Analytical Studies Branch Research Paper Series
A Long-run Version of the Bank of Canada Commodity Price Index, 1870 to 2015

Acknowledgements

The author would like to express his gratitude to all those who helped in the creation of this paper. Discussions with Ilan Kolet during earlier work provided inspiration, while John Baldwin provided guidance and support, without which the project would not have been started. Don Coletti graciously shared data he had collected for an earlier project, which proved invaluable in constructing the index. Danny Leung, Mark Brown and the participants of an internal presentation at Statistics Canada provided valuable feedback on the early stages of the index’s creation. And, most recently, comments from Pierre Duguay and Ian Keay improved the analysis and exposition within the paper.

Abstract

This paper develops historical estimates of the Bank of Canada commodity price index (BCPI) and links them to modern estimates. Using a collection of historical data sources, it estimates weights and prices sufficiently consistently to merit the construction of long-run estimates that may be linked to the modern Fisher BCPI. The resulting long-run Fisher BCPI estimates show a repeated pattern of rapidly changing prices followed by periods of calm. This pattern is most pronounced since 1900.

Keywords: commodities, super cycles, price index

Executive summary

Canada is a trading nation that produces significant quantities of resource outputs. Consequently, the behaviour of resource prices that are important for Canada is germane to understanding the progress of real income growth and the prosperity of the country and the provinces. Demand and supply shocks or changes in monetary policy in international markets may exert significant influence on resource prices, and their fluctuations constitute an important avenue for the transmission of external shocks into the domestic economy.

The Bank of Canada commodity price index (BCPI) is used to analyze and better understand these types of links between global commodity markets and important macroeconomic variables and phenomena, such as inflation, the U.S.–Canada exchange rate, the allocation of resources in production, regional growth patterns, and real income growth from the terms-of-trade changes. Currently, the BCPI extends from 1972 to the present. It includes a range of commodities important for Canada and the provinces. But, given its relatively short time span, it cannot be used to examine the long-run effects of commodity prices or how current commodity cycles relate to previous ones.

Drawing on a collection of historical data, this paper fills these information gaps by providing estimates of production weights and prices suitable for creating a historical BCPI. Fisher chain indexes based on the historical data are calculated and linked to the BCPI to create a long-run narrative of the movements of commodity prices important for production in Canada that spans from 1870 to the present.

While historical values for the BCPI and its sub-indexes can be created using historical data, it is important to note that the methods employed create estimates. This occurs because the process of constructing the historical estimates relies on data sources that are not as coherent as modern data sources nor were the data collected for the purpose to which they are put in this paper. Although the historical estimates contain appropriate historical information for economic analysis, they may not be as representative as current estimates.

The historical calculations show a number of historical events, from the emergence of oil production to the electrification of the production process. Therefore, they contain information relevant for understanding processes related to emerging commodity production and structural change. They also contain significant information about the timing of pressure exerted by commodity prices on Canada. And, through time, they show a process in which a number of waves of price increase apply pressure for short periods of time, followed by periods of calm. This pattern is most pronounced since 1900.

1 Introduction

Canada is a trading nation that has a comparative advantage in the production of commodities (Coletti 1992). Consequently, the behaviour of resource prices that are important for Canada is germane to understanding the progress of the U.S.–Canada exchange rate (Cayen et al. 2010; Maier and DePratto 2008; Issa, Lafrance and Murray 2008; Bailliu and King 2005; Powell 2005; Amano and van Norden 1993), of real income growth (Brown and Macdonald 2015; Baldwin and Macdonald 2012; Macdonald 2010, 2008, 2007; Duguay 2006; Kohli 2006; Francis 2008; Coletti 1992), of prices for intermediate inputs into the production system (Hirsch 2003), and of inflation (Kolet and Macdonald 2010; Cheung 2009). Because commodities are traded on global markets, demand and supply shocks in these markets can exert significant influence on the allocation of production across industries and regions within Canada. The fluctuations of commodity prices constitute an important avenue for the transmission of external shocks into the domestic economy.

Currently, the Bank of Canada commodity price index (BCPI) extends from 1972 to the present. It includes a range of commodities whose production is important for Canada and the provinces (Kolet and Macdonald 2010). The commodities cover the major agricultural, mineral, energy, forestry and fisheries outputs from the Canadian economy. These outputs are, subsequently, important inputs into upstream production within the Canadian economy. They are also important as exports to world markets that facilitate the purchase of machinery and equipment and consumer goods or that finance repayment of debts to external creditors. However, given the relatively short time span of the index, it is of limited use for understanding long-run relationships between commodity prices and variables of interest, such as inflation or the exchange rate, or for facilitating econometric estimation of the parameters of these relationships. It is also of limited use for examining how current commodity cycles relate to previous ones.

This paper aims to fill this information gap by providing historical BCPI estimates that may be used for this type of inquiry. Estimates of production weights and prices suitable for linking with the BCPI are created by drawing on a collection of historical data. Fisher chain indexes based on the historical data are calculated and linked to the BCPI to create a narrative of the movements of commodity prices important for the Canadian economy that spans from 1870 to the present.

Because the goal is to extend the BCPI, only those commodities currently in the BCPI are included in the historical calculations. These commodities include major agricultural outputs, major mineral products, major forestry products, major energy products and fisheries products. Owing to the breadth of commodities in the BCPI, the decision to maintain the current commodity bundle over historical periods does not cause significant measurement issues. This is because over most of the history of Canada, the same types of products are important for production and for export, but their relative importance changes (Baldwin and Macdonald 2012). Consequently, applying a chain-index formula allows the weights for the index to be continuously updated, and this makes the index relevant for the entirety of the period from 1870 to the present.

Nevertheless, it is important to note that the further one goes into the past, the less representative the index is likely to be. This is the case because the BCPI mixes Canadian production values with global prices for commodities that are measured in U.S. dollars. At present, it can be argued that the U.S. price represents the world price and, thus, the price reflecting supply and demand conditions on global markets that affect Canadian producers.^{Note 1} However, for earlier periods, this may not be the case. In particular, prior to the Second World War, prices quoted in British pounds may be equally or more representative of world prices. Consequently, it is not clear a priori that continued use of U.S. prices is appropriate for gauging the effect of changes in global commodity markets on the Canadian economy. Moreover, in the period between 1870 and the First World War, an important convergence in commodities between the Old World and the New World occurred (O’rourke and Williamson 1994). This convergence raised the price of commodities in the United States relative to the United Kingdom. Therefore, it is expected that different price pressures could be found by using U.K. prices rather than U.S. prices during the early part of the sample period.

Here, the assumption is made that prices in U.S. dollars are sufficiently representative of the price changes relevant to Canadian producers in global commodity markets that they may be used to calculate historical index values. This assumption is derived from the fact that the convergence between New World and Old World prices stemmed from dramatic changes in transportation costs. As a result, the same type of convergent forces should be present for both the Canadian economy and the U.S. economy with respect to Europe. The applicability of this assumption is borne out in the results section below, which shows that historical BCPI estimates are consistent with other data sources.

The remainder of this paper is structured as follows. Section 2 describes the data sources used, while Section 3 describes the methodology. Section 4 describes the results both in terms of the production shares and the chained indexes themselves. Section 5 concludes the paper.

2 Data

The data collected here, which are used to compile estimates of the historical BCPI, are drawn from a variety of sources with different methodologies. Great effort has been made to find source data that are consistent through time and that are consistent with the composition of the underlying commodities.

In the subsections that follow, the source data for values and prices across the commodities used to calculate the historical BCPI are discussed. The discussion follows the major groups for which commodity data are collected: energy, metals and minerals, forestry, agriculture, and fisheries. The data are grouped in this fashion to match the structure of the BCPI and its sub-aggregate indexes, and because the source of historical data is often the same for similar types of commodities. In the discussion, the span of data differs from that in the tables. The data discussions indicate the full span of the data available while the tables indicate only the data employed in the index. Differences can occur, for example, because estimates for the value of production extend farther into the past than do the price series, or vice versa.

2.1 Energy

Data for historical energy prices are drawn from two sources (Table 1). Estimates of the U.S. price of oil are taken from the U.S. Energy Information Administration’s (EIA) U.S. Crude Oil First Purchase Price (EIA 2017), measured in U.S. dollars per barrel. The EIA provides a time series spanning 1870 to 1972. For coal and natural gas, prices are taken from Manthy (1978). The coal price corresponds to U.S. dollars per ton of anthracite. The natural gas price is in U.S. cents per million cubic feet. The coal and natural gas prices cover the 1870-to-1972 and 1919-to-1972 periods, respectively.

Data on production values for energy are drawn from two sources. Production values from 1961 to 1972 are taken from the input–output series used in the currently published BCPI (Kolet and Macdonald 2010). Historical production values are taken from the Historical Statistics of Canada (Statistics Canada n.d.a) for the three energy commodities.

Values for coal extend from 1870 to 1961. Values for 1871 to 1873 are estimated based on interpolated estimates of the price per ton of coal from Series Q1-5 of Historical Statistics of Canada multiplied by the tons of production from Series Q6-12 (Quirin n.d.a). The interpolation is applied either because of missing data points or because of unexplained outliers in Series Q6–12 in the years 1871, 1872 and 1873. In addition, the value for coal production in 1945 contains a typo. The average of 1944 and 1946 is used in its place.

Values for natural gas span 1892 to 1962, while values for crude oil span 1870 to 1962. Oil production values from the Historical Statistics of Canada (Quirin, n.d.b) span 1886 to 1962. For the years 1870 to 1885, the volume of production (thousands of barrels per day) is multiplied by the price per barrel from the EIA to create a nominal dollar projector series. The growth rate of this series is then used as a basis to back-cast the historical growth rates for the value of production. However, unlike other series where the projector is taken “as is,” the variance of the projector series for oil is larger than the variance of the production series. As a result, a regression is used to predict the production-value growth rate based on the growth rates of the projector.

2.2 Metals and minerals

Historical estimates of prices for metals and minerals are drawn from three sources (Table 2). Gold, silver and nickel prices are taken from the data collected for Coletti (1992). Prices for copper, zinc and lead are taken from Manthy (1978), while prices for aluminum and potash are taken from the U.S. Geological Survey (Kelly and Matos 2014). Data for iron prices come from the U.S. Geological Survey for the period from 1900 to 1972. For the years 1870 to 1900, the pig iron price index from Manthy (1978) is used to back-cast the price. Although the pig iron price represents a more manufactured price for iron than the ore price from the U.S. Geological Survey, there exists a correlation in the growth rates of the two series for the period from 1901 to 1912 (0.73) reasonable enough to form a basis for historical linking.

Historical production values are derived from three sources. For the period from 1961 to 1973, production values are taken from Statistics Canada’s input–output system (Statistics Canada, n.d.c) and CANSIM table 152-0005 (Statistics Canada, n.d.d). For years prior to 1961, production values for gold, silver, nickel, copper, zinc, lead and iron are taken from the Historical Statistics of Canada (Quirin, n.d.a), Series P1-26, “Canadian production of principal metallic minerals, 1886 to 1975”. Aluminum is modeled based on exports from Series P27-58, “Canadian exports of principal metallic minerals, 1968 to 1977.” Estimates for potash production come from CANSIM table 152-0005 (Statistics Canada n.d.d) and 512-0002 (Statistics Canada n.d.e).

2.3 Forestry

Historical price estimates for forestry commodities are drawn from a number of sources (Table 3). Lumber prices are taken from Manthy (1978) for the years 1870 to 1973. Pulp prices are taken from Manthy (1978) for the years 1916 to 1973. To increase the range of the price data, Urquhart’s (1993) price for pulp is converted from Canadian dollars to U.S. dollars and used to back-cast the price series between 1875 and 1915. Prices for newsprint are back-cast using the Bureau of Labor Statistics (BLS) newsprint price index (BLS n.d.b) between 1947 and 1971. For the years 1890 to 1946, the National Bureau of Economic Research newsprint price (NBER n.d.a, b, c, d) is used to project prices historically.

Historical production values for forestry commodities are drawn from Statistics Canada’s input–output tables for the years 1961 to 1972 (Statistics Canada, n.d.c). Estimates from the Historical Statistics of Canada (Osborn n.d.) are used between 1908 and 1960. There is no overlapping period for lumber or newsprint values in 1961, so the levels from the input–output tables and the Historical Statistics of Canada (Osborn n.d.) are used as published. For the years before 1908, estimates of forestry production from Urquhart (1993) are used to back-cast the levels from the Historical Statistics of Canada (Osborn n.d.). For pulp, export prices and production values are used to interpolate for 1961 and production values are then used to back-cast to 1917. From 1880 to 1917, estimates from Urquhart (1993) are employed to back-cast production values.

2.4 Agriculture

Historical agriculture prices are predominantly derived from Manthy (1978) (Table 4). Prices for cattle only are used as a proxy for prices for cattle and calves. Price series for cattle, hogs, wheat, barley, corn and potatoes extend from 1870 to 1973. U.S. dollar prices for canola are derived from Statistics Canada data (Statistics Canada, n.d.b) by multiplying the average farm price in Canadian dollars by the U.S.–Canada exchange rate.

Historical values for agricultural production are derived from a number of sources: Statistics Canada ( n.d.b, c), the Historical Statistics of Canada (Trant n.d.), and Urquhart (1993).

Values for hogs and cattle and calves come from linking input–output estimates to estimates from the Historical Statistics of Canada (Trant n.d.) and Urquhart (1993). For each commodity, the growth rates of nominal estimates are used to back-cast the level of production from the 1972 value used to calculate the current BCPI.

For wheat, an index of the nominal value of wheat production is calculated by multiplying wholesale market prices for wheat by the thousands of bushels produced from the Historical Statistics of Canada (Trant n.d.). The growth rates from the index values are then used to back-cast the value of wheat production found in the input–output tables (Statistics Canada n.d.c).

For barley, estimates of the number of tons produced are multiplied by the price per tonne, to estimate the nominal value of production from 1908 to 1973. Growth rates from nominal farm sales of barley from Urquhart (1993) are then used to back-cast production values to 1870.

Estimates for the value of canola production for the years 1949 to 1973 are taken directly from the Historical Statistics of Canada (Trant n.d.), while estimates for corn are derived by summing the value of production of corn for grain and corn for fodder. For each type of corn, the number of tonnes produced is multiplied by the average farm price per tonne.

Potato gross farm values are from the Historical Statistics of Canada (Trant n.d.) for 1908 to 1973. Growth rates from Urquhart (1993) are used to back-cast the 1870-to-1907 period.

2.5 Fisheries

The fisheries component of the historical index comprises only finfish (Table 5). Prices are based on the BLS finfish price index (BLS n.d.a), which is the same index used in the BCPI. The annual values span the period from 1947 to the present, and values from the earliest data to 1972 are used for the historical BCPI. Estimates for the value of finfish production come from the Historical Statistics of Canada Morse (n.d.) for the years 1906 to 1972.

3 Methodology

The methodology for the historical BCPI has two components, which are discussed in the following two subsections. The first outlines the choice of index used in the calculation. The second details the choices of commodities to include in the index in a particular year.

3.1 Index calculations

The BCPI uses a chained Fisher index formula (Fisher 1922). The index was chosen for the BCPI to ensure conformity with other indexes currently in use (Kolet and Macdonald 2010) and because of its preferred properties (Diewert 1992). For consistency, the historical estimate of the index that is presented here uses the same Fisher formula as the modern BCPI. For the i = 1,...,N commodities used in the historical index calculation, the Fisher price index between periods t and t–1 is defined as

$$BCP{I}_{t/t-1}\equiv {\left(L{P}_{t/t-1}\times P{P}_{t/t-1}\right)}_{}^{1/2},\text{(1)}$$

Where LP is the Laspeyres price index and PP is the Paasche price index.

