Altered Carbon, Neo-Feudalism, Redistribution and the Nice Life

“Neo-feudalism” is one of those terms that sounds vaguely profound while doing very little actual analytical work. It’s typically deployed to gesture at everything people dislike about capitalism: inequality of income and asset ownership, dynastic wealth … the creeping suspicion that effort does not reliably map to outcomes.

When people talk about “neo-feudalism”, I sometimes think what they really have in mind is something like the world of Altered Carbon. Not because anyone believes we are about to upload our consciousness into cortical stacks, but because it captures (albeit grotesquely) the fear: wealth that compounds forever, elites that never die, a social order that hardens into permanence.

As caricatures go, it’s very effective. Elegant, even. As a guide to policy however, it has some fairly big gaps.

Let’s start with accumulation.

The horror scenario assumes that wealth is infinitely durable and politically untouchable. But in reality, it’s quite fragile. It withers away through inflation, bad decisions, competition, taxation, and (most reliably) heirs who are not as competent as their parents. Preserving capital across generations is hard work, growing it is even harder. The idea that capitalism would naturally produce immortal rentiers if it weren’t constantly beaten back by the state works well as clickbait, but it’s not especially accurate.

The more interesting blind spot though is moral, rather than mechanical.

The asymmetry.

We are constantly told that inheritance is dangerous because it corrodes ambition. Leave your children too much money, and they may fail to strive, compound skills, achieve their potential. They’ll fail to become what they otherwise could have been. This argument is treated as self-evident, so much so that it rarely needs defending.

What is not explained though is why this logic apparently stops at the boundary of the welfare state. If unearned money rots incentives, why does that only apply when the money comes from parents rather than the state? If the receipt of assets weakens ambition, why is long-term, low-conditionality income support assumed to be motivationally benign, or at the very least neutral?

Trust funds: create moral hazard.
Benefits: create dignity.

One dulls the drive, the other simply “supports”. It’s not obvious why incentives would care about provenance.

This asymmetry is not just theoretical. It plays out, rather messily, in real families.

The family.

Consider a familiar archetype. A cousin with a respectable, cushy job in IT at a company widely regarded as “nice to work for”. He has been there north of twenty years. Not because he could not earn more elsewhere, but because he chose not to. He values predictability, he likes the people, “they treat their staff well”, the hours fit nicely around his family life and interests. And so for over twenty years he has been — consciously or otherwise — prioritising his mental and physical health over income.

His wife, trained as a teacher, stopped working after their only child was born and never returned to the labour market, not when the child went back to school, and not even later when she went off to university. This too was a choice. Entirely legitimate, but a choice nonetheless.

They recently completed a self-build house, funded to a significant extent by my mother. Their daughter’s first year of university expenses (so far) was also covered.

For context, I grew up comparatively poor, raised largely by a single parent in a community that was … not deprived. We could afford shoes and clothes, we could not afford foreign school trips, which I sat out while others went. My mother inherited money when I was in my early teens. The inheritance itself was not unexpected, the amount of it very much was. Some of it paid for my education, leaving me without student debt, which was helpful.

I was also told to expect nothing further and to make my own way in life. I did not seriously dispute this. I accepted (if not always enthusiastically) the necessity of earning my keep. Until recently.

Enter stage left: my cousins and their entirely unobjectionable offspring. They’re nice people, however much it pains me to admit it. In fairness, the girl is both clever and a sweet kid.

The explanation offered to me for the gifting was simple: life had been harder for them. They married young, lived on a single income, and have not been on a foreign holiday in years. This house was a massive project for them. Meanwhile, I apparently have too many foreign holidays, which is taken as evidence of excess rather than effort.

What goes unmentioned is the other side of the ledger.

My holidays sit on top of 60-hour weeks, give or take. A career optimised for income and resilience rather than comfort, because in London a low-stress working environment is not a luxury universally available. Niceness, on its own, does not compound.

Had the money been given to me ten years ago, it would have made a genuine difference. I had to push through some real shit to get to where I am. It wasn’t given to me then. Instead, it was given recently to those who were judged to “need” it more, because they had spent decades prioritising their own wellbeing over money, and therefore ended up with less of it.

This is not altruism so much as retroactive insurance.

It’s a de facto underwriting of past lifestyle choices at the expense of those who made different ones. The signal is pretty clear: effort that converts into resilience is taken for granted; effort that converts into dependency is rewarded.

Exactly the same moral logic underpins much of modern state redistribution, only scaled up and layered with bureaucracy. Resources flow not to those who made difficult trade-offs early, but to those who deferred them the longest. The criterion is not effort, foresight, or contribution, but demonstrated need — even when that need is a predictable outcome of past choices. At the family level it is called “kindness”; at the state level it is called “justice”. In both cases, the incentive signal is the same.

The moral hazard of niceness

Choosing a comfortable “nice” employer, predictable hours, and low stress is increasingly treated not merely as a lifestyle preference, but as a moral good. Choosing geographic mobility, income optimisation and overtime compensated via a “bonus” scheme is treated rather ambivalently. It’s tolerated, but ultimately regarded as unnecessary or even faintly suspect.

Tax policy reinforces this judgement. Those who optimise for output are taxed on the assumption that their effort is discretionary, that they could slow down, stop, or relocate without meaningful consequence. Income shortfalls resulting from comfort-first choices are treated as structural and therefore deserving of relief.

But guess what, only one of these paths reliably generates a surplus. Modern economies depend disproportionately on people willing to accept volatility, sacrifice time, and prioritise earning capacity over short-term comfort for significant stretches of their working lives. Yet the system treats these choices are optional, while insulating their opposite from consequences.

This is not system neutrality. This is an incentive structure.

When the returns to effort are socialised downward, while the costs of opting out are socialised upward, it is only rational to adjust your behaviour. All this political hand-wringing about people choosing to work less, retire early, and how to get the “lazy” retirees back to work? That’s not a “phenomenon”, that’s a very predictable and rational erosion of the surplus that the system presumes would always be there, playing out right in front of us.

Also: this is not about condemning comfort. It is about recognising that a society which systematically rewards the right to stress-free existence over the production of value should not be surprised when value creation slows.

This is neo-feudalism’s missing middle.

The system does not protect the super-rich — that’s a myth. The super-rich need no protection; they can look after themselves pretty well. Instead, the system relies on a broad, productive middle that is expected to carry both the fiscal burden and the moral burden: taxed because it is efficient to do so, and guilt-tripped into accepting it because someone else is always worse off.

The irony is that Altered Carbon is not really about inequality per se. It is about ossification. The tragedy of Harlan’s World is not that some people are rich, but that nothing moves: elites do not fall, outsiders do not rise, effort has no dependable pathway to change outcomes.

If we genuinely feared neo-feudalism, we would obsess less about redistribution and more about mobility: whether effort is rewarded at the margin, whether competence beats entitlement, and whether the rules assume that some people will always produce while others are just fragile enough to required permanent bubblewrap.

When effort stops buying meaningful optionality, feudal systems are not overthrown. They stabilise.

For avoidance of doubt: this rant was inspired by, and is downstream of, the comments section on this Monevator article. Any confusion is my fault rather than theirs.

Don’t Panic, Part III: The Planet-Sized Brain Error of Treating Bonds Like Stocks

The frozen glaciers of the moon of Jaglan Beta

(For those who find acronyms and jargon mildly offensive to the soul, a glossary awaits at the bottom of the post. Proceed with caution.)

Investors are compensated for carrying risk, which in fixed income is mostly interest rate risk, credit risk and liquidity risk.

Interest rate risk is structurally slow moving in normal conditions, as central bank policy rates are largely steered by macro indicators and communicated via forward guidance and a team of highly caffeinated individuals who ruthlessly wordsmith every comma and semi-colon in the MPC minutes until all nine members can claim plausible deniability. Of course the market, an equally caffeinated entity, prices in future expectations of these moves immediately, sending yields a-swinging on the back of unexpected data, a whiff of an economic shock, or just because a central bank said they might do something upsetting in the future (see: taper tantrum).

But even though the policy rate moves slowly and the market’s expectations of it thrash around, interest rate risk isn’t normally amplified by the design of the products retail investors hold. There’s no hidden “duration trapdoor” lurking in the documentation — no gated redemptions, no off-the-run drift. Interest rate exposure is mostly transparent: you either own duration or you don’t. So pick your duration (long, short, or none at all) and accept the consequences like an adult.

And that’s all I have to say on the matter … for now. We’ll get to structured products, callable features, negative convexity, and extension risk later. Maybe. If I can be bothered. (This series is beginning to feel a bit like an airport security queue, if I’m honest.)

Anyway…

Credit risk, mercifully, comes with ratings agencies (yes yes, imperfect, but imperfect triage is better than none).

Liquidity risk comes with nothing. No standardised metric, no neat alphanumeric grade, no Moody’s “L-5 (Negative Outlook)”. The illiquidity premium is slippery because it depends on market conditions; it is not a stable attribute of an asset class but a situational one.

The market that everyone thinks trades but actually doesn’t

Equity investors inhabit a world of tick-by-tick quotes, tight spreads, and oceans of exchange liquidity. It looks like a high-pressure fire hose: constant flow, instant execution, anonymous matching.

Credit1 investors inhabit something closer to a Regency salon, where transactions occur by appointment only, intermediated by a dealer whose actual inventory is a rumour, and who may or may not be out to lunch.

And yet retail investors buy fixed-income funds as if they were a sensible, liquid, low-volatility counterpart to equities. This illusion survives because:

  • bond prices are reported infrequently,
  • funds often mark these assets to models,
  • … and no one thinks too hard about what happens when markets actually move.

Let’s ruin that for you.

Credit market liquidity is not bad, it’s structurally fictional outside of benign conditions.

Why is credit market liquidity fragile? Short answer: the market was designed that way

First, bond markets never centralised. Equities marched onto exchanges, adopted continuous order books, and developed transparent price discovery. Bond trading, however, remained predominantly OTC: a negotiation between you, a dealer, and the dealer’s balance sheet.

Issuers insist on continuously flooding the world with new CUSIPs / ISINs. A single corporate issuer routinely has a dozen of outstanding bonds, each with its own coupon, maturity, call option, covenant pattern, and issuance date. Multiply that across countries and sectors and you get a combinatorial swamp. No exchange is ever going to maintain a deep central book for 40,000+ line items that trade three times a month.

Second, dealers historically acted as warehouses, holding inventory so they could buy from you or sell to you. Nowadays that capacity is … diminished.

And third, the largest bond investors are insurance companies and pension funds — institutions that match long-term liabilities with long-term assets and consider “trading” a slightly vulgar pastime. Their financial accounts contain long disclosure sections that say they might sell if needed; in reality they rarely need to. Their balance sheets behave like a minor black hole where bonds enter, never to be seen again. They’re the Disaster Squid of Kakrafoon Kappa2: a creature that ingests any fixed income instrument within reach returning only paperwork, and even that on a five-year delay.

Add passive trackers, private wealth accounts and some niche hedge funds that do value strategies or whatever, and the slice of the market that actually trades bonds (retail funds, RV/macro hedge funds, a few active managers) is surprisingly narrow. The illusion of liquidity rests on a sliver of the market while the rest of the paper lurks in a buy-and-hold vortex.

Because of these reasons, the credit trading ecosystem is largely:

  • bilateral,
  • over-the-counter,
  • inventory-driven,
  • dependent on dealer balance sheets that Basel III shrank to the width of a Ryanair seat3.

Most corporate bonds and munis trade like antiques: occasionally, in small size, and often only after someone contacts a specialist dealer who last saw the item six months ago4.

A quick tour of the dysfunction

A. Market structure

There are thousands of individual bonds per market segment (IG, HY, EM):

  • The vast majority trade less than once a day.
  • Many trade less than once a week.
  • Some trade monthly, if the stars align.

Major exchanges (LSE, LuxSE, Euronext, SGX, HKEX, JPX) list bonds5, but do not trade them in any meaningful size, hence even exchange-listed bonds mainly trade OTC. The public quote is a decorative suggestion, not a firm price. Often not even indicative6.

B. Price transparency

UK/EU: minimal. Dealer inventories are private. This lack of pre-trade transparency means nobody knows the full range of available securities or the best price. For thinly traded or HY bonds, transparency is even worse.

Dealers quote different prices depending on client type, size, relationship, and whether they’ve had their coffee yet. Indicative electronic quotes disintegrate in volatile conditions; the real price only emerges after a round of apologetic exchanges.

Regulators have implicitly acknowledged this. Recent FCA and ESMA reforms have reduced pre-trade transparency obligations for bond systematic internalisers7. The dream of “stock-like transparency for bonds” quietly died in committee.

US: TRACE helps, but only with post-trade transparency, and it doesn’t solve liquidity.

C. Dealer inventories (wool in a hot wash)

Dealer inventories today are roughly a third of their 2007 size. Less warehousing capacity means less willingness to facilitate trades, which translates to more spread widening in stress.

Post-GFC rules (Basel III, leverage ratios, the Volcker Rule in the US; the UK/EU equivalents via CRR/CRD, the Fundamental Review of the Trading Book) made it expensive for banks to hold large inventories. Market-making capacity is now something banks ration.

In equities, a lit exchange book internalises liquidity: buyers meet sellers. In credit, the dealer’s balance sheet is the liquidity. When that balance sheet shrinks, the “market” shrinks with it8.

D. Indexation (a rant)

Passive credit is one of the most successful bad ideas in modern finance9.

Indices create the illusion of a liquid, representative, standardised market. If there’s an index for it, then the market must be deep and safe. Sadly, this does not apply to corporate credit indices10.

For example, many corporate bond indices are market-cap-weighted11. Meaning: they allocate more to the most indebted issuers. This is the opposite of an equity index, where the market cap reflects the value of the company. A sort of anti-meritocracy.

Sector weights is another clusterfuck. In equity indices, the sector split reflects a cross-section of economic activity. In credit, it tells you who issues debt, not who creates economic value. Here’s a shorthand:

  • Short-term AA: banks
  • 5-year IG: banks and consumer defensive
  • 10-year IG: banks and utilities with some industrials
  • HY: telecoms, property, and anybody mid-M&A12
  • EM: sovereigns spending the week arguing with the IMF

And then there’s the small matter of valuation. Most corporate bond indices are stuffed with off-the-run paper that barely trades, which means the index isn’t a reflection of market prices so much as whichever pricing model the index provider felt comfortable with that day. This is nothing like equity indices, which at least pretend to track actual trades in actual markets. In credit, the “price” of the index is usually a theoretical construct, so any tracking error between a passive bond fund and its benchmark is just one model disagreeing with another model about an imaginary number.

