The evolution of the Bank of Canada's balance sheet

Jim Hamilton has a couple of recent posts ([1], [2]) at Econbrowser documenting the remarkable changes in the Fed's balance sheet during the financial crisis, so I decided to take take a closer look at the effects of the Bank of Canada's activities on its balance sheet.

The asset side seems straightforward enough, and is indeed pretty dull when compared to the Fed's kaleidoscope of asset types:

 
When the credit crunch hit, the Bank acted to increase liquidity by putting half of its holdings of T-bills back on the market, and it made available another $30b worth of liquidity by means of resale agreements. Since September, the Bank's balance sheet has increased by about a bit over 50%. (The Fed's has increased by more than 100%).

On to liabilities:

It would have been nice to come up with a Canadian counterpart to Jim Hamilton's striking graph of the jump in the US monetary base, but I couldn't. The reason for this is that Canada doesn't have a monetary base - at least, not the sort of monetary base that you see defined in the old textbooks. In 1994, the requirement that charted banks hold reserves in the form of deposits at the Bank of Canada was abandoned, so the usual definition - currency plus chartered bank reserves - lost its meaning.

As Jim Hamilton notes, most of the expansion of the Fed's balance sheet has been made possible by an increase in reserves. For now, the banks are content to leave those reserves idle, now that the Fed is paying interest on them. But the possibility that they may choose to redeem those reserves at an inopportune moment could complicate matters for the Fed.

But once again, this seems to be another example of a US problem that Canada doesn't have. The Bank's balance sheet expansion was made possible by the federal government's borrowing and then depositing the proceeds in its account at the Bank of Canada. The Bank of Canada will be able to scale back its balance sheet when it wants, and at the speed it wants.

Comments

[Phil]

Thanks! Follow- up question: you say that the Bank of Canada can shrink its balance sheet when it wants and at the speed it wants. But doesn't that depend a lot on the nature of the securities purchased under resale agreements? If the Bank cares about the asset mix on its balance sheet and wants to return that asset mix to "normal", doesn't it largely have to sell the assets under repurchase agreements? And if those assets are "toxic", it seems like the ability to contract the balance sheet is more limited, particularly if those assets are losing value, so that they can only be sold at a loss (and therefore extract less liquidity than was created when they were bought).

[brendon]

I thought about doing something like this as well but as you put it, there isn't an interesting bank reserves story to tell due to the workings of Canadian monetary policy. Interesting nonetheless, I hadn't really thought about how the asset side expansion was essentially funded by government borrowing - so is this fiscal or monetary policy?

[Aaron Holtzman]

Can anyone explain the two multi-billion dollar blips in the amount of bank notes outstanding? At least for the time period in the plot, they both occur during the last weeks of the year.

[alou]

Its for Christmas shopping.

[Phillip Huggan]

M.Carney was musing last fall what that BoC should target in addition to inflation. There is a voxEU paper that suggests targetting financial "confusion" too. That is, when there is financial portfolio opaquity or perhaps overexposure to a recession in Canadian financial company balance sheetd, that it should be treated as inflationary or as a currency depreciation. Not that we've had that problem this time around. Obviously other ways to put the brakes on finacial "innovations" than BoC.

[Paul Friesen]

I don't understand the bit about chartered banks reserve requirements. I presume they still have reserve requirements, but what forms are they in? Do they still hold Bank of Canada deposits even if they don't have to?

[jp]

The way I see it, the possibility that the Federal government calls in deposits at the BoC and starts spending them will complicate things for the BoC. Given the huge budget and political pressure to spend on all sorts of projects + collapsing tax revenue, it's only a matter of time before the Feds take that money out.

Once this happens the liability labelled government deposits will convert into circulating currency, and if it isn't withdrawn/sterilized in some manner you'll see a large drop in the purchasing power of the c-dollar.