The Laspeyres index holds quantities fixed in the first period and allows prices to change. It is defined as

$$L{P}_{t/t-1}\equiv {\displaystyle \sum \frac{p_t{q}_{t-1}}{p_{t-1}{q}_{t-1}}}\text{(2)}$$

The Paasche index holds quantities fixed in the second period and allows prices to change. It is defined as

$$P{P}_{t/t-1}\equiv {\displaystyle \sum \frac{p_t{q}_t}{p_{t-1}{q}_t}}\text{(3)}$$

The value for the Fisher index for each pair of years, beginning with 1870 and 1871 and concluding with 1971 and 1972, is calculated. The link values are then used to chain the BCPI historically back to 1870 as

$$\frac{1}{BCP{I}_{t/t-1}}\times BCP{I}_t^{LevelIndex}\text{(4)}$$

The same calculation is applied to each of the sub-indexes to create historically chained versions of the BCPI for energy, for metals and minerals, for forestry and for agriculture. The chain values for the BCPI for fisheries consist only of the price series for finfish from the BLS. Consequently, the finfish price index growth rates are used to historically chain the modern series without the need for a Fisher index calculation.

3.2 Sample size

Through time, the composition of Canada’s commodity output changes, as new commodities are extracted, grown or produced. As a result, the commodity basket used to calculate the historical BCPI evolves over time. The practice adopted here for the historical BCPI is to add new commodities as they enter into production. To include a new commodity in the index, it must be in production for two consecutive years, as this is the span necessary to calculate the index.

There are two other factors that cause additional complications for constructing the index.

The first relates to data constraints. To include a commodity in the index requires two consecutive years of data on its price and the value of its production. In some cases, prices exist for the whole 1870-to-1972 period, but data on the value of production are lacking, even though production is known to have occurred. This is particularly the case for mining. Data on the value of mineral production used in the index begin between 1886 and 1889 for most mineral components. Prior to 1886, there do not appear to be measures of mineral production even though historical events in mining can be traced back to at least 1604 (Cranstone 2002). As a consequence, particular periods show an increase in the number of commodities that does not correspond with an onset of production, but rather with the advent of the publication of data.

The second complicating factor is iron ore. Iron ore production in Canada goes through a number of periods of production followed by periods when production stops. This on-again, off-again feature of iron ore production means that it repeatedly enters and exits the commodity bundle used to calculate the historical BCPI.

The pattern of expansion and contraction in the number of commodities in the commodity bundle used for the historical BCPI means that one of two strategies can be used for the index calculation.

One strategy is to use only those commodities for which there are data for the entire 1870-to-1972 period. This strategy greatly reduces the available commodity bundle and is not adopted here.

The alternative strategy is to adjust the commodity bundle used to calculate the index as new commodities are developed or as new data sources become available. The historical BCPI and its sub-indexes for energy, for metals and minerals, for forestry and for agriculture therefore have commodity bundles that expand and contract as commodity production evolves. The sub-index for fisheries does not evolve, as it is based only on finfish prices from the BLS (BLS n.d.a). However, when linked with the BCPI, the bundle expands in 1972/1973, when shellfish is added.

The process used to adjust the commodity bundle is shown in Table 6 for the historical BCPI and its sub-aggregate indexes. The calculations begin with a restricted set of commodities in 1870, and then are adjusted through time. The early period, particularly around 1886, has a more rapid expansion of the commodity bundle, as data on metals and minerals become available. Additions of commodities such as aluminum, potash and canola occur once production begins. Iron ore production in Canada goes through a number of starts and stops, and iron is the only commodity to repeatedly enter and exit the commodity bundle.

4 Historical estimates

The data on the historical BCPI provide information on two dimensions pertinent for calculating the historical index: information on production shares and the index values. Both dimensions are presented below. In all cases, historical BCPI values are linked to the BCPI to create a long-run BCPI that covers the period from 1870 to 2015. The information presented on production shares below focuses only on the data employed to create the long-run BCPI. In some cases, additional data exist on production shares, but it is not reported as it does not contribute to the index calculation.

4.1 Bank of Canada commodity price index shares

One major advantage of using a chained index formula is that it constantly updates itself for changes in the production composition and, thus, includes information on the extent of these changes. In broad terms, the composition of the commodities included in the long-run BCPI, in terms of the value of production in Canadian dollars, has changed greatly over the 145 years for which data are available (Chart 1).

Initially, commodity production was dominated by agricultural output. This is the case both because the majority of commodity production was derived from agricultural products at the time, and because there is limited information available for the first 15 years, particularly about mining and mineral commodities. The other major source of commodity production in the early period was forestry.

Data table for Chart 1

Data table for Chart 1
Table summary
This table displays the results of Data table for Chart 1. The information is grouped by Year (appearing as row headers), Energy, Metals and minerals , Forestry, Agriculture and Fish, calculated using percent units of measure (appearing as column headers).

Year	Energy	Metals and minerals	Forestry	Agriculture	Fish
	percent
1870	1.15	0.00	10.53	88.31	0.00
1871	1.35	0.00	10.58	88.08	0.00
1872	1.64	0.00	11.23	87.14	0.00
1873	1.49	0.00	16.13	82.37	0.00
1874	0.94	0.00	13.04	86.02	0.00
1875	0.90	0.00	9.80	89.30	0.00
1876	1.40	0.00	10.22	88.37	0.00
1877	1.34	0.00	11.48	87.19	0.00
1878	1.27	0.00	9.60	89.13	0.00
1879	1.35	0.00	9.96	88.68	0.00
1880	1.36	0.00	11.45	87.19	0.00
1881	1.02	0.00	10.79	88.19	0.00
1882	1.22	0.00	13.59	85.19	0.00
1883	1.45	0.00	17.59	80.96	0.00
1884	1.70	0.00	16.03	82.27	0.00
1885	1.76	0.00	15.18	83.06	0.00
1886	1.94	1.05	16.66	80.35	0.00
1887	1.95	1.01	14.58	82.46	0.00
1888	2.22	1.33	15.55	80.90	0.00
1889	2.38	1.70	16.42	79.50	0.00
1890	3.04	1.90	18.96	76.10	0.00
1891	3.44	2.57	18.98	75.02	0.00
1892	3.92	2.32	20.44	73.33	0.00
1893	4.44	2.91	21.35	71.30	0.00
1894	4.23	2.78	19.59	73.40	0.00
1895	4.57	4.01	19.33	72.10	0.00
1896	4.95	5.40	21.32	68.33	0.00
1897	3.93	7.42	18.26	70.39	0.00
1898	3.60	9.89	14.95	71.55	0.00
1899	4.36	13.31	16.91	65.41	0.00
1900	5.29	16.66	18.25	59.79	0.00
1901	4.02	14.12	14.57	67.29	0.00
1902	4.19	10.74	14.96	70.11	0.00
1903	4.36	10.02	16.17	69.46	0.00
1904	4.72	8.99	16.48	69.80	0.00
1905	4.59	10.64	17.19	67.58	0.00
1906	4.56	10.93	15.74	68.77	0.00
1907	5.33	10.52	20.07	64.09	0.00
1908	5.36	10.33	18.41	65.91	0.00
1909	4.20	8.74	17.15	69.91	0.00
1910	5.44	10.54	20.30	63.72	0.00
1911	3.60	7.74	16.27	72.40	0.00
1912	4.89	10.39	16.02	68.71	0.00
1913	4.83	10.52	14.35	70.30	0.00
1914	4.24	9.32	14.40	72.04	0.00
1915	2.99	8.75	11.21	77.06	0.00
1916	2.91	9.82	10.67	76.60	0.00
1917	2.79	8.60	13.35	75.25	0.00
1918	3.34	8.46	14.65	73.55	0.00
1919	3.75	5.69	17.31	73.24	0.00
1920	4.95	5.38	23.91	65.75	0.00
1921	6.75	4.81	22.97	65.48	0.00
1922	5.93	5.82	22.17	66.09	0.00
1923	5.80	7.07	24.45	62.69	0.00
1924	4.15	8.36	24.40	63.09	0.00
1925	3.75	7.98	21.57	66.71	0.00
1926	4.40	7.97	23.10	64.53	0.00
1927	4.26	7.93	22.03	65.78	0.00
1928	4.23	8.81	22.52	64.44	0.00
1929	5.11	11.48	27.74	55.68	0.00
1930	5.80	13.80	29.76	50.65	0.00
1931	6.58	16.92	30.04	46.46	0.00
1932	6.49	17.10	23.86	52.55	0.00
1933	6.29	21.68	20.44	51.59	0.00
1934	5.31	23.45	22.73	48.51	0.00
1935	5.42	24.43	21.87	48.28	0.00
1936	4.76	24.22	22.59	48.43	0.00
1937	4.89	28.40	25.79	40.92	0.00
1938	5.77	30.64	22.72	40.86	0.00
1939	4.49	26.65	20.35	48.51	0.00
1940	4.32	26.64	24.48	44.55	0.00
1941	5.04	31.05	28.18	35.73	0.00
1942	4.20	24.89	22.87	48.03	0.00
1943	4.43	23.93	22.30	49.34	0.00
1944	4.09	19.74	22.88	53.29	0.00
1945	4.64	17.34	28.21	49.81	0.00
1946	4.16	13.39	32.70	49.76	0.00
1947	3.70	15.15	36.06	45.09	0.00
1948	4.81	15.46	34.20	45.53	0.00
1949	5.14	16.20	32.86	44.58	1.23
1950	5.05	16.27	33.29	44.01	1.38
1951	4.77	15.35	32.35	46.30	1.23
1952	5.32	15.25	30.78	47.46	1.20
1953	6.81	16.37	32.99	42.63	1.20
1954	9.50	19.07	35.77	34.20	1.47
1955	8.55	21.16	33.46	35.74	1.10
1956	10.11	22.78	32.38	33.48	1.25
1957	11.74	22.39	32.66	32.06	1.16
1958	10.49	19.75	31.51	36.74	1.51
1959	10.41	21.81	31.26	35.32	1.19
1960	10.27	23.48	31.73	33.52	1.00
1961	12.21	23.20	33.40	29.99	1.20
1962	12.50	22.02	30.31	33.92	1.25
1963	12.41	21.47	29.28	35.78	1.06
1964	12.59	22.75	30.35	33.20	1.11
1965	12.59	23.71	29.96	32.67	1.08
1966	12.26	22.60	28.18	35.80	1.16
1967	13.69	25.65	28.73	30.90	1.03
1968	13.85	26.02	29.91	29.12	1.10
1969	14.07	24.52	31.33	29.17	0.91
1970	15.79	27.83	29.40	25.98	1.01
1971	17.45	24.22	28.94	28.46	0.93
1972	17.78	20.77	30.20	29.86	1.39
1973	17.46	19.34	28.30	33.54	1.37
1974	21.85	19.63	27.51	30.03	0.98
1975	26.29	18.62	23.02	31.18	0.89
1976	27.77	17.61	25.97	27.53	1.12
1977	31.04	17.80	27.13	22.78	1.25
1978	30.37	16.17	27.16	24.76	1.55
1979	30.97	17.20	26.98	23.28	1.57
1980	31.31	20.83	23.58	23.22	1.05
1981	32.24	19.00	23.09	24.44	1.24
1982	39.27	14.04	20.79	24.59	1.31
1983	41.27	14.70	21.25	21.60	1.18
1984	41.74	16.26	21.80	19.15	1.06
1985	42.70	15.04	22.22	18.69	1.34
1986	30.36	18.87	28.02	21.01	1.73
1987	30.74	18.91	30.97	17.40	1.98
1988	26.14	23.25	31.44	17.27	1.90
1989	27.20	22.48	30.49	18.18	1.64
1990	31.80	19.51	27.91	19.07	1.70
1991	31.08	20.84	27.25	19.02	1.80
1992	31.80	20.04	27.71	18.69	1.76
1993	31.31	18.05	28.88	20.13	1.62
1994	30.07	18.53	31.23	18.41	1.76
1995	26.17	19.20	34.92	18.05	1.65
1996	30.88	17.74	30.18	19.84	1.35
1997	31.77	18.38	30.82	17.66	1.37
1998	29.35	19.42	31.10	18.63	1.50
1999	34.66	16.20	30.52	17.01	1.62
2000	45.49	14.10	25.54	13.48	1.39
2001	46.92	13.95	23.16	14.57	1.41
2002	45.69	14.76	23.88	14.10	1.56
2003	53.81	12.82	19.37	12.59	1.41
2004	54.49	13.87	18.68	11.70	1.26
2005	60.60	13.41	15.33	9.70	0.96
2006	59.15	17.00	13.80	9.22	0.83
2007	57.10	18.97	12.18	10.90	0.85
2008	62.52	16.58	8.66	11.54	0.70
2009	59.47	16.11	8.89	14.59	0.94
2010	59.05	18.52	9.26	12.30	0.87
2011	58.99	19.28	7.62	13.26	0.85
2012	58.92	18.02	7.69	14.49	0.88
2013	60.63	16.08	8.15	14.30	0.84
2014	63.41	15.41	7.79	12.42	0.97
Note: BCPI stands for Bank of Canada commodity price index. Source: Statistics Canada, author's calculations.

Over time, production shifted toward new commodity groups: first forestry, then metals and minerals, and finally energy.

Forestry production gained importance through the first 80 years of the sample, as new publishing industries in the United States, particularly newspapers, created increased demand for pulp and paper, and as new production techniques, transportation infrastructure and trade agreements permitted the increased demand to be met from Canadian supply. The share of forestry in long-run BCPI commodity production peaked just after the Second World War, after which the contribution from forestry products began a slow decline. Other commodities, particularly oil, gained prominence. Forestry experienced a brief resurgence in its share of long-run BCPI commodity production in the 1980s and 1990s, but fell away rapidly after 1999, as demand for paper products in the United States declined with the widespread market penetration of electronic media.

Data on mining and mineral commodities begin to enter the long-run index calculation in 1886. Initially, there was little impact on the composition of production from metals and minerals. By 1891, the value of mining had reached 2.57% of production and proceeded to rise rapidly as the Klondike Gold Rush in Yukon led to an increase in gold extraction. As the gold fields played out, the relative value of metal and mineral production declined, even as additional commodities were added to the long-run index. The next large adjustment came during the Great Depression. A combination of weak agricultural prices and, in the mid-1930s, an attempt by the Federal Reserve to jump-start inflation by revaluing gold helped increase the share of metal and mineral production in the index. After the recovery in agricultural prices, the share of production in the long-run BCPI commodity bundle derived from metals and minerals declined by around half. The share of production then expanded through the 1950s and 1960s to reach a high point in 1970, before tending to decline until the end of the period.

Energy products have the most dramatic effect on the overall composition of commodity production in the latter half of the 20th century, across all long-run BCPI subgroups. After rising modestly through the first 80 years, the share of energy production rose rapidly after 1947. Initially, the increased share of energy production in the BCPI commodity bundle coincided with a decline in the share of agricultural products. As time progressed, however, the share of production from energy began to erode the influence of forestry, and then metals and minerals.

4.2 Bank of Canada commodity price index subgroup shares by commodity

4.2.1 Energy

Across energy commodities, there is a transition from coal towards oil and natural gas over the 145 years presented (Chart 2).

At the start of the period, production is almost equally split between coal and oil. However, oil had a high price over the first 15 years of the sample that declined as production rose, so, while its production share is large, the quantity of oil produced was minor.^{Note 2} During this part of the sample period, the oil industry in Canada was centered in Ontario, where, for example, the area around Oil Springs experienced an oil boom in the mid-1880s. As the oil played out, production and its share of output in the BCPI declined until the 1920s, when oil production began around Norman Wells in the Northwest Territories. But, it was not until 1947, when oil was discovered at Leduc, Alberta, that oil became the major source of energy production in Canada.

