Natural Gas and Oil Forecast: Middle East Risks Elevate Bullish Oil Trends

A resurfaced speech from Chinese President Xi Jinping suggests policymakers may start trading government bonds to regulate liquidity in the market, pushing the nation toward strategies used by the Federal Reserve and other major central banks around the world.

Xi’s call for the People’s Bank of China to “gradually increase the buying and selling of government bonds” in its open market operations sparked a frenzy of speculation among traders last week. The remarks — made in October but publicized recently in a new book and newspaper article — may hint at a policy pivot for a central bank that hasn’t made a significant bond purchase since 2007.

“Central banks in other countries generally use government bonds, or sovereign credit, as a basis to issue money,” said Liu Lei, a researcher at the National Institution for Finance and Development, a state think tank advising government agencies in China. “This is a necessary path for China’s central bank and monetary system to move into modern times.”

The vague comments from the Chinese leader led some traders to initially argue that Beijing may be considering quantitative easing — an unconventional form of stimulus that involves purchasing sovereign bonds and other assets to push down yields and boost economic activity. First adopted by the Bank of Japan more than two decades ago, the tactic was later used by the Fed and other policymakers after the global financial crisis and the coronavirus pandemic.

China’s economic woes have stirred debate in recent months about whether the world’s second-largest economy would consider drastic policies to shore up some sectors, like property. The PBOC has already been using targeted lending programs that some analysts liken to QE because they expand the central bank’s balance sheet.

Several economists demurred from interpreting Xi’s appeal for government bond trading as a revolutionary shift in policy.

For one thing, Xi specifically mentioned both buying and selling — a notable distinction from QE, which generally involves buying and holding government bonds and other assets, especially at a large scale. Interest rates in China are also still well above zero, giving less reason for the PBOC to consider a tactic that’s usually regarded as an emergency tool to spur demand when short-term rates have flatlined.

The PBOC didn’t respond to a faxed request for comment from Bloomberg News late last week about Xi’s speech. In the past, it has signaled disapproval over QE: Former governors have pointed to potential risks from US-led QE and warned that asset purchases would damage markets, hurt the reputation of central banks and create “moral hazards.”

Instead, the trading of sovereign bonds may be best viewed as an additional tool for the PBOC to pump liquidity into the market and ensure rates are stable.

The central bank already has several ways to provide money to the economy. It can inject funds through its monthly medium-term lending facility to support commercial bank lending, or lower the amount of cash banks need to keep in reserve.

Those methods have shortcomings, though. Economists see shrinking space for further cuts to the reserve requirement ratio. Loans need to be renewed. And any misjudgment of demand for liquidity risks leading to a serious cash crunch.

The PBOC “needs more flexibilities in managing liquidity and more tools to expand its balance sheet,” UBS Group AG economists Nina Zhang and Wang Tao wrote in a Thursday note. Because the size of the government bond market has expanded over the years, central bank bond trading has become “more necessary and feasible,” they added.

While the directive from the most important man in China certainly indicates the central bank may start buying up bonds, the actual timing of any purchase remains up for debate.

Central bank bond buying will likely be a “very slow process,” said NIFD’s Liu, adding that the shift is still in its “design phase.”

Others suggest forthcoming fiscal stimulus may mean the PBOC pulls the trigger on purchases sometime this year. That would help alleviate liquidity pressure from an upcoming surge in bond supply resulting from the planned issuance of 1 trillion yuan ($138 billion) worth of special sovereign debt in 2024, according to a report from Goldman Sachs Group Inc. economists on Thursday.

It’s also not clear how ramped up bond purchases would impact Chinese yields. They’d likely fall in the short term, according to Citigroup Inc. strategist Philip Yin, since the central bank’s debt purchases “should help improve market confidence about liquidity and digesting future government bond supply.”

The longer-term effects may be mixed. If China combines the use of more liquidity tools with further policies to stabilize economic growth, investors may shy away from safe haven assets in favor of riskier ones.