The sheer number of competing indices (Bloomberg, ICE, LSEG, Markit, Solactive…) is itself a sign of fragmentation. There is no FTSE All-Bond equivalent of the FTSE All-Share. There cannot be.

E. Settlement Lags

Corporate bond settlement: T+2 in the UK and EU, T+1 in the US, Canada, Mexico and a few others.

Hard currency EM bonds: mostly T+2, sometimes T+3, sometimes “at your earliest possible convenience”13. Local currency EM bonds: T+2 to T+whenever, depending on the jurisdiction. Timelines also differ depending on whether the trade is on- or off-market14.

Bank loans: T+10 to T+20 after KYC / assignment paperwork is complete and if the administrative agent is awake.

And this is the market that retail investors access through daily dealing funds and ETFs15.

Sleep well.

Why this matters: open-ended funds are not designed for what they hold

The retail market is stuffed with daily-dealing credit funds. Daily dealing is a nice fantasy until you realise:

  • investors can redeem instantly,
  • portfolio managers cannot sell instantly,
  • bonds cannot settle instantly.

This gap creates the liquidity mismatch, the original sin of fixed-income retail products. In normal times the mismatch is camouflaged because flows are modest.

In stress, it becomes the entire story:

  1. NAV drops slightly
  2. Investors panic (“something something recession fears”)
  3. Fund sells its most liquid assets first
  4. Portfolio quality deteriorates
  5. NAV drops more
  6. More redemptions
  7. Eventually, forced selling of illiquid paper
  8. Dealers step back
  9. Fund gates, swing prices, or sells to vultures at whatever bid comes back with a pulse
  10. Newspaper headline: “Investors shocked to find bond fund contained… bonds”

HY and EM funds can and do gate, usually after selling down every remotely liquid line item. Private credit and bank-loan funds use “interval fund” structures in the US and LTAF wrapper in the UK to slow redemptions to (usually) quarterly windows16. Structured credit is particularly vulnerable: bid lists evaporate, dealers refuse balance-sheet usage, and tranches go “no-bid” regardless of model value.

Even IG bonds can go no-bid when:

  • dealers pull balance sheet,
  • bid–asks widen to “don’t even bother”,
  • real money accounts (pensions / insurers) are full,
  • prop desks are banned or balance-sheet constrained.

If nobody is structurally permitted to catch the knife, the knife falls until someone is permitted.

Why ETFs are not a free lunch either

Credit ETFs are clever pieces of plumbing. They rely on authorised participants (APs) creating/redeeming units in exchange for a basket of bonds or collateral. ETFs can trade even when the underlying cannot.

But in stress:

  • APs step back,
  • the ETF becomes the only price discovery venue,
  • prices gap below stale NAVs,
  • retail investors complain that the ETF is “broken”.

In reality, the ETF is the only thing working. It’s the bond market that’s broken.

The ETF becomes a price discovery venue for bonds too illiquid to trade, which is a way of saying the ETF is pretending to be a market.

ETF prices incorporate the illiquidity discount the NAV refuses to admit. March 2020 proved this: ETF “discounts” were accurate marks. Secondary liquidity is not real liquidity. It is a shadow puppet of what the market would be if the market existed.

The risks

Let’s list some of the structural risks:

A. Liquidity Cascades: Selling begets selling. Funds dump what they can sell (usually the good stuff: senior IG, big issue sizes). That suppresses good assets first. The junk remains until later. Europe, with its bank-centric system, greater fragmentation and more constrained dealer balance sheets, is more vulnerable than the US.

B. Redemption Mismatch: Forced selling or gating. Pick your poison.

C. Herd Behaviour: Flows all go the same way across:

  • IG credit
  • HY credit
  • EM debt
  • bank loans
  • CLO liabilities
  • AT1s (remember Credit Suisse?)

Fundamentals do not matter when flows set the price.

D. Funding Risk: If a fund uses derivatives (CDS, TRS) or repo or financed positions, it inherits counterparty and margin-call dynamics. In stressed credit, margins widen at the exact moment liquidity disappears.

E. Pricing Risk: Model-based marks become fiction when markets stop trading. Evaluated pricing vendors extrapolate from the last trade, a few quotes, and the tremor in the analyst’s voice. NAV becomes a suggestion, not a fact.

  • Funds rely on evaluated pricing when there are no trades
  • This smooths volatility
  • Smooth volatility leads investors perceive stability
  • Perceived stability leads confidence
  • That’s the classic liquidity illusion

When your fund is quoting a price but nothing has traded in two days, the fund is basically marking to mythology.

Practical implications

ETFs: Best price discovery and intraday exit, but expect discounts in stress. Useful for DM government debt or diversified IG. Bad for deep HY or EM local.

Direct bonds: Fine if you can hold to maturity. Bad if you need liquidity in a hurry17.

Bond ladders: Useful for predictability, but in my view only practical in government debt.

Active credit funds: Only sensible if they publish explicit liquidity buckets (e.g., % of portfolio liquid within 1 day, 5 days, 30 days). In practice: very few do. Also: high fees, and you’re on a chain gang with a bunch of people you haven’t met who might lack cojones in stressed markets, start redeeming, thus driving the fund into a liquidity spiral18.

Private credit / loans: These belong in closed-end or interval structures. Anything else is a structural joke.

The government bonds exception

Government bonds are different. The major ones (USTs, Bunds, Gilts, JGBs, OATs) actually trade electronically, in size, with repo markets behind them. Dealer balance sheet still matters, but the market structure is recognisably modern.

But even here: fragmentation matters. Bunds are not the same as BTPs. Gilts are not the same as GGBs. Liquidity is not universal, it is sovereign-specific.

Still, for core DM sovereigns, open-ended funds basically work.

Mandate features (or: choose your poison and read the prospectus)

A brief public-service announcement regarding mandate funds. Some come with auto-eject clauses. These are contractual tripwires that say: if a bond falls out of the index or drops below Investment Grade, we sell it. Immediately. No questions, no timeouts.

In theory, this keeps the portfolio pure. In practice, it means forced selling at precisely the worst moment, because downgrades don’t tend to happen during boom times when liquidity is sloshing around. They happen in panics, when everyone else with the same mandate is also stampeding for the exit. Prices gap, liquidity evaporates, and the fund dutifully yeets the bond into the void19, locking in a loss.

Then there’s the opposite flavour: no auto-eject at all. Very zen. Very “long-term investor”. Very “we don’t let the rating agencies tell us what to do”. This lets the fund hold fallen angels, workout cases and the occasional financial crime scene long after everyone else has sold. Sometimes that’s deliberate (the manager genuinely thinks there’s value). Other times it’s just inertia.

The result can be a misleading form of stability: the fund reports a suspiciously smooth NAV while the underlying bonds haven’t traded in weeks. Until someone tries to redeem20.

Either way, pick your poison: Forced selling that crystallises losses now, or a possibility of a slow-motion rot dressed up as calm. Only one way to know which you’re buying: read the prospectus. Or don’t, and enjoy the surprise.

The mildly alarming effect of interest rates and inflation on liquidity

Interest rate moves don’t just change valuations; they change how easy it is to get anything done. A sharp policy rise tends to make dealers widen quotes and reduce balance sheet usage.

Inflation adds another layer. Markets don’t wait for the central bank to make a decision; they begin adjusting the second the data suggests that a decision might be required. Yields shift, curves reshape, flows rotate, and liquidity drains from the asset classes investors no longer wish to be seen holding. Dealers respond in the usual way: by doing slightly less of everything until the dust settles.

Rate cuts can also be awkward but in a different direction (everyone wants the same duration at the same time) but the effect is similar: the underlying market steps back to let prices sort themselves out.

Liquidity / depth of the market problems aren’t always about sellers vanishing. Sometimes it’s the buyers who get trapped because the supply disappears. This happens in those brief windows when central banks hint at mercy and everyone decides yields probably have peaked. Suddenly every bond that looked like toxic waste a month ago becomes “strategic”. Owners stop offering, dealers widen their offers or remove them completely, trading screens show prices that exist in theory but not in size.

This is still a liquidity failure, just the pleasant, upward version21. ETFs flip from “discount” stories to “premium” stories, and retail investors believe they have discovered a structural anomaly rather than the absence of sellers. This happened in April 2020, when we found out with some relief that not everyone was going to die and also (bonus feature) central banks were printing money again.

When inflation, growth or policy expectations change, spreads in HY, EM, and anything even slightly exotic can reprice faster than the underlying market can catch up. Market depth in these asset classes is problematic whether rates are going up or down. In a risk-off moment, bids vanish; in a risk-on moment, supply does. In both cases the mechanism is the same: market participants try to anticipate the central bank, and the market accommodates by being unavailable. Liquidity depends on someone taking the other side of the trade. When nobody volunteers, the market moves until someone is forced to.

TL;DR

Credit markets are structurally illiquid because they were built that way. OTC trading, fragmented issuance, balance-sheet-constrained dealers, and the surge of indexation all mean that liquidity is a fair-weather feature.

Retail investors accessing credit via daily-dealing funds or ETFs are often unaware of:

  • the liquidity mismatch,
  • the mechanics of stress periods, and
  • the speed with which the bid can evaporate.

This is not an obscure footnote, but the central truth of the entire asset class, which governs everything from yield spreads to fund design to systemic risk.

For anyone hoping to retire early and live off their investments, the greatest risk in markets isn’t derivatives or securitisation, but the belief that you can sell anything whenever you want at a predictable price. Credit markets regularly remind us that we cannot.

Part IV will be about currency hedging. Spoiler: I don’t hedge.

Notes:

  1. When I talk about the more interesting parts of “credit markets,” I don’t usually mean DM government bonds, supranationals, or provincial/state paper from the likes of Canada and Australia. Except, obviously, when I do. For example: parts of the DM government markets in smaller or more fiscally expressive EU members can trade with less depth than the latest Apple or Microsoft issue. Most of the time, DM sovereign markets are far deeper and more stable than corporate or EM credit. Unless the sovereign in question is having an episode, or happens to live uncomfortably close to Russia during one of its periodic reminders that international borders are, in its view, more of a suggestion than a rule.
    If this feels unhelpfully vague, congratulations: you are beginning to appreciate the structural fragmentation of the credit universe. The taxonomy is less “equity sectors” and more “subspecies of beetles” — numerous, overlapping, oddly regional, and frequently indistinguishable without specialist equipment. ↩︎
  2. Please note: The Hitchhiker’s Guide to the Galaxy did not, in fact, feature a Disaster Squid. But blimey if I wasn’t going to call a pension fund one anyway. I’m not passing up this opportunity. ↩︎
  3. For non-Europeans: this is slightly narrower than an A4 sheet of paper and roughly as comfortable. ↩︎
  4. Electronic bond trading exists. Platforms like Tradeweb/MarketAxess plus all-to-all RFQ networks have made the most liquid corporate bonds easier to trade. But the vast majority of issues still trade OTC, episodically, and with dealer balance sheets acting as the source of liquidity, especially outside the big benchmark, on-the-run, or money market names. Electronic execution has improved workflow far more than it has changed market microstructure. ↩︎
  5. Primarily to provide transparency, visibility, and regulatory oversight. It helps issuers attract a wider investor base. For example, some institutional investors have mandates or internal rules that only allow them to invest in securities that are publicly listed, or the portfolio has to have a certain percentage of securities that are publicly listed on a recognised exchange. In some jurisdictions listing can provide regulatory or tax advantages for the issuer or the investor. For example, in the UK, “quoted eurobonds” do not require the issuer to deduct tax on interest at source. ↩︎
  6. This makes audit time, when auditors try to price the holdings, extra fun. The usual process involves several rounds of explanations, beginning with very polite, and ending with a junior auditor sitting beside an irritated trader, sending RFQs and muttering about ruining every dealer relationship they have left. ↩︎
  7. Investment firms that execute client orders on their own account, doing so on a frequent, systematic, and substantial basis outside of a regulated market, multilateral trading facility (MTF), or organised trading facility (OTF). ↩︎
  8. Research from the Fed (e.g., Adrian et al.) and ICMA repeatedly confirms this. Turnover ratios decline when dealer leverage is constrained. In Europe (an even more bank-dominated financial system) this effect is even sharper. Capital-market intermediation never replaced dealer intermediation, even after MiFID II. During stress, dealers don’t tighten their belts; they vanish into the handlebars. Spreads widen, sizes fall, and quotes turn “indicative”. ↩︎
  9. Second only to volatility-targeting strategies, which maintain the bold tradition of selling low, buying high, and calling it a “discipline”. ↩︎
  10. Govies are mostly fine though. ↩︎
  11. Market cap” for bonds = par outstanding × clean price. ↩︎
  12. There is no better predictor of future distress than “transformational acquisition narrative”. ↩︎
  13. Usually interpreted as “this week, probably”. ↩︎
  14. For example, in South Africa, JSE trades settle between T+1 and T+3 and off-market trades will settle anywhere up to T+6. ↩︎
  15. Industry jargon for this is “democratisation”, stemming from Greek dēmokratia (δημοκρατία), composed of demos (δῆμος), meaning “people”, “the common people”, or “citizens” and kratos (κράτος), meaning “power”, “rule”, “authority”, or “strength”. It implies that a market or an asset class that was previously accessible only to large institutional investors (the “elite”) is now available to a broader range of retail investors (the “people”). In the context of an ETF, the operational mechanics and ultimate authority within the fund structure (such as the creation/redemption process handled by APs and the management decisions made by fund managers) rests with the same institutions that the term “elite” was meant to refer to. Also see: irony. ↩︎
  16. These structures have notice periods, redemptions windows and typically a limit (cap) on how much can be redeemed during any one window. You simply cannot have continuous liquidity when the asset class is designed to be traded like a Persian rug. For anyone pretending otherwise, here are a couple of examples. ↩︎
  17. This excludes govies, these you can usually sell. Unless it’s the beginning of COVID and everyone suddenly remembers what happened to Kate Winslet in Contagion. ↩︎
  18. It’s the classic “first-mover advantage” problem in open-ended funds. In a market downturn, some investors, anticipating potential losses or redemption gates, try to redeem their shares first. To meet these redemptions, the fund manager must sell assets, often the most liquid ones first. Forcing sales in illiquid credit markets during stress drives prices down further, penalising remaining investors. ↩︎
  19. Not an official industry term (yet), but widely understood. ↩︎
  20. NB: These mandates are fine for very short-term credit. Even if the paper is downgraded or falls out of the index for whatever reason, the fund manager will simply hold it to its (imminent) maturity. ↩︎
  21. Mechanically, it’s the same problem: there’s no two-way market, just everyone trying to transact in the same direction at once. ↩︎

Jargon Buster

All-to-All Trading — A supposedly democratic innovation where everyone can trade with everyone else. In theory, Tinder for bonds: you broadcast your interest and wait for a match. In practice, the same five dealers still do all the heavy lifting, while everyone else lurks anonymously hoping not to get shown a silly price.