After the expansion of oil production that occurred in Alberta and Saskatchewan in the 1950s and 1960s, the share of energy production from natural gas began to increase. This was a return to a 50-year trend that was interrupted by the rapid expansion of oil production in the middle of the century. During the resource boom of the 2000s, the trend was again interrupted as the share of oil in production increased rapidly.

Data table for Chart 2

Data table for Chart 2
Table summary
This table displays the results of Data table for Chart 2. The information is grouped by Year (appearing as row headers), Coal, Oil and Gas, calculated using percent units of measure (appearing as column headers).

Year	Coal	Oil	Gas
	percent
1870	56.39	43.61	0.00
1871	57.96	42.04	0.00
1872	64.09	35.91	0.00
1873	72.77	27.23	0.00
1874	82.95	17.05	0.00
1875	78.55	21.45	0.00
1876	65.03	34.97	0.00
1877	66.29	33.71	0.00
1878	76.96	23.04	0.00
1879	73.97	26.03	0.00
1880	82.55	17.45	0.00
1881	82.18	17.82	0.00
1882	85.61	14.39	0.00
1883	79.55	20.45	0.00
1884	82.22	17.78	0.00
1885	80.56	19.44	0.00
1886	83.47	16.53	0.00
1887	85.05	14.95	0.00
1888	82.57	17.43	0.00
1889	84.41	15.59	0.00
1890	82.89	17.11	0.00
1891	83.42	16.58	0.00
1892	82.39	17.61	0.00
1893	85.91	14.09	0.00
1894	86.57	13.43	0.00
1895	81.77	18.23	0.00
1896	81.90	18.10	0.00
1897	83.94	16.06	0.00
1898	84.87	15.13	0.00
1899	86.10	13.90	0.00
1900	89.64	10.36	0.00
1901	90.11	9.89	0.00
1902	92.05	7.95	0.00
1903	91.65	8.35	0.00
1904	92.78	7.22	0.00
1905	93.67	6.33	0.00
1906	94.93	5.07	0.00
1907	94.34	5.66	0.00
1908	96.05	3.95	0.00
1909	96.97	3.03	0.00
1910	98.29	1.71	0.00
1911	98.17	1.83	0.00
1912	98.69	1.31	0.00
1913	98.52	1.48	0.00
1914	98.60	1.40	0.00
1915	98.71	1.29	0.00
1916	98.61	1.39	0.00
1917	98.28	1.72	0.00
1918	97.80	2.20	0.00
1919	89.28	1.65	9.07
1920	92.32	1.28	6.40
1921	91.06	1.13	7.82
1922	88.71	1.16	10.13
1923	89.19	0.91	9.90
1924	85.39	1.14	13.48
1925	81.70	2.91	15.40
1926	83.26	2.54	14.20
1927	82.64	2.82	14.54
1928	81.48	3.63	14.90
1929	77.11	6.37	16.52
1930	71.57	9.53	18.90
1931	69.42	9.93	20.65
1932	69.40	7.94	22.66
1933	69.34	8.30	22.36
1934	68.06	9.22	22.72
1935	70.28	8.25	21.47
1936	68.17	7.86	23.97
1937	65.99	10.98	23.03
1938	62.77	16.79	20.45
1939	58.47	18.60	22.92
1940	59.00	19.25	21.75
1941	59.06	22.10	18.84
1942	58.76	22.81	18.43
1943	62.25	21.27	16.48
1944	65.38	20.15	14.47
1945	66.98	17.60	15.42
1946	66.78	18.59	14.63
1947	62.77	22.35	14.88
1948	60.89	27.83	11.28
1949	53.38	39.36	7.26
1950	47.59	48.81	3.60
1951	39.82	56.80	3.38
1952	35.44	60.64	3.91
1953	26.64	69.69	3.67
1954	19.00	77.79	3.21
1955	17.90	78.35	3.75
1956	14.10	82.58	3.32
1957	12.14	84.09	3.77
1958	11.90	81.73	6.37
1959	10.45	82.08	7.47
1960	10.29	80.12	9.59
1961	8.19	81.14	10.66
1962	6.45	80.90	12.65
1963	6.04	80.12	13.85
1964	5.52	79.87	14.61
1965	4.88	80.08	15.04
1966	4.64	80.06	15.31
1967	4.89	79.53	15.59
1968	4.49	79.20	16.32
1969	4.03	78.64	17.33
1970	5.50	76.69	17.81
1971	6.34	77.18	16.49
1972	6.60	77.07	16.33
1973	5.75	80.62	13.62
1974	5.89	80.62	13.49
1975	8.92	67.06	24.02
1976	7.97	60.35	31.69
1977	7.34	58.51	34.15
1978	7.04	59.66	33.30
1979	6.32	59.92	33.76
1980	5.51	58.79	35.69
1981	5.98	59.29	34.73
1982	5.89	63.19	30.92
1983	4.92	69.40	25.68
1984	6.13	68.10	25.77
1985	5.99	67.82	26.18
1986	8.44	59.86	31.71
1987	7.83	66.08	26.09
1988	9.68	57.08	33.25
1989	8.84	62.42	28.75
1990	8.41	65.32	26.27
1991	9.36	61.85	28.79
1992	7.44	63.00	29.56
1993	7.65	56.12	36.24
1994	6.58	54.79	38.63
1995	7.55	63.81	28.65
1996	6.54	64.11	29.34
1997	6.58	58.35	35.07
1998	7.15	51.13	41.72
1999	4.16	55.27	40.57
2000	2.47	51.82	45.70
2001	2.84	42.40	54.76
2002	2.96	54.91	42.13
2003	1.70	48.92	49.39
2004	1.88	53.32	44.79
2005	2.71	50.75	46.54
2006	2.55	58.54	38.91
2007	2.21	61.19	36.59
2008	3.29	65.82	30.89
2009	4.68	71.89	23.42
2010	5.01	76.07	18.93
2011	6.05	79.62	14.32
2012	4.88	85.32	9.79
2013	3.57	84.37	12.06
2014	2.60	82.79	14.61
2015	4.70	78.82	16.47
Note: BCPI stands for Bank of Canada commodity price index. Source: Statistics Canada, author's calculations.

4.2.2 Metals and minerals

Production shares for commodities in the long-run BCPI commodity bundle evolve over time as production starts or stops, and as data sources become available. Data for metals and minerals begin in 1886 with gold and copper, and expand to include silver, lead and iron in 1887 (Chart 3). Nickel enters the index in 1890, zinc in 1903, aluminum in 1905 and potash in 1967. Iron ore exits the index due to production cessation in 1900 and 1924, and re-enters the index in 1907 and 1940.

Production shares for metals and minerals exhibit features similar to those of production shares for agricultural commodities. There is an expansion through time in the number of commodities that affects the relative importance of each particular commodity. This occurs as new technologies, infrastructure, mineral discoveries and shifts in demand lead to changes in the composition of production. For example, transforming bauxite into aluminum requires significant quantities of electricity. As a result, the production share of aluminum increases as power sources become available to production.

Production changes affect relative shares, as in gold production, for example. Gold production constitutes a high proportion of overall production in the early part of the period. In the 1886-to-1890 period, this is the case partly because there is data on fewer commodities, but also because the Fraser River Gold Rush increased production. A similar event occurred in the late 1890s and early 1900s, as the Klondike Gold Rush drew people to the gold fields of Yukon. In the 1930s, the decision by the U.S. Federal Reserve to revalue gold from US$20.67 to US$35.00 per ounce led to a gold boom in Canada, and the share of gold production rose accordingly.

Through the period of stability in the 1950s and 1960s covered by the Bretton Woods period, gold production as a share of minerals and metals production declined to its lowest point. This period imposed price controls and trade restrictions on gold, and its cessation in the 1970s led to rapidly rising gold prices. This change is reflected in a rising share of gold in metals and minerals production, while the period of weak prices through the 1990s sees a relative decline. The share of gold production then rises through the 2000s as gold prices once again increased sharply.

Data table for Chart 3

Data table for Chart 3
Table summary
This table displays the results of Data table for Chart 3. The information is grouped by Year (appearing as row headers), Gold, Silver, Nickel, Copper , Aluminum, Zinc, Lead, Iron and Potash, calculated using percent units of measure (appearing as column headers).

Year	Gold	Silver	Nickel	Copper	Aluminum	Zinc	Lead	Iron	Potash
	percent
1870	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1871	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1872	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1873	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1874	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1875	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1876	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1877	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1878	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1879	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1880	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1881	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1882	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1883	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1884	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1885	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1886	78.33	0.00	0.00	21.67	0.00	0.00	0.00	0.00	0.00
1887	57.05	17.01	0.00	18.48	0.00	0.00	0.16	7.30	0.00
1888	41.32	15.77	0.00	36.54	0.00	0.00	0.42	5.95	0.00
1889	39.90	11.29	13.64	30.23	0.00	0.00	0.07	4.87	0.00
1890	32.27	12.00	23.28	27.87	0.00	0.00	0.05	4.53	0.00
1891	18.86	8.47	43.59	26.05	0.00	0.00	0.03	2.99	0.00
1892	25.35	7.75	34.76	23.98	0.00	0.00	0.34	7.82	0.00
1893	22.15	7.64	41.75	20.73	0.00	0.00	0.67	7.06	0.00
1894	25.57	12.34	37.67	17.50	0.00	0.00	1.58	5.35	0.00
1895	36.78	18.55	21.35	15.47	0.00	0.00	3.48	4.37	0.00
1896	36.51	29.08	14.01	14.20	0.00	0.00	3.55	2.65	0.00
1897	46.75	26.31	9.65	12.22	0.00	0.00	4.02	1.05	0.00
1898	65.95	12.67	7.75	10.72	0.00	0.00	2.14	0.76	0.00
1899	74.41	7.26	6.43	9.74	0.00	0.00	1.27	0.88	0.00
1900	73.63	7.38	7.81	8.48	0.00	0.00	2.70	0.00	0.00
1901	62.23	8.59	10.54	16.49	0.00	0.00	2.15	0.00	0.00
1902	64.33	6.89	13.47	14.26	0.00	0.01	1.05	0.00	0.00
1903	60.28	5.58	14.22	18.95	0.00	0.06	0.91	0.00	0.00
1904	54.10	7.81	14.03	20.81	0.96	0.03	2.25	0.00	0.00
1905	41.71	10.89	19.78	23.16	1.38	0.15	2.93	0.00	0.00
1906	29.49	14.81	20.40	28.82	2.12	0.02	2.94	1.39	0.00
1907	20.95	21.30	21.19	29.87	2.55	0.05	2.36	1.74	0.00
1908	24.89	30.16	18.51	22.31	0.93	0.00	1.70	1.50	0.00
1909	22.52	34.73	20.19	17.15	2.03	0.22	1.51	1.65	0.00
1910	21.40	37.62	20.85	15.60	2.24	0.09	0.95	1.25	0.00
1911	21.55	39.02	20.04	15.91	1.52	0.08	0.68	1.20	0.00
1912	20.79	32.61	19.66	21.92	3.03	0.13	0.97	0.89	0.00
1913	25.69	30.08	20.51	19.08	2.51	0.11	1.01	1.02	0.00
1914	27.42	27.30	20.83	18.53	3.74	0.17	1.04	0.97	0.00
1915	25.65	18.24	24.62	24.67	4.15	0.28	1.30	1.09	0.00
1916	18.59	16.49	24.95	32.29	4.63	1.07	1.27	0.72	0.00
1917	14.55	17.59	28.56	29.65	6.69	0.93	1.28	0.75	0.00
1918	13.12	19.16	29.85	27.83	6.64	0.96	1.60	0.84	0.00
1919	21.59	24.74	21.58	20.03	8.34	1.19	1.54	0.98	0.00
1920	20.92	18.16	28.85	19.76	8.52	1.50	1.58	0.71	0.00
1921	43.51	19.68	13.64	14.19	3.12	2.08	3.23	0.54	0.00
1922	46.72	22.96	9.80	10.76	3.65	2.14	3.86	0.11	0.00
1923	33.48	16.17	21.40	17.25	5.69	1.95	3.89	0.16	0.00
1924	35.13	14.99	19.29	15.89	6.23	2.59	5.88	0.00	0.00
1925	36.12	14.35	14.27	16.52	7.00	3.11	8.64	0.00	0.00
1926	35.81	14.00	12.62	18.11	8.34	4.07	7.05	0.00	0.00
1927	35.32	12.06	12.51	16.63	14.37	3.47	5.64	0.00	0.00
1928	30.88	10.29	15.68	23.70	11.91	2.98	4.56	0.00	0.00
1929	27.95	8.77	16.90	31.91	7.40	2.77	4.30	0.00	0.00
1930	31.90	7.56	15.96	29.21	9.18	2.63	3.57	0.00	0.00
1931	49.17	5.31	11.49	21.40	8.45	1.90	2.28	0.00	0.00
1932	65.66	5.45	5.86	14.73	5.04	1.41	1.84	0.00	0.00
1933	59.70	4.15	12.67	16.06	4.07	1.68	1.67	0.00	0.00
1934	57.28	4.42	15.96	15.62	3.08	1.88	1.75	0.00	0.00
1935	55.94	5.32	15.21	16.39	3.46	1.78	1.91	0.00	0.00
1936	54.68	3.52	16.25	17.25	4.28	1.71	2.32	0.00	0.00
1937	46.34	3.40	17.10	23.36	5.08	2.18	2.53	0.00	0.00
1938	52.54	3.12	15.15	18.74	7.43	1.38	1.64	0.00	0.00
1939	54.11	2.81	13.31	18.78	8.20	1.32	1.34	0.12	0.00
1940	52.81	2.40	13.74	17.81	9.96	1.39	1.52	0.38	0.00
1941	47.30	1.95	14.03	15.52	17.98	1.49	1.32	0.40	0.00
1942	41.47	1.98	13.85	14.09	25.14	1.63	1.42	0.41	0.00
1943	31.01	1.77	14.06	15.54	33.72	2.00	1.36	0.55	0.00
1944	28.12	1.49	15.37	17.09	33.87	2.20	1.27	0.58	0.00
1945	33.08	1.98	17.56	19.82	20.15	3.94	2.05	1.42	0.00
1946	36.97	3.80	14.33	17.37	16.57	4.85	3.15	2.96	0.00
1947	27.69	2.37	16.18	24.73	17.41	4.46	4.23	2.93	0.00
1948	26.37	2.63	16.49	23.99	18.61	5.17	4.78	1.95	0.00
1949	27.97	2.52	16.61	20.69	18.46	5.34	3.53	4.88	0.00
1950	27.90	3.16	16.45	21.33	17.50	6.01	2.93	4.73	0.00
1951	22.59	3.11	18.77	21.81	18.35	7.03	3.02	5.31	0.00
1952	20.91	2.93	18.36	20.98	21.85	6.57	2.77	5.63	0.00
1953	18.61	3.23	19.01	21.10	23.62	4.75	2.48	7.19	0.00
1954	18.10	3.22	19.49	22.42	22.69	4.07	2.63	7.39	0.00
1955	14.97	2.41	18.33	24.01	21.13	4.19	2.07	12.89	0.00
1956	12.56	2.16	16.43	25.55	21.31	3.87	1.81	16.30	0.00
1957	13.34	2.31	20.65	19.46	20.88	3.33	1.69	18.34	0.00
1958	15.61	2.77	17.34	18.38	25.37	3.45	1.58	15.49	0.00
1959	12.88	2.45	19.56	20.92	19.68	3.08	1.26	20.16	0.00
1960	12.02	2.36	20.10	21.24	23.57	3.08	1.25	16.37	0.00
1961	13.18	2.36	24.41	20.91	21.56	3.04	1.37	13.18	0.00
1962	11.72	3.74	23.66	18.97	21.10	2.93	1.12	16.77	0.00
1963	11.22	4.38	21.75	17.65	21.48	2.95	1.08	19.49	0.00
1964	9.16	3.65	21.02	18.45	22.22	4.14	1.17	20.18	0.00
1965	8.09	3.67	21.52	19.42	20.61	4.77	1.74	20.18	0.00
1966	7.05	3.37	18.22	21.80	20.24	5.23	1.61	19.33	3.15
1967	6.25	3.87	18.53	23.34	19.20	5.16	1.43	19.07	3.15
1968	6.05	5.47	19.05	21.77	18.72	4.80	1.34	19.96	2.83
1969	6.33	3.94	18.15	22.57	21.31	5.32	1.40	17.98	2.99
1970	4.39	3.46	24.30	22.65	17.51	4.87	1.50	17.48	3.83
1971	4.83	3.19	22.24	21.49	18.26	5.53	1.45	18.11	4.89
1972	6.03	3.77	23.28	20.75	16.95	6.42	1.53	16.05	5.23
1973	6.07	4.46	22.77	22.54	13.76	7.72	1.44	16.49	4.75
1974	6.70	5.34	19.40	23.68	14.94	7.09	1.10	15.28	6.46
1975	6.55	5.40	21.25	15.65	15.24	7.03	1.26	19.58	8.05
1976	5.22	5.41	21.84	16.00	12.28	7.82	1.23	23.36	6.83
1977	5.66	6.00	16.36	13.37	19.22	7.16	1.66	23.43	7.14
1978	9.43	6.95	10.24	13.58	23.08	6.94	2.20	19.70	7.89
1979	9.48	8.47	10.58	13.27	17.07	7.32	2.83	21.16	9.82
1980	19.47	9.07	11.73	12.03	16.59	5.28	1.68	14.62	9.54
1981	20.17	5.79	10.34	10.55	17.88	6.69	1.62	16.61	10.35
1982	22.79	5.65	7.35	9.27	20.45	7.40	1.41	16.95	8.73
1983	24.58	7.07	8.17	10.62	21.70	7.21	1.02	12.50	7.13
1984	25.41	5.39	9.44	9.01	21.24	8.42	1.10	11.46	8.53
1985	25.42	4.76	10.62	9.93	20.66	8.09	0.95	13.20	6.38
1986	34.67	4.22	7.65	8.94	21.84	5.02	0.95	11.41	5.29
1987	27.05	4.06	9.19	10.18	25.27	5.58	1.49	10.79	6.37
1988	20.75	3.73	15.31	10.82	25.78	6.55	1.03	8.36	7.67
1989	19.18	2.63	17.19	11.51	23.90	8.76	0.89	9.02	6.90
1990	22.40	3.01	14.25	12.49	22.11	8.15	1.00	9.31	7.27
1991	23.59	3.36	13.78	11.88	21.88	6.75	1.03	9.89	7.84
1992	22.87	2.11	12.33	12.31	24.22	7.70	1.07	9.10	8.29
1993	27.49	3.19	10.40	11.44	25.02	6.34	0.49	7.89	7.73
1994	22.95	2.90	8.67	13.37	28.31	6.75	0.64	7.92	8.49
1995	18.16	3.08	11.70	14.57	29.12	6.86	0.77	7.38	8.35
1996	21.97	3.17	11.99	11.68	26.91	6.92	1.08	8.65	7.63
1997	21.59	2.99	10.72	11.09	27.24	7.70	0.61	9.28	8.78
1998	20.67	4.12	8.70	9.30	29.44	6.33	0.49	10.32	10.62
1999	18.56	4.82	10.39	9.30	31.44	6.85	0.52	7.96	10.17
2000	15.76	4.21	13.15	9.34	33.52	6.04	0.37	8.58	9.03
2001	15.30	5.49	10.38	9.51	36.40	5.68	0.44	7.68	9.13
2002	17.59	4.74	11.31	8.56	36.68	4.52	0.29	7.06	9.25
2003	17.57	4.82	12.75	7.41	36.79	3.59	0.27	7.72	9.08
2004	17.08	6.01	15.92	8.96	31.72	3.37	0.28	6.50	10.16
2005	14.22	5.28	15.43	10.61	29.82	3.10	0.26	10.77	10.51
2006	12.08	3.65	20.91	13.46	28.82	5.04	0.27	8.57	7.20
2007	11.52	2.64	29.37	11.74	24.92	4.23	0.40	7.28	7.90
2008	12.35	2.26	15.94	10.48	26.01	2.68	0.45	9.59	20.25
2009	23.76	3.22	9.77	10.59	23.20	3.80	0.42	10.67	14.58
2010	21.16	2.77	11.48	11.19	18.65	3.02	0.31	16.24	15.18
2011	20.67	3.81	12.46	10.91	16.86	2.27	0.29	14.10	18.63
2012	25.55	4.20	10.17	11.08	15.51	2.29	0.25	13.76	17.20
2013	27.65	3.14	9.73	11.75	15.16	1.59	0.10	15.17	15.73
2014	31.40	2.54	11.21	11.61	14.90	1.46	0.02	15.13	11.73
2015	32.16	2.34	8.76	10.41	14.83	1.45	0.02	16.67	13.37
... not applicable Note: Data available starting in 1886. BCPI stands for Bank of Canada commodity price index. Source: Statistics Canada, author's calculations.

