Skip to content 
How apparently separate parts of the financial system can become exposed to the same stress without looking connected.
GLOBAL · PRIVATE MARKETS · FINANCIAL STABILITY
THE THESIS
For much of modern finance, diversification has been treated as a defence.
Different institutions hold different assets. Different funds pursue different strategies. Banks, insurers, pension funds and private-market investors operate under different mandates and regulatory frameworks.
On paper, those differences imply separation.
But separation of ownership does not necessarily mean separation of risk.
A bank can finance a private-credit fund. That fund can lend to a company owned by a private-equity sponsor. An insurer can hold debt originated within the same private-market ecosystem. Another lender can finance the fund itself rather than the underlying company.
Each exposure may appear distinct.
The economic risk beneath them may not be.
This is one of the harder problems now emerging across global finance:
Correlation can build before it becomes visible.
The concern is not that every participant owns precisely the same asset.
It is that apparently different positions may depend on the same conditions remaining favourable:
Cheap or available refinancing. Stable collateral values. Reliable liquidity. Low defaults. Continued investor confidence. And valuations that remain credible despite limited market price discovery.
When those assumptions move together, diversification can become less protective than it appears.
THE CONNECTION IS OFTEN INDIRECT
Private markets illustrate the problem particularly clearly.
Private credit has grown rapidly, providing companies with an alternative to conventional bank lending and public bond markets.
That diversification of funding can be beneficial.
But the system has not simply shifted credit risk from banks to non-banks.
Banks and non-bank financial institutions increasingly operate within the same financing ecosystem.
Banks provide credit facilities to private funds. They finance acquisitions. They lend against fund assets and investor commitments. They participate in securitisations and provide other forms of leverage.
Insurers and pension funds supply capital to private-market vehicles.
Private-equity sponsors can sit at the centre of networks involving multiple borrowers, funds and financing arrangements.
Risk has therefore not necessarily disappeared from the regulated banking system.
Some of it may simply have changed form.
WHAT LOOKS DIVERSIFIED CAN SHARE THE SAME DRIVER
Imagine three separate exposures.
A bank lends to a private-credit fund.
An insurer invests in debt originated by a private lender.
A pension fund invests in another private-market vehicle.
The counterparties are different.
The instruments are different.
The legal structures are different.
But all three may ultimately depend on highly leveraged companies remaining able to service and refinance their debt.
That is not simply conventional asset correlation.
It is structural correlation.
The exposures become connected through a common dependency.
When conditions are benign, those connections can remain largely invisible.
Loans perform.
Funds raise capital.
Valuations remain stable.
Refinancing remains available.
Different parts of the system therefore appear independently resilient.
Stress changes the picture.
Deteriorating corporate credit can weaken loans held by private-credit funds. Falling valuations can affect leverage and investor confidence. Banks financing those funds can become more cautious. Institutional investors can reduce allocations. Refinancing conditions can tighten further.
The original credit problem can begin to reinforce itself.
CORRELATION DOES NOT REQUIRE A CRISIS
The process does not require a conventional financial panic.
A prolonged tightening can be enough.
The Bank of England reported in December 2025 that approximately 20% of the UK private-credit debt captured in its analysis was due to mature by the end of 2026, with 42% due by the end of 2028.
That creates a common test.
Businesses that borrowed under easier conditions eventually have to refinance in whatever environment exists when their debt matures.
The outcome does not have to be immediate default.
Borrowers and lenders can extend maturities.
Loan terms can be amended.
Interest can sometimes be capitalised rather than paid immediately in cash.
Assets may remain within private structures for longer.
These mechanisms can reduce immediate pressure.
But they can also delay the point at which deterioration becomes clearly observable.
Apparent stability and underlying resilience are not necessarily the same thing.
THE SAME RISK CAN APPEAR IN SEVERAL PLACES
Financial claims can also be layered.
A company has private debt.
A fund owns that debt.
A bank finances the fund.
An insurer invests elsewhere in the same private-market ecosystem.
Another institution holds structured credit exposed to similar borrowers or economic conditions.
From the perspective of each balance sheet, the positions look different.
From the perspective of the system, several institutions may ultimately depend upon the performance of the same underlying corporate sector, the same refinancing market or the same collateral assumptions.
