Global foreign exchange markets Report Interpretation

Overview

This global FX handbook explains the structure of currency markets, the forces that drive exchange rates over different horizons, evidence on FX investing, and risk-management approaches for investors and corporates. Its central conclusion is that effective FX analysis requires matching valuation, macroeconomic, flow and regime-change tools to the relevant horizon.

Core views

FX is the world’s largest and most liquid financial market, but it is fragmented and largely over the counter rather than centrally traded. The report cites BIS estimates of $9.5 trillion in daily FX turnover in 2025. FX swaps were the largest instrument at $4 trillion per day, or 42% of turnover; spot accounted for $3 trillion, or 31%; forwards accounted for $1.8 trillion, or 19%; and options represented around 7%. The book distinguishes market-making and transactional activity from return-seeking activity: under Deutsche Bank’s realistic scenario, approximately 40% of adjusted turnover is profit-seeking and 60% is liquidity-seeking or transactional. Its upper-bound scenario produces a 60% profit-seeking share. This composition matters because corporates, institutional investors, reserve managers and other non-speculative users may trade for operational, hedging or policy reasons, leaving prices influenced by flows and intermediary balance-sheet capacity as well as fundamentals. The report traces FX market structure from voice dealing and bank-centred inter-dealer trading to electronic execution, internalisation, prime brokerage and a broader set of non-bank liquidity providers and venues. Dealer-reported spot trading has declined from roughly 70% in the 1990s to around 40% recently as internalisation expanded; for major pairs, the largest dealers are estimated to internalise 80% or more of trades. Prime brokerage now accounts for around 40% of spot trading, while principal trading firms have become important electronic market makers. Electronification has generally improved transparency, operational efficiency, spreads and intraday volatility, but fragmentation means displayed depth across venues can overstate executable liquidity. The report calls this a liquidity mirage and argues that effective spreads, slippage, fees and price impact provide a more realistic assessment. It also highlights settlement risk, noting that CLS payment-versus-payment settlement netted transferred funds to 4% of gross traded value; more than $14 trillion in gross obligations were settled on an average day in April 2025, with 36% settled through PvP systems, predominantly CLS. For exchange-rate determination, the report rejects a universal model and organizes analysis by horizon. In the short run, from intraday to several weeks, order flow, positioning, options flow, sentiment, risk appetite, macro surprises and technical behaviour dominate. In the medium run, from months to a couple of years, relative interest rates, monetary-policy cycles, growth, current accounts, capital flows, hedge ratios and yield curves move currencies around longer-run anchors. Over several years, PPP, productivity, terms of trade and net foreign asset positions shape equilibrium real exchange rates. The report’s practical sequence is to establish a strategic valuation anchor over three years or more using PPP, BEER and FEER; align the cyclical view over one to two years using interest-rate differentials, policy and external balances; use flows and positioning as a tactical overlay; and account for regime changes such as pegs becoming floats, major trade-policy shifts or financial crises. PPP is described as a powerful long-run valuation anchor but a poor timing tool. The report notes the PPP puzzle: deviations can have a three-to-five-year half-life. Deutsche Bank’s cited back-testing found that a cross-sectional strategy going long the most undervalued and short the most overvalued currencies on a 12-month holding period generated a 0.60 Sharpe ratio after costs, with stronger results in G10 currencies. Performance improved as holding periods lengthened and peaked at around 24 months in the cited study, but valuation models underperformed in the two years through early 2025 as cyclical forces and unusually divergent monetary tightening dominated. The report therefore favors PPP for G10 valuation and Deutsche Bank’s DBeer behavioural-equilibrium model for emerging markets, where terms of trade and country-specific dynamics are more important. The medium-run framework combines monetary, external-balance and financial-market mechanisms. Nominal rate differentials affect carry and near-term policy pricing, while expected real rate differentials are more relevant, though imperfect, for medium-run valuation. UIP has been persistently rejected: high-yield currencies have historically tended to appreciate rather than depreciate enough to offset their yield advantage. The report emphasizes that carry returns may compensate for negative-skew and crash risk, since high-yield currencies can fall sharply when volatility rises, liquidity deteriorates and leveraged positions unwind. Current accounts, the composition of capital inflows and net foreign assets also matter: deficits financed by long-horizon FDI differ from deficits financed by reversible portfolio flows. Changes in foreign investors’ hedge ratios can be a first-order FX driver because the stock of cross-border holdings is large; hedge costs are approximated by the front-end interest-rate differential plus the cross-currency basis. Financial-intermediation and dollar-centric theories supplement conventional trade-weighted models. The report argues that constrained intermediaries must