Since 2018, the financial lines market has offered the purest textbook case of the underwriting cycle. A cleansing of supply collided with the fear of a wave of Covid-related claims to produce the most brutal hard market of the century.
Financial lines bring together the classes of insurance that cover not physical damage but legal, financial and reputational risk. Foremost among them is directors' and officers' liability, known as D&O, which protects corporate officers against the financial consequences of management failures. To this are added professional indemnity, which covers the errors and omissions of service providers, fraud and crime insurance, the guarantees specific to financial institutions, and transaction liability, which secures mergers and acquisitions. These classes have in common that they indemnify the cost of litigation, an investigation or a claim, rather than the destruction of an asset.
This nature gives financial lines a particularly pronounced cyclicality. Because their loss experience depends on the behaviour of claimants, judges and regulators rather than on physical hazards, it is hard to anticipate and prone to abrupt surges. Because the capital required to underwrite these risks is highly mobile, supply floods in when returns are high and withdraws when they collapse. The financial lines market thus oscillates between hard-market phases, in which rates rise and capacity becomes scarce, and soft-market phases, in which competition compresses prices. The period that opened in 2018 offers an illustration of almost pedagogical clarity.
A technical feature aggravates this cyclicality, that of the long development of claims. Financial lines are most often written on a claims-made basis, which means that cover is triggered when the claim is made, sometimes several years after the alleged wrongdoing. A decade may elapse between the collection of the premium and the final settlement of the loss, the time it takes for the dispute to be investigated and then adjudicated. This latency makes the assessment of an underwriting year's true profitability extraordinarily uncertain at the moment it is written, because the insurer prices today for losses whose cost it will only learn much later.
This latency has a perverse effect on the cycle. During a hard market, insurers set prudent reserves which, if loss experience remains moderate, can later be released and flatter the accounts of subsequent years. Conversely, the rate inadequacy of a soft market reveals itself only with delay, when late-emerging claims surface and force a strengthening of reserves. Long-tail development thus desynchronises the perception of profitability from the reality of risk, and it is precisely this optical illusion that drives the market to prolong its excesses, upward as well as downward.
The hardening did not begin with the pandemic, it preceded it. Its source lies on the supply side, and more precisely within the Lloyd's market, whose results had deteriorated to the point of recording a net loss of two billion pounds in 2017, for a combined ratio of 114 %, followed by a further loss of one billion in 2018. Faced with this drift, Lloyd's launched its Decile 10 initiative in 2018, which required syndicates that had been loss-making for three consecutive years, as well as all syndicates for the least profitable 10 % of their portfolio, to produce a remediation plan or to cease the activity concerned1. The effect was immediate and massive, with around a fifth of the market having to submit a full plan, and nearly seven billion pounds of premiums, or 20 % of the total, having been remediated. The least profitable lines, among them certain financial lines, had on their own eroded 162 % of the 2018 result.
This restored discipline triggered a rate rebound that the rating agency Fitch put at nine consecutive quarters of increase. The withdrawal of capacity from the most loss-affected segments, in particular directors' and officers' liability, created a scarcity that pushed prices up well before the word pandemic entered everyday vocabulary. The hard market in financial lines is therefore first and foremost a supply phenomenon, a capacity correction orchestrated by the most influential market regulator in the world.
This contraction of supply met a demand for cover made more anxious by a loss experience already trending upward. Securities class actions in the United States had reached high levels over the 2017 to 2019 period, with more than four hundred annual filings, and so-called social inflation, that is, the tendency of juries to award ever heavier damages, was inflating the cost of claims2. The combination of contracting supply and anxious demand already constituted, by the end of 2019, the conditions for a hard market. The pandemic did not create this market, it brought it to its climax.
In the spring of 2020, the industry feared that the pandemic would trigger a surge of claims on financial lines. It anticipated a wave of securities actions against companies whose share price was collapsing, a multiplication of insolvencies leading to claims against directors, and abundant litigation around the handling of the health crisis. Lloyd's itself acknowledged its Covid exposure across several classes, namely event cancellation, business interruption, directors' and officers' liability and trade credit3. This anticipation justified an even sharper acceleration of rate increases, with average quarterly changes in D&O rates reaching 14 % in 2020 and 2021, a pace without equivalent over the century2.
The feared surge did not, however, materialise to the extent dreaded. The disruption of markets and the potential Covid-related claims did not have the catastrophic effect on balance sheets initially anticipated, and the class that truly suffered was event cancellation, not directors' and officers' liability3. Securities actions quickly returned to more moderate levels after the pandemic, far from the peaks of the late 2010s. The financial lines market thus found itself in a paradoxical situation, having raised its rates to historic levels to face a loss experience that, in the end, remained contained. This divergence between the price collected and the risk realised is the breeding ground of the next phase.
A cyclical factor amplified the movement, the explosion of initial public offerings and above all of special-purpose acquisition companies, the SPACs, in 2020 and 2021. Each new listing creates a need for D&O cover, and the boom of this period swelled the market's premium base. The abrupt reflux of IPOs from 2022, then the collapse of the SPAC market, made this additional demand disappear, contributing to the contraction in premium volume observed thereafter. The financial lines market therefore suffered a double demand shock, first upward then downward, which widened the amplitude of the cycle.
The exceptional profitability of the hard market produced its expected effect, the massive arrival of new entrants. Attracted by high rates and favourable underwriting conditions, new players and new capacity flooded into the directors' and officers' liability segment from 2022. The broker Aon estimated in early 2025 that the US D&O market was oversupplied, with more than one billion dollars of excess available capacity, a situation that mechanically intensified competition and pulled prices down4. The logic of the cycle thus reversed, the capital that had withdrawn in 2018 returning in force once the rent had become visible.
