I spent a few days in London last week and could feel something I had not felt in a while. Call it a feel-good factor. Some of it was the weather, most of it was the football. England opened their World Cup with a 4-2 win over Croatia, Harry Kane with two and Jude Bellingham running the game, and Thomas Tuchel’s side now sit among the favourites. The mood in the City always lifts when England win, and it lifted last week.

I watch the England game as a Scot, and not one of those Scots who reflexively backs whoever they are playing. This is our first World Cup since 1998, which is reason enough to be cheerful, and the great thing about following Scotland is that we travel with almost no expectation. We go for the experience and the fun, “No Scotland, No Party”, and the Tartan Army has duly gone down a storm in Boston. Tomorrow we play Brazil in Miami, which will be more of the same: a great day out and, whatever the football does, no real disappointment. It still took me straight back to the summer of 2002.

Morning kick-offs and an app on the desk

That tournament was in Japan and South Korea, so the games were on in the morning our time. I was at Credit Suisse First Boston, on the corner of Cabot Square, and the firm gave us an app to watch the matches at our desks. There was a pub on that same corner, close enough that without the app the floor would have emptied by mid-morning. This was years before MiFID and the unbundling of research from trading, when entertaining clients was actively encouraged, most of it done in the evening. For the big one, the quarter-final against Brazil, we laid on a large client breakfast.

That match kicked off at 7.30am BST on Friday 21 June 2002 in Shizuoka. Michael Owen put England ahead, Rivaldo levelled before half-time, then Ronaldinho floated a free-kick over David Seaman for the winner. England could not break down ten men after Ronaldinho was sent off, and the run that had carried the mood all tournament was over by mid-morning. Brazil went on to win the tournament.

“RSA is to insurance what Brazil has been to world football”

Our morning meeting started at 7.10am, as it always did, twenty minutes before kick-off that day. The second floor at One Cabot Square was a single open-plan trading floor so vast it seemed bigger than a football pitch, and you presented from a podium to the whole room. I stood up to give a bearish note on Royal & Sun Alliance, and I opened with the line: RSA is to insurance what Brazil has been to world football. The weight sat on those two words, because the joke was that a revered name could be a has-been, and the sales desks got it instantly. They started banging on the underside of their desks, the drum-roll a trading floor reserves for a call it enjoys. Twenty minutes later England walked out against the actual Brazil and lost.

The formal rating was only a Hold, but in that era a Hold was effectively a sell. Our team carried no Sell ratings at all, and hardly anyone did, because almost everything was a Buy and going negative on a household name, even by implication, was a lonely place to be. Everyone who owned the shares was unhappy with you, and at a firm with a sales force the size of CSFB’s, a call like that genuinely moved the stock. It aged well, even so. RSA went on to slash its dividend, bring in Andy Haste as chief executive, and raise £960m through a deeply discounted 2003 rights issue, as it tried to rescue a business widely seen as being in crisis. Being bearish on a blue-chip is uncomfortable in the moment, and a good deal more comfortable eighteen months later.

While we watched the football, the balance sheets were burning

The FTSE 100 had almost touched 7,000 at the end of 1999 and kept falling until it bottomed at 3,287 in March 2003, a drop of more than half. With-profits funds, the savings policies that smooth stock market returns into annual bonuses, had ridden the long bull market with very high equity weightings, in many cases 60% to 70% of the fund, and Prudential’s main with-profits fund was 72% in equities at the end of 1999 (see table below).

Source: Prudential

On the way up that was the engine of the proposition. On the way down it was a trap, and the trap is worth taking slowly, because it is the heart of the story.

The rule that did the damage was the resilience test. It required an insurer to stay solvent not just today, but after a further sharp fall in equities, typically another 25% on top of whatever had already happened, and to set aside an extra reserve sized against that hypothetical further drop. Now watch what a falling market did. Two things hit at once. The equities the fund already held were worth less, so its cushion shrank. At the same time, the test kept asking what would happen if those equities fell another 25% from here, and the more equities you held, the bigger that hypothetical loss was, so the bigger the reserve you had to find. The cushion was shrinking exactly as the reserve it had to cover was growing.