absorb currency imbalances created by cross-border portfolio flows, so exchange rates can move materially even when goods-trade fundamentals change little. It also describes persistent post-2008 deviations from covered interest parity as a cross-currency-basis problem linked to dealer balance-sheet costs and structural demand for dollar hedges. The dollar’s safe-asset convenience yield, dollar funding role and dominant invoicing position reinforce its systemic importance. The report cites BIS data that 88% of FX transactions have a USD leg and states that the USD accounts for 50% of SWIFT cross-border-payment messages. At shorter horizons, the report views order flow as the immediate mechanism through which information, risk appetite and hedging demand enter prices. Deutsche Bank’s CORAX indicators are used to distinguish client flows; cited findings show discretionary hedge-fund flows in G10 can predict spot for about ten days before reversing at longer horizons, while systematic flows provide more mean-reversion information. Positioning is treated as a measure of vulnerability rather than a mechanical timing signal: crowded positions can amplify reversals. Deutsche Bank’s EPIC+ positioning framework combines CFTC/IMM, CORAX, order-book, risk-reversal and DTCC options information; cited live results since January 2021 show a position-fading portfolio Sharpe ratio of 0.54 after costs, or 0.75 excluding EUR, CNH and THB, while a cross-sectional fading portfolio produced 1.2 after costs. The report cautions that contrarian performance weakens in highly volatile, strongly trending dollar markets. Options convey both expectations and market mechanics. Implied volatility prices protection, and risk reversals indicate skewed demand rather than pure directional forecasts. Dealer gamma hedging can dampen moves when dealers are long gamma or amplify them when they are short gamma. Deutsche Bank’s cited DTCC-based indicators distinguish investor transactions nearer at-the-money from hedger transactions in out-of-the-money options; the reported sample shows a momentum strategy using investor flow with a 0.83 Sharpe ratio, a contrarian hedger-flow strategy with a 0.67 ratio, and a blended strategy with a 0.96 ratio. Technical analysis is also presented as relevant where trends, positioning, herding and stop-loss activity drive prices beyond what standard macro models capture, although the report reiterates that exchange-rate forecasting remains difficult and random walks are hard to beat at horizons below one year. FX can be treated as an asset class through diversified, systematic factor exposure rather than forecasts of individual exchange-rate levels. The report identifies carry, momentum and value as the three canonical factors. Deutsche Bank’s DBCR index equally weighted transparent carry, momentum and valuation strategies across liquid developed-market currencies; its 1980–2006 back-test produced approximately 4% annualised excess return, 5.2% volatility, a 0.77 Sharpe ratio and an 11% maximum drawdown. The report says the combined factor approach remained resilient after the global financial crisis and may diversify conventional equity and bond portfolios, but stresses that returns are cyclical and exposed to crash risk, crowding, monetary-regime shifts and implementation costs. Cited CFTC evidence finds profit seekers made billions trading FX forwards at the expense of hedgers in most years since 1993, while the report estimates average discretionary macro-manager returns of 2.7% annually since 1995. For hedging, the report argues there is no universally optimal hedge ratio. Investors should assess FX as part of total portfolio risk, alongside hedging cost and expected spot direction, rather than treat it as an isolated transaction. Cited Deutsche Bank research finds dynamic hedging can deliver stronger risk-adjusted outcomes than static ratios, with volatility reduction generally more valuable than maximizing FX carry; tactical carry, momentum and valuation signals can reduce opportunity cost and preserve participation in favorable moves. Corporates face a different problem because they generally hedge known or forecast foreign-currency cash flows, revenues, costs, assets or liabilities. The report distinguishes cash-flow, balance-sheet and deal-contingent hedges, and recommends a two-step logic for multi-currency exposure: manage risk across the full currency basket to retain diversification benefits, then use currency signals to reduce hedge opportunity cost. For restricted emerging-market currencies, NDFs, onshore-versus-offshore forward costs and cross-currency basis must be considered; the report states that onshore forward curves have averaged around 40 basis points below offshore curves over the past decade. Finally, the report sees technology, regulation and geopolitics as forces reshaping FX infrastructure and currency demand. Distributed ledgers, automation, AI and machine learning may improve efficiency, fraud detection and settlement, but fragmentation also raises complexity and cyber risk. Machine learning has shown some promise for directional and volatility forecasts and high-frequency alternative data, but the report says improvements are typically small, regime-specific and often fail to beat a random walk on strict out-of-sample RMSE tests. Dollar stablecoins may extend dollar demand, particularly in markets with weak currencies or capital controls; the report states that 99% of outstanding stablecoins are USD-linked. Tokenisation and programmable settlement could make payment-versus-payment processes more synchronised, but the macro effect of CBDCs on exchange-rate determination remains speculative as of 2026.