The fall in rates proved as rapid as it has been lasting. Quarterly changes in D&O rates, which peaked at 14 % of increase in 2020 and 2021, fell to an average of 0.1 % in 2023 and during the first three quarters of 2024, before turning clearly negative2. In the second quarter of 2025, directors' and officers' liability recorded its sixth consecutive quarterly decline, and in the first quarter of 2026 the Marsh market index noted a seventh consecutive quarter of decline in commercial rates, with financial and professional lines down 5 %5. The broker Lockton noted that insurers were beginning to show fatigue in the face of the declines and considered that rates had reached their floor4.
The combined effect of falling rates and the reflux of listings translated into a contraction in turnover. Direct US D&O premiums fell to 10.8 billion dollars in 2024, down 6 % from the 11.5 billion of 2023, and declining for the third consecutive year1. The financial lines market thus experiences a rare situation, that of high profitability on a declining volume, a configuration that cannot continue indefinitely and that raises the question of the sustainability of the model once the benefit of past increases is exhausted.
The deepest lesson of this sequence lies in the desynchronisation of two clocks. The price of financial lines follows the capital cycle, which is by nature a mean-reverting movement, money flowing in and out according to returns. The underlying risk, for its part, follows the drift of the legal system, which knows no mean reversion but an underlying upward trend. Social inflation, the multiplication of so-called nuclear verdicts, that is, awards of extreme magnitude, and the rise of third-party litigation funding feed a structural increase in the cost of liability claims6. The number of core securities actions, moreover, remains above its long-term historical average. The danger arises when these two clocks diverge, when the price falls at the very moment the risk rises.
This is precisely the diagnosis that part of the market now makes. The rating agency AM Best openly wondered whether rates had not fallen too far and too fast, to the point of becoming inadequate in view of a loss experience whose fundamentals remain unfavourable2. The contrast with other classes is telling. While financial lines continue their decline, general liability insurance is on the contrary experiencing sustained rate increases, precisely because of nuclear verdicts and social inflation, as the Marsh index shows5. That two classes exposed to the same underlying drivers should move in opposite directions reveals that the rate of financial lines is governed, for now, by the abundance of capital more than by the measurement of risk.
The long-tail development discussed above makes this decoupling all the more insidious. The rise of third-party litigation funding, which allows investors to finance lawsuits in exchange for a share of the damages obtained, lengthens the duration of proceedings and increases their financial stakes, feeding the upward drift of awards. Yet the rate inadequacy that results will only materialise in the accounts several years from now, when these late claims emerge. The market is therefore living through a period in which falling rates sit alongside a flattering reported profitability, without the losses corresponding to the contracts written today having yet appeared. This accounting calm is misleading, because it measures yesterday's loss experience and not the one that today's loosening is in the process of manufacturing.
The soft-market phase unfolds even as new sources of loss appear whose pricing remains uncertain. Securities actions linked to artificial intelligence, targeting companies accused of having oversold their capabilities in the field, have risen significantly while litigation relating to SPACs and crypto-assets has receded6. Climate litigation is also developing, with actions targeting directors for their handling of environmental risks, and event-driven disputes following catastrophes such as wildfires. These new frontiers share a characteristic with the cyber risk analysed elsewhere in this series, the absence of sufficient history to estimate their loss experience reliably, which makes them particularly dangerous to underwrite in a market that is compressing its margins.
The soft market does not only benefit buyers in the form of lower rates, it also alters their long-term behaviour. Taking advantage of favourable conditions, many companies, particularly those with strong risk profiles, negotiate broader cover and explore alternative risk-transfer solutions such as self-insurance and captives, those insurance subsidiaries that groups create to bear part of their own risks5. This movement has a lasting consequence, because a company that has internalised a risk during a soft market is less inclined to return to the traditional market when rates rise again. The soft market therefore does not merely lower prices, it also erodes, potentially irreversibly, the base of risks that insurers will be able to price at the next hardening.
At the close of this analysis, the financial lines market post-Covid reads as a demonstration of the underwriting cycle in its most accomplished form. A supply correction met a demand fear to produce a rate peak, whose profitability attracted the capacity that then caused the collapse of prices. The only uncertainty bears not on the existence of a next turn, which appears inevitable, but on its trigger. A wave of claims linked to emerging litigation, a late materialisation of social inflation in reserves, or simply the exhaustion of insurers' patience in the face of rates judged inadequate could initiate the next hardening phase. The lasting lesson of this period is that a market which prices legal risk to the rhythm of capital condemns its participants to oscillate permanently between overvaluation and undervaluation, without ever truly coinciding with the risk they carry. For the observer, and for the informed policyholder, the issue is therefore not to predict the exact moment of the swing, but to recognise where in the cycle the market stands in order to arbitrate between transfer to the traditional market and retention of risk, for it is in the clear-eyed reading of the cycle, and not in the illusory promise of its disappearance, that the lasting advantage lies.
Solvency II rests on two postulates inherited from classical actuarial science, the ability to diversify weakly correlated risks and the ability to estimate their distribution from historical data. Cyber risk contradicts both.
Revisiting the Lloyd's LMA 21-042 directive and its implications for Property & Casualty reinsurance treaties
NIS2 does not mention insurance even once, and it does not directly regulate insurers. Yet by raising the cybersecurity baseline of fifteen to twenty thousand French entities and by making executives personally liable, this directive profoundly reshapes the risk that insurers underwrite.
A structure falls in a town you have never heard of, and a retiree's savings in Tokyo wobble that same month. A hidden wire runs between the two, and that wire is reinsurance.