RBS and Lloyds Bank are 84% and 43% owned by the state respectively. It seems the time has come for HM Treasury to work out how we unravel ourselves from this embarrassing state of quasi-nationalisation. These are the most talked about options:

1. Sell. A good old privatisation is surely the default option for HMT. One of the problems here is how do we know when to sell? If we get it wrong, we might lose out. Indeed, the experience of selling Northern Rock suggests its impossible to tell. Nevermind, Policy Exchange argue that “even if the loss runs into a few tens of billions from the initial purchase prices it could well be argued that it was price worth paying". This may be true. But it is highly unlikely that HMT will be content to sit back and take this hit on behalf of the taxpayer.

2. Free give away or mutualisation. Nadhim Zahawi MP has suggested we “simply give away the shares as soon as possible”. The problem here for HMT is that it has to take a £45 billion hit to the national accounts. When UKFI looked at Northern Rock mutualisation for HMT, it ruled it out on the basis that this “would involve gifting some of the existing shareholding to mutual members”. The Treasury doesn’t like giveaways.

3. A mixed model. Portman Capital, Stephen Williams MP and Policy Exchange have described models where some shares are given away for free, to all British adults on the electoral register, or to those who apply. But with a "floor price" (the level at which the shares were bought by the government) which will return proceeds of sales to the Treasury. Shareholders will only gain on profits above this floor. The problem here is that this model incentivises a quick sale once the floor price is breached. As Stephen Williams points out: “An essential element of the model is to minimise the incentive for an individual to immediately dispose of their shares…” But once the share price rises, this danger reappears.

So it seems that each plan is flawed in some way. Good luck HMT.

Meanwhile, other factors are at play. The Banking Standards Commission, Nigel Lawson, Mervyn King, and the Archbishop of Canterbury are interested in breaking up RBS. Labour wants a new regional banking model based on the German Sparkassen. More widely, some observers worry about the need to reconnect people with our financial institutions. Others are concerned about the absence of real competition in the market, that it’s too hard to switch accounts within a cosy cartel. John Kay wants to overcome short-termism in financial services.

But from the perspective of HMT, the real deal is simply how to get £45 billion back. The biggest flaw with the Williams and the Policy Exchange models is the downside risk remaining on the national accounts. Peston argues that even with a sell-off, the “prospect of the Treasury making a fat profit on the privatisation is probably slim to none.” If the banks aren’t worth very much, then whichever way you look at it, you can’t conjure value out of thin air (that’s how we got in this mess in the first place).

So here’s an idea for HMT to create the conditions to get its money back:

1. Break up RBS’s retail activity into smaller banks, with a strong regional identity.

2. Gift a new class of shares to local people in each area.

3. Crucially, these shares can pay a dividend on profits but cannot be sold for a generation, or are non-transferable.

4. HMT retains a conventional shareholding in each regional bank.

The combination of regional identity, potential financial gains and locking people in, creates long-term incentives for people to think about switching to these new institutions, to buy their products, to develop an attachment to them and a vested interest in their success. Whatever its flaws and naivety, this plan has one significant advantage over each of the others – it creates a new reason why we may one day see a successful institution - or institutions - which can be sold at a profit on behalf of the taxpayer.

At the other end of the economy to the lofty world of financial services is the community shop. Community-owned shops have a staggeringly low failure rate of just a few per cent. This is because of the creation of a virtuous circle of ownership and custom, where a shareholding strengthens the incentive to shop local.

Some might argue this flies in the face of homo economicus, promoting irrational, lazy and inefficient behaviour. But if it can be harnessed to create institutions we care about, a more diverse banking system, more competition and more long-term thinking, not to mention the small matter of £45 billion, then perhaps HMT should give it some thought.