Whatever China’s strategy, there’s a reason Xi’s comments have created such a stir: Once a central bank decides to start trading government bonds, things can snowball quickly.

The Bank of Japan, for example, intended to limit the scope of its initial program in 2001 before ramping up the amount of bonds it bought. When it embarked on a second round of QE in 2013, Japanese policymakers folded their regular bond trading program, called rinban, into their massive new asset purchasing plan.

In other words, there’s sometimes little real difference between buying bonds as a liquidity tool versus doing so to stimulate the economy.

The minutes of the late-March meeting of the RBA Board (the last in the Martin Place building) recorded that the data had turned out broadly as expected, and that this supported keeping the cash rate on hold. Unlike the February minutes, there was no mention of multiple policy options. While the Board endorsed the language of not ruling anything in or out, it seems that policy actions other than keeping rates unchanged were not on the table at this meeting. The current level of the cash rate is assessed as being just right, at least for the time being.

The minutes noted that the staff assessed that demand still exceeded supply, but the gap was diminishing quickly. A slowing in labour demand was called out, and growth in labour costs was assessed to have peaked. The Board nonetheless remains concerned that domestic costs could continue to rise quickly. The recent turnaround in productivity was noted, as was the role of the pandemic and economic cycle in driving the recent slump. However, the Board is uncertain whether this turnaround will continue. The possibility of a faster snapback in productivity than expected was not mentioned.

The ongoing decline in inflation was highlighted, with the three-month-ended rate for the monthly indicator used as a timelier metric. Monthly inflation was expected to rebound in coming months as special factors such as electricity rebates unwind. The pace of decline in goods prices was also expected to slow, though the minutes did not elaborate on the reasoning for this view. The minutes highlighted the bumpiness of the decline in inflation in other countries and noted that something similar could happen in Australia.

Policy was still assessed as restrictive, so at some point the policy rate will need to decline to prevent inflation from declining so far that it starts undershooting the target. But unlike its peers overseas, the Board is not ready to talk about this decision yet. The minutes again highlighted that interest rates had peaked at a lower level than in some peer economies, and that this reflected the Board’s intention to preserve the gains on employment made since the pandemic period.

A related difference that the minutes did not highlight is the varying squeeze on household sectors across peer economies. Debt-servicing as a share of household income is well above average in Australia at present. By contrast, the financial stability section of the minutes noted that it is below historical averages in many peer economies – despite higher policy rates.

The global outlook was seen as supporting risk sentiment. The chance of a significant downturn had fallen, while interest rates were still expected to decline later in the year. This combination is seen as boosting prices of risk assets and contributing to a more ‘risk on’ tone in markets. This has also been supportive of the Australian dollar exchange rate despite narrower interest differentials and a decline in key commodity prices. Improved risk sentiment also underpins Westpac Economics’ expectation of further upward pressure on the AUD/USD exchange rate later this year; the declines in commodity prices were largely expected and so already priced in.

We expect the RBA to reach the required level of assurance about the path of inflation later in the year, after the full suite of data for the first half of 2024 are released. We continue to expect the first rate cut to occur at the late-September Board meeting.

The other main decision recorded in the minutes related to the operational arrangements for monetary policy. This decision was further elaborated in a speech this morning by Assistant Governor (Financial Markets) Chris Kent.

The background to this decision is that the policy interest rate that the RBA focuses on is the interest rate that banks (and other deposit-taking institutions) charge each other to borrow unsecured overnight in the cash market. The asset that is being borrowed and repaid is exchange settlement funds – that is, banks’ deposits with the RBA, also known as reserves. These deposits are remunerated at a rate that is set below the policy target rate; currently, the spread is 10 basis points.

Prior to the pandemic, the RBA ran a ‘scarce reserves’ regime. Its balance sheet was small, and the staff needed to forecast daily liquidity flows into and out of the system – for example from tax payments and government spending – to keep the amount of reserves at whatever level would keep the cash rate at target. When the pandemic hit, banks wanted more liquidity, so the RBA switched to a regime of excess reserves.

