Yves here. Due to having closely chronicled the 2007-2008 crisis, and then learning even more afterward by continuing to dig into what happened, yours truly has a strong point of view about what happened. In his post below, Richard Murphy correctly points out that private debt excesses and not public debt, are what triggers financial crises. He points to a strong parallel to the runup to the crisis, that of widespread overvaluation of real estate. Recall that William White and Claudio Borio of the Bank of International Settlements were sounding warnings starting in 2003 about frothy residential real estate prices in many markets, particularly in the Anglosphere.
However, the reason no one anticipated the near collapse of the global financial system in September 2008 was that that was not a real estate bubble implosion but a derivatives crisis. As we explained long-form in ECONNED, the use of a statistical arbitrage strategy led to the creation of CDOs whose assets were primarily credit default swaps referencing particularly rancid tranches of subprime mortgage bonds (the inclusion of ~20% of securities, as opposed to CDS, greatly increased the market for these CDOs). This structure was extremely leveraged and in and of itself drove the demand for >50% of subprime mortgage bond origination during the so-called toxic phase of the subprime mortgage lending.
More generally, the use of credit default swaps created risk exposures greatly in excess of the value of subprime mortgage bonds. Most experts estimate the derivative wagers were 4-6 times the value of real economy exposures.
And as we further explained in ECONNED, those CDO and CDS exposures wound up disproportionately at systemically important, undercapitalized financial institutions.
So we are not disputing Murphy’s thesis that a crisis of some sort looks likely, and will become even more so with no course correction. However, what made 2008 distinctive, in a bad way, was the degree of gearing, as in leverage on leverage. Consider the impact of the CDO strategy we sketched out above:
If you look at the non-synthetic [non-CDS] component, every dollar in mezz ABS CDO equity that funded cash bonds created $533 dollars of subprime demand.
And there were other market-goosing effects, such as artificially lowering credit spreads (finance speak for making subprime mortgage loans cheaper than they would otherwise have been).
However, the unwind of a mere real estate bubble can be pretty bad, as the example of Japan and the US S&L crisis demonstrate.1 And unlike 2007-2008, where many expert observers such as the Bank of England, were warning of the dangers of credit default swaps, CDOs, as well as the position of roughly sixteen “systemically important” institutions, there is much less understanding of where leverage on leverage bombs lurk today.2
So if the debt-bomb is largely real estate and the scale is destructively large, the most likely result is zombification rather than a meltdown.
But an open question is whether a big unwind in such an economically important sector as real estate will trigger the explosion (or perhaps more accurately, implosion) of leverage-on-leverage plays, greatly amplifying the damage and potentially blowing back to the systemically critical payments system. One area of known leverage on leverage, although no one has a good grasp of the severity, is in private equity and private debt funds. For instance (as we decried when we first learned of the practice), private equity firms are not only borrowing heavily at the level of the companies in their funds. Many, if not most, now use “subscription lines of credit,” which allows them to borrow at the fund level with the unused commitments of the limited partners as the collateral. A 2025 note by law firm Mayer Brown confirms that credit funds are similarly engaging in fund level borrowing.
Crypto is another possible venue for leverage on leverage. The crypto-valuation plunge in late 2022 led to the collapse of Silvergate Bank, whose “deposits” were 98% crypto, and had also engaged in lending against crypto holdings. But I have yet to see a good decomposition of the role of the lending by Silvergate as opposed to merely a traditional bank run in its demise. US regulators have been making some attempts to make sure that stablecoins are properly backed by real assets, which largely undermines the attraction of being in that business at all. Another big crypto worry is that proposed rules put no limits on rehypothecation, while there are strict limits for US securities. Rehypothecation means that the collateral a customer posted for a loan can be repledged by the lender to obtain another loan. The fact that the UK did not have limits on rehypothecation and Lehman exploited that amplified the damage done by Lehman’s demise.3
These actual and potential bubbles are growing even as the growth prospects in the Collective West are poor. The UK faces a food crisis this winter. Budget crises are underway or loom in much of Europe due to the economic drag of Russia-sanctions-induced high energy prices, which will be made worse by US tariffs, running into the perceived imperative of greatly increased military spending. In China, as we have discussed, the government is seeking to tamp down an overproduction crisis, which is deflationary. Deflation makes debt overhangs even worse, since prices, particularly for labor falls while the value of borrowings and interest payments do not. Whether China will succeed is an open question. but we do see price anecdata that China is exporting deflation to Southeast Asia.