That left one escape route. A falling market hands you no new money, and the only lever you control is the size of that hypothetical further loss, which you shrink by holding fewer equities. So insurers sold equities and bought bonds, which the test treated as safe. Picture a fund that owes policyholders £100 and holds £130 of shares. The test asks whether it would still cover the £100 after a 25% fall: the £130 becomes about £98, which fails, so the fund sells shares and buys bonds until a 25% equity crash can no longer push it under. The falling market was forcing insurers to sell into the falling market.

Now make it systemic. Every insurer faced the same test at the same time, so they all sold equities together, and that wave of selling pushed the market down further, which shrank everyone’s assets again and made the test demand even more, which forced the next round of selling. Their own selling was feeding the very fall that was putting the next one in trouble. They had the weight to move the whole market, too: according to the ONS in 2000 insurers owned around 21% of UK quoted equities, against just 1.2% by 2024, so when they all sold at once the index had no choice but to feel it. This was real pressure inside the with-profits funds, not a paper one. The inherited estate, the surplus pot a with-profits fund builds up over generations and shares between policyholders and shareholders, was being eaten into, free asset ratios were collapsing, and Equitable Life had already closed to new business in December 2000. The weakest names were in serious trouble.

The note I wrote that July

This is where it helped to be an actuary. The headline gauge of strength in those days was the free asset ratio, which applied only to with-profits funds and measured the spare assets the fund held over and above its liabilities, expressed as a percentage. As the market fell those ratios collapsed, from around 9% to 12% at the end of 2001 to 5% to 7% by the middle of 2002. The trouble with the ratio was that it squeezed business mix, bonus policy, investment strategy and the valuation basis into a single number, which is why I spent much of that period warning investors not to read one company’s against another’s.

The credit rating agencies mattered even more, because they acted as pseudo-regulators, setting solvency requirements pitched well above the regulatory minimum, yet they were painfully slow to move their ratings as the market dropped. So I built our own tool, the CSFB capital adequacy model, that could read off an insurer’s implied financial strength at any level of the FTSE 100, mapping a capital adequacy ratio onto the familiar rating bands, AAA above 175% down to BBB at 100%, and updating the moment the index moved rather than months later.

The capital adequacy model did not come out of nowhere. Our team was built around proprietary models, the best known being the Value Tracker, a framework that let investors value insurers on a consistent basis and compare them like for like rather than taking each company’s own numbers on trust. It was the sort of tool clients genuinely relied on, and it was a big part of what carried the team towards the number one spot in the Institutional Investor rankings that year. The capital adequacy model was cut from the same cloth, a proprietary lens trained on the one question the market kept asking and the rating agencies were too slow to answer.

The chart it produced told a clear story.

RSA sat at the bottom of the pack, and on my numbers was close to dropping below 100%, beneath even the BBB floor, which is to say below investment grade on implied strength. That, more than the line about Brazil, was what sat behind the call.

The note made a subtler point too, and it is the one I am most pleased with looking back. The statutory minimum capital an insurer had to hold, the EU requirement that was in force in the UK that summer, was about 4% of reserves on business where the insurer carried the investment risk, with-profits included, and 1% on unit-linked, where the policyholder carries it. For the with-profits business that margin could largely be met from within the fund, because the inherited estate was sitting right there. So even though the with-profits funds were under genuine strain, I argued that the headline panic overstated the threat to formal statutory solvency for most names, and that the binding pressure for shareholders sat outside the with-profits fund, in unit-linked, international and general insurance, where the trapped policyholder estate was no help and which the free asset ratio, being a with-profits measure, did not even capture. The note even said, in as many words, that we expected insurers to lower their equity weightings gradually. That is exactly what they did, and it is the single biggest reason the sector is so safe today.