This was also a by-product of the various asset purchase programs introduced during that period. The RBA’s balance sheet expanded, and the actual interest rate banks transacted at in the cash market drifted below the published cash rate target, while remaining above the remuneration rate on exchange settlement balances.

Other central banks, including the Federal Reserve, Bank of Canada and RBNZ, have decided to stick with the excess reserves model even as they wind down their asset purchase programs. Others, including the Bank of England and European Central Bank, have instead settled on a happy medium of ‘ample’, but not excess, reserves. A key distinction between this operating model and its alternatives is that it is managed with full-allotment repo at a pre-specified interest rate, which fixes the price (interest rate) and accepts whatever quantity of reserves is needed to achieve this.

By contrast, in both the ‘scarce’ and ‘excess’ reserve regimes, it is up to the central bank to determine the quantity of reserves that it thinks will achieve the desired price. Still to be determined is the composition of assets the RBA will hold under repurchase agreements on the other side of its balance sheet.

As both the speech and minutes emphasise, this is an operational decision with no implications for the stance of monetary policy. Relative to reverting to the pre-pandemic scarce-reserves regime, though, the RBA’s balance sheet will be larger, with implications for the average size of future earnings to be distributed to the government. It will also mean that the RBA continues to hold some fraction of the government bonds on issue. These will not be available to banks and other deposit takers to meet their prudential liquidity requirements (the Liquidity Coverage Ratio).

As of 1:45 p.m. ET, the value of Bitcoin (CRYPTO: BTC) has dropped 2.8%, Ethereum (CRYPTO: ETH) is down 5.2%, and Bonk (CRYPTO: BONK) has fallen 9.4%. The move was impacted by a long holiday weekend in parts of the world, but 24/7 trading in crypto kept market speculation moving.

The biggest data point impacting the market today is the ISM factory index, which improved by 2.5% in March to 50.3%. A reading over 50% indicates that manufacturing managers think the manufacturing market is expanding. This followed 16 straight months of expected declines.

Investors take this reading as a sign the economy is doing fine, despite higher interest rates over the past few years. If inflation is under control and the Federal Reserve doesn't need to cut rates to help the economy, it may not cut rates at all.

Bond traders are now pricing odds at a rate cut in June 2024 at less than 50%, which would have been a shock coming into the year. Investors were expecting as much as six rate cuts and they may get through two quarters with none.

There's a correlation between technology and high-growth stocks and cryptocurrencies and part of the move higher over the past six months has been an anticipation of lower interest rates. As that thesis falls apart it's not unusual to see cryptocurrencies drop.

Bitcoin exchange-traded funds (ETFs) were another catalyst and that brought billions of dollars into the industry. But last week they started to see outflows, although those turned into inflows again late in the week.

Ethereum has also gotten a boost from speculation that it will get ETFs approved over the next few months. But that's not certain in the current regulatory environment.

If ETFs aren't a catalyst because more aren't approved, crypto values may slide.

Bonk's move is just a mirror of the others with some extra volatility added in. This is a meme coin that's become a popular alternative to Dogecoin, but its main value is speculation rather than being a blockchain currency like Bitcoin or Ethereum.

I think the major catalysts of 2024 are already behind us. Bitcoin's ETFs have helped increase awareness and investment in crypto and that pulled the entire industry higher. And speculation based on the economic environment and a potential cut in rates seems to have been overdone.

That leaves fundamentals driving the crypto market and that's not a great place to be. Bitcoin may have value as a "digital gold," but I think it's more of a speculative asset than anything. Ethereum's promise was always that it would provide real-world use cases in finance and digital assets, but it's so slow and so much more expensive that other blockchains are gaining popularity.

Bonk's meme pop has been nice, but the meme coins never seem to last. And without a catalyst, I could see all three cryptos falling over the next few months as investors look for more fundamental value on the market.