Note that Murphy had a second post, Could 2008 happen again? Part 2, describing frothy real estate lending in the UK.
By Richard Murphy, a chartered accountant and a political economist who has been described by the Guardian newspaper as an “anti-poverty campaigner and tax expert”. He is Professor of Practice in International Political Economy at City University, London and Director of Tax Research UK. He is a non-executive director of Cambridge Econometrics. He is a member of the Progressive Economy Forum. Originally published at Tax Research UK.
This is from the FT this morning:
Almost three-quarters of HSBC’s Hong Kong commercial property loan book was flashing warning signs by the end of June, as a prolonged slump in retail spending and sluggish demand for office space weighed on Europe’s largest bank.
HSBC is Hong Kong’s largest lender and the territory is the single largest source of income for the bank by geography. HSBC has made $32bn of commercial real estate loans in the territory, out of a $234bn Hong Kong loanbook.
Why flag this particular article on a morning when there is so much else that is going so wrong in the world? My reason is to highlight the fact that whilst the world’s media obsesses about government debt, there is, quietly, but undoubtedly inevitably, a growing debt crisis developing in the world banking systems.
Throughout the world, vast quantities of bank lending are secured on domestic and commercial properties. In other words, banks make loans because they can place mortgages over properties that they believe they can sell in the event of their customer defaulting, thereby providing them with the security to make the loan that the customer in question wants, whether that is to buy the property in question or not. This is, of course, how the domestic mortgage market works. It is also how much of commercial bank lending works.
The reason for highlighting what is happening to HSBC in Hong Kong is that the basic assumption made by all bankers worldwide — that property is their security — is something that can go horribly wrong. It did, of course, in the US domestic mortgage market in 2008, and the global financial crisis followed as a consequence. There is every reason to think that this might happen again.
In Hong Kong, the risk arises because there has been a collapse in the rental market in that territory, with the FT noting that rents have fallen by approximately 20 per cent since 2022, whilst the vacancy rate is running at something like 19 per cent, meaning that the properties in question are earning nothing at all.
Elsewhere, the risks differ. For example, in the UK, where approximately 85 per cent of all loans by banks are secured by mortgages, the most significant overall risk comes from the uninsurability of many properties in the future because of the risk that they might be flooded by rising seawater levels. This is true for many domestic properties. However, the problem might be even bigger in the commercial sector. It is thought that up to 80% of all commercial bank loan portfolios are secured on properties that might be subject to this risk. Just look at the flood risk map for central London and you will see precisely why this is the case.
The point I am making, therefore, is a straightforward one. Whilst politicians, neoliberal commentators, the mainstream media, and others wish us to be distracted by the supposed risk that government debt poses to us, and to our grandchildren – the lucky ones of whom will inherit a share of it – the real likely debt risk that will create the next global financial crisis is almost certainly already on the balance sheet of most of the world’s major bankers. That is because it is represented by overvalued property where the chance of loan repayment is low precisely because the properties in question will, at some point in the not too distant future, become unsaleable, either because of changes in the market demand for property within the commercial sector, or because flood risks will mean that the buildings in question will be uninsurable, creating a crisis for the banks because the security that they have for their lending will no longer be of any value.
To put this another way, the assumption that 2008 could not happen again is wrong. It could, because the next global financial crisis might well be precipitated by overvalued bank balance sheets, as was the case in 2008, even if the precise reasons for the overvaluation might change.
Bankers never, it seems, learn, and as fools in charge of money, they appear to be all too readily parted from it. We will, of course, all end up eventually paying the price for that.
____
1 The reason the damage from the S&L crisis was not as severe as most anticipated was that Greenspan was able to engineer a very steep yield curve, with cheap short-term rates but hefty intermediate and long-term yields. So banks were able to rebuild their balance sheets much faster than anticipated by engaging in simple-minded, traditional “borrow short, lend long” practices.