Analysis framework

The report moves from FX instruments and market participants to market structure, then applies a horizon-based framework for exchange-rate determination. It combines BIS turnover data, CFTC positioning data, DTCC options data, Deutsche Bank proprietary flow and positioning indicators, academic evidence, historical back-tests and practical hedging examples. It then applies the same valuation, cyclical and tactical concepts to systematic FX investing, institutional hedging, corporate hedging and future payments infrastructure.

Methodology notes

• MacroeconomicsPurchasing power parity

  Purchasing power parity valuation

  PPP compares relative price levels to identify long-run currency misalignment. The report uses it as a strategic valuation anchor, especially for G10 currencies, while cautioning that deviations can persist for years.

• MacroeconomicsInterest rate parity

  Covered and uncovered interest parity

  The report uses covered parity to explain forward pricing and cross-currency basis, and uncovered parity to frame the carry-trade anomaly in which high-yield currencies have historically not depreciated as theory predicts.

• Industry AnalysisSupply-demand framework

  Flow, external-balance and intermediary-capacity analysis

  Currency demand and supply are assessed through current accounts, capital flows, hedge ratios, order flow and the ability of intermediaries to absorb imbalances.

• Quantitative, Factor, and Portfolio TheoryMulti-factor model

  Systematic currency carry, momentum and value factors

  The report describes diversified, rules-based portfolios that combine carry, momentum and valuation signals rather than relying on a forecast for one currency pair.

• Quantitative, Factor, and Portfolio TheorySharpe and Information Ratios

  Risk-adjusted performance evaluation

  The report uses Sharpe ratios after costs to compare PPP value, positioning, options-flow and systematic-factor strategies on a return-per-unit-of-volatility basis.

• Other

  Total Portfolio Approach to FX hedging

  Currency risk is evaluated alongside asset exposures, correlations, hedging carry and portfolio objectives, rather than through a fixed standalone hedge ratio.

Key data

• Global daily FX turnover$9.5 trillionBIS estimate for 2025.
• FX swap turnover$4 trillion per day; 42%Largest FX instrument by global turnover in April 2025.
• Spot FX turnover$3 trillion per day; 31%BIS estimate for 2025.
• Profit-seeking share of adjusted turnover40%Deutsche Bank realistic scenario; upper-bound scenario is 60%.
• PPP value portfolio Sharpe ratio0.60 after costs12-month long-undervalued/short-overvalued cross-sectional portfolio; reported to work best in G10.
• DBCR back-testApproximately 4% annualised excess return; 5.2% volatility; 0.77 Sharpe ratio; 11% maximum drawdownOriginal 1980–2006 back-test of equal-weighted carry, momentum and valuation strategies.
• USD role in FX turnover88%Share of FX transactions with a USD leg according to the BIS.
• USD-linked stablecoins99%Share of outstanding stablecoins stated by the report.

Impact & implications

The report’s framework implies that investors, corporates and policymakers should not rely on a single FX signal. Long-horizon valuation can identify misalignment, cyclical macro and external-balance variables can guide medium-run direction, and positioning, options and flows can identify shorter-run vulnerabilities. Market participants must also account for fragmented liquidity, settlement infrastructure, hedging costs and structural changes in dollar funding and payments systems.

Risks

• FX fragmentation can create a liquidity mirage, where aggregated displayed depth is not genuinely executable and large trades generate greater price impact than expected.
• Carry strategies are exposed to negative skew and crash risk when volatility rises, liquidity deteriorates and leveraged positions unwind.
• Crowded positioning can turn modest negative surprises into sharp reversals through stop-losses, margin constraints and systematic deleveraging.
• Settlement, counterparty, operational, market, liquidity and regulatory risks remain material in the largely OTC FX market.
• Factor and back-tested strategy returns can be affected by crowding, regime shifts, transaction costs and differences between hypothetical and live implementation.
• Electronification and decentralised infrastructure can increase operational complexity and cyberattack vulnerability.

What to watch

• Whether fragmented venue liquidity remains executable under stress, including effective spreads, slippage, fees and price impact rather than displayed depth alone.
• Changes in intermediary risk-bearing capacity, bank dealer margins and the market-making role of principal trading firms.
• The evolving location of FX price discovery across primary venues, futures markets and alternative platforms.
• Cross-currency basis, front-end rate differentials and changes in foreign-investor hedge ratios as drivers of dollar flows.
• Regime changes in monetary policy, exchange-rate arrangements, trade policy and global financial conditions.
• The development of stablecoins, CBDCs, tokenisation and programmable settlement in cross-border payments.