Author: admin

The widest gap for more than two decades has opened up between the forecasts of the three main international organisations that track the world’s crude markets, causing confusion about global oil demand.

The Paris-based International Energy Agency (IEA), the Opec group of oil producers and the US Energy Information Administration (EIA) all reported their monthly oil statistics this week. The divergence between their forecasts for oil demand in 2026 reached 1.8mn barrels of oil a day (b/d), the equivalent of France’s oil consumption and the biggest gap since 2002.

The gap has gradually widened over the past two years, raising questions about the trajectory of the world’s oil demand as countries transition to clean energy.

The oil market has been caught this year between fears of a supply crunch — stoked by US sanctions on two of Russia’s largest oil companies in October — and warnings of a longer-term glut driven by rising US shale output and an economic slowdown.

Brent crude touched $82.63 a barrel in January before resuming its downward trend, then spiked again in June after tensions flared in the Middle East. The benchmark was trading at about $64 a barrel on Friday.

The IEA is the most bearish about demand in 2026, putting it at 104.7mn b/d. Opec is the most bullish, with a forecast of 106.5mn b/d. The EIA is between the two, at 105.2mn b/d.

Traders had been puzzled by the divergence, said Tamas Varga, an analyst at PVM, an oil broker. “These quite significant differences in views caused confusion, in that traders just simply did not know who to believe, did not know which number is accurate,” he said.

He added that there had traditionally been a strong correlation between oil stock levels and prices, so if the forecasts of stock levels diverged, “you can’t really figure out what the price could be in the future”.

There are multiple possible causes for the differences in the forecasts, including a greater share of the world’s oil falling under sanctions, and a lack of visibility about how much oil is being put into strategic storage by China, amid an off-again, on-again trade war with the US.

One oil market observer, who asked not to be named, said that Iran, under US sanctions, might be trying to hide a share of its oil production, causing problems with the data, while Giovanni Staunovo, an analyst at UBS, said that since the first Trump administration, China had held back some data, concealing the true size of its stocks, making it harder “to have the full picture”.

Adding to the confusion, said Martijn Rats, an analyst at Morgan Stanley, there were now 1.13mn b/d of oil in the IEA’s model that were “unaccounted” for. These are barrels that have been produced, but do not appear to have been consumed or stored.

“Historically, nine out of 10 times when you have unaccounted oil, you see revisions to demand as more data comes out,” he said.

He noted, however, that the missing barrels had grown from 110,000 b/d in 2024 to nearly 2mn barrels in August before falling back. The IEA said the unaccounted barrels were the result of a time lag in the data, but Rats suggested that the numbers were now so large, they might be distorting the view of demand growth.

“There’s lots of conflicting data, it’s a fog, but the fog is telling you these numbers are not jiving and the underlying demand might be better than we thought,” he said.

He added that if demand growth was 200,000 b/d a year stronger than previously thought, it would have big implications for the sector.

“Then you are back to the historical trend and you are thinking very differently about the impact of the energy transition on growth. Maybe less is changing than you thought,” he said.

The IEA said that while it was confident of its forecasts, its demand figures may be subject to revision because the number of countries that regularly published data on oil supply and demand had fallen in recent years, “to the detriment of market transparency and stability”.

There was also an underlying political divide among the institutions, said David Wech, chief economist at Vortexa, an energy data company.

“There is, overall, a certain trend in terms of energy transition and related aspects of climate change . . . it’s highly likely that Opec is wanting to give a positive outlook on the market, while it is possible that the IEA is, perhaps deliberately or emotionally, more on the conservative side.”

Unlock the Editor’s Digest for free

Roula Khalaf, Editor of the FT, selects her favourite stories in this weekly newsletter.

Private markets have been attracting increasing regulatory scrutiny, and for good reason.

It’s not just that the level of disclosures is less than what is required in public markets. Or even that valuations of private market assets are more based on models rather than public pricing, robbing regulators of market signals that could inform their work. It’s because private markets have become so big.

By value, global private assets under management came to just over $13tn in 2023, having more than doubled in size over the previous five years, according to a recent report by financial data firm Preqin. It estimated that this figure is on track to almost double again by 2030. The vast majority of these holdings are private equity assets, though the growth of private credit has been explosive. To put these numbers into perspective, Bain & Company estimates that the total assets under management in the global asset management industry for 2023 totalled $115tn.

But while the lack of data transparency makes it hard to say anything with confidence, it’s not obvious that this growth poses immediate direct risks to the banking system. Good data as to how much banks lend to private market entities in general, or even private credit firms specifically, is scarce. However, the IMF’s Global Financial Stability Report from April 2025 highlighted a Moody’s report that estimated bank exposure to private credit funds was $525bn at the end of 2023.

That $525bn sounds a lot. But global banks are huge. As Moody’s puts it, exposures are moderate with private credit loan commitments about 3.8 per cent of total loans on average in 2023. So sure, private credit managers could start making terrible loans that default but this scenario doesn’t look like it would immediately blow up the banking system — although the use of “synthetic risk transfers” which enable banks to transfer credit risk on diverse loan pools to investors (typically credit funds and asset managers) may complicate the picture.

If there is a problem, it looks more likely to crop up in the insurance sector. Private credit now accounts for more than 35 per cent of total US insurer investments and close to a quarter of UK insurer assets.

Private equity-owned insurers in particular have proved effective at improving capital efficiency — taking more risk with each dollar of capital. This could come through some combination of regulation-shopping the jurisdiction of their reinsurance operations, lending to affiliates, or engaging in wholly-owned portfolio securitisation.

It could also come from building out investment operations so they can take more substantial exposure to private credit — which can offer substantial illiquidity premia, the additional return that can come when taking on the risk of holding an asset that is not easily sold.

How risky is this? Well, the private credit holdings of insurers overwhelmingly carry investment grade credit ratings — signalling exceptionally low prospective default risk. However, it is an open question whether ratings by different credit rating agencies all deserve equal trust — especially those ratings that are issued privately without public disclosure. Colm Kelleher, chair of UBS, has accused insurers of ratings shopping, calling the phenomenon “a looming systemic risk”.

Moreover, a recent analysis from Moody’s shows that while a ninth of US life insurers’ fixed income holdings by value carry private ratings, this share jumps to more than half of their so-called Level 3 holdings. Level 3 holdings are assets that are the least liquid, hardest to value and priced using models that rely on internal assumptions. And, according to the credit rating agency, US insurers’ less-liquid private asset portfolio was skewed to lower-rated holdings at year-end 2024.

Recent high-profile company failures like First Brands and Tricolor — characterised as credit cockroaches by Jamie Dimon — provide a warning of the potential downside. The JPMorgan chief executive quipped that if you see one cockroach, there are probably more.

But given the wave of new money that has moved into loans and bonds, borrowers have been able to raise finance in both public and private markets at attractive rates. Default rates have been low — even taking into account that some financially stretched businesses have found private lenders willing to restructure their credit into cash-lite payment-in-kind loans.

Market prices suggest that there is little prospect of an economic downturn sufficient to impede debtors’ timely payment of principal and capital any time in the foreseeable future. If correct, insurers will continue to profit from their greater capital efficiency. So will ordinary people. Increased competition has pushed insurers to offer better terms for those seeking fixed or variable rate annuities for retirement income.

So tilting the scales away from resilience and towards profitability could prove to have been exactly the right thing to do for insurers. But we’ll have to see how they fare in the next credit downturn to find out.

toby.nangle@ft.com