The Reserve Bank will shift to having “ample reserves” as the balance sheet shrinks back toward pre-pandemic levels, Kent said in a speech at the Bloomberg Australia Briefing on Tuesday.

“I want to emphasize that this decision has no implications for the current or future stance of monetary policy,” said Kent, who oversees financial markets at the RBA.

“Rather, it is only relevant to the way in which we will achieve the desired stance of monetary policy through our operations. Nor does it have a bearing on the board’s current approach to quantitative tightening.”

The Bank of England, the European Central Bank and the Swedish Riksbank have announced plans to make a similar pivot of operational frameworks as bonds they purchased during the pandemic to support financial markets mature. The ECB said its review of operational framework preserves its current system of steering interest rates while giving banks more say over how much liquidity they need to operate.

Inflation has fallen sharply since the fever-pitch reached in the summer of 2022. However, progress in fully quieting inflation is further along by some measures relative to others. The Consumer Price Index in the 12 months through February rose 3.2%, a full percentage point faster than its average pace from 2000-2019. Excluding the volatile categories of food and energy, the core CPI has made less progress in settling down, having increased 3.8% year-over-year in February. In comparison, the core PCE deflator has fallen back below 3%. The more marked improvement in PCE inflation over the past year relative to CPI has led to an unusually wide gap between the two most prominent measures of U.S. inflation (Figure 1). Since 2000, the core CPI deflator has run an average of 0.3 percentage points higher than the core PCE deflator, but the gap currently stands at 1.0 percentage point—near the top end of its range in recent decades. What’s driving the wedge, and what could it mean for the path of interest rates and broader economy ahead?

There is no shortage of ways to look at inflation in the U.S. economy. Government agencies publish an array of price indices for goods and services consumed by different end users (e.g., businesses, households and the government), as well as sold by different types of producers (e.g., domestic or foreign). Among the various gauges of inflation, the Consumer Price Index (CPI) and Personal Consumption Expenditure Price Index (PCE deflator) are the most closely watched.

The CPI has a long history of being in the limelight and tends to be the most familiar measure of inflation. The CPI measures price changes for a representative basket of items purchased directly by consumers. Its focus on consumers’ out-of-pocket costs leads it to align more closely with consumers’ perceived inflation experience. The CPI is also widely used in wage contracts and by the government. For example, Social Security’s annual cost of living adjustment (COLA) is based on the CPI while income eligibility for many government programs and federal tax brackets are adjusted with the CPI. Further, the nearly $2 trillion market for Treasury Inflation-Protected Securities (TIPS) adjusts principal returns for inflation using the CPI, tying the CPI directly to a large segment of the financial market.

The PCE deflator also has a key tie to financial markets and the economy writ large: the FOMC benchmarks its 2% inflation target to the PCE deflator. As the preferred inflation gauge of the Fed, the PCE deflator is more relevant for the path of U.S. monetary policy. While over time the CPI and PCE move closely together, unusually wide divergences, like that of today, may send different messages about price pressures in the economy and what may lie ahead for the setting of interest rates.

The Fed’s preference for the PCE deflator can be chalked up to two primary attributes. First, it is wider in scope than the CPI and as a result, generates a more encompassing picture of inflation in the U.S. economy. Whereas the CPI measures the change in out-of-pocket costs to consumers, the PCE price index includes the cost of items purchased on behalf of households, such as medical care paid for by the government or employer-sponsored health insurance. The PCE deflator also incorporates implicit prices of items not directly purchased on the market or “furnished without payment.” Examples include the cost of financial services like “free” checking accounts, brokerage fees, or nonprofit services.

Notably, the wider scope of the PCE can lead to significant differences in categorical weightings. Items not purchased directly by consumers or items that are also purchased on behalf of households carry a bigger weight in the PCE (e.g., financial services and medical care). In contrast, goods and services purchased directly by consumers bear a higher weight in the CPI (e.g., gasoline, vehicles and shelter).

