The Federal Reserve's renewed focus on price stability and a shift towards less frequent communication under its new leadership, combined with Wall Street's reassessment of the interest rate path, have driven the US dollar to its strongest monthly performance in nearly a year.

Consequently, as the dollar regains its strength, traders in emerging market foreign exchange, bond, and equity markets are increasingly turning to currencies from developed nations such as the euro and the Australian dollar to fund their "carry trade" bets in developing Asian economies.

From established asset managers like Invesco to global giants such as AllianceBernstein, top firms indicate they are reducing reliance on the world's primary currency, the US dollar, to finance positions in higher-yielding currencies.

Foreign exchange traders at Morgan Stanley are advising clients to express bullish exposure to developing economy currencies through a broader basket, including not just the dollar but also the euro, yen, and Australian dollar.

Analysts at Citigroup recently recommended betting on the Brazilian real strengthening against the euro and the Australian dollar.

Previously, when the dollar was weak, it was actively used as a funding currency; however, its current resurgence, fueled by the Fed's communication shift, rate hike expectations, US interest rate advantages, and large capital inflows into safe-haven and AI sectors, means continuing to short the dollar for funding would expose investors to exchange rate losses from dollar appreciation.

As a result, capital is shifting its funding leg from the dollar alone to other developed market currencies like the euro, Australian dollar, Canadian dollar, and yen.

The Bank for International Settlements defines a carry trade as a leveraged cross-currency position that exploits interest rate differentials and low volatility, typically involving buying a high-yielding "target currency" with a "funding currency."

Currently, low-cost funding currencies for carry trades are increasingly leaning towards developed market currencies other than the dollar.

Financial market traders are actively borrowing or shorting these low-cost funding currencies to buy emerging market currencies like the Brazilian real, Colombian peso, and Turkish lira, which offer high-yielding government bonds and strong short-term appreciation potential, aiming to capture significant interest rate differentials and potential currency gains.

This carry trade strategy essentially involves borrowing low-interest or weak funding currencies to purchase high-yield, high-growth, or high-beta target risk assets.

As the dollar strengthens and emerging market capital shifts its funding leg from the dollar to other developed market currencies, it reduces the risk of funding losses and forced liquidations due to dollar appreciation.

This approach remains highly conducive to directing global capital towards markets like South Korea's stock market, seen as a global barometer for AI computing investment, and other Asian equity markets with high exposure to the AI computing supply chain.

South Korea is not a typical "high-interest currency carry trade target" like Brazil, Colombia, or Turkey; what truly attracts capital is the stock market beta returns driven by the "memory super-cycle" fueled by explosive demand for AI training and inference.

An investment strategy team led by Goldman Sachs derivatives expert Brian Garrett noted in a recent report that investors are reducing their exposure to the "Magnificent Seven" tech giants, while the market continues to reward tech companies profiting handsomely from AI capital expenditure, such as semiconductor firms and other beneficiaries of the AI computing infrastructure chain.

Goldman Sachs' mid-year investment outlook predicts that "tech giants will continue to fall out of favor, while the semiconductor sector continues to reign supreme."

Dollar's Resurgence Alters Carry Trade Dynamics

"We are not actually funding high-yield or high-carry currencies through the dollar, but through currencies from the rest of the world," said Christian DiClementi, Head of Emerging Market Debt at AllianceBernstein.

"We don't currently have a high-conviction directional view on the dollar index."

Wim Vandenhoeck, Co-Head of Emerging Market Debt at Invesco, is adopting a similar carry trade model.

He stated that the firm has diversified its funding mix beyond the dollar in certain strategies to include the euro, Canadian dollar, Australian dollar, and to a lesser extent, the yen.

Following the Fed's first policy decision under its new leadership, major global currencies weakened significantly against the dollar since the central bank's overall hawkish pivot in June.

Last year, a weaker dollar injected new bullish momentum into emerging market currencies.

In recent months, the strong dollar storm has returned, disrupting currencies in developed markets like Europe and some of the most favored currency trades in the developing world, while also eroding bond market gains and briefly pushing an MSCI index tracking emerging market currency strength into negative territory for the year.

The renewed bullish sentiment for the dollar stems partly from the new leadership at the Federal Reserve.