AP — Authorised Participant. The ETF middleman who creates and redeems units. Present in good times. Less so in bad times.

AT1 — Additional Tier 1 capital. Contingent convertible bank capital instruments. The ones that occasionally get written down to zero over a weekend.

Benchmark Issues — Big, glossy, billboard-sized corporate bonds (Apple, Microsoft, JPMorgan) issued in sufficiently large size so that dealers can’t misplace them. While on-the-run, they are liquid, famous and widely held. Very unrepresentative of the rest of the credit market. Eventually they go off-the-run, at which point they become far more representative.

Big Issues — Aka “jumbo”. A bond issued in such a large size that dealers can trade it without breaking into a stress rash. Generally $10bn+ (the kind of line where you can move $25m and not ruin someone’s day). In contrast, the typical $250m “boutique artisanal bond” trades biannually at best, usually when someone retires or dies.

BTPs — Buoni del Tesoro Poliennali. Italian government bonds. Famous for generous yields, sporadic volatility, and the market’s long-running attempt to price the difference between “political drama” and “political crisis”.

Bunds — German Federal Government Bonds. Eurozone safe asset cosplay. The reference rate for Europe and the instrument through which Germany expresses its deep commitment to low yields.

CDS — Credit Default Swap. An insurance contract on credit. The financial equivalent of selling someone your umbrella in a drought.

CLO — Collateralised Loan Obligation. A securitisation of leveraged loans. Comes in tranches ranging from “fine” to “probably fine” to “pray for carry”.

CRR / CRD — Capital Requirements Regulation / Directive. EU/UK bank capital rules derived from Basel III. Also known as “why your dealer no longer holds inventory”.

CUSIP — Committee on Uniform Securities Identification Procedures. The US security identifier. One issuer can spawn dozens of CUSIPs, all subtly different, ensuring you must pay attention to what you’re actually buying rather than what you think you’re buying.

DM — Developed Markets. Countries with functioning capital markets, central banks, and usually fewer than three currency crises per decade.

EM — Emerging Markets. Countries with higher yields, lower liquidity, and a statistically non-zero probability of telling the IMF to get lost.

ETF — Exchange-Traded Fund. A structure that pretends bonds are liquid by letting you trade the wrapper instead.

FCA — Financial Conduct Authority. UK regulator. Occasionally optimistic about transparency reforms; rarely rewarded. A successor to the old FSA, which lost supervision of all the interesting systemic stuff when that was relocated to the Bank of England after the GFC, leaving the FCA with the leftovers and a clipboard. I don’t mean to cast aspersions but the term “rejects” has occasionally been mentioned.

GFC — Global Financial Crisis. The time in 2007–2009 when credit markets went on strike.

GGBs — Greek Government Bonds. Once the poster child for “things that probably shouldn’t yield 3%”.

Gilts — UK Government Bonds. Sterling sovereign bonds. Named for their “gilt-edged” history, which is quaint given the last decade of LDI chaos, mini-budgets, and “market turbulence”.

G-SIB — Global Systemically Important Bank. A bank so large and interconnected that regulators have formally acknowledged its failure would be inconvenient. Comes with higher capital requirements, closer supervision, and a general expectation that it should please try not to explode.

HY — High Yield. Bonds issued by companies with higher credit risk. Also known as “speculative grade” or “why your fund gated”.

IG — Investment Grade. Bonds blessed by the rating agencies as unlikely to default. Some do anyway.

IMF —International Monetary Fund. The global lender of last resort for countries that have, shall we say, misjudged their macroeconomic vibe. Dispenses balance of payments support, policy advice, and the occasional lecture about fiscal discipline.

ISIN — International Securities Identification Number. A 12-character code that uniquely identifies a security, so that everyone can at least agree which bond they’re failing to trade. Begins with a country code, ends with a check digit, and does absolutely nothing to guarantee liquidity or pricing transparency.

JGB — Japanese Government Bond. The global home of yield curve control, where bonds trade more on the Bank of Japan’s mood than on anything resembling supply and demand.

LSEG — London Stock Exchange Group. Also an index provider. One of many determined to convince you that corporate bond indices mean something.

MiFID II — Markets in Financial Instruments Directive II. A sprawling EU/UK regulation. Gave us lots of reporting, little liquidity, and enduring industry trauma.

MPC — Monetary Policy Committee. The Bank of England rate-setting panel that meets to decide whether to raise, cut, or do nothing with interest rates, armed with data, forecasts, and an institutional fear of surprising anyone on purpose.

Munis — Municipal bonds. Debt issued by states, cities, and assorted local governments in the U.S. and Canada to fund things like schools, sewers, and occasionally a stadium nobody remembers asking for. Often come with tax perks; sometimes with plot twists. The UK, Australia and much of Europe don’t really “do” munis, preferring a more dignified arrangement in which the national / state government borrows on everyone’s behalf, thus sparing local councils the awkwardness of discovering what their credit spread actually is.

NAV — Net Asset Value. The fund’s official price. May or may not correspond to any level at which the underlying bonds could actually be sold.

OAT — Obligations Assimilables du Trésor. French government bonds. Considered safe, liquid, and generally civilised (apart from the occasional political hiccup when France remembers it is France).

OTR — On-the-run. The bond equivalent of “the new season just dropped”. The latest issue, not yet subsumed into the pension fund / insurer singularity. Dealers care deeply about it before quietly ghosting it six months later. OTR bonds have actual prices and actual trading. Everything older is “off-the-run”.

Off-the-run — Bonds that were once exciting but have since been adopted by pensions and insurers, never to be seen again. Probably stored in a terrarium on Mars. They trade occasionally, in apologetic sizes, at prices derived largely from astrology. Dealers quote them the way an estate agent quotes a haunted basement flat. The market pretends these bonds exist only because the index says they do, which is why your passive credit ETF ends up holding stacks of this dust. If it can get them. Otherwise it samples and swallows the tracking error… which is absurd, because if most of the index hardly ever trades, the index itself is marked to fiction.

OTC — Over-the-Counter. The world of bilateral trading, negotiated prices, and dealers who may or may not respond to your RFQs.

Prop desks — Proprietary trading desks. The bit of the bank that trades its own money rather than clients’, often for fun, profit, or quietly testing how much risk the regulators will let them take before someone notices.

RFQ — Request for Quote. The polite way of asking dealers, “How much would you like to sell me this thing for, or how much would you pay to take it off my hands?” Think of it as shouting into the void and waiting for a baritone voice to respond with numbers you may or may not like. 

SI — Systematic Internaliser. Firms required to quote prices under certain conditions. A regime recently “reformed” to lighten transparency. (Please contain your shock.)

T+1 / T+2 / T+3 / etc. — Settlement timelines. The number of business days between trade date and settlement. “T+whenever” applies to certain EMs and bank loans.

TRS — Total Return Swap. A derivative that lets you receive the total return of a bond or index without owning it. Useful. Until margins widen.

UST — US Treasury. Dollar-denominated sovereign bonds issued by the U.S. government. The global risk-free benchmark, except during debt-ceiling season, at which point they become a political hostage with a coupon.

Don’t Panic, Part II: The Liquidity Illusion

From thence the tide of fortune left their shore,
And ebb’d much faster than it flow’d before
.

A surprising number of UCITS fund investors appear to believe the acronym stands for Unlimited Cash In Troubled Situations. It does not, though one imagines the regulators would have much quieter lives if it did.

But before we get to UCITS, we need to talk about liquidity: the thing everyone thinks they have until the moment everyone tries to use it all at once. Apologies in advance, feel free to skip to TL;DR1.

Price volatility shouts. Liquidity clears its throat quietly

After Part I, we’ve established that the GFC was not chiefly about delinquent American homeowners, but about global liquidity evaporating with the kind of urgency normally seen when everyone within badge-swipe distance detects sunshine on a Thursday2 afternoon and evacuates to the nearest pub.

This matters, because most private investors learn to think of risk as price moves. Prices go up and down; it’s intuitive and plots nicely on a chart. Unfortunately, markets, and credit markets especially3, don’t care about charts. They care whether someone, somewhere, will take the other side of your trade.

Investment committees talk about liquidity a lot, usually while pretending they understood it long before the PRA’s latest “Dear CFO” letter arrived, because it determines:

  • whether you can exit,
  • at what price you exit,
  • … and whether that exit results in your risk officer turning up “for a quick chat”, having just re-run his VaR with “updated assumptions”4.

What credit market liquidity actually is (and what it isn’t)

Liquidity is not:

  • tight bid-offer spreads,
  • lots of trading volume last week,
  • the mystical protection conferred by the letters U-C-I-T-S.

Liquidity is:

  • the ability to transact at a vaguely sensible price when everyone else also wants to,
  • a function of dealer balance sheet capacity,
  • a variable that collapses faster than credibility at a crypto conference.

A helpful model:

Credit market liquidity is a public good generated by dealer balance sheets. Dealer balance sheets today are not what they used to be.

Years of capital requirements, “enhanced” risk management, and stern regulatory architecture5 mean that the market-making capacity that made pre-2008 credit feel deceptively liquid… is gone. Mostly replaced by optimism, hopeful glances and smaller tickets.

The European retail liquidity blind spot

Retail investors in the UK and EU love UCITS in the same way people love tap water: it’s ubiquitous, regulated, and assumed to be safe. But UCITS regulates portfolio construction and disclosure6, not the day-to-day tradability of whatever is inside the wrapper.

This means:

  • A corporate bond fund dealing daily is not necessarily holding assets that trade daily.
  • A high-yield fund promising daily liquidity is performing a sort of financial alchemy that only works if flows remain gentle and predictable.

European regulation doesn’t conjure liquidity; it only requires that the fund believes it can meet redemptions.

The market disagrees strongly during stress7.

Below are a few illustrative case studies. They vary in regulatory species — UCITS, NURS, and the odd free-range item — but they all share the same theme: at one time or another requiring investor or funding liquidity that refused to cooperate, citing non-receipt of the memo.

Exhibit A: The great UK property fund gating championships

Let’s begin with a classic: open-ended UK commercial property funds8. Over the last seven years, these funds have gated more often than the Ashtead level crossing.

Brexit? Gated.
Covid? Gated.
Interest rate spike? Gated.

The underlying assets are immovable, expensive to transact, and valued according to appraisal schedules that move with the speed of a rural parish council. They are fundamentally incompatible with daily dealing.

Yet open-ended commercial property funds were9 very popular “low-volatility income” products. Low volatility is easy when you aren’t marked to market between valuation reports. Problems begin when more than three people try to redeem at once.

This is the kind of liquidity mismatch that reminds you the fund may be “open-ended”, but the buildings definitely aren’t.

Exhibit B: Bond ETFs in March 2020

March 2020 gave us a nice demonstration of how modern liquidity works.

Corporate bond ETFs — IG and HY — traded at discounts of 5–10% to their NAVs. Retail investors were shocked. Institutional investors were… relieved to finally see a price.

The ETF price was real. The NAV was an assortment of indicative quotes from dealers who had absolutely no intention of honouring them10. This was not dysfunction. It was the first honest mark the market had seen that week.

Did this lead to some margin calls? Yes11.

Did anyone enjoy explaining that the NAV had been … aspirational? Less so.

But it was a useful reminder that ETFs provide liquidity when the market has some, and price discovery when it does not.

Exhibit C: The UK gilt crisis (LDI meltdown), or “how to break a market in 48 hours”

In 2022, LDI strategies — the pride of British pension engineering beloved by pension schemes and consultants alike — imploded when long gilt yields jumped 50–100bps in two trading sessions following the Truss / Kwarteng “fiscal event”.

Why did LDI strategies blow up? Because: they’re leveraged, collateralised duration exposures.

LDI funds use interest rate swaps and gilt repos to create synthetic long duration. This requires collateral. When gilt yields rise:

  • the value of the liability hedge drops,
  • collateral requirements spike,
  • fund must sell gilts to meet margin calls,
  • selling pushes yields up further,
  • which triggers… more margin calls;
  • repeat until the Bank of England switches on the sprinklers12.

This was not a solvency crisis. It was a funding liquidity spiral: a phrase that makes everyone involved insist it was a “technical issue” while discreetly updating their risk frameworks and asking no further questions. The Bank of England intervened not to bail anyone out but to prevent forced selling from turning the entire gilt curve into a liquidity crater.

And so the UK got its very own two-day system-wide shock. This was probably not the historic milestone anyone wanted, but one takes what one can get.

Exhibit D: Woodford, the canonical mismatch

I can’t discuss liquidity without mentioning Woodford.

The Woodford fund began as a UK equity income product, then wandered into less liquid terrain with the confidence of someone who believed daily dealing was a minor clerical detail13.

It was not.

Different asset class, same principle: selling the illusion of liquidity while owning assets that do not cooperate. It’s an all-time classic, and rest assured it will be staged again.

If your underlying assets require strategic patience, your fund should too. Retail funds with illiquid holdings rarely follow this rule.

The liquidity mismatch hall of fame

A brief catalogue of products whose liquidity promises come with long footnotes:

  • High-Yield Bond Funds: HY bonds trade infrequently14. HY bond funds deal daily. This is a structural impossibility disguised as a consumer product.
  • Emerging Market Debt Funds15: Liquidity ranges from “thin” to “ring back after the next election”.
  • Bank Loan Funds (a US speciality): Because nothing says “daily liquidity” like assets that take 10–20 days to settle. Investor redemption settlement? ETFs: T+116, Mutual funds: legally within a week, emotionally whenever the back office stops crying. Best of luck.
  • Synthetic ETFs: Collateral baskets can become illiquid, and swap counterparties can pull lines in stress. Otherwise they’re entirely satisfactory, according to the relevant Vogon paperwork, Form PF-38/B(7) Assessment of Collateral Integrity During Minor or Major Universal Catastrophes, which has never been wrong.
  • Absolute Return Multi-Strategy Funds: Often full of OTC derivatives and Level 3 marks wearing Level 2 makeup.
  • UK Open-Ended Property Funds: See above: champions emeritus.
  • Anything with the words “Income Fund” and “Unconstrained”: Often translates to “illiquid stuff with a daily dealing promise”; constrained by the inconvenient need to sell things.