Second, the PCE deflator and CPI are constructed using different statistical formulas. The PCE formula better accounts for the tendency of consumers to substitute away from products whose prices rise relatively quickly. A common example of this dynamic is if the price of beef is rising more quickly than chicken, consumers will substitute chicken for beef. This substitution effect leads to overall living costs rising more slowly than if measured by the fixed consumption basket in the CPI. As a result, policymakers view the PCE deflator as a more accurate barometer of the overall change in the economy’s price level. The CPI “basket” is updated annually based on the Consumer Expenditure Survey to account for shifts in consumption, and the relative importance of CPI categories adjust monthly based on comparative price changes; however, the PCE method captures substitution effects more quickly by leveraging the expenditure data available for the same period. This “formula effect” leads to the CPI typically running higher than the PCE deflator.

The unique scope, weighting and formulas of the CPI and PCE all influence the gap between the two inflation measures. The formula effect tends to be a persistent, but relatively small and stable source of the wedge (Figure 2). Over the past year, the formula difference has contributed 0.1 percentage point to the 0.7 point gap between the year-over-year rates of CPI and PCE inflation, fairly similar to its average since 2003 (Figure 3). Weighting has played by far the largest role in keeping the CPI running hotter than the PCE the past year, primarily due to the ongoing strength in shelter inflation. Other differences, such as seasonal adjustment or source of pricing data, historically have had little bearing on the gap, but over the past year have been another meaningful contributor (+0.3 points) to the wedge between CPI and PCE. The broader scope of the PCE deflator, in contrast, has helped to narrow the gap with CPI inflation over the past year by a historically large degree.

To help think about how the gap between the CPI and PCE deflator will evolve, we find it useful to also look at it on a categorical basis. We focus on the core indices given FOMC officials place greater emphasis on price growth excluding food and gas when considering the near-term outlook for interest rates.

As shown in Figures 4 and 5, housing more than accounts for the 1.0 percentage point gap between core CPI and core PCE the past year. The PCE uses the CPI’s primary rent and owners’ equivalent rent measures as source data, so housing’s more sizable contribution to core CPI boils down entirely to its larger weight (43% versus 18% in the core PCE index).

Rapidly rising costs for motor vehicles have also contributed to the relatively hotter rate of core CPI inflation recently. Motor vehicle insurance carries nearly four times the weight in the core CPI (3.5% versus 0.7% in PCE). In the CPI, insurance costs reflect the premiums paid by consumers, but not claims paid out by insurers on behalf of motorists. The PCE’s net-revenue approach has contributed to motor vehicle insurance costs rising less dramatically since insurers’ costs to cover claims have risen rapidly alongside higher vehicle repair and replacement costs (Figure 6). This pricing difference (part of the “other” category in Figure 5) in conjunction with a lower weighting, has led motor vehicle insurance over the past year to add less than 0.1 points to core PCE, compared to 0.7 points to the core CPI. Health and home insurance are also measured differently in the CPI and PCE but, in contrast to motor vehicle insurance, have narrowed the core CPI–PCE gap by a tenth over the past year.

Further, financial services usually narrows the wedge between the two inflation measures, but recently this category has limited the wedge by less than normal. Financial services account for more than 5% of the core PCE deflator, compared to just a few tenths in the CPI as the cost of many financial service products are not paid for out of pocket. Over the past year, the PCE index for financial services has outpaced the broader core PCE, but by a smaller degree than exhibited over the past decade (Figure 7). Accordingly, it has provided less of a leg-up to PCE inflation than usual.

Healthcare services costs (ex-insurance) have grown more quickly in the CPI than in the PCE index over the past year. However, the PCE’s greater weighting nevertheless leaves healthcare providing a bigger lift to the year-over-year rate of core PCE inflation, and thus has helped to narrow the current gap with the CPI. Healthcare services historically have contributed 0.1 point more to the year-over-year rate of core PCE than core CPI, but over the past year have added 0.2 points more.