4.2.3 Forestry

Across forestry commodities in the long-run BCPI, the largest source of adjustment comes from the emergence and evolution of publishing industries and their demand for newsprint and pulp and paper (Chart 4). Beginning around the end of the First World War and stabilizing by the late 1930s, a transition in the composition of production away from lumber towards pulp and paper and newsprint occurred. The production of newsprint and pulp and paper was the majority of production of forestry products until the 1990s, when the relative share of lumber began to rise, and the share of newsprint began to decline quickly.

Data table for Chart 4

Data table for Chart 4
Table summary
This table displays the results of Data table for Chart 4. The information is grouped by Year (appearing as row headers), Lumber, Newsprint and Pulp, calculated using percent units of measure (appearing as column headers).

Year	Lumber	Newsprint	Pulp
	percent
1870	100.00	0.00	0.00
1871	100.00	0.00	0.00
1872	100.00	0.00	0.00
1873	100.00	0.00	0.00
1874	100.00	0.00	0.00
1875	100.00	0.00	0.00
1876	100.00	0.00	0.00
1877	100.00	0.00	0.00
1878	100.00	0.00	0.00
1879	100.00	0.00	0.00
1880	99.75	0.00	0.25
1881	99.49	0.00	0.51
1882	99.34	0.00	0.66
1883	99.18	0.00	0.82
1884	98.80	0.00	1.20
1885	98.36	0.00	1.64
1886	98.12	0.00	1.88
1887	97.87	0.00	2.13
1888	97.66	0.00	2.34
1889	97.41	0.00	2.59
1890	91.64	5.85	2.51
1891	90.74	6.12	3.14
1892	88.19	7.59	4.22
1893	87.37	7.95	4.68
1894	86.01	8.91	5.09
1895	84.05	9.95	6.00
1896	85.18	9.08	5.74
1897	83.78	9.24	6.98
1898	81.66	9.94	8.40
1899	82.00	9.41	8.58
1900	82.11	9.24	8.65
1901	81.89	9.32	8.80
1902	81.14	9.63	9.23
1903	82.02	9.54	8.44
1904	80.95	10.54	8.52
1905	80.74	10.77	8.48
1906	78.53	12.49	8.98
1907	82.13	10.74	7.13
1908	80.24	11.95	7.81
1909	80.03	11.86	8.11
1910	80.03	11.58	8.39
1911	82.69	10.36	6.95
1912	77.12	14.29	8.59
1913	77.77	13.14	9.09
1914	69.80	16.87	13.32
1915	67.99	18.77	13.24
1916	54.04	25.09	20.87
1917	53.43	24.83	21.74
1918	56.55	25.21	18.24
1919	56.88	25.37	17.76
1920	52.13	25.07	22.80
1921	40.83	39.01	20.16
1922	41.31	37.12	21.57
1923	42.81	36.85	20.35
1924	41.50	39.85	18.65
1925	38.64	41.18	20.18
1926	35.84	42.93	21.22
1927	33.71	45.73	20.56
1928	33.34	46.39	20.27
1929	34.22	45.53	20.24
1930	31.07	48.24	20.68
1931	22.82	55.31	21.87
1932	18.43	58.63	22.94
1933	21.65	52.32	26.03
1934	24.31	52.09	23.61
1935	26.95	49.75	23.30
1936	28.80	48.90	22.30
1937	30.67	46.85	22.48
1938	32.23	47.50	20.27
1939	31.37	48.41	20.22
1940	31.00	46.35	22.65
1941	34.08	41.89	24.03
1942	37.77	37.07	25.16
1943	37.42	37.68	24.90
1944	38.22	37.17	24.61
1945	36.91	38.53	24.56
1946	34.85	42.52	22.63
1947	36.29	40.06	23.65
1948	34.24	40.39	25.37
1949	32.38	45.26	22.37
1950	35.48	42.58	21.93
1951	35.00	38.91	26.08
1952	33.99	42.25	23.76
1953	34.04	43.61	22.35
1954	32.60	44.39	23.01
1955	34.06	43.29	22.66
1956	32.84	44.80	22.35
1957	29.84	46.67	23.49
1958	30.15	45.89	23.96
1959	30.50	45.42	24.07
1960	30.18	46.16	23.66
1961	31.86	45.59	22.54
1962	33.58	43.05	23.36
1963	34.90	40.87	24.22
1964	34.92	40.35	24.73
1965	34.24	40.31	25.45
1966	32.90	41.70	25.39
1967	33.60	40.43	25.97
1968	37.24	36.59	26.17
1969	35.74	36.69	27.57
1970	32.86	37.14	30.00
1971	38.06	34.36	27.59
1972	44.82	29.90	25.28
1973	48.26	26.47	25.27
1974	33.52	30.50	35.98
1975	30.66	33.38	35.96
1976	35.61	31.66	32.73
1977	40.36	31.65	27.99
1978	43.73	31.10	25.16
1979	43.92	28.49	27.59
1980	35.73	30.94	33.33
1981	31.54	35.45	33.01
1982	30.25	37.52	32.23
1983	39.80	32.67	27.53
1984	36.29	33.82	29.89
1985	38.44	34.49	27.07
1986	38.18	33.53	28.29
1987	38.17	29.90	31.94
1988	35.56	29.55	34.90
1989	35.28	26.88	37.84
1990	36.80	26.67	36.53
1991	38.64	28.13	33.23
1992	44.86	24.58	30.57
1993	53.41	22.88	23.70
1994	53.03	20.71	26.26
1995	39.50	24.68	35.82
1996	48.13	25.60	26.27
1997	52.47	22.17	25.36
1998	51.82	23.22	24.96
1999	53.97	20.89	25.13
2000	47.80	21.25	30.95
2001	49.41	23.99	26.60
2002	53.37	20.50	26.13
2003	50.82	20.48	28.70
2004	53.86	19.12	27.02
2005	54.41	20.70	24.89
2006	54.13	19.52	26.36
2007	49.07	18.41	32.52
2008	45.80	20.38	33.81
2009	40.67	18.94	40.39
2010	40.34	17.80	41.86
2011	40.38	16.37	43.25
2012	48.18	15.39	36.43
2013	52.47	13.05	34.48
2014	49.61	12.68	37.71
2015	45.58	13.32	41.10
Note: BCPI stands for Bank of Canada commodity price index. Source: Statistics Canada, author's calculations.

4.2.4 Agriculture

The agricultural commodities included in the long-run BCPI are a select number of agricultural outputs that have particular economic significance (Chart 5). Nevertheless, the major events that affected agricultural production over the last 145 years are encapsulated in the commodities included here. Events such as the Wheat Boom of the late 1800’s and early 1900’s or the increasing importance of oilseeds (represented by canola) can be seen in the share of agricultural commodities included in the historical index. Similarly, the importance of animal production, which typically makes up nearly 40% of the weight in the BCPI for agriculture, can be seen.

While particular products, such as wheat or cattle, have an important place in agriculture over the entire time period, their relative size is not fixed. While year-to-year fluctuations in prices or production can lead to changes in the share of a particular commodity, a more important change has been the evolution that agriculture has undergone in Canada. In response to new developments, from infrastructure that opened the Prairies to research and development that made new crop types available, agricultural production moved first towards wheat and later towards oilseeds.

Data table for Chart 5

Data table for Chart 5
Table summary
This table displays the results of Data table for Chart 5. The information is grouped by Year (appearing as row headers), Cattle, Hogs, Wheat, Barley, Potatoes, Corn and Canola, calculated using percent units of measure (appearing as column headers).