Liquidity tools (swing pricing, anti-dilution levies) exist, but if you’re seeing them used, the horse has not merely bolted; it has stolen the trailer and joined a rival stable.

As an aside:
Before anyone misinterprets me: this isn’t a tirade against the products I’ve listed. Far from it. I’d love for the retail crowd to have access to as many investment vehicles as possible, and to make their own choices17. The exasperation (mine, but let’s call it collective) is aimed at the mismatch in attention spans. People will meticulously research a £200 hotel room for a night or a £400 rental SUV they’ll drive for a week, but will barely glance at the prospectus of a financial product they are considering parking thousands of pounds in. Most tick the box “I’ve been provided with a KID” with approximately the same diligence they apply when agreeing to Google’s ever-changing privacy policy.

As self-directed investors, it’s our job to do the homework — properly. That means actually reading and understanding the products we invest in: what’s inside the wrapper, what “normal range of conditions” really means, the exact ways the product might throw a small tantrum, whether it’s compatible with our personal financial sitcom, and whether the allocation fits within the portfolio as a whole.

As a DIY investor, if you can’t fully understand the product, it’s not appropriate for you. End of story.

(Rant over).

The mechanics: why liquidity vanishes exactly when you want it

Liquidity behaves like the market’s emotional support animal: very friendly in good times, nowhere to be found in bad. It is the first derivative of credit18. A simplistic way of looking at it:

Liquidity is hypersensitive to changes in credit availability. When credit expands quickly, liquidity is abundant even if absolute credit levels are low. When credit contracts quickly, liquidity evaporates even if credit levels are high.

Most awkwardly though, liquidity can also collapse even if credit isn’t shrinking; it might simply not be expanding enough to absorb redemptions, flows, or whatever panic the day has chosen to serve.

Liquidity is also pro cyclical, which is why markets overshoot both on the way up and down.

When lenders are confident:

  • Balance sheets grow
  • Banks extend loans
  • Dealers provide market-making
  • Margin requirements ease
  • Collateral is treated as high-quality
  • Leverage is encouraged

Liquidity everywhere. Markets cheerful.

When credit tightens:

  • Funding becomes scarcer.
  • Lenders demand higher quality collateral.
  • Leverage must be unwound.
  • Dealers downgrade their quotes from firm to indicative to ‘spiritual guidance’.

Liquidity gone. Prices fall, yields jump, everyone suddenly remembers they “meant to reduce risk anyway”. Markets become… educational.

Liquidity does not decline gradually. It vanishes abruptly, usually after saying it “just needs a moment”.

Why this matters (again, sorry)

Liquidity is not a property of assets; it is a property of market conditions. In credit markets, short term price movements often reflect changes in (interest rates, yes) and liquidity, not fundamentals.

Today:

  • dealers have less balance sheet capacity19 (Basel III)
  • ETFs and index flows concentrate liquidity in the same trades
  • many funds own things that settle T+whenever
  • volatility-targeted strategies trade pro-cyclically
  • everyone uses the same models and therefore panics in unison
  • … and your ability to find a buyer remains inversely correlated with your enthusiasm for finding one.

Liquidity is not a constant property. It is a conditional one: it exists when you don’t need it, and evaporates when you do.

This is why, in calm markets, bond ETFs appear to add liquidity, and in stressed markets they simply provide the price nobody wants to admit is real. The difference is roughly the width of Canary Wharf.

Risk models underestimate the speed of liquidity contraction because they assume continuous markets. Many investing blowups happen because people mistake a credit-dependent liquidity regime for a stable one20.

TL;DR

If you understand liquidity, you can understand nearly every modern market accident.

Liquidity remains invisible right up until it isn’t, at which point it becomes the only topic anyone emails about for a fortnight. Professionals fret about it. Private investors largely ignore it. Neither group enjoys dealing with it.

And now we’re ready for Part III, where we explain why the quiet illiquidity of credit markets makes the equity market look like Glastonbury — crowded, loud, messy but functioning. Credit is more like the car park afterwards: everyone wants to leave, nobody moves.

Notes:

  1. I once tried to explain this at a dinner party. I am yet to be invited back. ↩︎
  2. It used to be Fridays, but nowadays everyone works from home on Fridays, especially if it’s sunny, or cloudy, or windy, or if the dog looks a bit sad. ↩︎
  3. We’ll get to that later. Liquidity sagas, like Marvel movies, demand an extended universe. ↩︎
  4. This is roughly as welcome as finding the Head of Compliance waiting at your desk holding a printout, which is the financial services equivalent of your dentist starting a sentence with “Huh. That’s interesting”. ↩︎
  5. The “stern architecture” refers to the post-GFC regime (Basel III, capital charges, leverage ratios, LCR/NSFR, stress tests, resolution plans and such).
    The logic for implementing it was sound: some pre-crisis dealers were running enormous trading inventories on modest capital, levering up with the restraint of a labrador in a sausage factory. So regulators tightened everything. Hard. We now have stricter capital buffers, less leverage, and supervisors who react to the phrase “inventory build-up” the way a Victorian aunt reacts to the word “ankles”.
    Result: banks are better capitalised, less interconnected, etc, The next crisis will at least be slower and more puzzle-shaped.
    The trade-off: dealers now have far less balance sheet available for actual market-making, which is why modern credit liquidity sometimes resembles ordering a pint at 10:59pm — technically possible, practically improbable. ↩︎
  6. UCITS protects you from poor disclosure and concentration risk. It does not magically liquefy high-yield Canadian bonds on a Thursday afternoon. ↩︎
  7. Which is why UCITS liquidity tools include gating, suspensions, notice extensions, redemptions in kind, and side pockets. Such fun. ↩︎
  8. The 2020 NURS rule changes, the 2023 LTAF regime, and the recent migration of direct property managers toward more hybrid models all suggest that this particular flavour of mismatch is, if not being solved, then at least being tidied up. Which admittedly makes their inclusion here slightly opportunistic. Consider it a narrative convenience, and let’s not allow progress to spoil a perfectly good story. ↩︎
  9. For balance (yes, it happens): the aggregate NAV of the UK open-ended property funds has fallen by more than half since 2020. Many providers have either exited the market or are in the process of doing so, partly due to the fundamentals, partly because the NURS wrapper for direct property really doesn’t make sense, and partly (I suspect) due to the FCA’s habit of launching prolonged and intermittent consultations on potential reforms in the sector with the zeal of a springer spaniel chasing a squirrel. ↩︎
  10. ETF NAVs, during stress, can also rely on evaluated pricing models whose methodology sections are longer than heaven and duller than hell, and (some might say) equally as fictional. ↩︎
  11. This did in fact trigger collateral adjustments for leveraged credit funds, total-return swap strategies, and dealers running ETF-arb books. The people who blew up were not ETF holders, but those who incorrectly assumed NAVs were real prices. Spoiler: they were not. ↩︎
  12. LDI margin calls are pro-cyclical because swap BPV (DV01, for the Yanks) is linear but collateral flows behave like a light switch: off, off, off, THEN VERY MUCH ON. ↩︎
  13. Woodford Equity Income was marketed as a UK equity income fund. Then it drifted into small-cap, unlisted, and property exposures. The result: a daily-dealing fund with weekly-to-never-selling assets. ↩︎
  14. The average investment-grade corporate bond trades roughly once a week, give or take. High-yield: often less. The technical term for this is “fine, absolutely fine, don’t worry”. ↩︎
  15. Worth bearing in mind (for the lovers of even more exotic exposures): EM local-currency bonds are priced using a combination of quotes, spreadsheets from 2009, and the prevailing mood in Mexico City. But don’t worry — Mexicans are a very happy nation. ↩︎
  16. Yes, Authorized Participants can create or redeem ETF units in exchange for the underlying loans (or a representative basket). But the mechanism depends on APs being both willing and able to step in, which they are never obligated to do, and historically do not do when the market is on fire. ETF underlying liquidity is only as good as AP appetite, which vanishes at about the same speed as dealer balance sheet capacity. ↩︎
  17. Please don’t confuse this series with the investment equivalent of a middle-class parent demanding a public inquiry into why electric bikes weren’t banned last year, shortly before their offspring earned a provisional nomination for a Darwin Award by discovering a brand new way to collide with gravity. ↩︎
  18. Credit = the willingness of investors to extend funding or accept risk. Liquidity = how easily assets can be exchanged for cash without affecting price. Confusing the two = bad things happen. ↩︎
  19. If you’re thinking large asset managers have relationships with dealers — yes, they do. As they did in 2008. The relationship is cordial until the dealer’s risk officer logs in. ↩︎
  20. If you find this level of detail excessive, I regret to inform you that there is an appendix. ↩︎

Appendix: A Brief History of Liquidity Panic

(Compiled for readers who enjoy history and schadenfreude.)

Liquidity panics follow a reassuringly familiar pattern:

  1. Someone uses too much leverage,
  2. Someone else provides too little liquidity,
  3. Something important stops trading,
  4. A central bank appears, looking tired.

The details vary, the emotional arc doesn’t. Below is a non-exhaustive catalogue of such moments.

I. 1907: The Original “Funding Liquidity Spiral” (Before It Was Fashionable)

The panic of 1907 featured everything modern crises do: opaque securities, insufficient collateral, overconfident intermediaries, and a liquidity shock triggered by a rumour nobody could fully remember afterwards.

J.P. Morgan famously solved the problem by locking bank CEOs in his library until they agreed to provide liquidity. To this day, this remains the most successful use of executive sequestration in financial history. Modern regulators have replaced this approach with “moral suasion”, a concept about as effective as asking a cat to please reconsider.

II. 1929–1933: The Great Depression, Featuring Margin Calls (Before Email)

The defining innovation of this period was that leverage had escaped into the wild, and everyone was allowed to borrow 90% of the cost of shares because what could possibly go wrong.

When prices fell, brokers called in loans, forcing sales, driving prices further down, etc. Contemporary observers didn’t call this a “liquidity death spiral”. They called it “Tuesday”.

III. 1966: The First Modern Money-Market Squeeze

Bond dealers discovered that funding long-term Treasuries with overnight money had drawbacks. The Federal Reserve intervened using what would now be described as “macroprudential improvisation” and then tried not to talk about it for the next decade.

This event is mainly remembered by people who read vintage Fed minutes, a hobby widely regarded as a cry for help.

IV. 1970: Penn Central’s Commercial Paper “Surprise“

Penn Central, a railroad with the financial stability of a wet towel, defaulted on its commercial paper. This caused the entire CP market to freeze. Because: investors realised that short-term corporate IOUs were only as good as the corporations issuing them.

The Fed stepped in.
The disco era began.
These events may or may not be related.

V. 1987: “Portfolio Insurance” and Other Pranks That Got Out of Hand

Portfolio insurance allowed investors to replicate a put option by selling futures as markets fell. This worked well until falling markets required everyone to sell at the same time, which turned out to be … suboptimal.

Liquidity vanished, prices collapsed, and markets spent a day behaving like a flight simulator running on low battery. This was also when people discovered that volatility was a path-dependent experience.

VI. 1998: LTCM and the Birth of “Orderly Liquidation”

Long-Term Capital Management experienced a brief misunderstanding between:

  • how liquid it thought its trades were;

and

  • how liquid they actually were when Russia defaulted.

Rule of thumb: convergence trades converge until they don’t.

The Fed arranged a rescue, though “rescue” probably understates the level of swearing in the room. I’m only guessing. But: “firms disagreed over how much each firm should contribute to a rescue package and could not commit to such an effort on such short notice” and then later “with no other solution in sight, the talks resumed with more haste inside the New York Fed”? Imagine this in colloquial with expletives.

LTCM became mandatory reading for anyone who has ever said, “These correlations are stable”.

VII. 2007–2008: Subprime, Structured Credit, the Universe and Everything

It began with small declines in housing but escalated into a global liquidity crisis when investors realised that:

  • CDOs were complicated,
  • CDO-squareds were very complicated, and
  • SIVs were maintained in ring binders, apparently.

Then repo markets froze, money markets froze, term funding froze, and regulators developed a sudden interest in bank capital ratios.

Central banks responded with liquidity programmes of increasing creativity and decreasing dignity. The phrase “emergency facility” entered common usage, eventually becoming its own asset class.

VIII. 2012: The European Sovereign Debt Crisis

Peripheral sovereign spreads blew out when markets decided to investigate whether Greece really was of was not worthy of an A rating. Liquidity vanished from government bond markets that had previously been described as “deep and safe”, which turned out to be only partly true.

The ECB came up with “whatever it takes”, which remains the most effective monetary policy tool ever delivered in under six syllables.

IX. 2019: The American Repo Palaver (Technical Term)

A combination of heavy Treasury issuance, corporate tax payments, and balance sheet constraints caused repo rates to spike to 10%. NB: this is not normal.

The Fed intervened again. Everyone pretended not to be surprised.

X. March 2020: The Everything Liquidity Crisis

Markets in 2020 demonstrated a universal truth: when everyone wants cash at the same time, no asset is liquid, even if it is the U.S. Treasuries. (Aka: in stress, all correlations go to 1).

Bid-ask spreads blew out, bond ETFs traded at discounts, CTAs, RV funds, and dealers all hit the same liquidity exits simultaneously. The Fed reintroduced facilities last used in late Jurassic.

The Bank of England restarted gilt purchases at industrial scale, activated the CCFF, and generally made it clear that anyone attempting to price gilts without its permission was welcome to stop immediately.

The ECB relaxed collateral rules and began vacuuming up bonds with the resolve of an institution that had no intention of letting some paltry mass-depopulation event spoil its afternoon.

Money markets were stabilised.
Loo roll supplies were not.

XI. 2022: The UK Gilt Crisis (LDI’s 48-Hour Existential Event)

LDI strategies (leveraged, collateralised, and sensitive to rate shocks) discovered that gilts can move 50–100bps in two days. This surprised everyone, including the gilts.

Collateral calls escalated, funds sold gilts to meet margin, selling drove the yields up, which created more collateral calls, and so forth, until the Bank of England conducted a “decisive intervention”.

This is the only instance in modern finance where pension funds nearly broke the sovereign debt market of a G7 nation. Cool.

