The Bank of Canada and the exchange rate: the facts have changed, and so has its views

You will remember that the dollar went from 0.86 USD to 0.97 between July and October of last year, and that the question of what to do about the appreciating CAD took up a lot of our attention in the fall. (Sadly, much of my attention was diverted to how badly the issue was covered in the media: [1], [2], [3].) Back then, the Bank of Canada greeted this development with concern, and warned forex markets several times that it was prepared to intervene to slow the appreciation of the Canadian dollar.

After receding back to 0.93-0.94 USD level for a few months, the CAD is now trading in the 0.98-0.99 range, and people are talking of reaching parity and beyond within the upcoming weeks and months. But this time, the Bank isn't saying anything. Why not?

Commodity price movements are part of the story. Let's revisit the relationship between the CAD-USD exchange rate and the Bank of Canada's commodity price index. Here is a plot of the weekly data since 2000:

Comm_usx_2010_03
 

The exchange rate drifted above the usual range of variation last fall, but it appears to have moved back after a 10% increase in commodity prices. This re-alignment is likely one reason why the Bank is no longer talking publicly about an over-valued Canadian dollar.

The other is the inflation outlook. As Nick recently noted, exchange rate appreciation is a close (if not perfect) substitute for higher interest rates. An implication of this equivalence is that interest rates during the appreciation of the 2000s were generally lower than they were during the depreciation of the 1990s.

Here is the latest in my series of graphs of core inflation for various horizons:

Core_02_10

The recovery had barely begun in October, and not only was inflation below target, it was drifting down. This was most definitely not a situation that called for higher interest rates, and the Bank had every reason to be concerned when the forex market seemed determined to provide us with the equivalent effect in the form of an exchange rate appreciation.

These days, core inflation is at target and drifting higher. If the Bank hadn't already committed itself to hold interest rates until July, it would no doubt be thinking about starting to withdraw monetary stimulus very soon.

But if forex markets are willing to provide – in the form of an appreciation to parity and beyond – what works out to a reduction of monetary stimulus, then that suits the Bank's purposes fairly well.

For now, anyway.

9 comments

  1. Mike Moffatt's avatar
    Mike Moffatt · · Reply

    For what it’s worth, my exchange rate toy (MERT) predicts the value of the CDN$ to be 98.96 cents U.S. It’s currently at 98.34 cents. Given that interest rates are likely to go up first in Canada (rather than the U.S.), the Canadian dollar is undervalued given current oil prices.

  2. Stephen Gordon's avatar

    Interesting. Do you remember offhand what your model was saying in October? Was it saying that it was overvalued?

  3. Mike Moffatt's avatar
    Mike Moffatt · · Reply

    “Interesting. Do you remember offhand what your model was saying in October? Was it saying that it was overvalued?”
    Good question – I honestly can’t remember. The data is easy enough to come by, though. Predicted vs. Actual in 2009 would make for an interesting blog post.

  4. Kosta's avatar

    Thanks for the post Stephen, relating the inflation data to the exchange rate vs interest rates is very interesting.
    I want to quibble with the commodities story you’ve presented. While clearly CAD/USD is correlated to the price of commodities, I’m not so sure that this is because of Canadian dollar strength, as compared to US dollar weakness.
    If the increase in the price of commodities was related to increased strength in CAD, then it should be evident if one looks at other CAD/XXX currency pairs, such as the CAD/Euro exchange rate (particularly as Euro-land is a net exporter of commodities (or at least oil). I have a chart comparing CAD/Euro to both crude oil prices and the Dow Jones Commodity Index here http://special—-k.blogspot.com/2010/03/is-canadian-dollar-driven-by-price-of.html. If you take a look, you’ll note that there is no consistent correlation between crude oil prices and CAD/Euro (nor between commodity prices and CAD/Euro).
    To me this suggests that the relationship between commodity prices and CAD/USD is driven by USD, that is USD weakness increases both the price of commodities and CAD/USD in a proportionate manner (and keeps CAD/EUR relatively constant).
    This also suggests that looking at the relationship between commodity prices and CAD/USD doesn’t add a lot of extra information to the CAD/USD exchange rate. Rather it might be more informative to look at the US dollar index instead.
    On the other hand, and related to Nick’s point, if an increase in the exchange rate is to take the place of an increase in interest rates, does it matter if CAD strengthens or USD weakens? Maybe not to Canada, but it might to the US?

