Loosening the Bolts

There is a deep and recurring rhythm in financial history, and it goes like this. A crisis strikes; the system nearly collapses; in the aftermath, regulators and lawmakers tighten the rules — raising the capital banks must hold, restricting the risks they may take, building buffers against the next storm. For a while the new rules hold, the system is safer, and memory of the crisis is fresh. Then time passes. The crisis recedes into history; a new generation of bankers and politicians arrives who did not live through it; the tight rules come to feel like needless constraints on growth and profit; and the pressure to loosen them builds, dressed in the language of competitiveness and freeing the economy. The rules are relaxed. The system runs faster and returns more. And then, eventually, the next crisis strikes, and everyone discovers that the buffers they dismantled were the ones that would have mattered. The cycle is so reliable that it is almost a law, and in 2026 the United States is squarely in the loosening phase, with the bank stocks soaring on exactly the deregulation that the cycle says ends badly.

The rally is spectacular. America's six largest banks — JPMorgan, Bank of America, Citigroup, Wells Fargo, Goldman Sachs, and Morgan Stanley — have gained roughly $600 billion in market value, their combined worth climbing from about $1.77 trillion at the end of 2024 to around $2.37 trillion, a jump of more than a third in barely a year. The driver, by near-universal agreement, is the deregulation agenda of the Trump administration: a loosening of the capital requirements imposed after 2008, an easing of the rules that constrain how much risk banks can take and how much capital they must hold against it, and a friendlier posture toward the industry across the board. Freed from some of those constraints, the banks have unleashed a torrent of capital return — the eight largest spent a record $33 billion on stock buybacks in the first quarter of 2026 alone — and the market has rewarded the combination of looser rules, fatter returns, and a revived deal-making environment with one of the great sector rallies of the cycle.

What the bolts were for

To understand why this should give a thoughtful investor pause, you have to understand what the rules being loosened were for. The capital requirements that the post-2008 reforms imposed — the higher minimum capital ratios, the stress tests, the leverage limits, the rules governing how much of a buffer a bank must hold against its assets — were not bureaucratic whims. They were the hard-won lessons of the 2008 financial crisis, when banks that had operated with too little capital and too much leverage discovered, when the housing market turned, that they had no cushion to absorb the losses, and the entire global financial system came within days of collapse, saved only by the largest taxpayer bailout in history. The reforms were designed to ensure it could not happen again — to force banks to hold enough capital that they could absorb a severe shock without failing or needing rescue. Capital, in a bank, is the shock absorber, the buffer between a bad year and insolvency. The rules that mandate it are the bolts that hold the machine together when it is stressed.

Loosening those bolts does two things at once, and they are inseparable. It makes the banks more profitable and more generous in the near term, because capital that is not held as a safety buffer can instead be lent out to earn interest or returned to shareholders through buybacks and dividends — which is exactly why the stocks are soaring and the buybacks are at records. And it makes the banks more fragile in the event of a shock, because the buffer that was loosened is, by definition, the buffer that would have absorbed the next crisis. These are not two separate effects to be weighed against each other; they are the same effect, viewed from two angles. Every dollar of capital freed from the safety buffer to juice returns is a dollar that will not be there to absorb losses when losses come. The market is celebrating the first angle — the returns — and ignoring the second — the fragility — because the second only matters when something breaks, and at the moment nothing is breaking. The loosening feels purely good precisely because the risk it creates is invisible until it is realized.

We have unscrewed this machine before

History does not merely suggest this pattern; it screams it, with examples specific and recent enough that no one should be able to claim surprise. The savings-and-loan crisis of the 1980s and early 1990s followed a wave of deregulation that freed thrifts to take risks they could not manage, and ended in hundreds of failures and a large taxpayer cleanup. The 2008 catastrophe itself was preceded by a long bipartisan loosening — the repeal of the Glass-Steagall separation between commercial and investment banking, the light-touch treatment of derivatives, the permissive capital rules — that let banks build the leverage and the exposures that then detonated. In each case, the deregulation was justified at the time in the language of modernization and competitiveness; in each case, the loosened rules produced a few good years of higher profits and rising bank stocks; and in each case, the bill arrived later, larger than the profits, and was paid by the public. The pattern is not subtle. It is one of the best-documented regularities in all of economic history: financial deregulation is followed, with a lag of years, by financial crisis, because the loosening that boosts returns in the good times removes the buffers that would have contained the bad ones.