Their strong hawkish commitment to restoring price stability and a perceived more hawkish communication approach have reinforced market expectations for dollar appreciation and a prolonged period of tighter benchmark rates, meaning higher US interest rates will support the dollar.

However, despite factors including ongoing AI-related investments that have tempered dollar weakness and propelled the dollar index higher in June, Invesco's Vandenhoeck stated that "the emerging market carry trade theme continues to offer attractive financial trade opportunities."

A carry trade involves investors borrowing in a low-interest/low-yield region and investing in a high-yield region; such trades have been a significant source of returns and yield metrics for emerging market investments.

Since former US President Donald Trump announced a series of global tariff policies in April 2025, leading to a period of weakness for the dollar and US assets, borrowing dollars while using a "carry trade strategy" to buy a basket of high-yielding currencies including the Brazilian real, Colombian peso, and Turkish lira has generated a robust 26% return to date.

Colombia and Brazil boast some of the highest real interest rates/yields in emerging markets, with their currencies each gaining at least 5% against the dollar this year.

If the Fed begins raising borrowing costs this year as markets expect, yield differentials would narrow, reducing the actual returns investors can achieve from such carry trades.

Undoubtedly, demand for emerging market carry trades persists, but investors are growing increasingly concerned about the risks associated with the "dollar funding" leg itself.

Previously, if the dollar was weak, shorting the dollar and buying high-yield emerging market assets could profit from both the interest rate differential and currency appreciation.

Now, a stronger dollar has become the biggest downside risk for emerging markets.

A recent HSBC survey on emerging market sentiment showed respondents ranked "dollar strength" as the top downside risk at 29%, slightly above geopolitical risks at 28%.

This indicates the market's primary concern is shifting from "war, conflict, political risk" to "dollar appreciation squeezing emerging market liquidity, hurting local currency assets, and raising funding costs."

Even as investment interest in this carry strategy remains strong, institutional investors and professional traders are becoming more cautious about the significant risks of using the dollar as the primary funding currency.

A recent survey of 101 institutions by HSBC strategists showed dollar strength has replaced geopolitical risk as the top risk factor and concern for emerging markets; these institutions collectively manage $432 billion in assets within the asset class.

The report also indicated a "clear shift" in fixed income preferences towards hard currency bonds.

Cathy Hepworth, Head of Emerging Market Debt at PGIM Fixed Income, has adopted a pro-dollar stance in her portfolio, though she still sees value in some high-yield currencies.

In some cases, she is using the yen as a core funding currency to exploit the still very high interest rate differential between the US and Japan.

For many investors, moving away from a single dollar funding currency or a single carry trade currency is not primarily about predicting sustained dollar strength, but about managing potential market volatility—a volatility that resurfaced on Thursday when a weaker-than-expected US jobs report caused the Bloomberg Dollar Spot Index to drop as much as 0.7%.

Thierry Larose, Senior Portfolio Manager at Vontobel, stated that while a disorderly dollar appreciation is a key risk, increasingly "adversarial" US trade and geopolitical policies should continue to constrain the dollar, reducing the probability of such an extreme outcome.

For Bill Campbell, who oversees the global sovereign and emerging markets strategy team at DoubleLine Capital, founded and managed by Jeffrey Gundlach, diversifying funding sources for emerging market trades away from the dollar during the summer makes logical sense.

However, he warned against immediately unwinding all dollar-funded positions in the short term, as the Fed's policy path remains uncertain.

"It's not a foregone conclusion that the Fed is about to hike and that the dollar will necessarily go significantly higher from here," he emphasized. "What happens if they decide it's better to wait at least until December to hike?"

From Dollar to Multi-Currency Funding: Global Capital Continues Embracing the AI Theme

The dollar index posted its strongest monthly gain in nearly a year in June, making "shorting the dollar for funding" a high-risk exposure.

Therefore, emerging market capital shifting its funding leg to other developed market currencies is not about abandoning risky assets, but about reducing the risk of funding losses and forced liquidations due to dollar appreciation.

For the unprecedented global investment boom in AI computing infrastructure, this means capital is still seeking assets with "high certainty of orders, high industry bottlenecks, and high profit elasticity," but with a more diversified funding structure and greater emphasis on currency hedging.