XII. The Future (e.g., Next Year)

Future liquidity panics will follow the same format:

  • someone assumes liquidity is constant,
  • someone else discovers it is not,
  • models disagree with reality,
  • a central bank arrives with a firehose.

Prime candidates include:

  • private credit vehicles funded with flighty cash,
  • algorithmic strategies that all read the same signals,
  • whatever new structure marketing teams invent after deciding “daily liquidity” sounds much better than “twice a year with between 10 and 30 days’ notice”.

Conclusion

Liquidity panics are not anomalies; they are recurring reminders that:

  • markets are networks of confidence,
  • confidence is fragile,
  • liquidity is the first thing to disappear when frightened.

The history is incomplete, but then so is every liquidity model ever built.

The Unbearable Lightness of Working

Chancellor Rachel Reeves has recently unveiled a new contribution to political taxonomy: a “working person” now appears to mean “anyone earning below the 40% tax bracket”.

Finally, an objective measure. No more of this outdated fixation on hours, effort, or output. The government offers a cleaner test: is your income sufficiently taxable? Then sit down. You’re not working. You’re a Schrödinger’s labourer: simultaneously alive yet professionally inert, until observed by HMRC.

Useful! And it explains so much.

Now that’s what we call leisure

We’re having building work done at home. (Obviously. It’s the middle-class version of therapy: expensive, disruptive, and guaranteed to reveal structural issues you didn’t want to think about.)

The builder received one-third upfront. It was a radiant week in October, the kind that makes poets reach for metaphors and contractors reach for… time off. He appeared for three days, did four hours per visit (a total of 12 hours, for those who enjoy arithmetic). A whole 1.5-day week.

Then it rained. Apparently rain makes building work impossible (except for invoicing, which thrives in all meteorological conditions). Tools down at 3pm sharp, sometimes 2:58 pm if the drizzle is particularly morale-damaging. The project timeline has extended into something that feels more geological than contractual.

Meanwhile, my own contribution to the national project involves roughly 60 hours a week engaging in activities the Chancellor suggests do not constitute “work,” on account of my marginal tax rate. The builder, who works about the same hours as a French civil servant on strike, is firmly in the “working person” category. I, with my four-monitor prison setup, am not.

This is fine. Absolutely fine.

Chancellor’s CV (of changing definitions)

Reeves demonstrated aptitude for redefining “work” early in her career. Famously, she once reclassified her own postgraduate course as “working” at the Bank of England: an admirable early display of statistical creativity, which neatly foreshadows her current productivity strategy: improve national efficiency (output per unit of input) by redefining 15% of the tax base as “non-workers,” thereby reducing the denominator.

Elegant. Brutal. ONS will love it.

Guardian vs Gladwell

Just as this semantic revolution unfolds, The Guardian has helpfully explained that success is now largely an accident. Something between cosmic dice-rolling and a clerical error at birth. Gladwell’s Outliers can apparently be pulped: practice, focus, effort, the 10,000 hours — the whole Protestant work ethic — have been debunked and replaced by “vibes plus stochastic fortune”.

It’s a refreshing worldview. Everyone is absolved of responsibility except the people funding the Treasury. A perfect intellectual climate for redefining “work” around tax bands.

A public service glossary (revised for the new era)

  • Balanced Approach – More tax. Balanced chiefly on your wallet.
  • Brexit’s Impact on the Economy – See: “global factors”, “challenging headwinds”, and “please stop asking”.
  • Broad Shoulders – Anyone who has ever received a tax bill that induces mild nausea.
  • Challenging Headwinds – Meteorological phenomenon occurring whenever GDP numbers flatline. Typically used in place of the more accurate “we did something extremely stupid and would quite like you to stop bringing it up”.
  • Civic Duty – Paying more tax with a serene facial expression.
  • Consultation – A ceremony in which stakeholders are notified of the decision.
  • Economic Competence – A tone used when unveiling numbers with a short shelf life.
  • Efficiency Savings – Spending cuts, but with plausible deniability.
  • Fairness – Policy that benefits whoever is speaking.
  • Fair and Sustainable Settlement – More tax but with nodding.
  • Fiscal Responsibility – Announcing something you said last week you’d never do.
  • Global Factors – Everything except raising trade barriers against your largest trading partner.
  • Growth Plan – Tax rises with a motivational title.
  • Hard Work – Traditionally hours of labour; now a marginal rate threshold.
  • Honesty – A speech genre.
  • Industrial Policy – Nostalgia packaged as ambition.
  • Industrial Strategy – A document asserting Britain will become a world leader in something it currently does not make.
  • Integrity – Doing the right thing, once polling confirms it is safe.
  • Investment – Spending.
  • Listening Exercise – A consultation that produced an inconvenient answer.
  • Meritocracy – A belief system in which hard work is widely celebrated but officially irrelevant. See: “cosmic windfall,” “planetary alignment,” “favourable birth coordinates.” Effort is decorative; its outcomes are taxable.
  • Middle Class – Small enough to squeeze; too large to pity.
  • Modernising the State – Launching an app that doesn’t work.
  • National Conversation – An announcement.
  • NHS – Britain’s most cherished institution: part religion, part museum exhibit. An employer large enough to qualify as its own weather system. Known internationally for its cutting-edge queueing technology.
  • Office for National Statistics – The nation’s calculator, permanently updating its firmware.
  • Personal Responsibility – The unique privilege of the middle classes to financially support themselves, the old, the young, the infirm, and approximately 0.85 of an officially designated “working person.” See also: “civic duty” and “broad shoulders”.
  • Professional Experience (Bank of England edition) – A postgraduate course completed within the same postal district.
  • Public Sector Pension – “Long-term affordability concerns” followed by stasis.
  • Public Sector Reform – Hiring a consultant to implement a dashboard.
  • Public-Private Partnership – Combining the agility of the public sector with the accountability of the private sector.
  • Rental Licence – A mandatory permit issued by councils, governed by rules so opaque they must be deciphered using folklore. Commonly forgotten by landlords across the political spectrum, later discovered during a routine administrative embarrassment.
  • Spending – Someone else’s investment.
  • Stakeholder Engagement – A series of meetings where everyone agrees the problem is “complex” and then goes home.
  • Tough Decisions – Announcing something unpopular in lieu of something else even more unpopular.
  • Vision for Britain – Four bullet points, three slogans, and a photo of a warehouse.
  • The Wealthy – Individuals politicians threaten but cannot locate.
  • Working Families – Individuals who procreated without the means of supporting the offspring.
  • Working Person – Someone earning below the 40% bracket. Hours optional. Output negotiable. Hi-vis recommended.

Don’t Panic, Part I: The Restaurant at the End of the Repo Market

The patio, for those keeping score, remains defiantly moss-coloured

There’s an entire genre of YouTube explainers purporting to reveal What Really Happened in 2008. They tend to feature the same set pieces: collapsing grey rectangles, swooping arrows, sepia-toned skyscrapers toppling like dominoes, and a narrator who sounds under contract to say “greed” every 15 seconds.

It’s a soothing morality tale. A bit Aesop, a bit Adam Smith, with a cameo from the Big Bad Banker. The little guy was wronged; the suits were venal; the universe issued a cosmic slap.

It’s tidy, emotionally satisfying. It’s also, unfortunately, mostly wrong.

Yes, underwriting standards were poor. Yes, some CDOs were engineered with the ethical rigour of a toddler stacking Jenga blocks. But the story of 2007–09 is not, at its heart, a tale about credit blowing up. It was about funding drying up — first slowly, then suddenly — until institutions that were solvent on paper became insolvent purely because nobody would roll their borrowing1.

If that seems abstract, recall the very British sight of patient, polite queues outside Northern Rock in 2007. Britain was the first country to experience a GFC-style bank run, not because the bank’s loans were uniquely dreadful, but because its funding model evaporated.

Liquidity runs the show. Credit problems turn up afterwards and claim they were there all along.

A global margin call

The U.S. lit the fuse. The short version is unglamorous: a large chunk of the financial system had become dependent on short-term wholesale funding (repo, ABCP2, interbank lines). Mortgage securities weren’t funded with “money”; they were funded with borrowed money, rolled daily3, secured against collateral whose value needed to remain non-controversial.

When subprime delinquencies ticked up, nothing detonated immediately. Credit losses were containable. The real trouble started when:

  • the housing bubble began to deflate4,
  • confidence in the models cracked,
  • repo lenders increased haircuts,
  • ABCP buyers quietly left the building,
  • dealers declined to take collateral at par5,
  • and the shadow banking system revealed it had the structural integrity of a towel left out in a gamma-ray storm.

Suddenly everyone tried to delever at the same time. That is called a run. A run in wholesale markets behaves exactly like a queue outside a bank branch. Only quieter and much faster.

The global financial crisis was, above all, a modern-day bank run in the wholesale funding markets6.

The irony: even within the subprime structures, most AAA tranches never defaulted7. They were money-good if you could hold them to maturity. But you couldn’t hold them unless someone kept lending you the money to do so.

Which, as it turned out, they did not.

Northern Rock: when your funding model malfunctions

Britain provided the world with the first unmissable visual: actual queues around the block.

Northern Rock was not a uniquely rubbish lender. Its mortgages were not that much worse than anyone else’s. What was unique was its near-total reliance on securitisation funding and wholesale markets.

When those markets froze — not because Northern Rock was toxic, but because investors no longer trusted anyone’s mortgage collateral — the bank couldn’t refinance its short-term borrowing.

This was liquidity risk at its purest (and its purest form is always surprisingly boring):

  • assets: long-term mortgages
  • liabilities: short-term wholesale funding
  • outcome: insolvency, not because the assets blew up, but because the cash ran out

Replace “Northern Rock” with Bear Stearns, Lehman, Dexia — the mechanics barely change. These weren’t institutions felled by credit Armageddon. They were felled because their repo counterparties declined to extend the courtesy of tomorrow8.

Britain wasn’t collateral damage. Britain was Exhibit A.

The European chapter: banks that funded in dollars, lent in euros, and prayed in Basel

Europe’s banks entered the crisis holding large quantities of U.S. mortgage paper, not because they adored American suburbs, but because these assets were capital-efficient under Basel rules. Senior tranches of RMBS, CMBS, and other structured products attracted lower risk weightings, allowing banks to lever up more efficiently than with domestic corporate loans, bonds, and, in some cases, even certain sovereign debt.

Many European banks funded themselves in short-term US dollars (because: cheap9) via commercial paper, interbank lines, and FX swaps. Much of their collateral and assets were denominated in USD, but they also extended loans or held exposures in euros (or other local currencies), creating a subtle FX and funding mismatch. FX swaps were supposed to bridge the gap — until nobody wanted to lend dollars at any price.

So, European banks had:

  • dollar assets,
  • dollar collateral,
  • dollar funding,
  • …and no dollar printing press10.

When the US dollar markets froze, the ECB had to tap the Fed for swap lines just to keep European institutions funded.

The financial crisis hit Europe with a slight delay, which fooled some people into believing Europe had ‘different problems‘. It didn’t. It had the same problem: wholesale funding evaporated, liquidity disappeared, and solvent balance sheets were rendered temporarily insolvent because the plumbing failed, not because the underlying loans were toxic.

The unified story: liquidity → fire sale → credit losses

The proper sequence is:

  1. U.S. housing downturn (sharp by historical standards, but not civilisation-ending) begins to nibble at the edges of mortgage performance.
  2. Subprime RMBS delinquencies rise in the pools that were, let’s say, optimistically assembled.
  3. Ratings agencies rediscover arithmetic, downgrading the thin slices of structured credit they’d previously rated with the enthusiasm of a golden retriever.
  4. Mandate-bound credit funds11 start forced selling because their prospectuses say they must.
  5. Confidence in the models evaporates. When nobody believes the correlations, nobody believes the prices. Nobody knows what will get downgraded next.
  6. Liquidity drains from structured credit, as dealers decline to make markets in assets they don’t know how to value.
  7. Repo and ABCP funding disappears. Because: nobody wants to lend against collateral whose price is now suspect.
  8. Firms dump assets to raise short term funding.
  9. Mark-to-market accounting does its worst. Forced sales trigger fire-sale marks → marks erode capital → capital erosion triggers more forced sales. Counterparties start demanding even more collateral.
  10. Wholesale funding seizes, as everyone takes one look at everyone else’s balance sheet and quietly backs away.
  11. Monetary conditions tighten outside of policy control, because shadow-banking pipes have frozen.
  12. The real economy finally buckles, as credit creation slows to a crawl.
  13. Actual credit losses arrive, fashionably late.
  14. Everyone declares it was obvious all along.

Most of the true losses were the result of the liquidity run, not the cause. The devastation came from leverage and loss of confidence, not the face value of bad mortgages.

The insanity of financial markets is that perfectly good assets can only stay ‘good’ if enough people pretend they can sell them at any time. Once that illusion breaks, the entire edifice wobbles.

What hindsight actually showed

By 2008, the market reached the “sell first, value later” stage. AAA non-agency RMBS and CMBS tranches — those originally yielding maybe 30–60 basis points over Treasuries — still throwing off cash on schedule — traded for 20–40 cents on the dollar12. To buy them, you needed:

  • capital (unlevered, patient, no margin calls),
  • a deep understanding of the collateral and the structure, and
  • a psychological profile typically associated with bomb-disposal units.

That excluded 95% of the market at the time.

The remaining 5% were clinically unafraid of things labelled ‘super senior”. They were a mix of specialised hedge funds, distressed-credit investors, and a few deep-pocketed institutions that didn’t depend on short-term funding.

  • The hedge-fund crowd (think Pine River and Anchorage) snapped up super-senior RMBS at 30–50 cents and later collected near-par payouts. Some even ran the vintage “long cash, short ABX” trade when the synthetic indices overshot to the downside.
  • Distressed funds like Apollo, Oaktree, Cerberus and friends launched vehicles to hoover up legacy non-agency RMBS, CMBS and the occasional CLO tranche from capital-constrained banks, in some cases earning 3x returns as the bonds paid down.
  • Insurance companies, blessed with the closest thing finance has to permanent investable capital, quietly accumulated senior paper at deep discounts and simply held it to maturity.

The tragedy (or irony) of 2008 was that the people who understood these structures best were also the ones least able to hold them. Dealers and banks knew their senior AAA rated paper was money-good, but mark-to-market losses and repo haircuts made it impossible to keep it on their balance sheets.

Regulatory and funding constraints forced the smart money to sell good assets to the patient money at insultingly low prices. A wealth transfer, disguised as “systemic deleveraging”.

