Convoluted thoughts about regulatory frameworks

[fivemack]

Halifax will sell me a 3-month bond paying 6.85%, a 6-month bond paying 6.5%, a 1-year bond paying 6.45%, a 3-year bond paying 6.4%, or a 5-year bond paying 6.2%.

I thought that longer terms were supposed to be rewarded by better interest rates; on the other hand, why shouldn't I jump at this?

If I assume that Halifax would rather borrow money on the money-markets, where LIBOR is 6.61%, than through the time-consuming process of assembling small sums from thousands of tiny savers, then I conclude that the money markets want to charge Halifax perceptibly more than 6.85%. I think that this means that they believe there is a chance of perceptibly more than 0.89% [(6.85 / 6.61) ^ (3/12)] that Halifax will be unable to repay.

On the other hand, I'm a small saver, and the regulatory framework likes small savers. If Lloyds lend Halifax fifteen million pounds, and Halifax goes bust, Lloyds get nothing back. If I lend Halifax fifteen thousand pounds, and Halifax goes bust, I get £13,700 back. So I should be more willing to lend Halifax money than Lloyds is - indeed, since the money is otherwise making 5.8%, I should lend it to Halifax if I believe the odds of Halifax still being around in mid-February are better than about 20:1 for.

This seems somehow counter-intuitive; on the other hand, I will go off to the local Halifax tomorrow with a chequebook. There is then the question of hedging: I wonder what odds the local Ladbrokes will offer on Halifax being bust by mid-February?

Comments

[del-c.livejournal.com]

I thought that longer terms were supposed to be rewarded by better interest rates

Usually they are. When they aren't, it's called an inverted yield curve, and it implies that the people offering those bonds see bad times ahead.

[beingjdc.livejournal.com]

http://www.yieldcurve.com/marketyieldcurve.asp

[fivemack.livejournal.com]

I wonder why that looks different from the yield curve I see at http://newsvote.bbc.co.uk/1/shared/fds/hi/business/market_data/gilt/default.stm ? I don't recall ever seeing the UK Gilts yield curve other than inverted.

I presume the decline in the really long terms is pure demography - though I got the impression that the long-term gilts are basically a regulatory accident to do with pension funds wanting risk-free investments of arbitrarily low return but maturing when people who've just opened a pension upon leaving university get round to retiring at 75.

[beingjdc.livejournal.com]

Dunno, is it significantly different? The pretty graph is only updated on Monday, so the BBC data may have shifted after the BoE's little aberration.

[angoel.livejournal.com]

Or that they're urgently in need of cash now (say due to a credit crunch), but don't believe that they'll have as great a need to borrow that money in the future.

[hsenag.livejournal.com]

Didn't the government temporarily up the FSCS compensation to 100% up to 35K?

Bear in mind that Nationwide are still offering 6.70% for 1 year. Also, unless you've already filled your ISA limit, there are ISAs on offer that work out as a better deal after taking tax off.

Remember that LIBOR is an average; I don't know what the notional risk-free rate is, but it'll be less than 6.61%.

I don't really understand this stuff despite working for a group most of whose members have it as their job to know it; but I don't think your exponentiation is correct - the risk of them being bust in 3 months is more than the risk of them being bust in 1 month^3.

That rate may well also have an element of them drawing in money that will then be likely to be be left with them afterwards.

[fivemack.livejournal.com]

Is there a disadvantage that I haven't noticed, save in slain trees, in having the umpteen current accounts with balances of one pound that are required to get these various advantageous savings instruments?

The FSCS site says that they will compensate 100% up to 35k.

Northern Rock deposits appear to be 100% guaranteed to infinity, even newly-opened ones; on the other hand, to pick a 6.71% one-year bond from NR over a 6.70% one-year bond from Nationwide indicates a level of risk-return preference which would predict that I'd spend all my weekends skydiving and be lackadaisical as to whether I remembered the parachute or not.

[hsenag.livejournal.com]

You'll probably get credit checked each time you open one, so don't do too many in a short space of time.

[naath.livejournal.com]

Some banks require that your current account gets your salary (or at least someone's salary) or pension paid into it in order to allow you to access the better savings accounts.

[htfb.livejournal.com]

There can be good reason for Halifax to pay its retail depositors more than it would pay a commercial creditor.

In particular, issuing a bond, Halifax does not have the redemption risk that you might ask for the money back before the term is up. This increases the proportion of your loan that it can immediately lend on and the money it can make; you are being compensated, not for the negligible [excess] risk of Halifax going bust [over Lloyds], but for depriving yourself of the chance to do something else with the money before February. There's also the converse to the principle that if you owe the bank $100, it's your problem, but if you owe $100m it's the bank's---small creditors have much less opportunity to cause trouble for the bank, and their benefit from diversification of their book may exceed the cost of maintaining the retail account.

The inverted yield curve is a sign of expected future Bank of England rates, more than a market belief directly about the real economy. (The rather anomalous inversion between 10 and 30 years out is a consequence of demand from pension funds for long cash instruments exceeding supply.)

[vicarage.livejournal.com]

Companies buy their way into markets by offering good rates, then slacking off and hoping no-one will notice. ING were slapped around in the papers for doing that, and I shifted my savings out. Perhaps Halifax feel they want to increase their market share at the moment.