Investors aren’t stupid. As soon as Sir Keir Starmer said October’s Budget will be “painful” and that those “with the broadest shoulders should bear the heaviest burden”, shares in UK banks took a hammering.

For a government desperately rummaging between the cracks in the sofa for loose change to the deafening sound of pips squeaking, a windfall tax on banks seems like a no-brainer. Unfortunately that’s true in both possible interpretations of the phrase. 

Rachel Reeves said before the election that she didn’t want to hit the banks with higher taxes. But then again, she and Starmer said lots of things that appear to have been rendered void by the discovery of what Labour has calculated is a £22bn “black hole” in the public finances.

One former senior Whitehall official quoted by the Financial Times pointed out that banks have broad shoulders, have been making decent profits and no one likes them. True, true and thrice true. However, the unnamed official then added that a “sensibly crafted” levy could raise several billion pounds. 

And therein lies the rub; recent history is far from encouraging in this regard. Ask experts to name the worst tax in the UK right now and most would probably pick stamp duty. But, for those in the know, the bank levy would run it a close second. 

This was brought in following the financial crisis and taxes balance sheet liabilities at different rates according to their duration. The idea was to raise money, punish banks and discourage risky behaviour.

The trouble is, it’s based on a flawed (or, at least, incomplete) diagnosis of the causes of the financial crisis: most banks got into trouble because they had too many risky assets rather than too many short-term liabilities on their balance sheet. 

This means that the tax creates a perverse incentive. Imagine two banks borrow the same amount of money for the same length of time. The first uses it to invest in low-risk assets and earns a low return, the second invests in high-risk assets and earns a high return. The second bank will have a much lower effective tax rate despite it clearly being more at risk of going pop. 

What’s more, because the tax is on the balance sheets and not on profits, banks can in theory add the levy to their cost of lending, meaning it’s at least partly paid for by customers. And because it only applies to balance sheets of more than £20bn, it puts a glass ceiling on challenger banks and the UK branches of foreign banks. Needless to say, this hampers competition in the UK’s painfully sclerotic banking sector. 

Tax experts have long suggested it would be much easier to simply abolish the levy and increase the bank surcharge – which is an additional percentage the sector pays on top of corporation tax – back up to 8pc. 

Such a neat switcheroo would indeed be revenue neutral, remove unnecessary complexity, and have a better chance of promoting growth. However, there are two problems. The first is that “revenue neutral” isn’t going to butter many parsnips.

The second lies with the reason why Rishi Sunak cut the surcharge from 8pc to 3pc in the 2021 Budget. He was seeking to offset his increase to corporation tax. Without the surcharge reduction, UK banks would have been struggling under the highest tax burden anywhere in the rich world. Even with it, they still ended up paying more. 

There is a third option. This is the idea of “deposit tiering” – which basically means a reduction in the interest paid to lenders on their Bank of England reserves. The big four UK banks collectively earned £9.2bn on these deposits last year, a sum that is ultimately shouldered by the Treasury.