Year	Cattle	Hogs	Wheat	Barley	Potatoes	Corn	Canola
	percent
1870	31.93	21.51	10.32	24.55	11.68	0.00	0.00
1871	27.48	14.94	16.48	25.83	15.27	0.00	0.00
1872	30.41	16.14	18.13	24.45	10.87	0.00	0.00
1873	20.80	16.05	17.97	35.52	9.66	0.00	0.00
1874	20.80	16.22	13.54	37.95	11.49	0.00	0.00
1875	16.17	15.85	13.70	47.25	7.03	0.00	0.00
1876	20.79	19.39	14.20	35.04	10.57	0.00	0.00
1877	21.88	15.29	19.62	30.70	12.52	0.00	0.00
1878	21.01	11.67	16.84	33.10	17.38	0.00	0.00
1879	20.28	13.32	19.67	31.18	15.55	0.00	0.00
1880	19.22	13.23	18.33	43.22	5.99	0.00	0.00
1881	12.23	13.30	16.46	45.00	13.00	0.00	0.00
1882	23.52	14.31	21.01	30.07	11.09	0.00	0.00
1883	26.43	14.46	15.11	29.86	14.14	0.00	0.00
1884	26.76	14.05	20.85	29.81	8.53	0.00	0.00
1885	23.36	13.68	18.36	31.60	13.00	0.00	0.00
1886	20.71	15.76	16.50	34.23	12.81	0.00	0.00
1887	19.18	15.36	14.55	33.89	17.02	0.00	0.00
1888	20.56	17.15	14.55	39.04	8.70	0.00	0.00
1889	21.03	19.01	15.39	29.53	15.04	0.00	0.00
1890	27.62	17.03	19.08	20.77	15.50	0.00	0.00
1891	28.13	10.90	29.96	19.99	11.02	0.00	0.00
1892	23.49	22.65	26.03	11.65	16.18	0.00	0.00
1893	36.66	16.77	21.46	6.58	18.54	0.00	0.00
1894	27.75	29.31	17.30	9.19	16.44	0.00	0.00
1895	30.62	23.44	29.98	7.64	8.32	0.00	0.00
1896	31.62	25.33	20.73	9.22	13.10	0.00	0.00
1897	30.49	22.96	26.77	3.57	16.21	0.00	0.00
1898	27.65	22.93	31.34	4.40	13.69	0.00	0.00
1899	31.95	22.35	23.11	9.68	12.90	0.00	0.00
1900	35.05	23.28	23.29	10.69	7.69	0.00	0.00
1901	30.02	25.67	27.48	5.75	11.08	0.00	0.00
1902	31.03	26.31	24.96	4.99	12.71	0.00	0.00
1903	28.03	24.95	22.74	5.27	19.00	0.00	0.00
1904	29.59	26.08	24.23	6.04	14.06	0.00	0.00
1905	29.28	24.85	29.29	5.57	11.02	0.00	0.00
1906	22.00	26.84	33.28	5.10	12.78	0.00	0.00
1907	22.20	21.74	30.72	8.95	16.39	0.00	0.00
1908	18.93	21.37	37.11	8.81	13.79	0.00	0.00
1909	17.53	20.50	38.67	7.92	10.92	4.45	0.00
1910	20.15	29.06	32.98	5.22	9.51	3.08	0.00
1911	19.42	19.17	41.93	6.17	10.00	3.31	0.00
1912	22.26	21.19	38.37	5.85	9.28	3.06	0.00
1913	22.42	24.43	36.56	4.82	8.90	2.87	0.00
1914	24.82	22.15	36.23	4.92	9.23	2.66	0.00
1915	19.20	16.41	52.40	4.51	5.59	1.90	0.00
1916	19.99	17.00	51.14	4.57	6.31	0.98	0.00
1917	19.32	20.32	42.99	6.79	8.79	1.79	0.00
1918	21.72	22.38	33.80	8.41	10.66	3.03	0.00
1919	21.56	21.52	33.98	7.57	12.58	2.79	0.00
1920	16.86	17.36	43.44	5.90	14.05	2.38	0.00
1921	15.37	13.25	49.75	4.97	13.70	2.95	0.00
1922	13.32	17.84	53.16	5.39	7.92	2.38	0.00
1923	11.55	15.48	57.41	4.91	8.35	2.30	0.00
1924	12.43	18.52	49.82	9.51	7.08	2.65	0.00
1925	10.99	16.86	55.00	5.70	9.97	1.48	0.00
1926	10.66	17.54	55.47	6.63	8.44	1.26	0.00
1927	11.58	14.72	59.66	7.33	6.04	0.67	0.00
1928	13.10	14.95	58.01	8.68	4.41	0.85	0.00
1929	18.23	21.07	41.44	9.05	9.15	1.06	0.00
1930	19.39	24.46	41.26	5.51	7.96	1.40	0.00
1931	18.40	22.71	45.27	5.65	6.89	1.08	0.00
1932	16.01	14.60	54.96	6.02	7.45	0.96	0.00
1933	16.80	22.52	43.63	5.98	9.84	1.23	0.00
1934	14.72	21.96	46.61	8.58	6.43	1.70	0.00
1935	17.03	23.21	44.89	6.06	7.55	1.25	0.00
1936	14.94	20.90	42.78	10.86	9.39	1.13	0.00
1937	18.06	24.52	40.63	9.72	5.98	1.09	0.00
1938	18.92	24.44	40.44	7.14	7.84	1.22	0.00
1939	15.73	21.48	49.14	6.04	6.63	0.99	0.00
1940	16.86	24.20	47.17	5.48	5.46	0.82	0.00
1941	17.38	24.27	36.80	9.99	9.23	2.32	0.00
1942	12.60	18.56	46.29	14.05	6.93	1.57	0.00
1943	18.95	28.88	28.45	15.16	7.64	0.93	0.00
1944	16.73	25.34	36.67	13.75	6.22	1.28	0.00
1945	21.26	24.30	33.51	11.59	7.91	1.42	0.00
1946	18.91	20.22	41.63	11.13	6.79	1.31	0.00
1947	19.10	22.92	35.49	13.99	7.19	1.31	0.00
1948	19.27	19.70	43.32	11.01	5.27	1.43	0.00
1949	20.51	23.13	39.43	10.99	4.40	1.48	0.06
1950	21.51	19.93	41.61	12.05	3.31	1.59	0.00
1951	24.81	19.03	36.87	13.13	4.68	1.47	0.01
1952	21.75	14.34	43.67	14.19	4.55	1.48	0.02
1953	22.44	14.09	46.18	11.86	2.81	2.57	0.05
1954	26.26	23.99	29.77	11.04	5.29	3.58	0.07
1955	23.98	16.64	39.05	12.88	4.02	3.28	0.16
1956	23.76	14.41	41.66	12.42	4.20	2.93	0.62
1957	28.34	21.42	30.52	10.73	4.74	3.36	0.90
1958	30.39	24.84	27.31	10.19	3.68	3.04	0.54
1959	32.94	19.85	29.82	8.78	5.19	3.01	0.39
1960	32.39	15.16	35.54	8.69	4.55	2.65	1.01
1961	39.90	19.20	24.86	7.41	3.74	3.63	1.27
1962	33.43	14.83	37.18	7.38	3.35	3.26	0.57
1963	28.15	13.05	43.13	8.50	3.12	3.19	0.86
1964	28.68	13.33	39.96	6.92	5.39	4.23	1.49
1965	30.03	13.86	36.29	8.75	4.45	4.49	2.13
1966	27.14	13.79	40.29	9.84	2.40	4.52	2.03
1967	33.74	15.53	32.55	8.22	2.88	5.29	1.79
1968	33.46	14.41	32.58	9.66	2.90	5.68	1.31
1969	33.88	16.68	29.37	8.55	3.62	5.28	2.62
1970	38.84	18.83	13.22	11.42	3.86	7.61	6.23
1971	35.89	13.27	21.96	13.24	2.53	6.61	6.50
1972	36.92	14.49	16.56	16.86	3.76	6.60	4.80
1973	31.77	13.56	18.23	19.89	4.08	7.27	5.22
1974	29.37	11.06	31.35	13.36	2.09	7.33	5.45
1975	24.85	11.57	33.23	13.59	3.10	8.10	5.56
1976	23.31	11.74	39.07	12.52	2.62	7.73	3.02
1977	27.27	12.86	27.63	13.00	2.33	8.76	8.15
1978	30.72	14.60	23.40	9.55	2.06	8.42	11.25
1979	34.91	14.05	22.21	8.99	1.63	9.06	9.16
1980	29.74	11.55	27.45	13.14	2.92	9.50	5.71
1981	26.45	12.22	34.48	12.67	2.10	8.14	3.93
1982	24.99	14.93	35.33	10.45	1.87	7.76	4.67
1983	24.99	13.66	32.38	9.56	2.67	9.09	7.66
1984	26.18	14.80	27.83	10.05	2.31	9.45	9.37
1985	25.99	13.98	30.61	11.46	1.82	8.16	7.98
1986	26.78	16.98	29.87	11.31	3.04	5.89	6.14
1987	34.31	19.39	20.30	9.65	2.99	7.00	6.35
1988	34.78	15.03	22.93	8.16	3.50	5.82	9.79
1989	32.37	13.62	28.42	9.34	3.67	6.25	6.32
1990	31.03	14.98	30.11	8.79	3.04	6.03	6.03
1991	35.26	15.91	24.70	7.07	3.35	5.69	8.01
1992	38.25	15.02	24.20	6.91	3.27	4.61	7.74
1993	38.24	14.48	21.69	6.47	4.04	4.66	10.42
1994	35.75	13.84	19.41	6.23	4.00	4.96	15.81
1995	30.57	13.55	26.57	8.12	3.63	4.96	12.59
1996	24.53	14.72	31.77	10.55	2.85	6.51	9.07
1997	29.91	17.88	22.45	8.34	3.89	5.14	12.38
1998	32.30	13.32	21.89	7.53	4.43	5.52	15.00
1999	35.53	13.95	21.26	7.18	4.32	4.83	12.91
2000	40.10	18.58	18.21	6.89	4.00	3.74	8.49
2001	40.79	20.13	16.85	6.23	5.03	4.41	6.56
2002	40.39	18.69	16.63	5.20	5.63	5.62	7.84
2003	31.10	19.18	18.89	8.49	5.02	5.95	11.37
2004	28.24	22.99	19.54	7.16	4.56	5.39	12.14
2005	33.78	21.86	15.91	6.23	5.36	4.40	12.47
2006	34.48	19.27	17.11	5.46	5.42	5.45	12.82
2007	30.52	15.30	19.98	7.35	4.71	6.40	15.75
2008	21.52	11.28	28.51	7.13	4.49	5.90	21.16
2009	26.23	13.03	21.61	4.72	4.96	6.34	23.09
2010	26.77	15.50	17.98	4.24	4.43	7.50	23.57
2011	22.07	13.98	19.81	4.42	3.56	9.28	26.87
2012	23.97	12.86	18.07	5.27	2.77	10.47	26.59
2013	22.53	12.48	17.00	5.96	3.45	9.16	29.43
2014	34.69	15.81	16.65	3.13	3.10	5.73	20.88
2015	39.70	12.06	14.72	3.51	3.09	5.81	21.11
Note: BCPI stands for Bank of Canada commodity price index. Source: Statistics Canada, author's calculations.

4.3 Examining the utility of the long-run Bank of Canada commodity price index: 1870 to 2015

The historical chain index values that have been linked to the modern BCPI are shown in Chart 6, and the accompanying compound annual growth rates by decade are reported in Table 7. To examine the utility of the index calculations, two approaches are taken. In both approaches, the goal is to understand the characteristics of the long-run BCPI, to identify the types of uses it is most appropriate for.

The first approach examines the behaviour of the historical BCPI values with respect to major economic events.

The second approach compares the long-run BCPI values with two other indexes: an export index for Canada from 1870 to the present that is an updated version of the index from Baldwin and Macdonald (2012), and the General Wholesale Price Index (including gold) for the period from 1870 to 1975 from the Historical Statistics of Canada (Holmes n.d.). This comparison is performed only at an aggregate level, as experience has shown that this level has the greatest consistency when making comparisons across indexes or between values from different time periods and methodologies.

Data table for Chart 6

Data table for Chart 6
Table summary
This table displays the results of Data table for Chart 6. The information is grouped by Year (appearing as row headers), Total, all commodities, Energy, Metals and minerals, Agriculture, Forestry and Fish, calculated using log of index (1972=100) units of measure (appearing as column headers).

Year	Total, all commodities	Energy	Metals and minerals	Agriculture	Forestry	Fish
	log of index (1972=100)
1870	3.52	3.40	Note ...: not applicable	3.80	2.50	Note ...: not applicable
1871	3.39	3.45	Note ...: not applicable	3.64	2.52	Note ...: not applicable
1872	3.35	3.36	Note ...: not applicable	3.59	2.54	Note ...: not applicable
1873	3.42	3.16	Note ...: not applicable	3.68	2.56	Note ...: not applicable
1874	3.41	2.99	Note ...: not applicable	3.68	2.52	Note ...: not applicable
1875	3.35	2.97	Note ...: not applicable	3.63	2.43	Note ...: not applicable
1876	3.29	3.06	Note ...: not applicable	3.56	2.37	Note ...: not applicable
1877	3.19	2.89	Note ...: not applicable	3.46	2.32	Note ...: not applicable
1878	3.07	2.61	Note ...: not applicable	3.33	2.23	Note ...: not applicable
1879	3.08	2.51	Note ...: not applicable	3.35	2.25	Note ...: not applicable
1880	3.20	2.73	Note ...: not applicable	3.47	2.32	Note ...: not applicable
1881	3.37	2.99	Note ...: not applicable	3.65	2.38	Note ...: not applicable
1882	3.23	2.96	Note ...: not applicable	3.48	2.43	Note ...: not applicable
1883	3.18	3.02	Note ...: not applicable	3.42	2.40	Note ...: not applicable
1884	3.07	2.87	Note ...: not applicable	3.30	2.40	Note ...: not applicable
1885	3.07	2.97	Note ...: not applicable	3.31	2.35	Note ...: not applicable
1886	3.03	2.91	3.85	3.25	2.37	Note ...: not applicable
1887	3.09	2.93	3.85	3.32	2.38	Note ...: not applicable
1888	3.12	2.88	3.97	3.36	2.37	Note ...: not applicable
1889	2.92	2.68	3.92	3.12	2.37	Note ...: not applicable
1890	3.11	2.67	3.97	3.36	2.37	Note ...: not applicable
1891	3.02	2.63	3.85	3.25	2.35	Note ...: not applicable
1892	3.00	2.68	3.93	3.23	2.32	Note ...: not applicable
1893	3.02	2.71	3.76	3.25	2.33	Note ...: not applicable
1894	2.93	2.69	3.74	3.14	2.32	Note ...: not applicable
1895	2.84	2.70	3.57	3.03	2.28	Note ...: not applicable
1896	2.86	2.73	3.59	3.05	2.27	Note ...: not applicable
1897	2.97	2.68	3.58	3.22	2.26	Note ...: not applicable
1898	2.89	2.64	3.58	3.11	2.29	Note ...: not applicable
1899	2.93	2.72	3.63	3.13	2.36	Note ...: not applicable
1900	3.00	2.75	3.65	3.22	2.39	Note ...: not applicable
1901	3.06	2.83	3.67	3.32	2.35	Note ...: not applicable
1902	3.06	2.90	3.57	3.31	2.49	Note ...: not applicable
1903	3.07	3.01	3.58	3.32	2.47	Note ...: not applicable
1904	3.01	2.94	3.58	3.26	2.38	Note ...: not applicable
1905	3.04	2.88	3.63	3.27	2.50	Note ...: not applicable
1906	3.06	2.90	3.71	3.26	2.62	Note ...: not applicable
1907	3.18	2.93	3.74	3.41	2.70	Note ...: not applicable
1908	3.21	2.92	3.57	3.43	2.87	Note ...: not applicable
1909	3.25	2.89	3.53	3.49	2.90	Note ...: not applicable
1910	3.23	2.92	3.54	3.47	2.86	Note ...: not applicable
1911	3.28	2.94	3.53	3.52	2.95	Note ...: not applicable
1912	3.24	3.03	3.63	3.45	2.97	Note ...: not applicable
1913	3.31	3.04	3.61	3.52	3.06	Note ...: not applicable
1914	3.35	3.01	3.56	3.60	3.03	Note ...: not applicable
1915	3.32	3.00	3.62	3.53	3.12	Note ...: not applicable
1916	3.60	3.12	3.83	3.87	3.12	Note ...: not applicable
1917	3.90	3.33	3.86	4.22	3.32	Note ...: not applicable
1918	3.95	3.51	3.82	4.26	3.43	Note ...: not applicable
1919	4.05	3.70	3.75	4.36	3.57	Note ...: not applicable
1920	3.93	3.86	3.76	4.11	3.81	Note ...: not applicable
1921	3.57	3.89	3.56	3.65	3.63	Note ...: not applicable
1922	3.50	3.90	3.58	3.58	3.46	Note ...: not applicable
1923	3.50	3.96	3.60	3.57	3.49	Note ...: not applicable
1924	3.62	3.95	3.56	3.74	3.55	Note ...: not applicable
1925	3.77	3.93	3.60	3.95	3.62	Note ...: not applicable
1926	3.69	3.99	3.59	3.84	3.57	Note ...: not applicable
1927	3.66	3.96	3.54	3.81	3.54	Note ...: not applicable
1928	3.54	3.90	3.56	3.65	3.47	Note ...: not applicable
1929	3.63	3.89	3.61	3.77	3.51	Note ...: not applicable
1930	3.42	3.86	3.46	3.48	3.41	Note ...: not applicable
1931	3.13	3.76	3.32	3.01	3.29	Note ...: not applicable
1932	2.97	3.70	3.24	2.82	3.13	Note ...: not applicable
1933	3.17	3.62	3.29	3.24	2.96	Note ...: not applicable
1934	3.31	3.67	3.64	3.34	2.99	Note ...: not applicable
1935	3.41	3.62	3.65	3.53	2.99	Note ...: not applicable
1936	3.52	3.63	3.66	3.75	3.02	Note ...: not applicable
1937	3.51	3.57	3.74	3.66	3.07	Note ...: not applicable
1938	3.37	3.57	3.67	3.35	3.11	Note ...: not applicable
1939	3.41	3.51	3.69	3.43	3.12	Note ...: not applicable
1940	3.39	3.54	3.68	3.38	3.12	Note ...: not applicable
1941	3.55	3.62	3.68	3.73	3.20	Note ...: not applicable
1942	3.67	3.67	3.66	3.95	3.31	Note ...: not applicable
1943	3.81	3.75	3.66	4.14	3.51	Note ...: not applicable
1944	3.82	3.81	3.66	4.13	3.55	Note ...: not applicable
1945	3.85	3.84	3.67	4.19	3.56	Note ...: not applicable
1946	4.05	3.97	3.72	4.39	3.81	Note ...: not applicable
1947	4.26	4.09	3.84	4.63	4.03	3.36
1948	4.25	4.25	3.89	4.54	4.08	3.44
1949	4.19	4.26	3.88	4.42	4.07	3.39
1950	4.24	4.29	3.94	4.49	4.09	3.52
1951	4.34	4.32	4.04	4.61	4.17	3.53
1952	4.34	4.32	4.06	4.56	4.23	3.56
1953	4.29	4.37	4.10	4.44	4.24	3.48
1954	4.31	4.37	4.12	4.46	4.24	3.56
1955	4.29	4.36	4.22	4.32	4.27	3.49
1956	4.31	4.37	4.26	4.32	4.30	3.58
1957	4.32	4.47	4.22	4.37	4.29	3.62
1958	4.33	4.45	4.19	4.41	4.29	3.79
1959	4.33	4.42	4.23	4.37	4.30	3.84
1960	4.32	4.41	4.25	4.34	4.32	3.85
1961	4.33	4.43	4.25	4.36	4.30	3.83
1962	4.34	4.43	4.25	4.40	4.30	3.95
1963	4.32	4.44	4.25	4.34	4.30	3.92
1964	4.29	4.43	4.28	4.23	4.31	3.88
1965	4.31	4.42	4.31	4.28	4.33	3.94
1966	4.37	4.43	4.31	4.40	4.35	3.99
1967	4.35	4.44	4.35	4.30	4.36	3.98
1968	4.37	4.46	4.41	4.27	4.40	4.06
1969	4.43	4.50	4.47	4.34	4.46	4.27
1970	4.47	4.54	4.58	4.39	4.43	4.39
1971	4.49	4.60	4.57	4.36	4.52	4.38
1972	4.61	4.61	4.61	4.61	4.61	4.61
1973	4.87	4.71	4.89	5.03	4.79	4.79
1974	5.13	5.21	5.20	5.21	4.92	4.84
1975	5.16	5.50	5.06	5.08	5.03	4.85
1976	5.18	5.65	5.10	4.98	5.02	5.16
1977	5.18	5.73	5.13	4.84	5.08	5.30
1978	5.27	5.83	5.19	4.94	5.15	5.41
1979	5.45	6.01	5.51	5.08	5.27	5.55
1980	5.63	6.44	5.74	5.15	5.25	5.44
1981	5.69	6.69	5.57	5.18	5.27	5.48
1982	5.62	6.65	5.40	5.11	5.22	5.54
1983	5.62	6.59	5.53	5.12	5.25	5.49
1984	5.59	6.55	5.45	5.15	5.24	5.67
1985	5.50	6.45	5.35	5.03	5.20	5.73
1986	5.35	5.98	5.36	4.94	5.29	5.65
1987	5.46	6.06	5.55	4.92	5.45	5.91
1988	5.55	5.94	5.89	5.04	5.54	6.08
1989	5.59	6.08	5.82	5.11	5.57	5.84
1990	5.59	6.23	5.69	5.08	5.57	5.89
1991	5.49	6.14	5.58	4.99	5.45	5.88
1992	5.50	6.12	5.55	5.02	5.46	6.35
1993	5.51	6.07	5.46	5.06	5.60	6.38
1994	5.54	5.98	5.61	5.04	5.69	6.39
1995	5.60	5.96	5.73	5.09	5.81	6.50
1996	5.66	6.15	5.65	5.17	5.82	6.50
1997	5.61	6.13	5.63	5.05	5.77	6.75
1998	5.44	5.89	5.47	4.90	5.64	6.77
1999	5.53	6.10	5.49	4.86	5.70	6.75
2000	5.76	6.64	5.59	4.93	5.63	6.76
2001	5.64	6.44	5.48	5.00	5.58	6.76
2002	5.58	6.37	5.48	5.00	5.47	6.74
2003	5.77	6.68	5.59	5.03	5.58	6.78
2004	5.96	6.87	5.79	5.16	5.84	6.80
2005	6.20	7.24	5.92	5.15	5.80	6.94
2006	6.25	7.23	6.31	5.20	5.75	6.90
2007	6.32	7.29	6.45	5.34	5.69	6.89
2008	6.51	7.59	6.49	5.51	5.67	6.84
2009	6.11	7.07	6.34	5.25	5.48	6.72
2010	6.31	7.31	6.48	5.36	5.74	6.83
2011	6.48	7.46	6.65	5.64	5.76	6.82
2012	6.41	7.36	6.61	5.66	5.79	6.83
2013	6.41	7.40	6.44	5.60	5.88	6.98
2014	6.39	7.40	6.38	5.58	5.91	6.98
2015	5.95	6.75	6.27	5.45	5.78	6.98
... not applicable Source: Statistics Canada, author's calculations based on CANSIM table 383-0027.

