The financial supermarket was declared dead almost upon birth. Sears sold off Discover to its progenitor, Phil Purcell. American Express unloaded what had been Shearson plus Lehman Brothers plus E.F. Hutton, of sainted memory, on Dick Fuld and Sandy Weill. Discover and Purcell went to Morgan Stanley, which booted both of them. Weill's Citigroup, the biggest mega of all, has been declared on the wrong side of history more often than Prince Charles (and is now shuffling those Shearson and Hutton brokerage assets to Morgan Stanley). Lehman (Fuld again) failed. The "death of the financial supermarket" is a media staple as ubiquitous as the "death of newspapers." And yet the megafirm ("financial supermarket" feels too '90s) remains a massive reality. True, nobody's buying stocks at Sears anymore, except Eddie Lampert. Goldman, Sachs & Co., which attracts more attention than the sword swallower at the circus, is a wholesale "bank," though it's gargantuan and diversified, including wealth management (rich people are people too). But megafirms persist, many straddling wholesale and retail banking like nose tackles. Whether rooted in commercial banking, like Bank of America or Citi, or investment banking, like Morgan Stanley and Goldman, or both, like J.P. Morgan Chase & Co., they occupy more space than ever.

Are we stuck with them for eternity? It often appears so. Despite bailouts, they have not been broken up. Despite Arianna Huffington and Occupy Wall Street, and polls that suggest popularity on par with politicians, customers have not fled to local banks offering free mints. They are flush enough to lobby vigorously, and they still receive a TBTF subsidy from markets convinced they'd be bailed out again. And yet, they are complex products of a dynamic environment. What is remarkable about this is how poorly we understand that environment. We prattle on like geniuses about reform, and yet we're shooting blind. This explains, I think, our mania for rules, regulations and structural solutions: Break them up; nationalize them; toss bankers in the hoosegow. Do something! And yet, while cranking out solutions, we have no idea how to handle deeper issues -- size, risk, liquidity, interconnectivity -- without inflaming cankers of unintended consequences. We promise to stay awake. We swear on new incentives. We fall back on markets, thumbing our Hayek.

We were surprised when firms went jumbo; we'll be puzzled when they shrink. How did we get here? It wasn't that Purcell or Weill shared some Joseph Smith-like vision in the woods. The environment nudged them down a path and then rewarded them. That remains the case, despite The New York Times' concern that Wall Street lacks billionaires. In the '80s, after the '70s near-death experience, the economy snapped back, and markets revived. Competition was sharpening internationally, technologically, financially. New themes gained purchase: entrepreneurs, risk takers, innovators over ponderous corporate statesmen. There were competitiveness concerns; the initial deregulatory steps were cheered as progressive. There was pent-up demand for performance, which dovetailed with a zeal for share appreciation and, at the center, shareholders. Markets grew and proliferated. Junk! Size gave way to speed. Turnover mushroomed. Corporate assets were not viewed as organic combinations but as mechanical aggregations -- and thus capable of rearrangement through M&A. The corporation was imagined as the sum of cash flows or contractual obligations. So were we.

Of course, we have only the sketchiest of ideas how this mix of determinants shapes megafinance, just as we're still sorting out causes of the crisis. We are singularly ill-equipped to do this; math, which long ago trumped history, takes us only so far. How do you measure the impact of, say, Steve Jobs in the '80s, or the effect of anti-Goldmanism today? The fluctuations of interest rates or stocks reflect deeper Zeitgeistian changes. Quirky feedback loops proliferate. How do you explain, not to say measure, the aversion to stocks between 1929 and, say, 1959? Today, we see the obvious: the Volcker Rule, Basel III, Dodd-Frank, JOBS. These may be incoherent, contradictory or ineffective, but they annoy, hamper, harass. What's the long-term consequence? Will the centrality of stocks, shareholders and earnings per share survive governance failures and revived interest in private ownership? What effect will fragmentation, an invitation to regulatory arbitrage, have? The Volcker Rule may be a mess, but complying with it will shrink trading profits. Will that alter delicate intrabank political balances? Evolution proceeds incrementally, along margins, beneath surfaces. We dream of transformations and revolutions; we'll get the slow, silent drip of change. The megabank fortress may last a thousand years. Likelier, there's water in the dungeons and termites in the timbers.