When inflation becomes a persistent backdrop for the global economy, the foundation of a decades-old asset allocation theory is crumbling. Akhil Bansal, Head of Asset-Backed Finance (ABF) at Carlyle Group LP, warned in a Monday interview that the reliability of traditional fixed-income products as portfolio shock absorbers is declining, as their correlation with equities has experienced a structural shift.
"The diversification benefits that fixed income is supposed to provide simply don't appear to be working," Bansal stated bluntly. The math reveals the severity of this change: from 2010 to 2020, the correlation between public fixed income and equities was only 0.13, indicating they often moved in opposite directions, validating the classic logic of bonds and stocks as complementary portfolio choices. However, from 2020 to the present, this correlation has surged to nearly 0.63. This means bonds and stocks are increasingly moving in tandem, and the traditional "safe-haven" function of bonds is disappearing.
Inflation: The Structural Driver of Positive Stock-Bond Correlation
Bansal attributes this market breakdown to accelerating inflation. He stated that rising prices simultaneously impact both stocks and public fixed-income investments, erasing the diversification advantages that have supported the classic 60/40 portfolio construction for decades. This assessment aligns with broader market data. Analysis from Hitachi Research Institute shows that since mid-2025, the correlation between US stocks and long-term government bonds has been steadily strengthening. During the tariff shock in early April 2026, US stocks and long-term government bonds fell together, replicating the same market dynamics seen in 2022, when the 60/40 strategy experienced its worst performance in 150 years. Osaic has recently cut its fixed-income allocation within its 60/40 portfolio from 40% to 31%. As Bansal noted, this higher correlation represents a structural change. When the inflation rate remains persistently above approximately 2.7%, the stock-bond correlation tends to turn positive. Only during a recession can US Treasuries still provide a hedging function. Currently, the ongoing Iran war pushing up energy prices, demand growth driven by AI infrastructure investment, and the compounding effects of tariff policies are locking in inflationary pressures at a structurally high level.
The "Stealth Tech-ification" of the Corporate Bond Market: Concentration Risk Emerges
Bansal also pointed to a second concerning trend: concentration risk in the corporate bond market. Investors may find their fixed-income portfolios are becoming as tech-dominated as the S&P 500 index. This concentration risk was fully validated by data from 2026. By mid-July, six major technology companies had issued approximately $244 billion in bonds globally. Barclays estimates that in 2026, global hyperscale data center operators will issue $285 billion in investment-grade bonds, with a market concentration comparable to the "Magnificent Seven" tech stocks a decade ago. Bansal noted that companies like Alphabet and Meta are often classified as communication or consumer cyclical sectors, rather than technology, which obscures their true position in the AI and data center boom. This "mismatch" in industry classification further blurs investors' real perception of their tech risk exposure within their portfolios. UBS estimates that global tech and AI-related debt issuance in 2025 more than doubled year-over-year to $710 billion, with a potential to further approach $990 billion in 2026. This flood of AI-related debt is making the US corporate bond market appear safer, yet it conceals hidden risks. The trillions of dollars that bond investors are providing for AI buildout may ultimately have lower-than-expected profitability, potentially trapping even currently stable companies in difficulty.
Carlyle's Solution: A "Low Correlation" Portfolio of Private ABF and Energy Assets
Faced with the failure of traditional fixed income's diversification function, Bansal is not advocating for abandoning fixed income entirely. Instead, he recommends seeking assets that can generate income and stability while having a low correlation with corporate earnings. Bansal clearly stated that ABF is not intended to replace fixed income, but rather to play a role in "generating income and providing diversification, a role that public fixed income has historically played."
Carlyle's push into private ABF is accelerating. As of the first quarter of 2026, Carlyle Group LP's global credit business had $209 billion in AUM, with its asset-backed financing strategy exceeding $12 billion, growing over 30% year-over-year. Carlyle's ABF platform managed over $10 billion in assets as of December 31, 2025. In April of this year, Carlyle raised $1.5 billion in the first round of fundraising for its newly established asset-backed investment fund. Carlyle's credit business accounts for approximately 44% of the firm's management assets. KKR is also doubling down on this track. KKR points out that direct lending and ABF have historically low correlations with fixed income, helping to achieve better long-term investment outcomes. KKR's ABF platform already manages over $74 billion in assets.
Regarding AI-related ABF investments, Bansal opposes a blanket exclusion of all AI concepts. He stated that chip financing and data center transactions ultimately depend on cash flows and leases from investment-grade counterparties – areas where ABF can safely engage. "One of Carlyle's approaches is that we don't view AI as a single area, but rather look at diversified investments," Bansal said, specifically highlighting energy and natural gas investments as a unique alternative investment strategy within the AI ecosystem for Carlyle. This contrasts with BlackRock's recent bond issuance for Meta's data center. BlackRock, through a special purpose vehicle (SPV) named Sopaipilla Investor, issued $12.55 billion in bonds, backed by Meta's lease payments over a 20-year period starting in 2028. This off-balance-sheet structure allows Meta to access AI computing power infrastructure without directly increasing its debt burden, but investors bear the concentrated risk tied to the credit of a single technology giant.
This strategy has recently achieved material progress. In June 2026, Carlyle Group LP and Diversified Energy announced a $2 billion partnership to invest in proven developed producing (PDP) natural gas and oil assets in the United States. Commenting on the deal, Bansal said, "Diversified Energy is a leading operator of long-life energy assets and a pioneer in introducing PDP securitization to the institutional market."
Comments