Only one degree of freedom per Fixed Announcement Date

The Bank of Canada's announcement today, raising the overnight rate target from 0.25% to 0.5%, wasn't a surprise. The surprise was the kicker at the end:

"Given the considerable uncertainty surrounding the outlook, any further reduction of monetary stimulus would have to be weighed carefully against domestic and global economic developments."

Most people were expecting the Bank to continue to raise the overnight rate throughout the second half of the year. I interpret the Bank's statement as deliberately contradicting those expectations. And the forex market seemed to interpret the Bank the same way, since the Loonie fell over half a cent on the news (though it has since recovered most of that).

What the Bank says it expects it will do in the future matters as much, if not more, as what it actually does today.

It's tempting to think that this gives the Bank an extra degree of freedom or two when doing monetary policy. It can do something today; and it can say it will do something at the next Fixed Announcement Date, and the FAD after that... But that's not quite right. What it really means is that the Bank can borrow an extra degree of freedom from the future. But that "loan" must be repaid, by doing what it had previously said it would do, even if it doesn't want to. On average, there is only one degree of freedom per FAD. If the Bank uses two degrees of freedom at this FAD, it must use zero degrees of freedom at some future FAD. Otherwise, like any borrower against the future, it will lose credibility and be unable to borrow in the future.

But the ability to borrow degrees of freedom from the future can be useful. If you need extra degrees of freedom at one time more than others, it can be a good idea to borrow against future repayment. That's sort of what the Bank did just over a year ago, when it made a "conditional commitment" to keep the overnight rate target at 0.25% until July 2010. It felt it needed monetary policy to be extra powerful at that time. I say "sort of", because it's not at all clear that the Bank really committed itself to do anything that it wouldn't have wanted to do in future anyway. The condition on the conditional commitment made it less than a full commitment.

"With monetary policy now operating at the effective lower bound for the overnight policy rate, it is appropriate to provide more explicit guidance than is usual regarding its future path so as to influence rates at longer maturities. Conditional on the outlook for inflation, the target overnight rate can be expected to remain at its current level until the end of the second quarter of 2010 in order to achieve the inflation target. The Bank will continue to provide such guidance in its scheduled interest rate announcements as long as the overnight rate is at the effective lower bound."

But telling people what you currently expect you will want to do in the future is not the same as making a commitment to doing something in the future. You are forecasting, not committing, your own future actions. The kicker at the end of today's announcement should be interpreted as a forecast, not a commitment. The Bank thinks that demand is weaker, and monetary policy will need to set lower rates, than the market thought the Bank thought. And since the market presumably believes the Bank is a better forecaster of the Bank's future actions than the market itself, at least over the shorter term, the Bank's forecast can influence the market.

Forecast of your own future actions, as opposed to commitments, don't take away degrees of freedom from the future. They don't stop you doing what you would want to do in the future. But they do make what you do in the future less of a surprise. And it's always the "surprise" or unexpected component of the Bank's actions that move markets.

For what it's worth, I think the Bank is right. Even as the Canadian economy returns to normal, in terms of the outlook for inflation, there is no reason to believe that interest rates will return to what they were in the last decade. The natural rate of interest is not some invariable constant written in stone. The US economy is still weak. The Eurozone is still very scary. There's a lot of fiscal tightening in prospect, some good  and some ill, but it's happening nevertheless. My call is for Canadian interest rates to stay lower over the next 10 years than they were over the last 10 years. Let's see.

 

Comments

[Kosta]

A very interesting way of formulating the choices on the fixed announcement date; I quite like how your description of Mark Carney adding an extra degree of freedom to his decision set. You identified one of the costs in his choice, that being that the BoC has to give back this extra degree of freedom on a future FAD.

But isn't there another cost? Specifically, don't the markets like having a predictable monetary policy absent of surprises. Carney and the BoC are deliberately being ambiguous about their future policy decisions, to give themselves more room to act in the future. But the cost is that the markets are less certain about Carney's future decisions. Perhaps this is appropriate, given that global macroeconomic prospects look uncertain, but don't decisions like this automatically decrease them markets confidence in Carney and the BoC?

[Nick Rowe]

Kosta: Thanks!

I'm still getting my head fully around this idea.

Suppose, for example, that the Bank permanently borrowed one degree of freedom from the future, and kept rolling over the loan. It announces at this FAD what the overnight rate target will be at the next FAD. So changes in the overnight rate are never a surprise, at anything less than a 6.5 week horizon. But markets can still be surprised about the Bank's announcement of the future overnight rate.

Half of my mind thinks it would be silly for the Bank to do this. It's equivalent to throwing away the last 6.5 weeks of information. But the other half of my mind says it won't really make any difference, since the overnight rate 6.5 to 13 weeks in the future is probably about as important as the overnight rate 0 to 6.5 weeks in the future.

My brain hurts.

[Stephen Gordon]

I know just the doctor for that.

[Nick Rowe]

Exactly the sketch I had in mind, Stephen!

(I once heard Kevin Dowd say it in a Carleton seminar once, after answering a particularly difficult question!)

[Determinant]

Weren't the 2000's the decade with the lowest average nominal rates already? So are you predicting 3% rates for the 2010's then?

[RSJ]

I wonder how important these announcements are. I think as long as the credit markets know the relative priorities and rulesets of the CB, they can pretty much predict what will happen -- i.e. perhaps the degrees of freedom are actually much less.

Imagine a scenario where the CB announces that it will follow a certain reaction algorithm, in which case it only has freedom when there is ambiguity in the data. The CB then randomly picks some course of action to resolve the ambiguity. That creates entropy, or uncertainty equal to a small white noise term.

On the other hand, suppose the CB only made one decision each year, but it was completely random in both direction and magnitude.

I claim the second scenario would have more degrees of freedom and more uncertainty for the financial markets. And given the already inherent ambiguity in investing, I can certainly see how the white noise ambiguity of the first scenario is irrelevant to investment decisions.

Maybe the real informational content is to show shifting priorities, or possible changes in the ruleset.

[westslope]

Isn't the BoC simply saying that rates hikes will be slow and ponderous?

As for future interest rates, assuming the inflation target and range do not change, any prediction of lower rates implies that real rates will be lower. Real rates will be lower only if inflation expectations are low and firmly anchored, capital is readily available, and perceived risks of catastrophic events remain low and decrease.

In the meantime, the lead western super-power--the USA--continues to stumble from one hyper-vigilant reaction to another. The USA has lost all credibility on the issue of nuclear proliferation. Nuclear and ambitious colonial power Israel shows no signs of slowing down. There continues to be widespread support in North America for Israel using terrorist tactics to consolidate control over lands taken in the 1967 war of colonial conquest. (Stephen Harper simply reflects his constituents' views on these subjects.) People in North America have no understanding why many in Europe oppose Israeli colonialism. I believe the willingness to kill civilians for land and water will continue to be high.

So will the spectre of regional nuclear war push real interest rates up? Or can western nations continue to kill and take with impunity?

[Scott Sumner]

Nick, I think you will be right about interest rates. But here's the tougher forecast: Are we looking at a Japanese-style more than decade long period of sub-one percent interest rates? Will Keynes finally be right? I'd say there's at least a 25% chance of that in the US. I think it is a bit less likely in Canada.