2 We fiercely criticized Geithner, Paulson, and Bernanke at the time for not hauling the big players into a room and demanding that they ‘fess of about their CDOs and CDS positions and who the “protection writers” were. That might, for instance, given them an earlier warning of the critical position of AIG and the monoline insurers (the latter did not come into focus until early 2008).
Rehypothecation is a procedure by which one bank lends securities that its clients have pledged as collateral. In the US, rehypothecation is capped to a certain amount of a customer’s assets, but in Europe there is no such cap.
Two years ago this week, hedge fund managers discovered that assets they had posted as collateral with their prime broker were stuck somewhere within Lehman Brothers. Many of these managers had not even been using Lehman as their prime broker.
The assets wound up in Lehman through rehypothecation, often without the hedge fund managers’ knowledge. With some of their portfolio assets frozen in the collapsed US bank, the hedge fund managers had to work hard to maintain liquidity in their funds, a need that was particularly acute since, separately, many investors in hedge funds were seeking their money back.
25 comments
Comments are closed.
Thanks for the overview (as so often, Yves intro leverages the value of the article x3). I was trying to explain all this to an economic neophyte friend this weekend – I wish I’d had something succinct like this to hand so I wouldn’t have made a mess of it.
There are lots of bubbles in the world economy, the real question is whether they are interlinked and how much leverage is hidden behind them. I suspect that hidden leveraged debt not as bad as it was in 2007 – some lessons were learned (although this in turn can be problematic, as it has led in some economies in the EU to a lack of investment). But I think the real threat is of a domino of individually ‘manageable’ problems running out of control. The likeliest trigger in the short term would look to be a collapse in the AI bubble hitting at just the wrong time in the overall US investment cycle (i.e. exacerbating a cyclical downturn). The UK looks particularly vulnerable to a sterling crisis and I don’t think Japan is out if its immediate currency problems either. China’s attempt to export its way out of deflation is having a chilling effect on investment right across Asia – and there are multiple property bubbles across that region, from South Korea to Malaysia, all ripe for popping.
What might cause a collapse of the AI bubble, you might ask;
Google is warning it’s 1.8 billion gmail users of a new cyber threat posed by AI. It’s called an indirect prompt injection.
From Men’s Journal;
And;
The guys responsible for this want to spend a trillion dollars to make it more powerful.
W8 is the flashy end of London’s prime market. There are sub sectors of this postcode W8 8 for instance where the average selling price of property over the last 10 years has not only plummeted in real terms its fell by double figure percentage in nominal terms. Wealth based on property here is being destroyed. This effect, particularly for apartments, is now felt across London and increasingly the South East of England. This is resulting in much reduced sales in these areas, with a glut of overpriced property on the market.
The national stats showing continued small nominal growth in prices are heavily influenced by the continuing churn in the lower priced areas in the North and Midlands. This effect will soon be replaced by the overhang from the South.
I see a similar situation across metro areas in the US, with the sun belt leading and as yet the North East retaining value.
The UK economy is to a large extent dependent on the health of the housing market. Its an interesting dilemma for the BoE.
“the properties in question will, at some point in the not too distant future, become unsaleable, either because of changes in the market demand for property within the commercial sector, or because flood risks will mean that the buildings in question will be uninsurable”
Since Richard Murphy addresses the real-estate situation in Great-Britain, I would like to point out that there are other factors that are leading to buildings there becoming unsellable. A lingering one, which is not much discussed any longer, is the very large number of buildings that are unsafe with respect to fire hazards — most notably because they are clad in highly flammable material, as tragically demonstrated by the Grenfell Tower disaster.
Owners are faced with massive insurance premiums (when they are lucky enough to get insurance policies for their now officially dangerous buildings), huge transformation bills (when they can get competent personnel in time to perform such transformations), and can of course not sell anything (because banks do not offer mortgage to prospective buyers for flats in such unsafe buildings).
One more factor to those enumerated by Murphy increasing the pressure on the British real-estate market…
Will not happen for simple reason Central banks will spend unlimited money to save the market. Same thing in Europe central bank, you dont hear anything about Greece debt after Europe central bank guarantee their loan what was 36 billion when the crash happen now their debt is 160 billion but no one complain about their debt, in my opinion this is even against EU rule to guarantee Greece loan.
