Wednesday, 17 September 2008

Sorting out the credit crunch (3)

The Lloyds-TSB/HBOS merger is of course another way of doing it...

The problem with banks and financial institutions, as I explained here and here, is the double- and treble counting of losses. They all know that there will be losses, i.e. mortgages at vast multiples of income secured on houses that are falling in value, but because the mortgages have been repackaged and sold on so many times, nobody in the chain knows who end up taking it on the chin, so we end up with half a dozen different counter-parties all fretting about the same underlying loss, and all facing tumbling share prices and credit rating downgrades (outside investors in turn don't know which parties in the chain will be worst hit).

In part 2, I suggested that the various counter-parties divvy up the loss between themselves and get on with their lives, but of course the horse trading might drag on a bit. Especially if you end up with people who don't really have a culture of negotiating sensibly and honourably.

So here's Plan B - if all the banks, hedge funds and Sovereign Wealth Funds etc in the whole world were to merge into one mega-bank, they can net off the intra-group assets and liabilities (counter-party risk within a group is effectively nil) and there wouldn't be a "lack of trust" issue. The underlying losses - defaults by mortgage borrowers is the same, that won't go away - but there would be no double- and treble counting.

There'd still be bickering over who gets how many shares and bonds in the combined entity in exchange for shares and bonds in the entities being taken over, but that can be taken step by step. Clearly, a single global bank is not a good idea from a competition point of view, but I guess that you'd reach the same result with four or five.

I doubt that the various Monopolies and Mergers Commissions or protectionist gummints around the world would agree with this, but hey, they are part of the problem, not the solution.

2 comments:

Anonymous said...

I think you are forgetting leverage.

In a 10% fractional reserve environment, a £1 injecton creates 1/0.1=£10 more money.

Loosing £1 from a 10% fractional reserve means that £10 is removed from the system.

On your £150,000 house the drop of £60,000 wipes out the £15,000 capital set aside by the 1st bank, wipes out the £13,500 lent by the 2nd bank, all the way to the 5th bank who make a loss, but just keep afloat.

Think in % terms, not absolute terms.

Of course, the leaveraging in the system at the moment is likely to be far higher than 10/1.

Mark Wadsworth said...

Anon, nice try, but you're looking at the wrong end.

£1 capital (net asset) = £10 new loans (asset) minus £9 new deposits/bonds (liability).

So if £10 disappears from the assets side, that doesn't wipe out £100. It wipes out £1 capital and £9 deposits/bonds. So the bank loses £1 and its creditors lose £9.

Let's assume that the whole £9 was lent by Bank 2, that in turn wipes out 90p capital and £8.10 deposits/loans.

And so on.

The total real loss is whatever the original debt write down is plus transaction costs. Sure, the total figures on each side will be enormous, but once you've netted them all off, the real loss remains the same.