The lessons

Gary Gorton and Andrew Metrick later labelled it correctly: a run on the shadow banking system. No viral infographic required; no villain in a suit; no morality tale. Just plumbing failing at scale.

The cartoon version flattens this into “subprime bad → everything worthless”. It’s comforting. It suggests that next time, all we need is better ethics and fewer liars.

But the real lesson is darker:

  • Liquidity is a shared hallucination13.
  • Confidence is reflexive.
  • Leverage makes solvency conditional on mood.

This does not play well on YouTube, so the cartoon version persists.

Why this still matters

Because the same structural features remain:

  • Daily dealing funds holding illiquid property or private credit.
  • Corporate bond UCITS promising liquidity the underlying assets cannot provide.
  • European banks still dominating bond market-making with diminished balance sheets.
  • LDI strategies that went from “boring actuarial plumbing” to “full-margin-call Broadway musical” in 48 hours in 2022.
  • ETFs that rely on authorised participants, whose balance sheets conveniently disappear during crises.

We have built a financial system that promises liquidity while knowing full well it can’t provide it during a panic.

Yes, banks today are better capitalised than they were in 2008. But credit market liquidity is worse and dealer balance sheet capacity is structurally lower. Capital helps with solvency, not with forced selling.

Retail investors still overwhelmingly focus on price volatility, because it’s the only risk anyone ever talks about, and because their app displays it in neon green. But in credit markets, price is rarely the thing that kills you. Liquidity is.

And liquidity is the first thing to vanish when the lights flicker.

At this stage in the Guide, the large, friendly letters would say DON’T PANIC. Finance, regrettably, does not come with a cover.

TL;DR

The core GFC lesson is simple:

Risk = needing liquidity when everyone else does.

Misunderstand that and you will misunderstand:

  • leveraged investing
  • bond ETFs
  • synthetic ETFs
  • currency hedging
  • property funds
  • LDI-style structures
  • anything containing a swap, forward, repo line, or “it tracks the index, we promise” marketing line

Which brings us neatly to Part II, where we’ll explore why liquidity risk resides in almost every investment structure the retail market considers normal.

Notes:

  1. Subprime was garbage, yes. But most of RMBS/CMBS in issue as of 15 September 2008 was not subprime, and even in subprime most AAA tranches didn’t default. In fact, as of 2013, principal-weighted cumulative loss rates for AAA-rated subprime securities were under 0.5%. Even garbage does not spontaneously become a global extinction event unless the funding dries up. ↩︎
  2. Asset-backed commercial paper. If you’re currently winding up for a lecture in the comments about me not defining acronyms… don’t bother. I am avoiding a date with my jet washer and a patio full of grime. This is procrastination, not the drafting of a finance textbook. ↩︎
  3. For the purists: yes, some repos were 30–90 days. Congratulations. But nobody funded a trillion-dollar balance sheet with one repo line. The typical expiry profile meant something was always rolling — daily, weekly, whenever the moon was in retrograde. Add ABCP, FX swaps and tri-party resets and the effective maturity of the whole edifice was “perpetually next Tuesday.” So yes, I’m using a bit of artistic licence here. Please direct all objections to the nearest quant who insists term structure models are intuitive. ↩︎
  4. The U.S. housing market fell twice. The first dip (2006–07) was the subprime bubble deflating — something the models (the copula, the rating-agency assumptions, the CDO structuring logic) didn’t really contemplate, but still relatively contained and heavily concentrated in California, Florida, Nevada, and Arizona. That’s the housing downturn I’m talking about here.
    The second, far larger leg down came later (2008–11) and was the result of the liquidity shock itself: funding markets froze → the economy stalled → unemployment spiked → foreclosures surged. In other words, the major housing collapse was a consequence of the financial crisis, not its original cause. ↩︎
  5. Even good collateral becomes “bad” once the collective mood dips below “mildly anxious”. ↩︎
  6. Those who say liquidity runs are a symptom not the cause, miss the point. The proximate cause varies with fashion: tulips, railways, dot-coms, a housing bubble — we don’t know what the next trigger is going to be. But we do know this: liquidity is the oxygen, and when it disappears, even healthy assets suffocate. We can manage our exposure to liquidity risk. ↩︎
  7. Especially those issued before 2007. Senior subprime tranches were designed to be last in line for losses, cushioned by layers of subordinated paper. Mezzanine tranches took heavy hits, equity tranches were obliterated, yet the majority of senior pieces ultimately paid out in full.
    For context:
    – In 2008, non-agency mortgage-backed securities made up roughly a quarter of the total U.S. MBS stock, and subprime loans comprised about half of that (0.5 x 0.25 = 0.125).
    – Most defaults hit the mezzanine and equity layers, which were a mere 13% of subprime issuance (0.125 × 13% ≈ 0.0163). Senior AAA tranches saw minimal losses.
    – By the end of 2013, cumulative losses on subprime senior tranches were less than 0.5%, and if held to maturity, these securities delivered IRRs of roughly 3%.
    – At that time, across all subprime securities (including all tranches) average cumulative loss rates stood around 6%; likely lower today, as the economy improved.
    – But wait: the 6% includes lost principal and interest, and the subordinated paper had target yields of 10–20%+, and the bonds were not wiped out on Day 1 … Yes, you’ve guessed correctly, the overall IRR on the entire garbage subprime portfolio, principal-weighted and held to maturity, is (small, but) positive. ↩︎
  8. Richard Fuld was wrong about many things; the fragility of his funding model should not have been one of them. ↩︎
  9. Dollar funding in money markets was often cheaper than euro or sterling funding, though in truth, that only gets you halfway there. The other issue was that European banks had spent the better part of a decade hoovering up USD-denominated assets (RMBS, CDOs, leveraged loans, monoline-wrapped munis) and needed dollar liabilities to match. Lacking anything resembling a natural USD deposit base, they turned to the short-term wholesale markets, where U.S. money market funds were delighted to buy European bank paper for a few extra basis points. FX swaps made the whole thing look riskless, Basel made it look capital-efficient, and everyone agreed to pretend that rolling three-month funding for ten-year assets was a perfectly adult thing to do. It worked brilliantly until 2007, at which point MMFs fled, the cross-currency basis blew out, and the system reminded everyone that maturity transformation in a foreign currency is less a business model than a dare. ↩︎
  10. As every experienced hitchhiker knows, one should always bring a towel, and sometimes, a foreign currency swap line. ↩︎
  11. E.g. AAA-only mandates with auto-eject rules. ↩︎
  12. No one trusted any model. The market was frozen. Banks needed cash and were forced to sell. Mark-to-market losses meant more capital calls, more deleveraging, and so prices detached completely from expected cash flows. ↩︎
  13. It’s the Infinite Improbability Drive of finance: extremely powerful, highly unstable, and poorly understood. ↩︎

The Appendix of Truth

The 2008 financial crisis:

AspectLoreReality
TriggerMassive subprime defaultsLiquidity shock triggered by a collapse in confidence in structured credit collateral. Subprime delinquencies were in excess of what was modelled, but
he crisis trigger was the freeze in wholesale funding (especially repo and ABCP), not the magnitude of actual mortgage losses.
Nature of crisisCredit riskLiquidity + collateral valuation event that later produced credit losses. The system broke because intermediaries could no longer fund themselves against suddenly illiquid collateral; credit deterioration followed financial sector distress.
Who failedBad lendersHighly leveraged market-making and maturity-transforming intermediaries (banks, SIVs, monolines) that relied on short-term funding to hold long-term structured assets. Some mortgage originators failed too, but they were not systemically critical.
Senior tranche lossesWidespread defaults, AAA blowing up everywhereVery limited senior losses. Pre-2007 senior subprime tranches largely paid in full; cumulative losses were miles below the panic-implied marks. The mezzanine and equity absorbed the brunt.
ResolutionBailouts for “bad bets” *Liquidity backstop for systemically critical funding channels. Central banks provided term funding, collateral swaps, and guarantees; banks repaid these loans with interest. Capital injections were targeted at solvency concerns but were not the bulk of the intervention.

* I suppose this is the point where we must address the famous £33bn “cost to the UK taxpayer”, that ubiquitous Wikipedia number, lifted from a 2021 OBR estimate (Table B.1, for the hobbyists). A few clarifications, if I may:

  1. The £33bn is not the cost of “bailing out the banks”.
    It’s a 2021 snapshot of the OBR’s then-estimated net fiscal position across all crisis interventions. The headline turns negative only because the OBR layers on an estimated £46.5bn “Exchequer Financing” charge (essentially the interest cost of government borrowing used to fund the rescues).
  2. Much of the apparent damage came from a -£17.5bn unrealised loss on the government’s NatWest stake.
    This wasn’t a bill to taxpayers; it was an accounting estimate. Most of everything else in that table had already been realised by then, apart from UKAR (Northern Rock & B&B). UKAR subsequently wound down roughly to plan and delivered a +£2.4bn realised gain to the Treasury.
  3. When the Treasury actually exited NatWest in 2025, the picture changed.
    The government recovered roughly £35bn (share sales + dividends + fees) against £45.5bn injected. That’s a realised shortfall of ~£10.5bn , not −£17.5bn embedded in the OBR’s 2021 accounting mark. Swap the mark with the real number and the “£33bn cost” compresses to roughly £26bn.
  4. Then there are taxes, which the OBR headline conveniently ignores.
    NatWest has paid about £9.5bn in UK taxes in the past five years, Lloyds more than £18bn. The OBR treats them with the same intellectual honesty HMRC shows when it puts my private medical cover on my P11D while declining to credit me with the NHS savings created by me not using the NHS because I have private medical insurance.
    Add those tax receipts back into the accounting and the –£33bn becomes roughly a +£1.6bn fiscal profit.
  5. And since we’re already being petty, let’s finish the job and go full spreadsheet.
    Include the reduction in government borrowing costs attributable to those additional tax inflows and sale proceeds (+£2.6bn by my estimate). Once you do, the notorious £33bn loss turns into a £4.3bn gain.
    This still ignores PAYE and NICs paid by the banks’ employees, and the wider fiscal benefit of thousands of people in well-paid private sector jobs rather than on welfare — but frankly, we don’t need any of that to make the point.

Bottom line: the much-quoted £33bn “cost of the bank bailouts” is a zombie statistic: long dead, still staggering around the internet, and overdue for burial.

Don’t Panic: A Hitchhiker’s Guide to Modern Credit Markets, Financial Crises, and the Illusions We All Pretend Are Real

This photo is an (almost) faithful representation of the author’s patio, as it looks today

It began, as these things often do, with peak procrastination.

There I was on a Sunday afternoon, meant to be jet-washing the patio (a task I’ve been avoiding since approximately last spring), when I made the tactical error of replying to a comment on an investment blog1 — about a niche asset class that private retail investors2 are, as a rule, fenced out of, to their detriment in my view.

I wasn’t planning on enlightening anyone, mind you. I was simply searching for a just-plausible excuse to avoid manual labour for another half-hour.

To my surprise, some tinfoil hats soon appeared, explaining how private markets, and the crooked financial services establishment more broadly, are conspiring to defraud the little guy. It quickly became apparent that some commenters couldn’t distinguish private equity from private credit, and structured credit might as well have been witchcraft. Inevitably, someone invoked a YouTube “explainer” proving that private markets were the next financial scandal.

I clicked on the link. That was my second tactical error.

Disclosure: the video featured approximately five hundred ad breaks, which did not leave me predisposed to charitable interpretation.

Its diagnosis of private equity as a terminal misallocation mostly mistook market plumbing for pathology3. The presenter’s pub-grade anecdotes about “people not wanting to buy football clubs in Ireland anymore” were offered as evidence of institutional sentiment, which is a bit like inferring the global inflation outlook from the price of a ham sandwich. PE has its issues (exit timing, cost of debt, the occasional Rube Goldberg LBO), but none of them require a grand bubble narrative. What the video framed as systemic rot was mostly a grab-bag of liquidity premia, anecdotes and mark-to-model misunderstandings dressed up as revelation.

The algorithm, sensing weakness4, promptly offered me a sampler platter of additional “educational” delights, among them The Crisis of Credit Visualised by a bloke called Jonathan Jarvis. I clicked that, too. My third tactical error.

Suddenly I was listening to a man confidently explaining that securitisation was essentially a scam invented by greedy banks to blow up the world because Greenspan kept U.S. interest rates low.

Which is… not quite right.

Curiosity (or masochism) dragged me into the comments section, where someone declared: “This is how economics should be taught in schools.” For fuck’s sake.

Seeking solace, I fled to ZeroHedge, which usually relaxes me more than Private Eye. (For reasons I cannot explain, I can read their doomsday kooks5 with a sort of smug benevolence I seldom achieve in real life.) Unfortunately, instead of comfort I found a klaxon announcing that another global financial crisis was already under way because Jamie Dimon saw a cockroach in the JPM cafeteria6 and, you know, where there’s one —

I closed the tab.

At this point, the patio felt genuinely appealing.

But then I thought: this level of misunderstanding really does need correcting. Not because the internet deserves it (it doesn’t), but because these misconceptions keep leading self-directed private investors into the same predictable problems:

  • mispricing risk,
  • misunderstanding leverage,
  • buying products they shouldn’t,
  • avoiding products they should buy,
  • and generally approaching capital markets with the situational awareness of those blokes who go kitesurfing fly-fishing for great whites off Cape Town.

Many DIY investors tend to fall into two broad camps:

  1. The excessively cautious, who buy fixed-income products they don’t fully understand because someone on a forum told them volatility is evidence of moral decay; and
  2. The aggressively cavalier, who cheerfully buy crypto ETNs at lunchtime while waiting for their sandwich to be toasted.

The population in between is thinner than it should be.

So I began writing a brief clarification — just something to set the record straight about what private equity, leverage and structured credit actually do; how the 2008 crisis actually unfolded; how equity products differ from credit; and why liquidity and duration mismatch, rather than price volatility, is the thing to watch.

Modern finance runs on a small collection of comforting fictions: liquidity, diversification, ratings, hedging, and the notion that regulators have actually read the products they approve. Most of these illusions function perfectly well until someone tries to look directly at them, at which point everyone gets very cross and blames the nearest asset manager.

Predictably, the “brief clarification” mutated into something the size of a novella. It spawned footnotes, demanded a table of contents, and began arguing with itself.

I had to accept the obvious: this needs to be a series.

Hence Don’t Panic. Because: I love the Hitchhiker’s Guide and finance deserves at least one title pun not involving bulls, bears, or the occasional badger.

