The Relationship Between Exchange Rates and Commodity Prices

Over the last several years, the value of the Canadian Dollar (CAD) has been on an upward trend, greatly appreciating relative to the American Dollar. A rise in commodity pricesInterest rate fluctuationsInternational factors and speculation Many economic analysts believe that the rise in the value of the Canadian Dollar is due to a rise in commodity prices stemming from increased American demand for commodities. Canada exports a great deal of natural resources, such as natural gas and timber to the United States. Increased demand for those goods, all else being equal, causes the price of that good to rise and the quantity consumed of that good to go up. When Canadian companies sell more goods at a higher price to Americans, the Canadian dollar to gains in value relative to the U.S. dollar, through one of two mechanisms: 1. Canadian Producers Sell to U.S. Buyers Who Pay in CAD This mechanism is quite straightforward. To make purchases in Canadian Dollars, American buyers must first sell American Dollars on the foreign exchange market in order to  buy Canadian Dollars. This action causes the number of American Dollars on the market to rise and the number of Canadian Dollars to fall. To keep the market in equilibrium, the value of the American Dollar must fall (to offset the larger quantity available) and the value of the Canadian Dollar must rise. 2. Canadian Producers Sell to U.S. Buyers Who Pay in USD This mechanism is only slightly more complicated. Canadian producers will often sell their products to Americans in exchange for American Dollars, as it is inconvenient for their customers to use foreign exchange markets. However, the Canadian producer will have to pay most of their expenses, such as employee wages, in Canadian Dollars. No problem; they sell the American Dollars they received from sales, and purchase Canadian Dollars. This then has the same effect as mechanism 1. Now that weve seen how the Canadian and American Dollars are linked to changes in commodity prices due to increased demand, next well see if the data matches the theory. How to Test the Theory One way to test our theory is to see if commodity prices and the exchange rate have been moving in tandem. If we find that they are not moving in tandem, or that they are completely unrelated, well know that changes in currency prices are not causing exchange rate fluctuations. If commodity prices and exchange rates do move together, the theory may still hold. In this  case, such correlation does not prove causation as there could be some other third factor causing exchange rates and commodity prices to move in the same direction. Though the existence of correlation between the two is the first step in uncovering evidence in support of the theory, on its own such a relationship simply does not disprove the theory. Canadas Commodity Price Index (CPI) In A Beginners Guide to Exchange Rates and the Foreign Exchange Market,  we learned that the Bank of Canada developed a Commodity Price Index (CPI), which tracks changes in the prices of commodities which Canada exports. The CPI can be broken down into three basic components, which are weighted to reflect the relative magnitude of those exports: Energy: 34.9%Food: 18.8%Industrial Materials: 46.3%(Metals 14.4%, Minerals 2.3%, Forest Products 29.6%) Lets take a look at the monthly exchange rate and Commodity Price Index data for 2002 and 2003 (24 months). The exchange rate data comes from the St. Louis Fed - FRED II and the CPI data is from The Bank of Canada. The CPI data has also been broken down into its three main components, so we can see if any one commodity group is a factor in the exchange rate fluctuations. The exchange rate and commodity price data for the 24 months can be seen at the bottom of this page. Increases in the Canadian Dollar and CPI The first thing to note is how the Canadian Dollar, the Commodity Price Index, and the 3 components of the index have all risen over the 2-year period. In percentage terms, we have the following increases: Canadian Dollar - Up 21.771%Commodity Price Index - Up 46.754%Energy - Up 100.232%Food - Up 13.682%Industrial Materials - Up 21.729% The Commodity Price Index has risen twice as fast as the Canadian Dollar. The bulk of this increase seems to be due to higher energy prices, most notably higher natural gas and crude oil prices. The price of food and industrial materials has also risen during this period, though not nearly as quickly as energy prices. Computing the Correlation Between Exchange Rates and CPI We can determine if these prices are moving together, by computing the correlation between the exchange rate and the various CPI factors. The economics glossary defines correlation in the following way: Two random variables are positively correlated if high values of one are likely to be associated with high values of the other. They are negatively correlated if high values of one are likely to be associated with low values of the other. Correlation coefficients are between -1 and 1, inclusive, by definition. They are greater than zero for positive correlation and less than zero for negative correlations. A correlation coefficient of 0.5 or 0.6 would indicate that the exchange rate and the commodity price index move in the same direction, whereas a low correlation, such as 0 or 0.1 would indicate that the two are unrelated. Keep in mind that our 24 months of data is a very limited sample, so we need to take these measures with a grain of salt. Correlation Coefficients for the 24 months of 2002-2003 Exch Rate Commodity Index .746Exch Rate Energy .193Exch Rate Food .825Exch Rate Ind Mat .883Energy