4.3.1 Historical Bank of Canada commodity price index versus historical events

The first step in assessing the utility of the historical BCPI estimates is to examine the extent to which major historical events are captured. Because historical index values are estimated from a number of projectors, it is important to verify that major events important to the history of commodity prices are represented in the long-run estimates, to ensure that the back-casting methodology produces relevant results for the historical period. For example, the years 1929 to 1933 are expected to show deflation that is among the largest on record, with agriculture prices more affected than other prices.

In the first decade and a half, from 1870 to the mid-1880s, there is a general decline in the historical BCPI and its sub-indexes. This corresponds to a period of deflation in the United States and an appreciation of the Canadian dollar relative to the U.S. dollar, as the United States took action to return to the gold standard at pre-Civil War levels. It accomplished this in 1879 (Powell 2005).

The historical BCPI shows a rapid escalation of prices around the beginning of the First World War, followed by a reduction after the war (see Statistics Canada 2015; Bertram and Percy 1979). The onset of the Great Depression is similarly represented by a decline in historical index values between 1929 and 1932. The historical BCPI fell by 48.0% during these years, with the historical agricultural BCPI experiencing a particularly large decline of 61.4%.

The Second World War and its aftermath correspond with a period of strong price growth for the historical BCPI and all of its sub-indexes. Between 1939 and 1945, the historical BCPI rose by 55.5%, with the agriculture sub-index rising by 101.6%. Then, between 1945 and 1951, the historical BCPI rose by an additional 62.4%. These increases were then followed by a period of relative calm (the late 1950s) and slowly rising inflation rates (the 1960s). This “golden age” is also reflected in the historical index calculations.

Based on general conditions throughout historical periods, the historical BCPI reflects the major events that influenced commodities important to Canada during the 1870-to-1971 period. Consequently, it is possible to draw the inference that the historical values of the index contain information consistent with macroeconomic phenomena.

4.3.2 Long-run Bank of Canada commodity price index versus historical price indexes

While the historical BCPI captures major historical events, its utility is greater if it reflects changes at an annual frequency and if the magnitude of its changes reflects the relative importance of particular events. Consequently, a second step for assessing the utility of the historical calculations is to link them to the BCPI and compare this long-run BCPI with other sources of information.

Here, two price indexes are compared with the long-run BCPI. The first comparison price index is an updated version of the long-run export deflator from Baldwin and Macdonald (2012) (Chart 7). Its time span permits a comparison with the long-run BCPI over the entire 1870-to-2015 period, and thus covers the BCPI and its historical estimates. The second is the Generalized Wholesale Price Index from the Historical Statistics of Canada (Holmes n.d.) (Chart 8), which covers historical estimates and the linking period of the early 1970s.

Data table for Chart 7

Data table for Chart 7
Table summary
This table displays the results of Data table for Chart 7. The information is grouped by Year (appearing as row headers), Export deflator and Bank of Canada commodity price index, total, all commodities, calculated using log of index (1926=100) units of measure (appearing as column headers).

Year	Export deflator	Bank of Canada commodity price index, total, all commodities
	log of index (1926=100)
1870	3.894	4.438
1871	3.949	4.302
1872	3.970	4.261
1873	4.007	4.330
1874	4.027	4.326
1875	4.074	4.267
1876	4.104	4.207
1877	3.998	4.107
1878	4.025	3.982
1879	3.930	3.996
1880	3.953	4.119
1881	3.991	4.288
1882	4.086	4.141
1883	4.109	4.092
1884	4.064	3.985
1885	4.009	3.989
1886	3.978	3.941
1887	3.980	4.003
1888	4.032	4.031
1889	4.031	3.835
1890	4.052	4.024
1891	4.057	3.935
1892	4.047	3.918
1893	4.024	3.930
1894	4.027	3.847
1895	3.982	3.757
1896	3.945	3.773
1897	3.922	3.885
1898	3.987	3.806
1899	3.970	3.842
1900	4.016	3.911
1901	4.041	3.973
1902	4.059	3.979
1903	4.091	3.985
1904	4.101	3.928
1905	4.069	3.951
1906	4.143	3.975
1907	4.187	4.092
1908	4.234	4.123
1909	4.228	4.164
1910	4.238	4.143
1911	4.230	4.192
1912	4.205	4.158
1913	4.220	4.227
1914	4.253	4.264
1915	4.327	4.230
1916	4.446	4.518
1917	4.797	4.814
1918	4.893	4.862
1919	4.938	4.960
1920	5.052	4.843
1921	4.719	4.488
1922	4.540	4.411
1923	4.534	4.410
1924	4.554	4.532
1925	4.637	4.689
1926	4.605	4.605
1927	4.582	4.574
1928	4.549	4.454
1929	4.526	4.540
1930	4.426	4.336
1931	4.243	4.043
1932	4.135	3.887
1933	4.154	4.086
1934	4.243	4.222
1935	4.264	4.320
1936	4.305	4.437
1937	4.394	4.422
1938	4.336	4.285
1939	4.305	4.326
1940	4.397	4.304
1941	4.437	4.468
1942	4.502	4.588
1943	4.553	4.725
1944	4.620	4.734
1945	4.653	4.768
1946	4.748	4.960
1947	4.863	5.171
1948	4.956	5.161
1949	4.930	5.104
1950	5.034	5.153
1951	5.140	5.253
1952	5.131	5.254
1953	5.107	5.208
1954	5.099	5.220
1955	5.138	5.200
1956	5.165	5.221
1957	5.159	5.236
1958	5.154	5.245
1959	5.170	5.241
1960	5.176	5.238
1961	5.191	5.243
1962	5.231	5.255
1963	5.239	5.236
1964	5.260	5.203
1965	5.277	5.229
1966	5.304	5.283
1967	5.322	5.261
1968	5.340	5.283
1969	5.360	5.347
1970	5.393	5.387
1971	5.395	5.409
1972	5.434	5.520
1973	5.566	5.785
1974	5.845	6.040
1975	5.963	6.072
1976	6.017	6.091
1977	6.101	6.099
1978	6.187	6.183
1979	6.354	6.361
1980	6.506	6.545
1981	6.582	6.603
1982	6.602	6.531
1983	6.617	6.533
1984	6.653	6.506
1985	6.674	6.417
1986	6.661	6.267
1987	6.679	6.376
1988	6.681	6.462
1989	6.702	6.504
1990	6.693	6.506
1991	6.657	6.408
1992	6.682	6.412
1993	6.725	6.427
1994	6.781	6.450
1995	6.841	6.513
1996	6.846	6.572
1997	6.845	6.523
1998	6.839	6.356
1999	6.850	6.440
2000	6.911	6.674
2001	6.924	6.551
2002	6.906	6.496
2003	6.890	6.684
2004	6.911	6.878
2005	6.938	7.111
2006	6.940	7.167
2007	6.948	7.232
2008	7.049	7.424
2009	6.947	7.026
2010	6.963	7.227
2011	7.030	7.391
2012	7.022	7.326
2013	7.033	7.324
2014	7.068	7.307
2015	7.036	6.860
Source: Statistics Canada, author's calculations based on CANSIM table 383-0027.

Data table for Chart 8

Data table for Chart 8
Table summary
This table displays the results of Data table for Chart 8. The information is grouped by Year (appearing as row headers), Generalized Wholesale Price Index (including gold) and Bank of Canada commodity price index, total, all commodities, calculated using log of index (1926=100) units of measure (appearing as column headers).

Year	Generalized Wholesale Price Index (including gold)	Bank of Canada commodity price index, total, all commodities
	log of index (1926=100)
1870	4.115	4.438
1871	4.135	4.302
1872	4.241	4.261
1873	4.245	4.330
1874	4.197	4.326
1875	4.156	4.267
1876	4.092	4.207
1877	4.038	4.107
1878	3.963	3.982
1879	3.927	3.996
1880	4.016	4.119
1881	4.025	4.288
1882	4.026	4.141
1883	3.995	4.092
1884	3.951	3.985
1885	3.894	3.989
1886	3.880	3.941
1887	3.902	4.003
1888	3.934	4.031
1889	3.936	3.835
1890	3.951	4.024
1891	3.951	3.935
1892	3.878	3.918
1893	3.892	3.930
1894	3.825	3.847
1895	3.806	3.757
1896	3.773	3.773
1897	3.787	3.885
1898	3.832	3.806
1899	3.850	3.842
1900	3.878	3.911
1901	3.900	3.973
1902	3.942	3.979
1903	3.955	3.985
1904	3.966	3.928
1905	3.998	3.951
1906	4.001	3.975
1907	4.077	4.092
1908	4.075	4.123
1909	4.092	4.164
1910	4.103	4.143
1911	4.135	4.192
1912	4.186	4.158
1913	4.163	4.227
1914	4.186	4.264
1915	4.256	4.230
1916	4.432	4.518
1917	4.731	4.814
1918	4.839	4.862
1919	4.889	4.960
1920	5.039	4.843
1921	4.694	4.488
1922	4.574	4.411
1923	4.580	4.410
1924	4.593	4.532
1925	4.625	4.689
1926	4.605	4.605
1927	4.582	4.574
1928	4.569	4.454
1929	4.560	4.540
1930	4.463	4.336
1931	4.284	4.043
1932	4.200	3.887
1933	4.220	4.086
1934	4.271	4.222
1935	4.297	4.320
1936	4.322	4.437
1937	4.429	4.422
1938	4.374	4.285
1939	4.346	4.326
1940	4.431	4.304
1941	4.506	4.468
1942	4.561	4.588
1943	4.600	4.725
1944	4.621	4.734
1945	4.633	4.768
1946	4.683	4.960
1947	4.845	5.171
1948	5.014	5.161
1949	5.039	5.104
1950	5.102	5.153
1951	5.231	5.253
1952	5.170	5.254
1953	5.146	5.208
1954	5.129	5.220
1955	5.138	5.200
1956	5.168	5.221
1957	5.176	5.236
1958	5.178	5.245
1959	5.190	5.241
1960	5.191	5.238
1961	5.202	5.243
1962	5.230	5.255
1963	5.249	5.236
1964	5.252	5.203
1965	5.272	5.229
1966	5.308	5.283
1967	5.326	5.261
1968	5.347	5.283
1969	5.393	5.347
1970	5.407	5.387
1971	5.419	5.409
1972	5.487	5.520
1973	5.681	5.785
1974	5.883	6.040
1975	5.947	6.072
1976	Note ...: not applicable	6.091
1977	Note ...: not applicable	6.099
1978	Note ...: not applicable	6.183
1979	Note ...: not applicable	6.361
1980	Note ...: not applicable	6.545
1981	Note ...: not applicable	6.603
1982	Note ...: not applicable	6.531
1983	Note ...: not applicable	6.533
1984	Note ...: not applicable	6.506
1985	Note ...: not applicable	6.417
1986	Note ...: not applicable	6.267
1987	Note ...: not applicable	6.376
1988	Note ...: not applicable	6.462
1989	Note ...: not applicable	6.504
1990	Note ...: not applicable	6.506
1991	Note ...: not applicable	6.408
1992	Note ...: not applicable	6.412
1993	Note ...: not applicable	6.427
1994	Note ...: not applicable	6.450
1995	Note ...: not applicable	6.513
1996	Note ...: not applicable	6.572
1997	Note ...: not applicable	6.523
1998	Note ...: not applicable	6.356
1999	Note ...: not applicable	6.440
2000	Note ...: not applicable	6.674
2001	Note ...: not applicable	6.551
2002	Note ...: not applicable	6.496
2003	Note ...: not applicable	6.684
2004	Note ...: not applicable	6.878
2005	Note ...: not applicable	7.111
2006	Note ...: not applicable	7.167
2007	Note ...: not applicable	7.232
2008	Note ...: not applicable	7.424
2009	Note ...: not applicable	7.026
2010	Note ...: not applicable	7.227
2011	Note ...: not applicable	7.391
2012	Note ...: not applicable	7.326
2013	Note ...: not applicable	7.324
2014	Note ...: not applicable	7.307
2015	Note ...: not applicable	6.860
... not applicable Source: Statistics Canada, author's calculations, CANSIM table 383-0027 and Statistics Canada, n.d.a, Historical Statistics of Canada.