What I’ll discuss (and why you may regret reading it):

  • Why the GFC was fundamentally a liquidity crisis7, not a morality play
  • How retail investors fixate on the wrong risks and ignore the right ones
  • How leverage quietly detonates portfolios
  • Why many ETFs contain structures that behave beautifully… until they don’t
  • Why FX hedging is often a complete waste of yield8
  • Why complexity in financial products is almost always camouflage, useful mainly for institutions dodging capital rules
  • And why, despite popular opinion, securitisation9 and structured credit are not the bogeymen. Ignorance is

The unifying theme:

Risk never disappears. It just changes shape, moves house, or hides behind a prospectus.

Investors only get paid for carrying risk; their job is to make sure the pay is worth the burden.

This should not be revelatory, yet judging by the internet, it apparently is.

Who this series is for

  • Anyone with a patio to wash and nothing left to read on a Sunday
  • Anyone who once Googled “CDO explained simply” and immediately regretted it
  • People who think liquidity and duration matching are boring
  • People who think FX hedging is free
  • Anyone who has ever said “but it’s UCITS so it must be fine”10
  • Those who understand that regulators, like Vogons, are by and large not evil, merely poetry enthusiasts with a troubling fondness for paperwork. What they lack in galactic foresight they more than compensate for with forms in triplicate11

Who this series is not for

  • People who believe YouTube finance channels are educational
  • Individuals currently holding 3x leveraged ETPs “for the long term”12
  • My neighbour, who keeps asking me if I think gold is “about to pop”
  • Anyone expecting something optimistic or motivational13 (wrong blog)

Coming next

Post 1: a diagram-free recounting of what the GFC actually was, and why popular explanations miss both the point and the learnings we should draw from it.

Followed by:

  • liquidity
  • leverage
  • ETFs
  • FX hedging
  • complexity
  • and the general tragicomedy of retail risk management14

The patio, I might add, remains unwashed.

Which feels appropriate.

Notes:

  1. An eminently respectable personal finance site, located in a reliably sensible corner of the internet, with a readership that is, on the whole, financially literate, level-headed and refreshingly immune to Reddit-style hysteria. Which made it all the more surprising when the comments section on that particular article began to feature contributions that appeared to have been written during a power outage by people wearing colanders. Must have been a freak atmospheric event. ↩︎
  2. By which I mean the self-directed, spreadsheet-armed, ISA-maxing, ETF-comparing private investors: sensible, generally curious, but still vulnerable to the occasional YouTube prophet. ↩︎
  3. London-listed PE trusts trading at discounts tell you far more about U.K. retail liquidity, closed-end structure and buyers’ risk budgets than about the intrinsic worth of the underlying companies. Treating that and the price volatility as evidence of deteriorating fundamentals is looking through the telescope the wrong way round. As a rule of thumb, PE trust discounts should not be interpreted as a sign of apocalypse unless accompanied by frogs, locusts, or a Goldman research note titled “This Time It Is Different”. ↩︎
  4. The algorithm has correctly inferred that if you click on one bad finance video, you clearly deserve twenty more. ↩︎
  5. ZeroHedge remains the only website capable of convincing readers that both hyperinflation and deflation are imminent, simultaneously. They are famously optimistic about only one thing, and it’s always silver. Everything else is a conspiracy by the lizard people. ↩︎
  6. Jamie Dimon’s cafeteria cockroach was, as later reported, actually a stale raisin. Markets moved anyway. ↩︎
  7. A liquidity crisis is like a fire drill: everyone pretends to know the exits, but half the building ends up in the wrong car park. ↩︎
  8. FX hedging is beloved by investors who enjoy paying for the illusion of safety, much like extended warranties for toasters. They claim they’re happy to take the cost until you remind them that compound interest works in reverse too, like a loyalty programme run by Lucifer. ↩︎
  9. Securitisation has been blamed for everything from the GFC to the extinction of the dodo. I’m not claiming both accusations contain roughly equal evidence, I’m implying it. ↩︎
  10. UCITS rules were designed to protect retail investors, not to protect retail investors from themselves. A subtle but important distinction. ↩︎
  11. I promise not to mention Vogons again until at least Post 3, although the FCA may appear sooner. ↩︎
  12. Under the FCA rules, anyone wishing to buy a 3× leveraged ETP must qualify as a professional client, meeting sector knowledge, transaction history or portfolio size criteria. Also: anyone holding these products “for the long term” should, strictly speaking, be supervised by a responsible adult. ↩︎
  13. Anyone seeking optimism or motivation in a finance blog is advised to reconsider their life choices immediately. ↩︎
  14. This series may run longer than expected. Much like QE. If you learn anything from it, please keep it to yourself. I wouldn’t want to set unrealistic expectations. ↩︎

The Governance–Industrial Complex

Ask not what pains; nor farther seek to know
Their process, or the forms of law below
.

There are many irksome figures in financial services — activist investors, General Counsels who require your input-slash-sign-off-slash-certification on every document they draft, COOs who treat resource allocation as a contact sport, ESG evangelists who’ve never opened a balance sheet, regulators sans a clue, auditors with God complexes (also sans a clue), and rating agencies whose models make astrology look data-driven.

Yet none inspire such helpless, weary exasperation as the Ex–Big Four Partner: freshly departed from the temple of best practice and now here to transform your hard-won, low-drama equilibrium into an agenda item.

They are the purest distillation of corporate life: bulletproof CVs1, impeccable diction, and an unwavering belief that the real problem is insufficient governance of the governance framework2.

A short taxonomy

You can recognise one instantly. They carry themselves with the calm of someone who has never met a spreadsheet cell they couldn’t delegate.

Their wardrobe is London business formal, 2012 edition.

Their aura is PowerPoint at rest.

When they join your firm, you’ll hear phrases like “raising the maturity level of the control environment”, which roughly translates as “we’re about to have 48 more meetings a month”3.

They arrive at your mid-sized investment committee meeting with the giddy energy of someone who’s just discovered “tone from the top” and intends to bring that tone to every top, including yours.

Their mantra? “We just need to formalise this“. You thought that meant “write it down“. It actually means: “We will now build a 47-tab Excel matrix documenting decision rights for every lunch order over £20”.

They can survive on Nespresso pods and a sense of procedural justice.

According to research4, 92 per cent of new “alignment initiatives” can be traced to a recent Ex–Big Four hire who has discovered a gap between “policy” and “practice” and intends to fill it with committees.

The bullshit premium

Confidence, in theory, should signal competence. In practice, the Ex–Big Four Partner’s confidence signals only that they once survived a meeting where someone used “synergy” as a verb.

They do not lead, per se — leadership would require the messy business of making decisions. Instead, they govern. From a distance. Preferably via a “cross-functional working group”.

If something breaks, they’ll suggest forming a subcommittee to “unpick the learnings”. If you’re smart, you’ll fix it yourself before the first meeting invite goes out.

To challenge one is futile. They do not argue; they intonate. Their tone implies the existence of supporting evidence somewhere, probably in a document marked Draft v5_final_FINAL.

And when cornered, they employ the Name Repetition Gambit5:

“Look, HoSimpson, what you need to appreciate, HoSimpson, is that the issue here, HoSimpson, is governance.”

Once your name appears thrice, the matter is settled. You have been risk-assured into silence.

No mud, please (we’re advisory)

Execution is for other people. To roll up one’s sleeves is a red flag: sleeves are for sign-off, not soil6. The Ex–Big Four Partner’s true superpower is staying above the fray while convincing everyone they’re right in it.

They lead through deck density, not direction. If a deliverable can’t be captured in a matrix, it’s probably not worth doing7.

The Ex–Big Four Partner is allergic to execution, but they will happily write a policy about someone else doing it.

Final thoughts

The market rewards risk when it works; the Board rewards governance when it doesn’t.

Somewhere between the two lives the Ex–Big Four Partner: consultant-smooth, feedback-resistant, and entirely certain that “this needs escalation at next month’s committee meeting”.

They are both our nemesis and our mirror: proof that in this industry, failure isn’t a risk — it’s a process.

Notes:

  1. They have spearheaded, led, initiated, implemented, overseen and optimised Every. Fucking. Thing. under the sun. ↩︎
  2. It’s frameworks all the way down. ↩︎
  3. Delegating the taking of the meeting minutes is one of their core competencies. ↩︎
  4. HoSimpson et al., Rectum-Derived Facts (2025 Edition), compiled from the author’s Outlook Deleted Items folder. ↩︎
  5. The technique is taught at Partner Course between “Strategic Gravitas 101” and “Advanced Stakeholder Management”. ↩︎
  6. See also: All Hands on Deck: A History of Naval Metaphors in Corporate Leadership (unpublished manuscript). ↩︎
  7. In extreme cases, they will commission a matrix about which matrices the other matrices should reference. For further reading, consult ISO 9001-M: Quality Management Systems for Matrices. ↩︎

Zen and the Art of Moat Maintenance

Blind in their fear, they bounce against the wall,
Or, to the moats pursued, precipitate their fall.

I’ve been contemplating barriers to entry recently.

As a phenomenon, they tend to get decidedly mixed reviews.

Sociologists think they’re evil incarnate. Patent office staff think civilisation would disintegrate without them. Business investors think they’re indistinguishable from ‘value-add’ — unless they’ve invested in a start-up, in which case they tend to agree with the sociologists. Economists, meanwhile, talk about them as if they’re something out of Ivanhoe: moats, drawbridges, flaming arrows of compliance.

Personally, I think they’re closer to a velvet rope outside an average Mayfair bar: part deterrent, part performance, mildly irritating, largely unnecessary. Designed to make you feel that something inside must be worth the hassle.

Take finance. (Please, someone, take it.) Here barriers come in flavours like an expensive suit, professional credentials that take a few years to obtain, and the ability to pretend you understand MiFID II (or is it III now? I lose track), Solvency II or Basel III without googling them under the table. The hardest part isn’t the skill — it’s surviving the process. No one really earns their place so much as outlasts HR.

In theory, they protect incumbents. In practice, they mostly protect the incumbents’ bonuses.

Outside work, life’s much the same. The gym with a joining fee. The ‘quick coffee chat’ that’s actually a third-round interview. The flat demanding three months’ rent upfront and two personal references. All micro-barriers — low walls quietly testing just how badly you want in.

But barriers aren’t all bad — or at least, not equally bad for everyone involved. They create scarcity, and scarcity creates fees. (See also: private equity, The Ned, sourdough.)

No, really. They have some uses. Without them, everything would be a race to the bottom — and the bottom, as we know, is already crowded. They also create structure, a sense of achievement. They make success feel earned, even when it’s really just restricted.

There’s a funny moment that comes — usually around year ten of one’s City career — when you realise the gate you fought so hard to get through was never the only gate. The architecture is concentric, as if designed by Jacques de Saint-Georges d’Espéranche1. Just as you reach the top of the ladder and squeeze between the merlons onto the battlements, you spot another wall beyond — with its own rickety ladder already waiting.

After a while, we see how the system works and, being practical creatures, we copy it. If you can’t beat them, start a subcommittee and install a rampart of your own.

And so we all become little Beaumaris Castles of self-preservation. Each password, qualification, ‘quick coffee chat’ and airport business lounge is another wall in the concentric fortress of adulthood.

The genius, as ever, isn’t in keeping others out — it’s in keeping ourselves busy defending the habits we’ve developed and the structures we’ve built to sustain them. And defend we do, at the expense of daylight, fresh air, and cardiovascular health.

But here’s the good news: the gates were never locked from the outside. You could, technically, just walk out. Of course, you don’t. It’s taken a decade just to get through the moat2 — and besides, you’ve already paid the joining fee.

Some of us, though, do start to wonder, usually after bonus season. We open a spreadsheet, mutter “financial independence”, and start modelling various scenarios of compounding3. Before long, there’s a sketchy plan, a blog, a pseudonym, and a readership of forty-seven. We call it an exit strategy.

Others aim for a gentler liberation: a sabbatical, a ‘consulting year’, a course in carpentry in Portugal4. Each is its own small rebellion — a temporary drawbridge lowered before (more often than not) being pulled back up with relief.

And so the barriers remain — only now they’re tidier, better landscaped, with a view and an annual fee. 5

Notes:

  1. Not to be confused with Georges de Saint-Jacques d’Espéranche, who, if he existed, would almost certainly run a boutique wealth-management firm in Knightsbridge. ↩︎
  2. The author writes from within the second curtain wall of the City. At this rate, the inner ward should be reachable by 2050, the keep by 2075, subject to regulatory approval. ↩︎
  3. Studies show that 87 per cent of Excel models for early retirement assume an annual return of 5 per cent, stable inflation, and zero existential dread (Source: M’earse et al). ↩︎
  4. At time of writing, carpentry courses in Portugal remain oversubscribed. Some restrictions may apply; flights not included. Please consult your financial adviser before attempting to build a chair, or an illusion of meaning. ↩︎
  5. Readers are reminded that this does not constitute career advice, investment advice, or psychological counselling. Any resemblance to actual colleagues, institutions, or moats — living or defunct — is entirely coincidental. Fortification strategies may go down as well as up. ↩︎

Trying to Stay Passive While the Market Goes Private

And when the sun restor’d the cheerful day,
He rose, the coast and country to survey.

As a passive investor, I’m supposed to do nothing. Ideally forever. Also: I dislike thinking too much.

But public markets keep shrinking — fewer IPOs, more delistings, and an ever-increasing proportion of capitalism conducted behind NDAs and closed doors. So, in a lapse of principle, I’ve been buying shares in listed private equity vehicles and management outfits — Oakley Capital Investments (OCI), Brookfield Corp (BN), and Blackstone (BX), to name a few1.

This is the private equity exposure accessible without the $5 million minimum investment, a Cayman lawyer and a relationship manager who calls you by first name and means it.

It is, admittedly, a semi-active decision2. But then, so is breathing.

The great vanishing

Index trackers only own what’s listed — which increasingly means the large, the dull, and the capital-intensive: banks, miners, utilities, telecoms, and, more recently, hyperscale data-centre landlords powering shrimp-brained chatbots whose job is to compliment you on the insightfulness of your questions.

Over the past two decades, the investable universe has been quietly hollowed out. Companies delist; buybacks reduce float; and private equity, sovereign wealth funds, infrastructure pools and family offices compete away an ever-greater share of corporate ownership. The index captures less of the real economy than it used to. And when companies do list, they tend to arrive pre-aged: sponsor-polished, late-stage, and priced to perfection.

