delivers a consistent result across fund types. It lowers annualised volatility by 0.61 percentage points for bond funds, 1.47 points for equity funds, and 1.03 points for mixed funds. Relative to observed volatility, each reduction is close to 10%. Adding the estimated return-seeking component barely changes these risk outcomes.

The extra trading cost is modest. Relative to observed trading, the optimal strategy adds about 0.4 basis points a year for bond funds, two basis points for equity funds, and 0.5 basis points for mixed funds. Costs therefore cannot account for the large gap between actual and feasible risk reduction. The more plausible constraints lie inside the investment process: simple currency-by-currency mandates, incomplete information about portfolio covariances, benchmarking concerns, or weak incentives to optimise a risk that may be less visible than asset selection.

For our sample period 2019-2023, however, the volatility gains would have come with lower realised returns. The euro depreciated against the US dollar over much of the sample, allowing unhedged dollar positions to benefit from the dollar’s appreciation. The fully optimised strategy would have reduced this exposure and therefore also these gains. A manager assessed against an unhedged benchmark may consequently avoid a hedge that reduces risk but creates visible relative underperformance. This helps explain why managers’ incentives may not align with the risk-minimising choice for investors.

Implications

The results suggest that international risk sharing remains incomplete even among sophisticated institutions with access to liquid derivative markets. The problem is not simply that many funds choose not to hedge. It is that funds with discretionary mandates often use derivatives in ways that are partial, focused on one currency at a time, or even risk-augmenting. Better integration of currency decisions with the full asset portfolio could deliver economically meaningful risk reduction at low cost.

For investors and regulators, the policy message is straightforward. Access to derivatives is not enough to ensure that currency risk is well managed. Reporting systems have made institutional hedging observable; the next challenge is for funds to bring derivative use in line with genuine portfolio risk reduction.  

Source : VOXeu

GLOBAL BUSINESS AND FINANCE MAGAZINE

Recent Posts

Gradually, then suddenly – the fiscal value of climate resilience

The human and economic costs of extreme weather events are already large and will grow…

18 minutes ago

Resilience without retreat: How EU firms are redesigning their supply chains

Geopolitical tensions are changing the nature of supply-chain risk. This column presents new survey evidence…

19 minutes ago

Updating what workers learn: Technology, training content, and the supply of expertise

Raising educational attainment is the canonical way skills keep pace with technology. This column argues…

21 minutes ago

The geography of financial frictions

Firms’ access to external finance differs widely across Europe. This column uses data from the…

23 minutes ago

The effects of inflation on euro area public debt

Whether inflation can ease public debt burdens is an increasingly important question as debt sustainability…

25 minutes ago

Currency risk drives remittance fees as much as markups do

Senders now have more ways than ever to move money across borders, yet costs remain…

27 minutes ago