Weirdness on bank taxes

It\’s odd, I find myself giving one and a half cheers to the IMF proposals for bank taxes. And yet I\’m sure that I\’m one of those who is supposed to oppose them.

The surprise from the banks has been palpable, and it will soon turn to bloodcurdling threats of the dire effects of such taxes (while the same lobby groups will simultaneously point out that consumers will anyway end up footing the bill in higher charges).

No, I realise that I don\’t shout loud enough (or have a tall enough soapbox) to be heard in the ivory towers whre Guardian editorials are composed. But I was certainly one of those shouting about the Robin Hood Tax and about how it would be consumers paying it, not banks.

But these IMF proposals do look very different to me.

Firstly, we\’ve the liabilities tax. Me, I support this. Yes, I\’m entirely aware of the moral hazard problems but we\’ve already got an admission that no big bank will be allowed to fail. Thus the Obama Admin suggestions of a tax to pay for the implicit insurance (it\’s already there, implicity, so yes, let\’s make it explicit and charge for it) makes perfect sense to me.

Note the detail, that it\’s on liabilities over and above capital….and liabilities over and above those liabilities which are already covered by such things as retail deposit insurance. In one sense it\’s a simple admission that banks without deposit insurance are subject to runs. The recent failures were runs in the wholesale markets….markets which don\’t have deposit insurance. The solution thus is (despite moral hazard) to have insurance which is charged for, as the FDIC does retail deposit insurance.

If this is passed on to consumers, well, who cares? The consumers are getting something of value from it: deposit insurance.

It\’s the second tax that has me puzzled. It\’s really a tax on gross value added (profits plus what is paid to staff), the corporate equivalent of how we calculate gross domestic product. I\’m not having all that much luck in getting a numerical example to work properly (and I\’m reasonably convinced that this is a product of my inability with numbers, not with my inability with the logical model) but I at least currently, believe that taxing GVA is likely to increase the bonus part of bankers\’s remuneration, not reduce it.

Let\’s say a bank makes $ 1 billion in GVA. Current split at someone like Goldman Sachs would be $500 million to the staff, wages and bonuses together, $500 million in profits (that\’s about right isn\’t it? 50% staff compensation in toto? Even if it isn\’t it doesn\’t change the logic.)

The tax is on the whole $1 billion. Now, if that $ billion is a number which will be static going intohhte future the bank doesn\’t care if the 50% is staff pay, a reasonably fixed cost, or bonuses, a much more variable cost. They\’ve still got to pay the same tax on it.

However, what if such profits are highly variable? Which, of course, they are. A rising bond market (falling yields) makes fortunes for bond market firms. All their trading stock, their market making stock, rises in value. Similarly, falling bond prices, rising yields, lose them fortunes for the opposite reason. OK, that\’s just an example. But investment bank profits are not a reliable and steady stream, to say the least.

OK, so next year GVA is $500 million. The GVA tax has to be paid on that $500 million whatever the split between staff, bonuses and profits. Or it\’s $2 billion, no matter. The bank is going to want to have as much of staff compensation in the variable part of their pay, not the fixed amount, to deal with this swing in GVA and the tax that must be paid on it.

So a GVA tax should increase the bonus portion of compensation and reduce the fixed.

Which isn\’t, I think, what people are actually aiming for, is it?