Public markets are the showroom. The warehouse is private. Larger, less transparent, by appointment only — and everyone inside insists the lighting is fine.

Management, unwatched (and allegedly wiser)

The narrative is that private companies perform better because management isn’t ‘distracted by quarterly results’. Translation: they don’t have to publish them at all.

It’s much easier to ‘focus on the long term’ when the short-term numbers are optional3.

The added benefit is discretion: problems can be managed quietly, earnings can be ‘normalised’, projections can be based on ‘professional judgement’, disappointments can be phrased in the future conditional, and valuations can be whatever the model says — or, if the output seems wrong, the model can be ‘improved’.

Still, the reported numbers look good. Emphasis on reported.

How the sausage gets structured

Private equity is simple if you describe it vaguely. Money flows from limited partners (named for the extent of their insight) to general partners (named for the general absence of such limits).

The GP buys businesses, adds leverage, deploys consultants to ‘create value’, and eventually sells them — ideally to another fund pursuing the same strategy. Private equity is essentially asset management, but with more leather chairs, better coffee and a slower feedback loop.

The ambition: triple your money in a decade. The method: spreadsheets, PowerPoint, ‘alignment of incentives’, and a fondness for 30-slide appendices4.

When a GP says they ‘partner with families to achieve generational objectives’, it usually means they bought the business and let the founders keep a minority stake.

The legal form of things

A private equity ‘fund’ is not a fund in the conventional, regulated, daily-valued sense. More often than not it is a limited liability partnership domiciled somewhere warm where the regulatory correspondence is cordial.

‘Investing’ means committing to provide money later — when the GP finds something to buy, needs fees, or both. It’s like a subscription with variable billing and no cancellation rights.

Distributions arrive when convenient, accompanied by notes written in the future perfect, using ‘carry’ as both a noun and a warning. The whole arrangement operates on trust and the comforting opacity of 300-page limited partnership agreements no one claims to have fully read.

Direct entry requires a minimum of $5m, often more, much more, and a tolerance for legal PDFs. It helps if you’re a pension fund or a sovereign wealth fund. Or a hedge fund like Berkshire Hathaway or Fairfax Financial Holdings.

Individual participation is restricted to ‘sophisticated investors’, loosely defined as people who already have too much money to care. Some PE funds offer monthly subscription ‘access vehicles’ designed to make private equity ‘inclusive’, in the same way first-class upgrades make aviation inclusive.

Small retail investors are not invited. Hence my workaround: listed vehicles like OCI, or shares in the management companies themselves. You get the mezzanine gallery: the polite approximation of being in the room.

Management companies get you exposure to the fee income, performance allocations, co-investment vehicles (both Brookfield and Blackstone invest quite heavily in the funds they manage, which is why I bought the shares), and the comforting resilience of being the house, not the customer. OCI and similar vehicles are effectively fund-of-funds: they invest in the underlying PE funds their affiliates manage, playing the limited partner role. The catch for retail investors is that your RIY will be worse than that of the LPs — you pay the underlying fund fees and the listed vehicle’s operating costs.

But… beggars can’t be choosers; small investors can’t be limited partners. It’s better than nothing.

“We’re not like the others”, claimed everyone

Every fund deck has a Differentiated Approach slide.

‘We work with founders’.

‘We use conservative leverage’.

‘We move quickly and confidentially’.

‘We avoid general auctions’.

Translation: we buy things quietly, using slightly less debt, before anyone else notices.

“Average leverage: 3.7× EBITDA versus a US buyout average of 5.8×.”

This is presented as restraint. It’s also still quite a lot of debt5.

Funds invest over five to seven years, hold positions for eight to twelve, and have a fund life of fourteen to sixteen. By the time they distribute the proceeds to limited partners, the macro backdrop has been entirely rewritten; the LPs will have lived through at least two economic cycles and three central bank regimes.

If you’re accessing PE through listed stock, you can sell any time. You shouldn’t — but you can.

Valuation: a creative pursuit, sensibly unhurried

Private equity valuations are where finance meets storytelling. With no market price, ‘fair value’ relies on:

  • What they last paid;
  • What someone else paid for something vaguely similar;
  • EBITDA multiples calibrated to how confidently the CFO can say the number aloud;
  • A junior associate’s estimate of ‘sustainable free cash flow’ derived through a process known as ‘sensible extrapolation’ (plus one slide marked Further Upside Potential); and
  • A discount rate selected by someone who owns two espresso machines and believes both are essential.

In public markets this is called guessing; in private equity, it’s marking to model, with context. Figures are refreshed quarterly, ensuring optimism remains current. Volatility is treated as a matter of personal taste: if it’s not removed altogether, it is scheduled. This is considered a feature.

Diversification, kind of

Private equity diversifies your portfolio — largely because it isn’t repriced often enough to correlate with anything else. When markets fall, private equity valuations ‘lag’. It’s the technical term for “refuse to move until audited”.

That is also considered a feature.

Continuation funds: selling to yourself, professionally

If an asset cannot be sold at a flattering price, the GP may sell it to… another fund they also run. This is known as a ‘continuation vehicle’, and it ensures:

  • No one takes a write-down; and
  • The GP earns fees a second time.

Everyone insists this is perfectly normal. And they are right — which is the interesting part6.

Fees: we all know, but try not to say out loud

The standard fee structure is:

  • 2% management fee (guaranteed);
  • 20% performance fee (if things go well);
  • plus transaction fees, monitoring fees, and advisory fees (because: why not).

The waterfall (the payment order) ensures the GP gets paid first, then more later, then sometimes again. Limited partners receive whatever can be described as a return.

The phrase used is ‘alignment of incentives’, which here means everyone is incentivised to sell the business before the music stops.

Case study 1: how the magic happens (and where it stops)

In 2015, Brookfield Business Partners bought GrafTech, a graphite-electrode maker, for $295 million. Three years later they floated it on the NYSE at $15 a share, sold about 13 per cent, and collected $885 million through the IPO, dividends and note repayments. Their remaining 30 per cent stake was then worth $1.8 billion.

Total: roughly a 9× return, or a 3× uplift just from the IPO.

Brookfield called it ‘value creation’. Everyone else called it ‘good timing, very politely executed’.

Fast-forward: GrafTech trades around $1.707, down from the post-IPO high of $15.50.

If you owned a broad U.S. market index ETF in 2018, such as ITOT or VTI, you probably bought GrafTech somewhere near the top — the part of the movie where Brookfield rolls the credits and you sit through the sequel.

So yes, private equity can produce spectacular returns. Just not necessarily for the people who eventually inherit the shares.

This is not a morality tale. It simply illustrates the timeline difference between private accumulation and public participation. The return profile is path-dependent, not symmetrical across holders.

Case study 2: NAV as a suggestion

In theory, London-listed private equity vehicles should hover somewhere around their NAVs8. In practice, 3i trades miles above and OCI trades meaningfully below. This looks odd if you imagine markets are coldly rational. They are not. They are perfectly capable of reading a valuation report and then pricing the shares on vibes.

3i’s vibe is very straightforward: Action. A single, large, fast-growing discount retailer that everyone can understand without turning a page. It is the Taylor Swift of portfolio companies. Investors see it, nod, and pay a premium. End of thought process.

OCI’s portfolio, by contrast, is mid-market tech, consumer and education businesses — solid, growing, but requiring slightly more attention span than the market typically deploys at 8:07am on results day. The value is real, but the narrative is less click-and-go. And so the shares trade at a discount, not because the NAV is wrong, but because exits are not yet packaged as headline events with stadium lighting and pyrotechnics9.

The point, such as it is

Public markets are shrinking. Capitalism is not.

More of the economy — infrastructure, renewables, logistics, software, even the data centres that host this paragraph — now exists in private structures where reporting is optional and liquidity negotiable. Owning shares in listed private equity vehicles is simply a way of remaining faintly connected to the parts of capitalism that don’t have to return analyst calls.

I don’t believe private equity is superior, only that access has stratified. Ignoring that is another form of concentration risk.

So I picked up shares in Blackstone, Brookfield, OCI, and a couple more. Owning them doesn’t make me a private equity investor — only a shareholder in those who are. It’s not conviction in the model, exactly. More a form of passive surrender, which, in portfolio terms, is often indistinguishable from prudence.

Notes:

  1. True, the index already includes these firms. But the listed vehicle only owns a portion of the private funds — the rest sits with LPs. So this is just me dialing up the fraction. I am overweighting the small slice of private markets the public markets already accidentally own. Think of it as nudging a rounding error until it starts to feel strategically meaningful. ↩︎
  2. Yes, you can buy private equity ETFs. I inspected IPRV. It holds some managers I’d rather not have dinner with, plus trusts priced like they’ve already won. Plus fees and spreads. I am not paying a premium to own other people paying a premium. ↩︎
  3. Private companies do submit quarterly results — just not to the exchange. They go to a PE associate who once modelled airports for Macquarie and is now tasked with ensuring nobody panics. Maths checked, nod delivered, surprises postponed until next quarter. ↩︎
  4. I’ve mocked the Appendix of Truth before. Yet here we are, footnotes nested in a blog post, performing the same function. Add a chart and it’s officially Pot-Kettle-Black. ↩︎
  5. Leverage is context-dependent — like calling a glass of champagne “modest” at the office Christmas party. In any case, EBITDA is a number that assumes productive asset depreciation is a myth and tax is optional; it exists mostly to make debt feel emotionally distant. Bonus: it’s a nice-sounding acronym that looks aesthetically pleasing on slides, particularly in Avenir or Proxima Nova. ↩︎
  6. A continuation fund (named for the continuation of fees) is essentially moving wine from one decanter to another and claiming the aroma has matured. Wine remains wine, fees continue. ↩︎
  7. All prices as of writing. Opinions subject to change, but probably won’t. ↩︎
  8. Private equity NAVs aim to approximate fair value. If we accept that premise, logic suggests the listed vehicle should trade a touch below it — a small adjustment for the privilege of paying two sets of people to manage the same capital, plus the usual pleasures of being a listed entity: exchange fees, auditors, regulators, and the production of annual reports largely read by no one but the administrator. ↩︎
  9. Disclosure: I own shares in OCI. Translation: I am wagering on fundamentals, compounding, and the eventual recognition of rational valuation. Bold stance, historically known as: waiting. ↩︎

Nosce Te Ipsum

Best laid plans of mice and men.

Remember the 4th of April?

There’s nothing like a bit of financial turbulence to reveal one’s true nature. Some people, in such times, discover they’re bold. They stride into the gale of uncertainty, clutching a smartphone and muttering ‘buy the dip’ with the zeal of a Viking ordering mead.

Others prove disciplined. They stick to their allocations, calmly rebalance, and talk about ‘staying the course’ while gently implying they saw it coming.

Then there are those who’ve quietly held gold bars in the shed for just such an occasion, or whose spouse once invested in ‘something called Treasury gilts’ after a long chat at a National Trust café with a man named Jasper.

And then there’s me.

The plan (in theory)

When Trump began his tariff theatrics back in March — the full we’re being treated unfairly by everyone routine — I knew what was coming. He’d embark on his gorilla-style chest-beating threat display, there’d be a sell-off in equities and a bit of USD devaluation, followed by some half-hearted deal he’d declare ‘the best ever’, after which markets would recover. The pattern was obvious.

I had a plan for this sort of thing. There’s a folder on my desktop called Contingency. Inside it: a tidy note to myself, a few charts, and an action plan clearly marked When to Sell.

The idea was simple. Take a bit off the table, wait for the inevitable overreaction, then buy back at a discount.

Of course, I didn’t.

I opened the app several times. I looked at the absurdly high prices, thought, ‘Yes, now would be sensible,’ and then made coffee. On one occasion, I sorted a drawer. On another, I re-read some 2008-era Deloitte emails about the Opes Prime insolvency, for atmosphere.

But sell? No. I couldn’t be bothered.

The psychology of the hoarder

This wasn’t indecision. I wasn’t paralysed. I simply didn’t want to sell.

I like my ETFs. I’ve had them for years. I’ve watched them grow from scrappy little allocations into something almost respectable. Some are slightly embarrassing hangovers from earlier experiments — the financial equivalent of novelty mugs one hides when polite company visits — but they’re mine.

Every time I hovered over ‘confirm trade’, a small part of me recoiled. Not out of doubt — I knew prices would fall — but out of attachment. These holdings had been good to me. They’d paid small, polite dividends. They’d kept me company. Selling felt wrong, like dropping an old dog off at the shelter because he’d gone grey around the muzzle and the vet’s just quoted £3,650 for his dental work.

I’ve realised that, when it comes to investing, I have the temperament not of a trader but of a hoarder. Not the tabloid sort with bin bags and concerned neighbours — more the tidy, middle-class kind who simply can’t bear to part with things.

A cat Investor in a dog market

Investors, I’ve decided, are either dogs or cats.

Dog investors chase things. They react. They bark at headlines and believe action itself has value.

Cat investors, meanwhile, watch. They sit on their portfolios and blink. They’re not brave or lazy — just oddly content to do nothing, even when they’re certain something should be done.

I am a cat.

When the April tariff panic arrived, I sat on my portfolio like a cat on warm laundry. I observed the market twitch, took note of the movement, and did precisely nothing. Not out of discipline. Out of habit.

Also, trading is effort. One must locate the post-it note with the current password, remember three digits — in the correct order — from the six-digit PIN, and pass a CAPTCHA seemingly determined to fail you unless you identify 43 motorcycles in a Slovenian car park. Honestly, at that point, I’d rather just take the drawdown.

Lessons (of a sort)

What I’ve learned from all this is that I am not, strictly speaking, a serene long-term passive investor of the textbooks. I am a collector.

I collect ETFs and ETCs (and the occasional listed private equity share. Yes, I know. Whatever). I like seeing them lined up across my ISA, SIPP and GIA — neat, familiar, domesticated. I don’t rebalance so much as tidy occasionally. I don’t adjust allocations so much as admire them.

It’s not ignorance. It’s not fear. It’s comfort.

So, in light of that, I’ve amended my ‘plan’. My new official FIRE date is 24 December 2027 — because nobody sensible wants to work the week between Christmas and New Year anyway. Until then, I’m hoarding cash.

If markets crash before then, I’ll stay in work another year and buy the dip (in theory). If they don’t, the six-year cash buffer should buy me enough time to come to terms with selling something.

Either way, I’ll be fine.

I like my investments. I like having them. And if the market chooses to have another fit while I’m walking the dog, that’s its problem.

I’ll deal with it later. Probably.