Morgan Stanley forecasts that global AI-related debt issuance will exceed $500 billion, nearing $570 billion by 2026, and states that hyperscale cloud provider capital expenditure will surpass $1 trillion by 2027.

While US markets were closed on Friday, global stock markets outside the US rallied strongly, with European benchmark indices rising for a fourth consecutive week to new record highs.

The South Korean won and stock market surged on Friday—the benchmark KOSPI index closed up nearly 6%, driven by strong rebounds in shares of major memory chipmakers SK Hynix and Samsung Electronics, highlighting the return of global capital to the AI computing investment theme and a rapid correction of overly pessimistic pricing based on fears of an "AI computing glut" following Meta's plans to sell excess capacity.

The KOSPI index has been one of the world's strongest benchmark indices this year amid the global AI frenzy.

The multi-currency carry trade theme significantly reduces the friction of "a strong dollar crushing dollar shorts" in funding trades, making it easier for global capital to continue flowing into South Korea's two major AI memory heavyweight stocks.

Unlike typical "high-interest currency carry trade targets" like Brazil, Colombia, or Turkey, what truly attracts capital to the South Korean market is the beta of AI tech stocks during the AI memory super-cycle.

Just as news that Meta planned to sell idle AI computing capacity sparked fears of an AI computing supply glut, hammering the AI investment theme and causing a global tech stock sell-off, Nvidia (NVDA.US), the world's most valuable company and "AI chip superpower," stated on Thursday it is entering revenue-sharing agreements with rapidly scaling AI startups.

This move allows these AI startup clients to exchange a portion of future profits for access to Nvidia's vast AI GPU computing clusters.

In the view of Wall Street analysts, Nvidia's move is the industry's strongest rebuttal from a leading company against the pessimistic narrative of "excess AI computing infrastructure demand," highlighting that as AI applications and agents proliferate globally, demand for AI computing power remains nearly insatiable.

Wall Street analysts unanimously argue that Meta selling computing capacity, coupled with SoftBank establishing SB Neo to enter the US AI cloud market and Nvidia's latest offer to AI startups, is not a bet on "computing surplus" but a bet on the long-term expansion of AI training and inference demand into the gigawatt-scale infrastructure era.

The increasing leverage and crowded positioning in the AI semiconductor trade theme, alongside rising pricing pressures for consumer electronics leaders like Apple, were accompanied by a single-day drop of 7.9% in the Philadelphia Semiconductor Index and multiple intra-month swings exceeding 5%.

This highlights that the AI computing supply chain linked to semiconductors has entered a phase of high volatility, extreme leverage and crowded bullish positions, and high pressure to meet expectations.

Coupled with Meta's pivot to selling computing resources, this is why some institutional investors have recently begun emphasizing overly pessimistic bearish narratives like "the AI semiconductor trade frenzy has peaked" and "the AI bubble is gradually bursting."

However, Nomura, a prominent Wall Street investment firm, released a report on Wednesday refuting the "semiconductor peak theory."

The key to Nomura's rebuttal is not simply stating that AI chips will keep rising, but pointing out that AI cloud infrastructure demand is spreading from isolated GPU shortages to systemic component mismatches.

According to Nomura's research framework, AI server revenue is forecast to grow 78% and 76% in 2026 and 2027, respectively.

The number of global data center projects is expected to increase from 240 to 280, with about 50 being gigawatt-scale projects.

New computing capacity deployment in 2027 is projected to reach 32GW, with visibility for 23GW already in 2028.

The real bottlenecks are now spilling over from GPU capacity and TSMC's CoWoS advanced packaging to wafer-level substrates, AI PCBs, copper-clad laminate (CCL), electronic fabric, MLCCs, glass substrates/ABF substrates, IC substrates, high-end capacitors, power management chips, and high-speed optical interconnect components for data centers.

In the view of SemiAnalysis, Meta's move to sell AI computing resources and Nvidia's new "AI computing for revenue share" model form a highly consistent positive signal for the AI computing supply chain: demand for AI computing power remains strong.

What is truly scarce are GPUs/ASICs/TPUs, DRAM/NAND/HBM memory components, power resources, liquid cooling equipment, network infrastructure, and even the complete package of data center campus delivery capabilities and low-cost financing channels, including data center CPUs, high-speed optical interconnect systems, and data center transformer systems.