Different methods for measuring airline fares, the lion’s share of public transportation, also have whittled down the gap in core CPI and core PCE. The CPI sample consists largely of discount fares for non-business travel scheduled for either the same month or a month or two out. The corresponding PCE index uses Producer Price Index data for domestic airfare, which measures the change in average revenue per passenger (including business travelers) for flights departing during the reference month. Airfares as measured by the CPI are down just over 6% year-over-year, but are essentially flat per the PCE deflator.

Finally, an array of other services have offered a significant counterbalance to the wedge that shelter and motor vehicle insurance have driven between the core CPI and core PCE. These include categories out of scope of the CPI, like social services or nonprofit institutions serving households, both of which are running higher than the broader core PCE and effectively bringing it more in line with the core CPI. Core PCE inflation also includes food services, which is up 4.4% over the past year and is included in the headline CPI but excluded from the core CPI. A summary of the differences in current category contributions versus their historical effects on the core CPI–PCE gap is shown in Figure 8.

Looking ahead, we expect the gap between core CPI and core PCE when measured on a year-ago basis to remain uncharacteristically wide through the first half of next year (Figure 9). On a quarterly annualized basis, the return to the ~0.3% realm should occur somewhat earlier; we estimate around the first quarter of 2025.

The timing and extent to which the gap between core CPI and core PCE settles back down hinges heavily on the path of shelter inflation given its current hefty contribution. While the moderation in official measures of shelter inflation has been painstakingly slow to materialize, we expect housing disinflation to continue this year with the year-over-year pace of primary shelter resembling its pre-pandemic pace of 3.0-3.5% by next spring. A host of private sector measures suggest apartment rent growth has already settled down, with rents for single family units having also slowed materially (Figure 10). That said, primary shelter has grown to account for an even larger share of CPI since 2019 (+2.5 pts), while the PCE share is little changed. As such, weighting effects are likely to continue to play somewhat of a bigger role on the CPI–PCE gap unless shelter inflation undershoots broader core inflation, an outcome that is plausible but unlikely in our view.

We also anticipate that the chasm stemming from motor vehicle insurance will narrow over the course of the year. Insurers have had to play catch-up to the spike in vehicle and repair costs following severe supply shortages in the auto industry, a process drawn out by the need for regulatory approval to raise premiums in most states. However, motor vehicle insurance premiums per the CPI have now risen more than the price of vehicles, equipment and repair services since the start of 2020, suggesting a diminishing need for premiums to continue rising at the past years’ eye-watering pace (Figure 11). Recent months’ data show tentative signs of the CPI for motor vehicle insurance already beginning to cool, with prices up 17% on a three-month annualized basis compared to 21% on a 12-month basis.

The expected narrowing of the inflation gap as shelter and motor vehicle insurance inflation slows is likely to be partially offset, however, by cooler price growth in categories currently providing an unusually-large boost to core PCE. As inflation pressures ease from more tepid demand and a balanced labor market, we expect slower price growth for healthcare, food services and other sundry services to reduce the outsized contribution “other core services” are currently providing to core PCE. Public transportation is also unlikely to reduce the inflation gap ahead as the CPI for airfares recovers from last summer’s swoon.

The relatively higher rate of CPI compared to the PCE deflator likely goes some way in explaining why consumers continue to reel from the current inflation environment, but Fed officials see enough progress to expect some policy easing later this year. Even as consumer sentiment has improved over the past year, it remains low relative to the strength of the jobs market as prices paid by consumers continue to rise at an unfamiliar speed.

Yet, while core inflation as measured by both the CPI and PCE price indexes remains too high, it has fallen more sharply over the past year when measured by the PCE deflator and is closer to the pace consistent with the Fed’s inflation target (Figure 12). Furthermore, although the gap currently remains large, it is not unprecedented. A similarly sized wedge opened up in the late 1990s and in 2001 and 2009. As the pricing environment cools from its extreme state, the gap should diminish as well.