  5. JP Koning's avatar

    I think Kosta is right. The chart just correlates weakness in the US dollar to weakness in the US dollar. And when you try to correct for this, the correlation between the CAD and oil prices will be close to non-existent.
    A while ago I tried the same exercise with the Swedish Krona and Norwegian Krone. Norway a huge oil exporter, Sweden an importer. Should be like US/Canada, right? According to theory, the Norwegian Krone has got to be a petrol dollar. But no, there is no correlation whatsoever of SEK/NOK to oil prices. Nor is there a correlation between EUR/NOK and oil. After seeing that I stopped using commodity prices to understand x-rates. Fine in theory, bad in practice.

  6. Unknown's avatar

    I’ve been using commodity prices as a proxy for the terms of trade. The fit is pretty good.

  7. westslope's avatar
    westslope · · Reply

    I have to agree with Stephen that commodity prices appear closely correlated to the terms of trade. That certainly applies to the past decade as Stephen’s graph dramatically illustrates.
    Wait till base metal prices firm up a bit more. The flow of foreigners lending money to Canadian-based operations will increase dramatically. (As an aside, it might be an opportune time to start shifting retirement savings into US equity.)
    Incidentally, some pundits are predicting that the BoC will raise overnight rates before the US federal reserve raises federal fund rates.

  8. Stephen Gordon's avatar

    While clearly CAD/USD is correlated to the price of commodities, I’m not so sure that this is because of Canadian dollar strength, as compared to US dollar weakness.
    I’ve been thinking about this, and I’m not sure that this distinction really matters that much. To a reasonably good approximation, the US is the Rest of the World.

  9. Kosta's avatar

    Kosta: While clearly CAD/USD is correlated to the price of commodities, I’m not so sure that this is because of Canadian dollar strength, as compared to US dollar weakness.
    Stephen: I’ve been thinking about this, and I’m not sure that this distinction really matters that much. To a reasonably good approximation, the US is the Rest of the World.
    I mostly agree with you Stephen, from Canada’s exchange rate perspective the vast majority of trade is in CAD/USD. So it doesn’t really matter whether CAD appreciates or USD depreciates, the net effect is the same.
    Then when considering exchange rate appreciation in lieu of interest rate increases, it again doesn’t matter whether CAD appreciates or USD depreciates, but only that CAD/USD increases, at least at the first iteration.
    But at the second iteration, it might be important whether CAD increases or USD decreases. Knowing which currency is moving would be helpful in predicting the future exchange rate path.
    For instance, consider the recovery in Nick’s model http://worthwhile.typepad.com/worthwhile_canadian_initi/2010/03/the-division-of-labour-between-interest-rate-and-exchange-rate.html
    Consider the point where the Canadian economy has stopped outperforming the US economy and that a US recovery is imminent. In Nicks’ model, CAD/USD has appreciated and is about to start depreciating. Nick’s model, if I understand it correctly, calls for the BoC to raise interest rates which would stop the depreciation of CAD/USD in addition to preventing the economy from overheating.
    But let’s think about why CAD/USD has appreciated to this point? Potentially it is because commodity prices have increased, increasing Canada’s terms of trade, which increases CAD. Alternatively, it’s because USD has depreciated as the US economic recovery has lagged.
    Now if the US economy now recovers, it is likely that commodity prices will by buoyed by the additional economic activity coming from south of the border. It is also very likely that Canada’s terms of trade will increase as the US imports more goods from Canada. It is also very likely that USD will appreciate as the US recovery takes hold.
    If CAD/USD is driven by commodity prices and the terms of trade, then one would expect the currency pair to appreciate as the US economy recovers and demand for Canadian commodities, particularly oil, increases. But if CAD/USD is driven by the US dollar, then one would expect the currency pair to depreciate as the US economy recovers.
    According to Nick’s model, the expected and actual movement of the currency pair should affect the BoC decisions on interest rates. It could be important to assess whether CAD/USD will follow commodity prices as the US economy recovers.

Leave a reply to JP Koning Cancel reply