You assume competence. As much as I was bitterly opposed to their policy decisions, Geithner, Paulson and Bernanke did a very good job of a financial system rescue after having failed to get sufficiently in front of the compounding problems.
There is no one even remotely at their level on Team Trump. Bessant is a failed hedige who even at his peak ran only a small shop. He knows the functional equivalent of bupkis about bank and payment system plumbing. The trading/market acumen of the New York Fed got a big downgrade due to the shift from using open market operations to manage short term rates to the unwarranted subsidy of paying interest on reserves. The 2017 repo panic was a direct result of Fed incompetence, as in not being able to manage liquidity in its new pet system in a tightening cycle.
And lets not forget that Trump is in a rush to replace FED board members. His only requirement will be that they are yes men and women. Expect even more competence erosion.
The political situation was radically different too. Every race was more or less set in stone, and Bush was out in front of cameras and made McCain show up too instead of letting him try a campaign stunt.
Well, the number “160” you mention is not billions but the percentage of GDP the total debt represents. It was 129% in 2010, when the Big Ones in the EU shouted crisis. At some point near 2020 the public debt to GDP shot to around 200%. Only after the inflation of 2021-2023 the percentage fell to around 160%.
Also Greece was obliged to have primary surplus budgets of around 3% and more. This is why Greece has practically flatlined ever since. The average GDP growth stands at about 2% on average for the last 10 years. A rate that has not compensated the 27% fall in the period 2009-2014.
Real Estate? How last Century!
“We” now face crypto!
An asset for the entire Millenium! Become a billionaire … soon. Very soon. Not a Ponzi. Absolutely not.
Anyone using their ‘coin’ as collateral?
Run!
Please get a grip.
1. The leverage in real estate ginormously exceeds that in crypto
2. The failures of FTX and Binance, despite their size, not even remotely being systemic events says that crypto is not important enough to the critically important payments system to pose a systemic threat.
More generally, big falls in asset markets do not generate crises unless leveraged back to major payments systems participants. See the 1987 crash and the dot-com collapse as proof.
Yves,
Doesn’t the increased intertwining of stable coins and treasury debt, and potential leverage based on that, make for a more systemically dangerous situation than during the FTX fiasco, with greater likelihood of contagion?
If not, why not? Genuine question, thanks…
What makes banking work is access to liquidity in order to make payments for its customers. You take out a car loan and the bank pays the car dealership. You buy groceries and use your debit card and the bank pays your receipt. A bank is continuously borrowing and paying… some of that “paying” is tied to loans and some of that paying is just depositors using their account balances.
As long as the bank can keep making its payments it can keep operating, even if it is technically from an accounting perspective insolvent. Lehman Brother’s didn’t go to bankruptcy because it was known to be insolvent, but because it couldn’t make it’s payments and it was not a US regulated depository institution which would have had access to unlimited payment funds via the FED Discount window.
Anyhow, the key item is that in a PAYMENTS financial crisis something occurs that causes INTER-BANK liquidity to evaporate faster than regulatory authorities can replace with CONFIDENCE.
Any financial asset class can experience a significant decline in value and it can cause all sorts of economic harm if it damages the banking sector sufficiently to reduce or stop productive lending. But relevant to damaging the payment system you need a situation that freezes inter-bank lending. Those are real OH-Sh*t moments. Trying to understand where those systemic weak points are requires knowing a lot of the ins and outs of a lot of different financial assets and their derivative products. Theoretically it’s the job of financial regulators to flesh out those potential concerns and shut them down before a full-blown payment crisis happens.
As Yves and many NC commentators have inferred… a popping asset bubble is in our future, but it might not be a remake of Lehman Brothers. That said, there is sufficient gray space and regulators long asleep at the wheel that there COULD be a systemic payment/liquidity problem that is not well understood just waiting to create an inter-bank liquidity crisis. The question then becomes what pops the bubble and what might triggers the liquidity crisis if such factors are in play.
A significant difference between 2008 and now is the level of inflation. On one hand it allows debt values, such as housing, to be inflated away without nominal asset price reductions. On the other hand, it restricts the ability of the Fed to drop interest rates.
I think that the US economy is heading for a major shock in 2026, coinciding with Trumps ability to replace Powell. He will appoint a Fed chair that is amenable to dropping interest rates, which will further fuel real inflation and asset prices.
