Since Antony Jenkins lost his job at Barclays last week, numerous hypotheses have been offered to explain his sudden departure.

Some say the bank’s erstwhile chief executive simply was not tough or talented enough for the post. It is claimed he lacked a decisive plan to cut the bank’s costs. Being a bit of a retail banking wonk, he could not get on with the group’s investment bankers.

But there’s another, deeper reason Mr Jenkins may have turned out to be surplus to requirements: Barclays simply is not that interested in running a retail bank.

That might seem an odd thing to say. The group is about to do the splits because of post-crisis legislation requiring UK banks to put their retail arms into new subsidiaries, that are “ringfenced” from the rest of the business.

Barclays owns one of the UK’s largest high street banking chains, as well as big mortgage and credit card operations. Safe within the ringfence, these businesses should be the crown jewels in the post-split group. Meanwhile, the “casino” side — principally an investment bank built up over two decades by the group’s last boss Bob Diamond — faces being remorselessly shrunk, or even sold.

Seems a rational plan? Well, apparently Barclays may have other ideas. The investment bank’s biggest problem is that capital is set to become more costly outside the ringfence. So, to magnify the charms of what used to be called Barclays Capital, the group wants to toss in not only the corporate banking business, but possibly the credit card and mortgage operations too.

That might leave the ringfenced bank as little more than a pool of insured deposits. Like the UK’s Trustee Savings Banks of old, this would invest in little other than gilts and perhaps some personal loans.

While Barclays was always expected to opt for a “narrow” ringfence, this approach would go further. It would help the group avoid turning the gleaming Barcap casino into what — to bulge-bracket eyes — would be little more than a provincial bookie’s. True, some retrenchment would still be needed, notably the dismantling of certain trading operations. But more of the Lehman franchise that Mr Diamond bought in 2008 would be left intact.

One might ask why — apart from a desire to preserve the pampered lifestyles of investment bankers — Barclays would defy the logic of ringfencing. Putting retail businesses such as consumer mortgages alongside Barcap may make it harder for those operations to compete with corralled competitors at other banks.

Well, there is a practical reason. The switcheroo would help the bank get around a deeper governance problem posed by the ringfence rule.

Ringfencing does not just force a bank to place its retail business in a new subsidiary. The unit has to be given its own independent board, potentially limiting the parent’s ability to drive strategy or draw dividends.

That might not matter too much if the ringfenced business is pretty much the whole group, as it is with Lloyds Bank, or a relatively trivial part, as with HSBC. But with Barclays, the split is pretty much down the middle. It raises the troubling possibility that half or more of the listed group might be beyond shareholders’ control.

A super narrow ringfence gets around this issue by stripping most of the profit out of the ringfenced side. It also leaves Barclays with a bigger investment bank. These operations may not be flourishing now. But, with others pulling capacity out of the market, there could be value at some point in being one of Europe’s “last men standing”.

It is easy to see why all this might appeal to the Barclays board — especially now Mr Jenkins is out. Such a structure would, however, pose real questions for the regulator.

Ringfencing was supposed to put banking into neat and hopefully safe boxes. While creating a super secure “narrow bank”, such a plan would subvert it. Indeed, you could say it involves bolting a Northern Rock on top (literally) of a Lehman. Funding mortgages through wholesale deposits did not end well for the former in 2007.

That need not be a show-stopper. Regulators have many more sophisticated and intrusive powers than eight years ago. They can put capital surcharges on risky business. In extremis, they can bail in the bondholders to recapitalise the bank. These tools should be enough to contain a super Barcap. But, if anything like it is to go ahead, the tools must not just be in place. They must be made to work, too.

Twitter: @Grepsul

Copyright The Financial Times Limited 2018. All rights reserved.