This is where hidden correlation becomes particularly difficult to measure.
The problem is not simply knowing who owns an asset.
It is knowing:
Who ultimately depends on it?
DATA DESCRIBES THE PARTS BETTER THAN THE NETWORK
Financial regulation has traditionally developed around institutions.
Banks report as banks.
Insurers report as insurers.
Funds operate under their respective regulatory and reporting frameworks.
That structure makes operational sense.
Systemic risk, however, does not respect institutional boundaries.
A regulator can therefore possess substantial information about individual organisations while retaining an incomplete picture of the connections between them.
The problem is especially difficult within private markets, where information can be fragmented across funds, financing vehicles, counterparties and jurisdictions.
Visibility at the institutional level does not automatically create visibility at the system level.
STRESS REVEALS CONNECTIONS
This changes what financial stress testing needs to ask.
The question is no longer only:
Can this institution survive the shock?
It must also ask:
What will the institution do because of the shock?
Will it withdraw financing?
Reduce credit lines?
Demand more collateral?
Sell assets?
Stop refinancing borrowers?
Alter its risk appetite?
And what will institutions on the other side of those transactions do in response?
The distinction matters because individually rational behaviour can produce collectively destabilising outcomes.
The IMF’s 2025 work on non-bank financial institutions found that stress within non-banks can transmit back into the banking system. In one modelled scenario, non-bank stress and full drawing of bank credit lines produced significant declines in capital ratios for portions of both the US and European banking sectors.
The Bank of England is now examining precisely these interactions.
Its Private Markets System-Wide Exploratory Scenario brings banks and non-bank institutions into the same analytical framework to examine how their actions could interact during a severe but plausible downturn and whether those interactions could amplify financial stress.
That represents an important change in perspective.
The relevant unit of analysis is increasingly not just the institution.
It is the network.
WHAT WE OBSERVE
01 — INTERCONNECTION IS INCREASING FASTER THAN VISIBILITY
Banks, insurers, pension funds and private-market firms are increasingly linked through lending, investment and financing relationships.
Information about the ultimate distribution of those risks remains incomplete.
02 — DIFFERENT ASSETS CAN DEPEND ON THE SAME CONDITIONS
Different instruments do not necessarily represent different economic risks.
They may all depend on refinancing remaining available, collateral remaining valuable, leverage remaining manageable or borrowers continuing to perform.
03 — STRESS CAN PROPAGATE WITHOUT FAILURE
Institutions do not need to collapse for financial conditions to tighten.
Several lenders reducing exposure simultaneously can be enough.
Several investors becoming more cautious can be enough.
Several sources of refinancing disappearing together can be enough.
The transmission mechanism is behaviour.
WHAT MATTERS
The conclusion is not that private credit is inherently unstable.
Private markets can broaden access to capital, diversify corporate funding sources and provide forms of long-term finance that banks or public markets may not supply as efficiently. The Bank of England itself recognises these benefits.
The structural issue is subtler.
Diversification protects a system only when the underlying risks are genuinely independent.
As finance becomes more interconnected, institutions can appear separate while remaining economically dependent on the same conditions.
When those dependencies are difficult to observe, measured diversification can overstate actual resilience.
The danger is therefore not merely concentration.
It is concentration that looks like diversification.
Financial instability often makes correlations obvious only after stress arrives.
By then, connections that previously appeared theoretical become operational.
The system discovers that institutions which looked independent were relying on many of the same assumptions all along.
ERTHS ASSESSMENT
The next major challenge in financial stability is not simply identifying weak institutions.
It is identifying common dependencies between institutions that individually appear strong.
The risk is not that everything is connected.
It is that we may not know which connections matter until they are tested.
SOURCES
For the published page, I’d show these as clean source links beneath the article:
International Monetary Fund
Global Financial Stability Report, April 2024 — Chapter 2: The Rise and Risks of Private Credit
International Monetary Fund
Global Financial Stability Report, October 2025 — Shifting Ground beneath the Calm
International Monetary Fund
Growth of Nonbanks is Revealing New Financial Stability Risks — October 2025
Bank of England
Financial Stability Report — December 2025
Bank of England
Private Markets System-Wide Exploratory Scenario — 2026
Bank of England
Financial Stability Report — July 2026