The main underlying problem remains that the economy has become fully bifurcated. Top 20% or so are responsible for 70 or 80% of consumer spending, so the economy is increasingly oriented to serve that market segment exclusively. This is a big driver of asset price inflation, because the rest of the economy has been cannibalized to fuel unsustainable income growth for the top of the economic ladder.
While this model is unsustainable long-term, I think it still has some legs short-term, because of the situation we are in. Dropping interest rates will provide a short-term boost to the system while making the long term consequences more severe, on par with this administration’s MO.
Until recently it was pretty much all young men into things crypto, and sooner or later the market for packaged air will collapse, and that’s the perfect demographic of an unruly mob, in my mind.
AirBnB’s and the like didn’t exist in the 2008 imbrogliou, and add an odd wrinkle to things financial in real estate, as there aren’t any real ties to the area aside from making money off a community.
From Yves’ description, you could call today’s economy (like that of 2008) the Arbitrage Economy.
More and more is “double-pledged” as collateral — the original collateral from hedge funds, the hedge-fund valuation, and on down the Ponzi line.
Sorry to be a stickler. Hedge funds trade nearly entirely in liquid securities and those are subject to strict limits on rehypothecation per SEC rules. Prime brokers also became more restrictive after the Lehman collapse.
The leverage on leverage is in private equity, private debt, and moving into crypto. But crypto may not blow back much to the payment system.
as i have glided through my myriad tasks this morning, a new word “rehypothecation” has been repeating itself over and over in me head…so,lol…thanks for that.
(of course, on track 2, and simultaneously, and for some damned reason, Paranoid Android has been on a loop for at least 2 weeks. unsure what to make of that…and it kinda bothered me enough that i dug into various analysis of that song on reddit, etc….will report back if anything actionable rears its head)
Thanks for your analysis at the start of the post, Yves. It’s not just real estate–it was that plus high leverage made opaque through derivatives.
But, right on cue, Bloomberg reports we now have PE CV-squared funds…continuation funds of continuation funds?
https://www.bloomberg.com/news/articles/2025-08-13/private-equity-continuation-vehicles-become-cv-squared-after-growth?srnd=homepage-americas
Yves, what is your take on Warren Mosler statement (a generally strong belief in MMT circle) “There is no financial crisis so deep that a sufficiently large fiscal adjustment cannot deal with it.”, also nicknamed “Mosler’s law”.
Thank You!
Of course what Mosler says is true: money exists only in our minds.
Socially, we agree to what it takes to balance assets and liabilities across innumerable interlocking balance sheets. Such social constructs have included debtors prisons, bayonet confiscations of gold, systemic financial collapses into deflation, and of course recently bailouts.
In our system the Fed/Treasury makes money, if they make enough of it they can solve any balance sheet problem. The issue, and issue Yves takes above (7:19 AM) with Paulson, Geithner and Bernanke is not that they could save the system by bailing it out, it is who’s interests the bailouts were structured to serve that displaced millions of home owners while enriching the criminals who created the crisis in the first place. A moral hazard of such YUGE proportions it’s now delivered us Trump, twice.
Thanks for your comment jsn.
I should have qualified my question a bit better, Mosler “law” seems to include any type of financial crisis, including the ones with a heavy deflationary overhang which can be extremely destructive and difficult to deal with. I guess the government can always buy stuff from the private sector and destroy it in the most extreme cases.
Yes, per MMT so long is the government is only using its own money it can do with it anything the society has the capability to do. Money in this sense is just a structured incentive.
What the government should do is the policy question and a truly democratic society would look for both just and effective solutions, rather than simple effective solutions for those with power/money.
Mr Bessent was on Fox today apparently saying that the US will consider allied assets inside the US as its own, to be used for the revitalisation of the US economy. So not only is any gold you ever sent to the US for safe keeping gone but those bonds you bought and kept at the New York Fed are gone as well.
On top of the export taxes being brought in, this points the way to capital taxes as well, in other words the trapdoor is going to shut. This points to a very different crisis to 2008, one involving the international financial system and who is going to bail that out under existing arrangements ? All the IMF, World Bank and SDR’s gone as well ? Perhaps they really are playing for all the marbles.