With a large number of headlines being spent discussing the Fed raising rates, and the possibility of the Bank of Canada cutting rates on Wednesday, I decided it was a good idea to examine the effects this will have on an important exchange rate, the CAD/USD. Etienne Bouchard and I built a model aimed at allowing investors to have an easier proxy for forecasting the CAD/USD exchange rate.
In my model, I used data starting in 2001 until present. I included four variables that I believe are the most important to this exchange rate: WTI spot price, the current account surplus or deficit as a percentage of GDP, the difference between the overnight rate set out by the Bank of Canada (BoC) and the Federal Funds Rate, and lastly the inflation differential between Canada and the United States. This has led to a regression equation equal to the following:
CAD/USD=β0+β1WTI+β2 CA%+β3 RateDiff+β4 InflationDiff+Ɛ
The fundamental reasoning behind why I think these variables will affect the CAD/USD are detailed below:
WTI Crude Oil Spot Price
Before beginning to theorize why the WTI Crude Oil Spot Price could have an impact on the Canadian dollar, it is important to note that the price is quoted in American dollars. This is one of the drivers behind the possibility that a link may exist between the price of oil and the Canadian dollar. When looking at the last 6 months, on average Canada exports approximately 115,660 thousands of barrels of oil to the United States per month (EIA, 2015). When they export this commodity, they get paid in American dollars; however, most of their expenses like their workers, rent, and so on need to be paid in Canadian dollars. Due to this, they need to exchange these American dollars in the open market for Canadian dollars.
By doing so, they are putting upward pressure on the Canadian exchange rate due to the increased demand they have for these dollars, and a downward pressure on the American dollar due to the increased supply. Since these 2 million barrels are sold per day, this is a large driver of the economic conditions within Canada, and something that can help keep the Canadian dollar stable. However, when the price of oil drops, Canadian oil companies are no longer getting as much money from their American counterparts that they are selling to. Consequently, they do not exchange as much in the open market. Due to this, there is no longer as much upward pressure on the Canadian dollar, which causes it to fall in value.
Current Account Deficit
A current account deficit occurs when a country imports more than it exports. This can affect a country's currency similar to the case explained above when looking at the WTI Crude Oil Spot Price. When a country is importing more than it is exporting, it is in the market purchasing foreign goods, more than it is selling domestic goods. To purchase these goods, Canadians must go into the market and sell Canadian dollars and purchase the foreign currency. When this is done, they are increasing the supply of Canadian dollars in the market, thus applying downward pressure on the exchange rate. When this is done repeatedly, it will affect the exchange rate. Therefore, we will aim to prove that when a country repeatedly has a current account deficit, it will cause the domestic currency to depreciate.
(Source: Y-charts Economic Indicators)
Interest Rate Differential
When the Bank of Canada cut the overnight rate in January, we saw the Canadian currency tumble by 5.778% in just 9 days. The reason that an interest rate cut can affect a country's currency is because of the way the health of the economy is perceived by foreign and domestic investors. More often than not, a central bank decides to cut the shortest-term interest rate, the overnight rate, in order to stimulate spending in the economy since there is a smaller opportunity cost for consumers to spend now instead of saving since the interest rate that consumers earn on their deposits is now smaller.
Furthermore, since all interest rates are based off of the overnight rate, it only makes sense that all other rates will fall in a similar pattern to the overnight rate. With all other rates falling, it makes borrowing cheaper for firms as well as consumers. With firms having the ability to borrow at a lower rate, they will be able to increase their investment spending, where they are able to finance this spending at a lower rate, which will benefit themselves and the economy in the future. Consumers follow the same mentality; they are able to have their borrowing rate on their mortgage and car loans be lower, which means they will be more willing to act in the present, instead of postponing consumption to the future. This gives the rationale behind the government dropping the interest rate to stimulate the economy.
However, when the interest rate differential widens, the domestic country will experience capital outflows since consumers can earn a higher interest rate in other countries. This will cause them to sell domestic currency in order to purchase foreign currency; due to the increase supply in the market place, the domestic currency will once again depreciate. This leads us to believe that it is relevant to our research to consider the difference in interest rates between Canada and the U.S. (the Overnight rate vs. the Fed Funds rate). As in economic theory, if interest rates are higher in Canada than in the United States (value above 1 in the graph below) there will be inflows of money by U.S. investors who are seeking higher returns. The opposite holds true, if the Fed Funds rate is higher than the Overnight Rate, outflows of money by Canadian investors will depreciate the $CAD.
(Source: YCharts Economic Indicators)
After correcting for autocorrelation, and heteroskedasticity, the following regression output table was obtained:
The impact of the independent variables on the dependent variable is clearly shown by examining the regression coefficients. With a coefficient of 0.0040, we can conclude that the price of WTI, has a fairly small impact on the CAD/USD exchange rate. This means that when WTI increases by 1$, the CAD/USD will increase by 0.0040 cents, appreciating the $CAD. We incorporated this variable due to the large drop in WTI that has been seen since October, but also due to the seemingly strong correlation between the two variables. That being said, while we obtain an extremely strong correlation coefficient of 0.91745, we are given empirical evidence that one correlation does not imply cause and effect relationships. That being said, if we get a 60% decrease in prices of WTI like in October and November, this will undoubtedly put pressure on the CAD/USD exchange rate.
The second explanatory variable is the current account (surplus or deficit) as a percentage of GDP. An increase of 1% of this value would lead to a 0.23 cent decrease in the CAD/USD exchange rate, depreciating the $CAD.
The third explanatory variable is the difference between the overnight rate and the fed funds rate (overnight fed funds). A 1% increase in the difference between these rates would lead to a 0.89 cent decrease in the CAD/USD.
The fourth and final independent variable is the difference in inflation between Canada and the U.S. Being a measure of the increase in prices, a higher inflation rate should negatively impact the value of $CAD. According to our regression coefficient of 0.51, this assumption holds true. For every 1% increase in the difference between the Canadian and US inflation rate, the CAD/USD will increase by 0.51 cents, appreciating the CAD.
Based on the R-squared value of 0.8523, we can conclude that the model created is able to explain 85.23% of the variation in the CAD/USD exchange rate, showing that the model fits the data nicely.
Using this equation, and having an opinion on where the above variables are moving, it is easier for you to forecast the CAD/USD exchange rate.
Based on the regression output, we have a model that is able to describe the data, and fundamentally fits our line of thinking. We are now able to use this model and equation to forecast an exchange rate based on the beliefs of certain investors, whether you think the BoC will cut rates on Wednesday, and if you believe the Fed will be raising rates in either September or December. Once again this model is built off of data since January 2001, and has been correcting for all data problems.
Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.
I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.