Spitballing the Evolution of the Bank of Canada’s Balance Sheet and Rate Sensitivities in Trade Wars

• The BoC’s balance sheet is projected based on a key speech and our assumptions

• The BoC seeks to retain bragging rights on global balance sheet leadership

• The BoC’s balance sheet will shrink to just over 5% of NGDP

• The BoC will hold a dwindling share of outstanding GoC bonds into uncertainties

• Settlement balances will shrink to a higher floor than previously guided

• There was another reminder not to assume rates will go down in trade wars

• Markets shrugged, but the speech could come to be highly meaningful over time

Deputy Governor Gravelle’s speech on balance sheet management changed the goal posts somewhat and put a bit more meat on the bone in terms of the parameters around how the BoC will seek to achieve longer-run goals on the size and composition of its balance sheet and its role in the bond market and payments system. Maybe they sought to provide a bit more certainty about their framework into a period of heightened macroeconomic uncertainty.

Gravelle’s prior guidance that Quantitative Tightening will end by 2025H1 was repeated but without any firmer guidance around timing that could be any moment between now and June and before a large maturity of GoC bonds in September that would otherwise drain settlement balances.

THE PROJECTIONS

Charts 1–4 show our projections based upon assumptions that are explained later.

The BoC’s assets will continue to decline until reaching a point of stabilization next year and then returning to mild growth. In dollar terms, assets may end the decade at around a quarter trillion Canadian dollars. As a percentage of nominal GDP, assets may decline to equal about 6% and therefore return to pre-pandemic norms (chart 1).

The BoC’s GoC bond holdings will register very mild growth over the back half of this decade when gross purchases return net or maturities of existing holdings (chart 2). Bond holdings will be around double the just over C$75B holdings that existed before the pandemic. Given years of deficit-financed bond issuance under the Fall Economic Statement’s rosy scenarios what this means is that the BoC’s share of GoC bonds outstanding will be in steady decline. That is likely to make the yield curve progressively more vulnerable to shocks. Their share could sharply decline if issuance surges by tens of billions in trade wars and they hold off on QE with inflation risk mounting.

The BoC’s assets will gradually normalize to pre-pandemic inclusion of bonds, bills and repos (chart 3).

Chart 4 shows what may evolve to be the composition of the BoC’s liabilities by 2030. Notes in circulation will dominate followed by settlement balances.

DON’T COUNT ON RATE CUTS

The audience Q&A was among the best I’ve seen in one of these events in quite a while. Maybe every BoC speech should be delivered before VersaFi group (formerly "Women in Capital Markets").

The money question for rate watchers was popped right off the bat. When asked if a trade war began will the BoC think about QE as a support mechanism, Gravelle said:

“You'll see a lot of our analysis in a couple of weeks. It's really hard to talk about scenarios. It depends on three key things. The size of the US tariffs and the breadth. If Canada retaliates. What do both countries do with the tariff income. A tariff would have a detrimental impact on our growth. Retaliation will have an offsetting effect through an inflation impact at the same time that we have a slowdown in the economy. That puts the central bank in a very challenging place. At the end of the day our mandate will still be 2% inflation whatever comes our way.”

That means one shouldn’t assume that the BoC will either cut rates or introduce QE. There is a very high bar set against going the QE route again at this central bank.

Reinforcing this are two other points. One is to recall that the BoC hiked despite a nascent recovery in demand as inflation soared significantly due to supply chain turmoil in the pandemic. The folks who said they wouldn’t hike in that environment got their forecasts wrong.

Second is to recall Governor Macklem's speech on trade and supply chain risks last September:

"But even with a better understanding and better information, trade disruptions may mean larger deviations of inflation from the 2% target. Supply shocks present central banks with a difficult trade-off—monetary policy can’t stabilize growth and inflation at the same time. That means we have to focus on risk management, balancing the upside risks to inflation with the downside risks to economic growth.

Most importantly, we must avoid adding uncertainty to an already uncertain environment. That means ensuring inflation is low, stable and predictable even as global trade is being rewired, recast and redirected."

Ergo, 2% inflation is job one. In a modestly sized open economy with a flexible exchange rate and high import propensities, tariff retaliation would be likely to motivate the Bank of Canada to offset inflationary pressures through higher borrowing costs unless there is no retaliation (chart 5). 