With respect to long-run export prices, a one-to-one correlation is not expected. This is because, while commodities are important for exports, exports also include finished and semi-finished products throughout the history of Canada (see Baldwin and Macdonald 2012), and because export prices are measured in Canadian dollars rather than U.S. dollars. For example, there is a significant increase in the long-run BCPI in the 2000s, as commodity prices increased. This translates into rising export prices, but not on the same scale as in the long-run BCPI, owing to the appreciation of the Canadian dollar (which offset some of the price increase in Canadian dollar terms) and to the effect of non-commodity exports, such as motor vehicles. Nevertheless, the movements found in the long-run BCPI do correlate with both the timing and magnitude of changes in export prices.

During the earliest part of the period, when the United States was deflating its economy after the Civil War, the price of exports shows little trend, while the long-run BCPI declines. This may reflect differences between the composition of commodities in the two indexes, differences between resource price changes and non-resource price changes, or differences in index construction, as the historical values for the export price index come from a fixed-weight series, while the long-run BCPI is chain-weighted.

Both the export prices and the long-run BCPI begin trending upward around 1895, and then follow very similar patterns over the course of the First World War, the 1920s and the onset of the Great Depression. The Gold Reserve Act of 1934 raised the value of gold from US$20.67 to US$35.00 per ounce, with the aim of increasing inflation in the United States. This change has a greater effect on the long-run BCPI than on the export price, because monitary gold is classified differently from commodity gold.

A minor difference occurs during the Second World War, as the long-run BCPI rises more quickly than export prices. The source of the difference is not immediately apparent, but may stem from the changing composition of trade towards munitions and armaments because of the war (Baldwin and Macdonald 2012) or from wartime government price controls that could have interrupted the usual peacetime links between supply, demand and price.

The stability of the 1950s and 1960s, as well as the rapid increase in prices of the 1970s and 1980s, are both reflected with similar timing and magnitude in export prices and in the long-run BCPI.

Finally, the latest commodity boom of the 2000s stands out because of the different paths of the two series, the first time this occurs on record. The export price index does not change its trajectory during the 2000s, while the long-run BCPI rises rapidly. This may reflect the path of non-resource exports or the effect of the floating exchange rate, as the commodity boom of the 2000s is the first in the history of Canada where there was no intervention from governments in the form of currency controls or price controls for energy products during a period of rising commodity prices.

The wholesale price^{Note 3} index exhibits a tighter relationship with the long-run BCPI than do export prices, but it is available only for a shorter time period. During the early part of the period, from 1870 to around 1895, the wholesale price index and the long-run BCPI both decline, with similar trends, and begin to rise at the same time after the mid-1890s. From this point to the onset of the First World War, they again share a similar trend. A difference then emerges in the year-to-year variation through the 1920s and 1930s. The long-run BCPI exhibits more short-term volatility, which may be caused by compositional differences or the use of an improved, chain-weighted index formula. From the First World War to the end of the wholesale index in 1975, the long-run BCPI and the wholesale index exhibit essentially the same relationship as the BCPI and the export price deflator.^{Note 4}

The comparison of the long-run BCPI with export prices and wholesale prices supports the use of the index at an annual frequency. The year-to-year fluctuations in the long-run BCPI are reflected in other series, the magnitude of changes is in line with other data sources, and they share similar trends. Moreover, because the long-run BCPI is calculated using a chain-weighted index formula, its weights are more accurate than earlier data sources.

Based on their relationship with other indexes, the historical values for the long-run BCPI are judged to be of sufficient utility to merit their use at an annual frequency for economic analysis. The estimates do not exhibit time-series characteristics that are out of line with other data sources, and, as a result, they are useful as inputs for statistical and econometric models and for trend-cycle decompositions.

4.4 The behaviour of the long-run commodity price indexes between 1870 and 2015

This section describes a number of characteristics of the long-run BCPI. The discussions aim to illustrate properties of the long-run BCPI series, rather than long-run relationships, which require the construction of more complex models. Across metrics, the complete available dataset is always used. As a result, the number of observations is lower for the metals and minerals and fisheries sub-indexes, owing to their relative lack of data points.

4.4.1 Descriptive statistics for growth rates of the long-run indexes

For the long-run BCPI and for each of the sub-indexes, there is a positive skewness to the distributions of annual growth rates and a degree of kurtosis, indicating that the distributions have thick tails relative to the normal distribution (Table 8). In other words, the series are periodically subject to innovations that are larger than expected (based on a normal distribution), and positive innovations are larger than negative ones. The highest average annual growth rate is for fisheries, for which data are only available after the Second World War. The second-highest is for energy, followed by forestry, metals and minerals and agriculture. In each case, the median lies below the mean, and the skewness is positive for all sub-indexes. The implication is that the more extreme periodic positive shocks that occur raise the arithmetic average of the annual growth rates relative to the mid-point of the distribution.

Similarly, the periodic innovations that create the positive skewness and kurtosis have an outsized influence for measuring long-run growth rates. The influential observations can also create an impression of larger-than-actual growth rates across time, because they tend to bias estimates of average growth upwards, relative to compound annual growth rates based on the first and last periods available. Across the sub-indexes, the effect is present in all commodity groups.

The effect of extreme events, particularly positive events, is sufficiently strong that the long-run growth-rate distributions do not appear to be normally distributed. Jarque–Bera tests for normality reject the null hypothesis of a normal distribution at conventional significance levels. The evidence against normality is weakest for the overall index, but quite strong for the sub-indexes

Across the sub-indexes, there is also a limited correlation in their annual growth rates (Table 9). The strongest correlation is between metals and minerals (0.42), while the weakest is between energy and fisheries (0.02). Agriculture and energy have the highest correlations with the long-run BCPI. Generally, annual frequency correlations show moderate to weak relationships between commodity types. In addition, the annual correlations are subject to influential data points, because of the same type of sensitivity the average estimates have, or that ordinary least squares estimates would have, owing to the skewness and kurtosis of the growth-rate distributions.

4.4.2 Periods of accelerated growth

While annual frequency correlations imply that commodity shocks are only modestly correlated across commodity groups, an examination of Chart 6 (the log levels of the indexes) shows a step-like path through time, with periods of rapid change interspersed with periods of calm. This pattern is present in the long-run BCPI and in all sub-indexes, except for fisheries. The pattern recurs repeatedly throughout the history of the long-run BCPI, but is most pronounced after 1900. Moreover, this pattern has a similar timing across all commodity types, except for fisheries. In other words, the pattern suggests that some form of common aggregate shock, rather than idiosyncratic commodity-specific shocks, is present, and that the effect is present over periods spanning multiple years.

Data table for Chart 9

Data table for Chart 9
Table summary
This table displays the results of Data table for Chart 9. The information is grouped by Year (appearing as row headers), Bank of Canada commodity price index, total, all commodities, Energy, Metals and minerals, Agriculture, Forestry and Fish, calculated using percent units of measure (appearing as column headers).

Year	Bank of Canada commodity price index, total, all commodities	Energy	Metals and minerals	Agriculture	Forestry	Fish
	percent
1870	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1871	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1872	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1873	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1874	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1875	-4.60	-8.90	Note ...: not applicable	-4.65	-2.70	Note ...: not applicable
1876	-1.73	-6.57	Note ...: not applicable	-1.52	-2.06	Note ...: not applicable
1877	0.77	-2.46	Note ...: not applicable	1.21	-1.59	Note ...: not applicable
1878	-1.55	-0.75	Note ...: not applicable	-1.55	-1.23	Note ...: not applicable
1879	-2.04	1.70	Note ...: not applicable	-2.19	-1.12	Note ...: not applicable
1880	-2.53	0.39	Note ...: not applicable	-2.93	-0.19	Note ...: not applicable
1881	-1.84	0.50	Note ...: not applicable	-2.13	-0.03	Note ...: not applicable
1882	-1.31	1.60	Note ...: not applicable	-1.67	0.67	Note ...: not applicable
1883	0.72	4.51	Note ...: not applicable	0.49	1.82	Note ...: not applicable
1884	0.87	5.08	Note ...: not applicable	0.73	1.45	Note ...: not applicable
1885	-2.57	0.20	Note ...: not applicable	-3.14	0.66	Note ...: not applicable
1886	-2.29	-3.07	Note ...: not applicable	-2.35	-0.08	Note ...: not applicable
1887	-1.72	-3.18	Note ...: not applicable	-1.69	-0.85	Note ...: not applicable
1888	-1.37	-3.36	Note ...: not applicable	-1.33	-0.83	Note ...: not applicable
1889	-0.12	-1.46	Note ...: not applicable	0.20	-0.71	Note ...: not applicable
1890	-1.04	-2.80	Note ...: not applicable	-1.06	-0.35	Note ...: not applicable
1891	-1.48	-2.08	-2.58	-1.55	-0.90	Note ...: not applicable
1892	-2.01	-1.95	-2.35	-2.14	-1.21	Note ...: not applicable
1893	-1.01	-2.01	-3.95	-0.49	-1.23	Note ...: not applicable
1894	0.13	-0.30	-3.40	0.66	-0.94	Note ...: not applicable
1895	-1.79	0.70	-3.39	-2.09	-0.07	Note ...: not applicable
1896	-0.05	1.37	-1.87	0.13	0.47	Note ...: not applicable
1897	0.86	1.80	-2.62	1.53	0.35	Note ...: not applicable
1898	0.79	2.22	-1.86	1.22	1.98	Note ...: not applicable
1899	1.74	3.69	-1.54	2.42	1.90	Note ...: not applicable
1900	2.09	2.83	0.14	2.95	1.35	Note ...: not applicable
1901	2.16	1.90	0.49	2.72	2.82	Note ...: not applicable
1902	1.12	2.70	1.67	0.57	4.42	Note ...: not applicable
1903	3.34	3.39	1.98	3.58	4.97	Note ...: not applicable
1904	3.28	2.40	-0.40	3.58	6.26	Note ...: not applicable
1905	2.96	1.78	-1.13	3.21	6.23	Note ...: not applicable
1906	2.01	1.17	-1.11	1.77	6.24	Note ...: not applicable
1907	2.51	0.57	-0.24	2.43	5.60	Note ...: not applicable
1908	2.06	0.29	0.86	1.62	5.96	Note ...: not applicable
1909	3.47	1.21	0.69	3.14	8.05	Note ...: not applicable
1910	3.64	1.50	-0.48	3.93	6.28	Note ...: not applicable
1911	2.99	1.24	-0.81	3.26	5.97	Note ...: not applicable
1912	5.32	2.27	1.52	6.06	5.12	Note ...: not applicable
1913	8.81	4.96	3.61	10.42	5.32	Note ...: not applicable
1914	8.88	7.46	3.65	10.14	6.44	Note ...: not applicable
1915	10.25	9.52	2.71	11.60	8.43	Note ...: not applicable
1916	8.47	11.21	2.94	8.60	10.39	Note ...: not applicable
1917	5.53	10.51	-0.23	5.23	8.46	Note ...: not applicable
1918	3.90	10.47	0.18	3.64	5.58	Note ...: not applicable
1919	3.47	11.48	0.98	2.65	6.25	Note ...: not applicable
1920	5.29	11.43	-0.09	5.50	5.89	Note ...: not applicable
1921	3.46	9.89	-2.24	3.57	6.61	Note ...: not applicable
1922	-1.26	7.87	-2.67	-2.23	3.69	Note ...: not applicable
1923	-2.14	5.44	-2.87	-3.00	2.02	Note ...: not applicable
1924	-4.54	2.44	-1.89	-5.81	-0.32	Note ...: not applicable
1925	-2.32	0.42	-1.37	-1.99	-2.88	Note ...: not applicable
1926	-1.05	-0.25	-0.90	-0.68	-2.14	Note ...: not applicable
1927	-3.04	-1.38	-2.65	-4.16	-1.66	Note ...: not applicable
1928	-4.64	-2.78	-3.68	-5.92	-3.56	Note ...: not applicable
1929	-3.64	-3.50	-2.69	-2.17	-6.04	Note ...: not applicable
1930	-3.91	-2.82	1.43	-3.70	-6.49	Note ...: not applicable
1931	-1.87	-3.97	1.77	-0.11	-5.97	Note ...: not applicable
1932	-0.16	-3.49	2.42	2.99	-5.38	Note ...: not applicable
1933	0.94	-3.60	3.10	3.56	-4.01	Note ...: not applicable
1934	-1.48	-3.46	1.72	-0.77	-3.98	Note ...: not applicable
1935	1.04	-3.76	3.43	3.02	-2.82	Note ...: not applicable
1936	3.61	-2.34	4.94	6.64	-1.60	Note ...: not applicable
1937	7.20	-0.70	5.72	13.19	1.00	Note ...: not applicable
1938	6.17	0.66	4.94	10.15	4.02	Note ...: not applicable
1939	6.19	1.04	0.37	11.25	6.11	Note ...: not applicable
1940	5.14	2.27	0.17	8.79	6.56	Note ...: not applicable
1941	4.15	2.43	0.12	6.65	6.44	Note ...: not applicable
1942	6.67	4.79	-0.17	10.24	8.95	Note ...: not applicable
1943	10.70	6.15	1.99	16.19	11.20	Note ...: not applicable
1944	10.12	8.78	2.39	14.26	11.60	Note ...: not applicable
1945	9.75	8.43	2.31	13.51	11.61	Note ...: not applicable
1946	8.33	7.82	3.01	9.69	10.88	Note ...: not applicable
1947	8.08	7.68	4.46	8.32	10.50	Note ...: not applicable
1948	6.46	6.75	4.55	5.62	8.83	Note ...: not applicable
1949	5.85	6.69	5.03	4.34	8.47	Note ...: not applicable
1950	5.61	6.32	5.20	3.96	8.30	Note ...: not applicable
1951	3.04	4.52	5.78	0.02	5.48	Note ...: not applicable
1952	0.65	3.27	4.90	-3.03	3.04	2.83
1953	0.94	2.49	3.87	-1.55	2.48	2.24
1954	1.66	2.22	3.57	0.30	2.50	4.82
1955	1.05	1.57	3.43	-0.93	2.41	3.88
1956	-0.15	1.09	2.38	-2.75	1.66	3.83
1957	-0.09	1.38	2.27	-1.98	0.79	3.41
1958	0.53	0.73	1.74	-0.25	0.69	5.55
1959	0.18	0.76	1.56	-1.10	0.70	4.36
1960	0.05	0.88	0.74	-0.94	0.50	4.69
1961	0.11	0.66	0.51	-0.33	0.31	4.27
1962	0.55	-0.45	1.06	0.67	0.62	4.43
1963	0.20	-0.07	1.85	-1.02	0.75	2.28
1964	0.51	0.43	2.02	-0.87	1.11	2.65
1965	1.27	1.03	2.50	0.35	1.60	5.15
1966	1.66	1.15	3.79	0.62	1.47	6.67
1967	1.79	1.92	3.70	-0.16	2.46	5.18
1968	3.29	1.92	4.10	3.57	3.48	8.39
1969	7.02	3.19	7.47	10.76	5.59	11.03
1970	9.95	10.57	11.09	12.28	7.04	10.93
1971	9.71	14.14	9.62	9.45	8.04	10.50
1972	10.16	15.82	9.61	9.56	7.83	14.66
1973	9.99	16.57	9.28	8.41	8.01	15.33
1974	10.24	17.18	9.25	8.64	8.17	13.98
1975	11.96	19.13	12.13	9.74	9.93	14.30
1976	13.94	24.30	15.25	10.98	8.69	13.36
1977	13.31	27.46	13.09	8.17	7.85	10.88
1978	9.17	25.82	7.64	1.44	5.09	9.29
1979	5.96	17.85	5.22	-0.46	3.83	8.20
1980	5.30	13.73	5.68	1.13	2.58	10.19
1981	4.14	10.84	4.26	0.93	2.17	6.88
1982	2.51	5.81	4.05	1.40	2.50	4.44
1983	2.81	5.58	5.77	0.12	3.63	6.41
1984	1.65	1.99	5.99	-0.17	3.25	6.86
1985	-0.12	-2.30	2.28	-0.21	3.84	5.54
1986	-0.77	-3.72	2.63	-0.88	3.66	5.70
1987	-1.03	-4.21	3.20	-1.04	2.95	4.97
1988	-1.00	-3.79	1.36	-0.83	2.77	12.14
1989	-0.54	-3.78	1.39	-0.64	4.33	10.29
1990	0.66	-3.69	4.11	0.38	5.97	9.76
1991	2.94	0.17	5.51	1.81	6.27	11.76
1992	2.34	1.59	2.25	3.04	4.46	8.55
1993	0.80	2.63	-2.36	0.37	2.84	9.66
1994	-1.38	-1.37	-3.29	-2.01	1.22	12.30
1995	-0.43	-0.48	-1.67	-2.07	1.82	11.49
1996	3.54	8.40	0.62	-0.32	2.39	11.66
1997	2.20	6.60	-0.19	0.14	1.61	4.98
1998	1.44	6.34	0.71	-0.37	-1.07	4.49
1999	3.49	11.39	0.28	0.16	-0.91	4.80
2000	5.14	13.96	1.23	1.25	1.04	3.82
2001	7.37	16.64	3.64	0.11	0.54	5.45
2002	8.55	16.75	9.09	1.94	0.51	1.83
2003	11.00	19.83	12.51	5.20	1.13	1.42
2004	12.39	21.05	12.69	7.64	0.35	1.16
2005	5.80	8.58	10.03	4.30	-0.92	-0.19
2006	9.57	13.52	12.85	4.72	3.04	1.14
2007	12.13	16.16	14.91	8.31	4.43	1.26
2008	9.14	10.94	13.22	8.22	3.39	0.79
2009	6.74	9.18	8.96	6.09	1.21	2.43
2010	3.63	4.08	6.82	5.95	1.88	0.78
2011	-1.01	-1.06	0.37	4.07	1.06	1.20
2012	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
2013	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
2014	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
2015	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
... not applicable Source: Statistics Canada, author's calculations.