Food .336Energy Ind Mat .169Food Ind Mat .600 We see that the Canadian-American exchange rate is very highly correlated with the Commodity Price Index over this period. This is strong evidence that increased commodity prices are causing a hike in the exchange rate. Interestingly enough, it appears that according to the correlation coefficients, rising energy prices have very little to do with the rise of the Canadian Dollar, but higher prices for food and industrial materials may be playing a big role. Energy prices hikes also do not correlate well with rises in food and industrial materials costs (.336 and .169 respectively), but food prices and industrial material prices do move in tandem (.600 correlation). For our theory to hold true, we need the rising prices to be caused by increased American spending on Canadian food and industrial materials. In the final section, well see if Americans are truly are buying more of these Canadian goods. Exchange Rate Data DATE 1 CDN = CPI Energy Food Ind. Mat Jan 02 0.63 89.7 82.1 92.5 94.9 Feb 02 0.63 91.7 85.3 92.6 96.7 Mar 02 0.63 99.8 103.6 91.9 100.0 Apr 02 0.63 102.3 113.8 89.4 98.1 May 02 0.65 103.3 116.6 90.8 97.5 Jun 02 0.65 100.3 109.5 90.7 96.6 Jul 02 0.65 101.0 109.7 94.3 96.7 Aug 02 0.64 101.8 114.5 96.3 93.6 Sep 02 0.63 105.1 123.2 99.8 92.1 Oct 02 0.63 107.2 129.5 99.6 91.7 Nov 02 0.64 104.2 122.4 98.9 91.2 Dec 02 0.64 111.2 140.0 97.8 92.7 Jan 03 0.65 118.0 157.0 97.0 94.2 Feb 03 0.66 133.9 194.5 98.5 98.2 Mar 03 0.68 122.7 165.0 99.5 97.2 Apr 03 0.69 115.2 143.8 99.4 98.0 May 03 0.72 119.0 151.1 102.1 99.4 Jun 03 0.74 122.9 16.9 102.6 103.0 Jul 03 0.72 118.7 146.1 101.9 103.0 Aug 03 0.72 120.6 147.2 101.8 106.2 Sep 03 0.73 118.4 135.0 102.6 111.2 Oct 03 0.76 119.6 139.9 103.7 109.5 Nov 03 0.76 121.3 139.7 107.1 111.9 Dec 03 0.76 131.6 164.3 105.1 115.5 Were Americans Buying More Canadian Commodities? Weve seen that the Canadian-American exchange rate and commodity prices, particularly the price of food and industrial materials, have moved in tandem over the last two years. If Americans are buying more Canadian food and industrial materials, then our explanation for the data makes sense. Increased American demand for these Canadian products would simultaneously cause an increase in the price of those products, and an increase in the value of the Canadian Dollar, at the expense of the American one. The Data Unfortunately, we have very limited data about the number of goods the American are importing, but what evidence we have looks promising. In The Trade Deficit and Exchange Rates, we looked at Canadian and American trade patterns. With data provided by the U.S. Census Bureau, we see that the U.S. dollar value of imports from Canada has actually went down from 2001 to 2002. In 2001, Americans imported $216 billion of Canadian goods, in 2002 that figure dropped to $209 billion. But by the first 11 months of 2003, the U.S. had already imported $206 billion in goods and services from Canada showing an increase year-over-year. What Does This Mean? One thing we have to remember, though, is that these are dollar values of imports. All this is telling us is that in terms of U.S. Dollars, Americans are spending slightly less on Canadian imports. Since both the value of the U.S. Dollar and the price of commodities has changed, we need to do some mathematics to find out if the Americans are importing more or fewer goods. For the sake of this exercise, we will assume the United States imports nothing but commodities from Canada. This assumption does not greatly affect the results, but it certainly makes the math much easier. Well consider 2 months year-over-year, October 2002 and October 2003, to show how the number of exports has increased significantly between these two years. U.S. Imports From Canada: October 2002 For the month of October 2002, the United States imported $19.0 billion of goods from Canada. The commodity price index for that month was 107.2. So if a unit of Canadian commodities cost $107.20 that month, the U.S. bought 177,238,805 units of commodities from Canada during that month. (177,238,805 $19B / $107.20) U.S. Imports From Canada: October 2003 For the month of October 2003, the United States imported $20.4 billion of goods from Canada. The commodity price index for that month was 119.6. So if a unit of Canadian commodities cost $119.60 that month, the U.S. bought 170,568,561 units of commodities from Canada during that month. (170,568,561 $20.4B / $119.60). Conclusions From this calculation, we see that the United States bought 3.7% fewer goods over this period, despite a price hike of 11.57%. From our primer on price elasticity of demand, we see that the price elasticity of demand for these goods is 0.3, meaning theyre very inelastic. From this we can conclude one of two things: The demand for these goods are not at all sensitive to price changes so American producers were willing to absorb the price hike.The demand for these goods at every price level increased (relative to former demand levels), but this effect was more than offset by the large jump in prices, so overall quantity purchased declined slightly. In my view, number 2 looks a lot more likely. During that period, the U.S. economy had been spurred by massive government deficit spending. Between the 3rd quarter of 2002 and the 3rd quarter of 2003, the U.S. Gross Domestic Product increased by 5.8%. This GDP growth indicates increased economic production, which would likely require increased use of raw materials such as timber. The evidence that increased demand for Canadian commodities has caused the rise in both commodity prices and the Canadian Dollar is strong, but not overwhelming.