We doubt the Fed is too bothered by the CPI–PCE gap. The elevated rate of shelter inflation is the largest contributor to the wedge at present, but officials continue to espouse optimism that shelter inflation will ease further ahead. In the March 20 FOMC press conference, Chair Powell said that “There’s a little bit of uncertainty about when that will happen but there’s real confidence that [lower market rents] will show up eventually over time.”

While the wedge between CPI and PCE inflation does not appear to be a major source of consternation among Fed officials, it could make the FOMC’s job of returning PCE inflation to 2% a little harder on the margin with cost of living adjustments and price contracts more likely to be based off of the stronger CPI inflation rate. In addition, the Fed does not solely look at the PCE deflator when assessing the current state of inflation. If the gap persists at recent levels and the drivers shift away from housing, it could contribute to some unease among policymakers about the risk of cutting the fed funds rate too early.

Nevertheless, we do not believe the current gap between the two most ubiquitous measures of U.S. inflation is causing any particular misgivings among the FOMC participants about the path of inflation ahead. With the Fed’s preferred inflation measure continuing to grind lower, some easing of policy this year likely remains in the cards.

The US dollar hovered near a 4-1/2-month high against major peers on Tuesday as traders rushed to push back bets for the Federal Reserve's first interest rate cut this year. The dollar held close to a six-week peak versus the euro and sterling reached on Monday, after US data unexpectedly showed the first expansion in manufacturing since September 2022.

Fears of intervention by Japanese officials limited dollar gains against the yen, even as long-term US Treasury yields - which the currency pair tends to track - jumped more than 14 basis points to a two-week top at 4.337 per cent overnight.

Gold, which performs best when yields are falling, was knocked back from a record peak. The US rate futures market now factors in 61.3 per cent odds of a Fed rate cut in June, down from about 70.1 per cent probability a week ago, according to the CME's FedWatch tool.

"The divergence of solid growth dynamics for the US and waning Fed rate cut risk against sluggish growth for other FX majors suggests that any DXY dips should be seen as buying opportunities," said Westpac's head of currency strategy, Richard Franulovich, referring to the dollar index.

The dollar index , which measures the currency against the yen, euro, sterling and three other peers, edged 0.02 per cent higher to 105.02, following a 0.51 per cent rally to as high as 105.07 on Monday.

The euro slipped 0.08 per cent to $1.0733, sticking close to the overnight low of 1.0731. Sterling was 0.04 per cent lower at $1.25455 after sliding to $1.2540 in the prior session. The Japanese yen firmed slightly on Tuesday to 151.565 per dollar , after dipping to 151.77 the previous day.

It reached a 34-year trough of 151.975 last week, spurring Japan to step up warnings of intervention. On Tuesday, Finance Minister Shunichi Suzuki reiterated that he wouldn't rule out any options to respond to disorderly currency moves.

Japanese authorities intervened in 2022 when the yen slid toward a 32-year low of 152 to the dollar. The yen's slide has come despite the Bank of Japan's first interest rate hike since 2007 last month, with officials cautious about further tightening amid a fragile exit from decades of deflation.

"Despite heightened risk of intervention, the BOJ's policy stance remains very accommodative and Japanese data continue to show the fragility of their 'virtuous cycle' economic recovery," said Westpac's Franulovich.

"If intervention occurs, resultant flushes in USD/JPY below 150.00 could be relatively deep given the recent surge in leveraged shorts in JPY. However, they are still likely to be seen as buying opportunities once positioning has become more balanced."

Elsewhere, the Australian dollar was flat at $0.6489, after skidding to a nearly one-month low of $0.64815 on Monday. New Zealand's kiwi dollar eased 0.07 per cent to $0.5949, edging back toward the 4-1/2-month trough at $0.59395 from overnight.

Spot gold edged up 0.11 per cent to $2,253.09, after dropping back from a record high at $2,265.49 in the previous session. Leading cryptocurrency bitcoin declined 0.87 per cent to $69,158, but was well within its relatively narrow trading range of the past week.