The straightforward strategy of buying companies that have recently been spun off from their parent has generated very

Embarrassing parents - swan duckling

good results. The forces of divestiture and conglomeratization eternally wrestling with each other like gravity and energy in the cosmos.

How about this for a rational explanation:

Spin-offs generally result in more “pure play” stocks which then become more accurately valued in the marketplace. When the pre-spun-off business are tied up in larger companies, that component of the parent co. share price tends to be undervalued (due to uncertainty, lack of transparency, their tendency to be used to fund weaker siblings etc.) as opposed to fairly or overvalued – thus the outperforming of both parent and especially child post spin-off.

The arguments for mega-banks like Citi are the semi-mythical appeal of “one stop shopping” to the consumer and economies of scale in general which only foreign banks would have if we aren’t allowed to build our own monsters. The brokerage arms of mega-banks now rely more on their parent co.’s (zero interest) deposit base for capital instead of the bond market so I doubt much potential would be unlocked by a spin-off for that particular part of the business. Jefferies, the last decent sized broker/dealer not married up with a deposit taking institution, just tied the knot with Leucadia largely for this very reason. Citi has more or less the same model as the other mega-banks, they just happen to suck at executing on it so it’s possible that the sum of their parts is more valuable than the current configuration only because of bad management. But generally bigger banks like JPM have higher IRRs than smaller banks these days and probably the only advantage smaller banks now have is a capability for better, less formulaic underwriting and more accurate, hands-on risk assessment – but that’s not moving the needle much right now.

TBTF banks do have natural advantages over smaller banks (the compliance burden alone!) but we’re better off when most S&L type banking is utilitarian, boring and pretty tightly regulated (a view expressed by Taleb, parts of Dodd-Frank and certainly many others). I would argue that TBTF is simply too big because of the systemic risk and moral hazard issues that are inseparable from that model (a view certainly not expressed in Dodd-Frank).