What makes the 2026 version particularly worth watching is the speed and the completeness of the loosening, and the political environment around it. The capital rules being eased — including the bank-specific buffers and the leverage constraints that were the centerpiece of the post-2008 safety architecture — are being relaxed by an administration explicitly hostile to the regulatory state, in a climate where the regulators charged with restraining the banks have been reoriented to encourage them. And the political dimension has grown unusually personal: the president has taken to publicly praising individual banks, with Citigroup's stock notably outperforming after a presidential endorsement, and the bank itself unveiling a fresh $30 billion buyback program — a degree of entanglement between the political leadership and the banks it is supposed to oversee that itself erodes the independence on which prudent supervision depends. When the supervisor becomes the cheerleader, the supervision is already gone.

Which bolts, specifically

It is worth being concrete about what is being loosened, because the abstraction "deregulation" hides how targeted the changes are. The post-2008 reforms layered several distinct safety requirements onto the big banks, and it is precisely these that are being relaxed. There is the capital ratio — the proportion of a bank's assets that must be funded by loss-absorbing equity rather than borrowed money — the single most important determinant of whether a bank can survive a wave of losses; proposals to raise it further (the "Basel III endgame") have been watered down or shelved, and existing requirements eased. There is the leverage ratio — including the supplementary leverage ratio that caps how much a bank can borrow relative to its capital regardless of how "safe" the assets look — now being relaxed in ways that let the biggest banks expand their balance sheets and hold more assets against the same sliver of capital. And there are the stress tests — the annual exercises that check whether a bank could survive a hypothetical severe recession, and that gate how much capital it is allowed to return — now being made less severe, more predictable, and more permissive, which directly unlocks the record buybacks.

Each of these is a bolt with a specific function, and each is being turned in the same direction: toward letting the banks operate with thinner cushions, larger balance sheets, and bigger payouts. The technical details vary, but the through-line is singular and unmistakable — the quantity of loss-absorbing capital the system is required to hold against a shock is being reduced, and the freed capital is flowing to shareholders and to expanded lending. By one estimate, the 2026 round of proposals would cut the core (CET1) capital requirements of the largest banks by something like 5 to 6% cumulatively — billions of dollars of buffer, per bank, reclassified from "safety" to "available for return." Defenders argue, with some merit, that the post-2008 rules were excessive, that they over-constrained lending and made U.S. banks less competitive than foreign rivals, and that a sensible recalibration is overdue. That argument is not crazy; rules can be too tight as well as too loose, and there is a genuine debate about the right level. But the direction and the timing are what matter for risk: the system is being made to hold less against a shock, deliberately and across multiple dimensions at once, at a moment when the potential shocks are visibly accumulating — and "the old rules were a bit too strict" is a thin justification for thinning the buffers of a too-big-to-fail banking system into the teeth of a cycle this late.

The buyback tell

There is a specific feature of this rally that deserves scrutiny, because it reveals what the deregulation is actually being used for. The record $33 billion of buybacks in a single quarter — and Citigroup's standalone $30 billion program, building on a $20 billion one before it — represent banks taking the capital that looser rules no longer require them to hold and handing it to shareholders. Buybacks are not inherently bad; returning excess capital to owners can be perfectly rational. But there is a crucial difference between returning capital you genuinely do not need and returning capital you are only permitted not to hold because the safety rules were loosened. In the second case, the buyback is not a distribution of surplus; it is a depletion of the buffer, converting the system's shock-absorbing capital into shareholder returns and a higher stock price. The bank looks more valuable — fewer shares, higher earnings per share, a soaring price — precisely because it is now holding less of the capital that would protect it in a crisis. The market reads the buyback as strength. It is, in part, the monetization of reduced safety.

This is the same dynamic, in a more systemically important guise, that ran through the bank-deregulation episodes of the past: the good times are spent consuming the buffers, paying them out as returns, so that when the bad times arrive the buffers are thin. The banks and their shareholders enjoy the higher returns now; the public bears the tail risk of a thinner-cushioned banking system later. And because the largest banks are the ones at the center of the system — the ones whose failure would be catastrophic, the ones implicitly backed by the taxpayer because they are too big to fail — the loosening of their capital rules is the loosening that matters most for systemic stability. The $600 billion of added market value is, in significant part, the capitalized value of the safety the system has agreed to give up.