I can't see this sterilization being a simple thing. The BoC no longer has as many government securities to sell off in order to reduce currency in circulation. And who knows what securities have been purchased from the big banks listed under "securities purchased under resale". It may not be economically or politically feasible to let these latter operations expire. Unfortunately, letting notes in circulation expand significantly may end up being the path of least resistance.

Note that Federal government deposits are funding 35% of the BoC's liabilities. At no point in history has this number been this high. Historically the highest it got was 5% and and over the last decade 3% has been the norm. With so much depending on the federal government, I can't help but think that this puts central bank independence to the test.

[Stephen Gordon]

I guess I was assuming that the feds wouldn't be out to make the Bank's life difficult.

And I'm not sure that I see why they would withdraw those deposits outside of the context of a co-ordinated reduction of the Bank's balance sheet as the SPRA program is wound down. T-bill yields are so low that the costs of borrowing are much less than the costs of provoking the risk of a spike in inflation.

[Nick Rowe]

If I am reading this right, there's one massive difference between the BoC and the Fed in one regard: the Fed's liabilities include bank deposits (reserves) about equal in magnitude to the currency; the BoC's liabilities must include a very small amount of banks' deposits under "other liabilities".

And why are "Federal government deposits" so big on the BoC's liabilities graph?

Paul: Chartered banks still hold (some small) deposits at the BoC, even though they don't have to. They earn interest on those deposits, at 0.25% less than the overnight rate target. (They pay interest of 0.25% above the overnight rate target on negative deposits, or loans from the BoC.)

[jp]

Hi Nick. As best as I understand things, government deposits are a way of sterilizing recent SPRA's. The $30 billion or so in SPRAs would have sent $30 billion in new currency into the economy ie. augmented that component of liablities by $30 billion.

This would have been inflationary, so some sort counterbalancing or sterilization was necessary. The BoC can sterilize by adjusting the allocation of Federal government deposits, ie short term funds the government holds at various banks. By moving existing and new Federal deposits away from commercial banks and allocating them into the Feds' account at the BoC, currency is withdrawn from circulation and the effect of the SPRAs are cancelled out. Because the SPRAs have been so big, the requisite government deposits amassed at the BoC have been big too.

[Stephen Gordon]

That sounds right to me. And the fact that the feds were so helpful in sterilising the SPRA-induced increase in the balance sheet was what leads me to think that they won't be trying to make trouble for the Bank.

[Nick Rowe]

jp: that sounds right to me too. But also wrong. As you say, they seem to be using the old "drawdowns and redeposits" mechanism (God, I thought/hoped I would never have to teach that mechanism again!) to prevent the money supply increasing (or sterilise the SPRA, as you say).

But to avoid inflation!!!! Inflation is the last thing we want to avoid at the moment. It's deflation that scares me (especially given the deflationary expectations revealed in the BoC survey, or at least, expected inflation well below the 2% target). The BoC would probably phrase it as preventing the overnight rate falling below the current 0.50% target.

In short, it means the SPRA etc. was qualitative easing, not quantitative easing (using Willem Buiter's terminology). They wanted to change the composition of assets, but not expand the total liabilities to the public. (Government deposits at the BoC are hardly the "hot potato" money we need right now).

I think they ought not be sterilising.

[jp]

Stephen: Well, I have a feeling that the SPRA program is mainly about solvency support and capital replacement, not liquidity support, and therefore won't be easily unwound.

I quite liked your charts and I do hope you update your BoC balance sheets every couple of months or so and put them on your blog, like Hamilton does at Econobrowser. Very interesting to read and we'll be able to see how things pan out. Thanks.

Nick: your qualitative vs quantitative point is a good way to put it.

If the BoC didn't have a protective monopoly I would be selling all its liabilities as we speak, due to its ratty looking balance sheet. 35% of its funding from one short term lender with the option to call is a recipe for disaster, as are 40% of assets in unidentified private securities. Abitibi's balance sheet looks better than that.