BALANCE SHEET ASSUMPTIONS

The assumptions to the earlier projections are based in part on Gravelle’s guidance but also some assumptions including on how this year’s risks to the economic outlook may affect things.

In a sense, it’s like an elementary school math problem in which the BoC is imposing three constraints on key components that play a big role in the outlook for normalizing its balance sheet by 2030 and we have to solve for the outcomes.

• First, they wish to target settlement balances equal to a range of $50–70 billion going forward. This is higher than the previous target range of $20–60B because they figure that precautionary holdings of reserves by depositing financial institutions are higher than previously judged which could reflect several sources of uncertainty toward markets, liquidity and the economy. Raising the range is consistent with my reaction when they first announced the targets. Included in this target range is an estimated $20–30 billion of settlement balances needed to support the payments’ systems needs. We can use this range as a plug on the balance sheet and adjust other components as needed and subject to the other assumptions.
• Second, they seek to normalize the balance sheet by having the stock of GoC bonds outstanding on their balance sheet roughly approximate currency in circulation in a more normalized setting.
• Third, to more closely match floating rate exposures they seek to have holdings of Treasury bills and repos roughly equal settlement balances, or Members of Payments Association deposits that are deposited by financial institutions at the central bank (the rough equivalent to reserves US banks hold at the Federal Reserve).

Additional guidance included a preference toward ramping up holdings of repos first when QT ends in 2025H1, and then t-bills by Q4 of this year while holding off on gross purchases of GoC bonds until sometime thereafter but no earlier than late 2026. Until then, maturities will continue to draw down the BoC’s GoC bond holdings.

Marry to all of that a few more assumptions of my own:

• that there is modest nominal GDP growth over the projection period. This will drive growth in the amount of currency in circulation to keep the wheels greased. The assumptions for NGDP growth and how they translate to growth in currency in circulation given payments innovation are both debatable.
• That the Government of Canada’s deposits held at the BoC will double in the near-term as tens of billions of deficit-financed stimulus proceeds get deposited before being deployed. This too is contestable; maybe there won’t be a need for any or much stimulus if my fears around tariffs subside, or maybe there will have to be an awful lot more (they went up by about C$160B in the pandemic).
• That markets will function reasonably well throughout the projection period and so emergency support programs are unlikely.
• That there will not be a need for outright Quantitative Easing that would return the BoC to the aggressive balance sheet management practices introduced at the start of the pandemic. There could be numerous motives to introducing another QE program in light of the fact that the BoC’s initial reasons evolved over time.

Any or all of these assumptions may be invalidated by what we experience over time.

THE BOC EXPECTS LOSSES TO DISAPPEAR

When asked during the Q&A about the implications of normalizing balance sheet for the BoC's carry profile (meaning their profit and loss and related effects on the balance sheet) Gravelle said:

“We bought bonds and earned interest [in the pandemic], we grew reserves and had to pay interest rate on that. As we raised our policy rate we raised the rate we had to pay on our deposits and so for a while we were losing money because the fixed rate on our bonds was lower than what we paid. So we lost up to $9B. Our forecast now is based on the market expectation of the policy rate will cure that loss by the end of 2029.”

That means that the roughly C$8.7 billion equity hit on their balance sheet should gradually disappear.

WHY SPEAK NOW?

Why deliver a speech on balance sheet management when it normally is delivered annually a couple of months from now or so and ahead of what could be a key week for Canada? One reason may have been to set expectations for the framework into such uncertainties both on rates and balance sheet management. Another was to explain that while they not fusses about the deviation of market rates from their policy rate (chart 6) and have not changed their reasons for why this persists, they wanted to refresh the framework ahead of large bond maturities on their balance sheet (chart 7) and given fresh thinking on optimal settlement balances. Plus, the BoC has led other central banks in and out of QE and so this is extending their claim to bragging rights.

OTHER DETAILS

I would encourage interest readers to review the whole speech for further details. One is that when asked how the BoC would balance demand between off- and on-the-run bonds, Gravelle showed no preference:

“We'll want to buy that mix of assets I spoke to. We're buying government bonds in the secondary market to have as minimal an impact as possible. We're not aiming for a specific level of liquidity. We're aiming to improve market liquidity. We're not aiming to one or the other type of bond.”