The regulator that kept moving the goalposts

The same mood, a very different sector

So I applied that test to today’s sector, and almost nothing about it resembles 2002. Back then the danger was on every screen, with the FTSE falling in real time and solvency draining away with it. Look underneath the sector now and you find the opposite, because the thing that nearly killed it has gone. I set this out at length last year in Haunted by the Dot-Com Crash, Built for Today, and the short version is that the modern UK life insurer barely resembles the one that nearly broke in 2002.

The names on that 2002 chart are, in the main, still here, just under different owners. CGNU became Aviva, which later swallowed Friends Provident as well, the Prudential arm became M&G, and Legal & General is still Legal & General. RSA’s life funds, the very business I was bearish on that morning, were closed and then sold to Resolution in 2004, the vehicle that became Pearl, then Phoenix, and today carries the Standard Life name. So the big names you know now, Aviva, Legal & General, M&G and Standard Life, all trace back to that chart, and every one of them now runs a de-risked, bond-driven balance sheet a world away from the equity-backed funds of that summer.

Equities now carry a punitive 39% capital charge under Solvency II, so insurers hold almost none of them directly, and their books are matched, long-dated bonds set against long-dated promises so that the two move together. Regulation and matching are really the same thing here, because the whole regime is built around matching assets to liabilities, which is exactly what makes holding equities, which match nothing, uneconomic.

Source: L&G Annual Report 2025

The numbers above show how complete the change is. A 25% fall in equity markets would move Legal & General’s solvency by only 5 percentage points. Its coverage stood at around 210% at the end of 2025, and the company now aims to run between 160% and 190%, and I would find even the 160% floor of that range perfectly comfortable. The sector’s real low point was 2003, not the 2008-09 financial crisis, and it has since taken Brexit, the pandemic, the 2022 gilt and LDI episode and the market turmoil of this year’s US-Iran conflict, without any of them threatening solvency.

The irony is that the old model was simple enough that an outsider like me could estimate a company’s capital at different market levels on the back of an afternoon’s work. Today’s Solvency II model is stochastic, 100,000 scenarios run across the whole business with every risk interacting, and no outsider can reproduce it, so any attempt would be spurious. The modern machine is vastly more accurate and far safer, but the price of that safety is visibility, because it is a black box. That opacity is a large part of why companies now run solvency above 200%, well beyond what is genuinely needed, where in 2002 the cover was far thinner. The safer the engine became, the less anyone outside can see into it, so investors demand a bigger cushion.

The surprise, for most people, is that a stock market crash, the very thing that defined 2002, would now barely register on a modern insurer’s balance sheet. Equity risk has gone from the force that nearly sank the sector to something close to an irrelevance, and that is not luck. It is the direct result of a capital regime built around matching, and it has left these balance sheets about as solid as anything in financial services.

Enjoy the football, and the balance sheets too

The draw has set up one last echo, putting Brazil in front of both home nations this summer: Scotland meet them first, tomorrow in Miami, and on the projected route England could meet them again in the quarter-final, the same giant waiting in the same round 24 years after Shizuoka.

Enjoy the football, because England playing well is good for the national mood and good for the City, and moments like this are rare enough. Hold both thoughts at once, though. The summer that felt best in my career was also the summer this sector nearly drowned, and it taught me to look underneath whenever the mood turns giddy. I have looked, and this time the news is good. The danger that was on every screen in 2002 has gone, and a falling market, the one thing that used to terrify anyone running an insurance balance sheet, would now barely move the dial. So enjoy the football, and for once you can enjoy the calm under the surface too.

Gordon Aitken runs Aitken Advisory, providing strategic advice on life insurers and pension funds. I work with investors, insurers and advisers on transactions, capital strategy and market positioning. As I did throughout my sell-side career, I meet fund managers and investors for one-to-one briefings on the sector. If you would like to arrange a briefing or discuss a potential engagement, click the button below to email me.