Heads they win, tails the public loses

The deepest reason bank deregulation is so seductive and so dangerous is the asymmetry of who captures the gains and who bears the losses, and it is worth stating plainly because it is the engine that drives the whole cycle. When the rules are loosened and the good times roll, the gains are private: they flow to the banks' shareholders as buybacks and a rising stock price, and to the banks' executives as compensation tied to that stock price and to the return-on-equity that thinner capital mechanically inflates. A bank that holds less capital earns a higher return on the capital it does hold — that is simple arithmetic — and executive pay is overwhelmingly geared to exactly that metric. So the people who decide how much risk to take, and who lobby to loosen the rules, are paid handsomely and immediately for the thinner buffers, in real money they keep.

When the bad times arrive, however, the losses are socialized. Because the largest banks are too big and too interconnected to be allowed to fail without catastrophe, the public — through the central bank and the taxpayer — stands behind them, implicitly guaranteeing their survival, as it did in 2008. This is the heart of the moral hazard: the banks and their stakeholders capture the upside of the loosening in good years, and the public absorbs the downside of the fragility in bad ones. "Heads they win, tails the public loses" is not a cynical slogan; it is an accurate description of the incentive structure that the implicit too-big-to-fail backstop creates, and it is precisely why the safety rules exist — to force the privately-benefiting risk-takers to hold enough capital that the publicly-borne tail risk is contained. Loosening those rules transfers value from the public (which bears more risk) to the banks' shareholders and executives (who capture more return), and dresses the transfer in the language of competitiveness. The $600 billion of added market value did not appear from nowhere. A meaningful slice of it is the capitalized value of risk shifted from the banks' balance sheets onto the public's implicit guarantee — a wealth transfer the market has priced as a windfall, because on a stock chart a wealth transfer toward shareholders looks identical to value creation.

The shocks are not hypothetical

The deregulation would be concerning in any environment; it is more concerning in this one, because the shocks that could test the banks' thinned buffers are not hypothetical — they are documented throughout this very series. The banking system is exposed, directly and indirectly, to the commercial-real-estate maturity wall and the office losses bleeding through the regional banks; to the private-credit stress and the leveraged-loan exposures; to the consequences of an oil shock and a possible recession; to a long-bond market straining under fiscal pressure; to the unwinding of any of the various leveraged and circular structures the AI boom has spawned. A banking system is supposed to be the shock absorber of the economy — the buffered, well-capitalized core that can take losses in a downturn and keep lending. Deregulating the banks at the moment when so many potential sources of loss are visibly accumulating is the opposite of prudent: it thins the absorber exactly as the shocks that would test it are multiplying. The system is being made more fragile and more leveraged at a moment when the case for resilience has rarely been stronger.

None of this is a prediction of imminent bank failure. The largest U.S. banks are, today, genuinely strong — far better capitalized than they were in 2008, profitable, well-managed by capable executives, and operating in an economy that is, for now, growing. The deregulation does not make them weak overnight; it makes them incrementally less buffered, and the danger is not this quarter but the cumulative effect over the years of loosening that history says tends to follow, as each relaxation makes the next one easier and the buffers thin step by step until a shock arrives to reveal how thin they have become. The bank rally may run for a long time; deregulation can juice returns for years before the bill comes due, and the investors riding it may do very well in the interim. That is precisely how the cycle works — the loosening pays, handsomely, right up until it doesn't.

But the forensic point is the one history keeps making and markets keep forgetting: that the capital rules being dismantled were written in the blood of the last crisis, that loosening them trades long-term safety for short-term return, and that the celebration of a deregulation-fueled bank rally is, viewed across the full arc of the cycle, the sound a financial system makes while it cheerfully unscrews its own safety equipment in the calm before the storm. The bolts came out of a lesson learned the hard way. They are being loosened, now, by people who did not pay for that lesson, in the name of returns that are real and a safety that is invisible. The machine runs faster with the bolts loose. It always does — until the day it is asked to hold together under load, and discovers that the very buffers that would have saved it were paid out, years earlier, as the buybacks that made everyone rich while the sun shone. The six hundred billion dollars is real. So is the reason it exists. And the bill, as it always does, will arrive later — addressed, as it always is, to the public.