What's an AAA rating worth?

Paul draws my attention to a daft question from Andrew Selous:

Does the Prime Minister agree that the United Kingdom’s retention of its triple A status, when France lost its triple A rating this week, shows that the UK retains the confidence of international markets because of the difficult but necessary decisions that we are taking?

I say it's a daft question because as Jonathan has repeatedly pointed out, low gilt yields owe more to fears about the weakness of the global economy than they do to "the confidence of international markets".

But let's look at this a different way. What is an AAA rating worth? Perhaps the cleanest simple measure comes from the US municipal bond market. Here, AAA-rated 10 year bonds yield 1.5%, whilst AA-rated ones yield 1.61% and A-raters yield 2.32%. This implies that a one-notch downgrade from AAA status would raise yields by only 0.11 percentage points; small wonder, then, that reaction to France's downgrade was so puny. However, a two-notch downgrade would add around 0.8 percentage points to gilt yields.

How much damage would this do to the economy? If we take the Bank's estimate (pdf) of the impact of QE as a guide, a 0.8 percentage point rise in gilt yields would reduce GDP by 1.6%; I suspect this is a maximum estimate. This is equivalent to 470,000 jobs, which is not much more than the fall in public sector employment since late 2009. 

However, such an impact is easily avoided. The Bank of England estimates that its first £200bn of QE reduced gilt yields by around one percentage point. This implies (with caveats) that the adverse effect on yields of a two-notch downgrade could be offset by another £160bn of QE.

On balance, I suspect that our AAA rating is worth something, but not very much, and it's benefit is probably not as great as the cost of securing it.

Put it this way. By 2014-15 the coalition will have tightened fiscal policy by 2.1% of GDP more than under Labour's plan (table 3.1 of this pdf). There are three conditions which, jointly, would justify this:

1. This is the minimum tightening necessary to prevent a two-notch downgrade, or worse.

2. The fiscal multiplier is low. If we assume a 0.8 per cent rise in yields would take 1.6% off GDP, then a multiplier of less than 0.76 (1.6 divided by 2.1) would mean that the tightening does less harm than the downgrade.

3. QE is for some reason infeasible or less effective in reducing yields than the Bank's estimate.

Personally, I suspect these conditions don't hold.Instead, there's a danger that our AAA rating today serves the same function as sterling did in the 50s, 60s, and 70s - as a national virility symbol, to which the economy is sacrificed.

Comments

[Luis Enrique]

I agree that low yields have more to do with the parlous state of everywhere else than admiration for UK fiscal discipline.

What could cause our bond yields to jump? I don't imagine we'd ever seriously be perceived as being at risk of choosing to default, but we might be regarded at risk of inflating our way out of trouble, in which case expected real returns on UK bonds could plummet causing nominal yields to rise (is that right?). I presume if that happens the rating agencies would slam the door after the horse has bolted.

What could we do to put ourselves in the set of basket case countries, rather than in the set of least-bad countries, which we currently reside in? The only thing I can think of is if we embarked on a really large "we are going to borrow and spend our way out of the mess" strategy, and it didn't work.

So I suppose the only thing that can be said of the Tory position is that at least we haven't done that. That is to say, there is at least one possible "bad" outcome that they have avoided, thanks to fiscal rectitude.

Of course the counter argument (which I think I agree with) is that if we did stop cutting and (horrors) even started splashing some money around, the multiplier is sufficiently high that we wouldn't actually need to borrow more. i.e. it would work.

[marc]

He didn't ask about gilt yields, he asked about ratings.

[gastro george]

A rating is almost irrelevant for a monetarily sovereign country. See Japan and the US.

[Tom]

"What could we do to put ourselves in the set of basket case countries, rather than in the set of least-bad countries, which we currently reside in? "

There's also 'declare you're going to cut the deficit, but in doing so also kill growth resulting find yourself with spiralling social costs, flat demand, less margin to cut/stimulate as time goes on, an insurgent population, increasingly nervous business being disinclined to inest as well as a big deficit'. I'd suggest we're well on the way to that one.

[Luis Enrique]

Tom - yes! after hitting 'post' I regretted not having added something like that. If they've avoided once possible bad outcome, they've put us in another one.

[chris]

@ marc - if anyone's daft enough to care about ratings other than because of their impact on yields, then they are guilty of an even worse form of the fetishism I deplore.

[Keith]

British politics seems to repeat the same errors as in the past on a predictable basis. Our obsession with confidence and impressing bankers is Philip Snowden of 1930- 31. As if the body politic was a very senile person with hard arteries unable to learn and repeating old habits no longer appropriate. Can we have a mutiny some where to derail the policy as in 1931?