One way to illustrate the pattern seen in Chart 6 is to calculate growth rates for the long-run BCPI by decade. During particular decades—1910 to 1920, 1940 to 1950 and 1970 to 1980 (Table 7) —there is a rapid increase in commodity prices, while in the decade or two that follow, there is retrenchment or only minor growth. Once again, a sequence of periods that exert strong pressure followed by periods of relative calm prevails through time.

An alternative method of illustrating the pattern of growth followed by calm is to apply a moving average to growth rates. While this is not a sophisticated way to filter the time series, it is useful for demonstrating the on-again, off-again growth found across commodity types. Here, a nine-period central moving average of year-to-year changes in the sub-indexes is shown (Chart 9). The moving averages imply that, since 1900, Canada has been subject to four waves of commodity price increases. In the first half of the period, there is a period of accelerating commodity price growth, culminating with the First World War, before a period of retrenchment in the 1920s and 1930s. Another cycle starts thereafter and grows through the 1940s, before a period of low growth rates in the 1950s and 1960s. The 1970s usher in the third increase in commodity prices, followed by a period in the late 1980s and 1990s of weak growth. The fourth, and latest, wave occurs in the 2000s.

Description for Chart 10

The title of the chart is "Box and whisker plots for the growth rates of the long-run Bank of Canada commodity price index, commodities by year."

This is a box and whisker chart.

There are in total 139 years on the horizontal axis, ranging from 1870 to 2008. The vertical axis starts at -80 and ends at 150, with graduation marks every 10 points.

There 135 box and whisker plots, one for each year, in this graph.

The vertical axis is "percent change."

The horizontal axis is "years."

Chart 10 shows box and whisker plots for changes in the prices of commodities included in the Bank of Canada commodity price index (BCPI). Each box and whisker plot comprises a tall, skinny rectangle (the box) and a line extending up and down from the rectangle (the whiskers). The minimum value of each box (the bottom of the rectangle) is the first quartile (Q1). The maximum value of each box (the top of the rectangle) is the third quartile (Q3). The length of each box represents the dispersion of the middle 50% of observations.

The whiskers extend from the box at the top and the bottom to a maximum possible length of Q1 minus 1.5 times the interquartile range or Q3 plus 1.5 times the interquartile range. The interquartile range is defined as Q1 minus Q3. For whiskers at the bottom of the box and whisker plot, the length of each whisker extends to the actual value for a price change that is below Q1, but closest to Q1 minus 1.5 times the interquartile range. For whiskers at the top of the box and whisker plot, the length of each whisker extends up from Q3 to the actual price change that is closest to, but still less than or equal to Q3 plus 1.5 times the interquartile range.

The box and whisker plots provide information about the shape of the distribution for price changes of BCPI commodities. If there is a general increase in commodity prices, then the box shifts up. If there is increased variance across commodity prices, the length of the box increases. If there is asymmetric growth in commodity prices, the box can shift up and elongate, or extend itself only in an upward direction indicating greater variance and a rising average. The whiskers accentuate these movements by providing additional information about changes in the tails of the distributions.

Over the 1870-to-2008 period, the time series of box and whisker plots oscillates up and down in a cyclical fashion. In some periods, such as around the First World War or the 1970s, the oscillation is larger, while in others, such as the 1950s, the oscillations are smaller. Outside of the First World War, the oscillations tend to be larger in the period after the early 1970s than in previous periods. The onset of the Great Depression coincides with a multi-year period with little positive price.

The length of the boxes also changes through time. Those periods with the largest oscillations also tend to be the periods with the greatest variability. This shows itself as boxes of increased length during the period after 1970, around the onset of the Great Depression and around the First World War. The 1950s and 1960s have small oscillations and short boxes.

The whiskers reinforce the changes in variability shown by the increasing or decreasing height of the boxes. However, they also contain additional information about price growth in particular periods because they do not expand and contract in the same manner from above and below. During periods of price increase, particularly around the First World War and the early 1970s, the whisker rising form Q3 elongates more than the whisker descending from Q1. The implication is the small or negative values are uncommon while larger than expected positive growth areas are more common. The reverse occurs during the onset of the Great Depression. During these years the whisker at the top of the plot is virtually non-existent while the whisker that descends from Q1 (in this case into negative growth rates) becomes elongated.

The source of Chart 10 is “Statistics Canada, author’s calculations.”

For additional information, please contact Statistics Canada at STATCAN.infostats-infostats.STATCAN@canada.ca.

A third way to examine the on-again, off-again growth is to look at the distributions in price growth across commodities. Chart 10 shows box plots for the growth of the individual price series that go into the BCPI. The box plots are drawn so that the length of the box covers the 25th to the 75th percentile observations of the distribution. The whiskers extend to the upper and lower adjacent values. The upper adjacent value corresponds to the sum of the largest observation below the 75th percentile value plus 1.5 times the interquartile range. The lower adjacent value corresponds to the smallest observation above the 25th percentile value minus 1.5 times the interquartile range. Outliers, of which there are some extreme values, are not shown.

Chart 10 shows that when periods of growth occur, the entire distribution of price growth tends to increase. This does not mean that every commodity rises by the same amount, or that the position of a commodity in the distribution is fixed through time. Rather, the up-and-down fluctuations of the distributions suggest a more broadly based source of price changes during the periods of stronger price growth, rather than commodity-specific, or commodity-group-specific, changes.

Additionally, the spread of commodity-price movements, which in the box plots is represented by the length of the box and whiskers, increases during periods of stronger price growth and contracts during periods of slower price growth. In other words, commodity price changes exhibit greater variation during periods of more rapid increase.

Finally, periods of heightened price growth exhibit a degree of serial correlation, so that the periods of accelerated growth are not randomly dispersed. Instead, the distribution of growth rates increases for a number of years in a row, in periods such as the First World War or the 1970s, which are then followed by a number of years where the distribution is lower.

4.4.3 Commodity price shocks

The source of the dynamics in commodity-price movements is of considerable interest, and two competing hypotheses are presented. The first is that changes in monetary policy regime are responsible. The second stems from the study of commodity super cycles and suggests that demand- and supply-based changes are primarily responsible. Rather than argue for one or the other, this paper will simply note that there is support for both theories and that it is difficult to gain sufficient degrees of freedom to disentangle them.

4.4.3.1 Monetary policy shocks

The notion that changes in monetary policy correspond with rapid growth of commodity markets could be interpreted in three ways. The first is that the periods of accelerated commodity price growth reflect changes in the overall price level. In this case the real price of commodities, that is, the long-run BCPI deflated by an aggregate price index, should be fairly flat, and should not exhibit large changes around monetary regime changes. A second is that changes in monetary policy elicit relative price changes (Balke and Wynne 2003), and that commodity prices in particular are more sensitive (Clarke 1999). Under this hypothesis, the real price of commodities increases sharply around changes in monetary regime. A third possibility is that the relationship is spurious. Major changes in economic order could be derived from various societal pressures, and when shifts occur, they affect many aspects of an economy or society, including the relative price of commodities and monetary policy.

In the period between 1870 and 2015, there were 4 (possibly 5) major monetary policy regimes. The first is the period under the classical gold standard (1870 to the First World War), the second covers the inter-war period where nations attempted to re-establish the gold standard, the third is the Bretton-Woods period from the end of the Second World War until the early 1970s, and the fourth is the period of fiat currencies in place from the end of the Bretton Woods period until the present (see Bordo 1993, 2008).^{Note 5}

The possible fifth regime, which is less clear, could come from a transition to a multi-polar monetary system. In this respect, the emergence of Brazil, Russia, India and China (BRIC nations), and particularly of China, had the effect of greatly changing international monetary policy, and so could be interpreted to have initiated a period of monetary policy regime change (see for example The Economist 2005). Similarly, the adoption of the Euro in 1999 could be argued to have changed international monetary policy arrangements. These events moved the global economy towards a multi-polar monetary system (Dailami and Masson 2009), and may therefore represent a major shift in monetary policy, but unlike previous shifts since 1870, one that occurred outside of the United States.

To examine whether or not periods of more rapid commodity price increases are consistent with changes in the overall price level, the long-run BCPI is deflated by the U.S. final domestic expenditure (FDE) deflator (Chart 11). The U.S. FDE deflator is used because the prices in the long-run BCPI are in U.S. dollars. The long-run index shows that over the 145 years covered, commodity price inflation accelerates at around the same time that major changes in monetary policy regime. These accelerations also occur when the long-run BPCI is deflated using a U.S. final domestic expenditure price index^{Note 6} (Chart 11). The rapid increases in the real long-run BCPI around the First World War, the Second World War, the end of the Bretton Woods period, and over the 2000s provide strong evidence that commodity prices adjust more than the overall price level. In other words, the deflated data do not support the notion that periods of accelerated commodity price change simply reflect periods of more rapidly rising overall price levels.

Data table for Chart 11

Data table for Chart 11
Table summary
This table displays the results of Data table for Chart 11. The information is grouped by Year (appearing as row headers), Average, Bank of Canada commodity price index / U.S. final domestic expenditure price index, First World War shock, Second World War shock, First oil shock, Second oil shock and 2000s commodities boom, calculated using index (1972=100) units of measure (appearing as column headers).

Year	Average	Bank of Canada commodity price index / U.S. final domestic expenditure price index	First World War shock	Second World War shock	First oil shock	Second oil shock	2000s commodities boom
	index (1972=100)
1870	109.29	122.53	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1871	109.29	108.07	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1872	109.29	106.99	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1873	109.29	115.51	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1874	109.29	119.10	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1875	109.29	117.98	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1876	109.29	115.48	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1877	109.29	105.28	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1878	109.29	98.72	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1879	109.29	103.86	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1880	109.29	111.75	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1881	109.29	133.75	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1882	109.29	113.30	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1883	109.29	111.71	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1884	109.29	105.17	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1885	109.29	108.32	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1886	109.29	102.94	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1887	109.29	109.22	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1888	109.29	111.02	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1889	109.29	90.50	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1890	109.29	112.02	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1891	109.29	102.92	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1892	109.29	102.19	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1893	109.29	102.64	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1894	109.29	99.57	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1895	109.29	92.81	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1896	109.29	93.82	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1897	109.29	106.76	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1898	109.29	97.30	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1899	109.29	99.32	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1900	109.29	104.77	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1901	109.29	111.02	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1902	109.29	110.26	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1903	109.29	108.32	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1904	109.29	101.35	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1905	109.29	103.10	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1906	109.29	102.46	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1907	109.29	110.39	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1908	109.29	116.65	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1909	109.29	121.81	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1910	109.29	114.68	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1911	109.29	119.91	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1912	109.29	112.88	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1913	109.29	120.28	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1914	109.29	123.51	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1915	109.29	116.96	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1916	109.29	144.91	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1917	109.29	164.65	164.65	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1918	109.29	147.41	152.14	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1919	109.29	141.20	142.61	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1920	109.29	107.67	135.25	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1921	109.29	87.02	129.50	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1922	109.29	86.41	124.97	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1923	109.29	83.64	121.38	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1924	109.29	94.50	118.52	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1925	109.29	108.21	116.22	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1926	109.29	98.59	114.37	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1927	109.29	97.31	112.88	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1928	109.29	87.28	111.67	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1929	109.29	95.16	110.69	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1930	109.29	80.87	109.90	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1931	109.29	66.90	109.25	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1932	109.29	64.78	108.72	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1933	109.29	81.30	108.29	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1934	109.29	88.86	107.94	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1935	109.29	96.18	107.65	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1936	109.29	106.61	107.42	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1937	109.29	101.12	107.23	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1938	109.29	89.99	107.07	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1939	109.29	94.74	106.94	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1940	109.29	91.92	106.84	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1941	109.29	101.77	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1942	109.29	106.17	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1943	109.29	116.23	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1944	109.29	114.71	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1945	109.29	115.66	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1946	109.29	123.66	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1947	109.29	137.49	Note ...: not applicable	137.49	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1948	109.29	128.74	Note ...: not applicable	131.26	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1949	109.29	121.39	Note ...: not applicable	126.36	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1950	109.29	125.88	Note ...: not applicable	122.48	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1951	109.29	130.04	Note ...: not applicable	119.40	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1952	109.29	127.52	Note ...: not applicable	116.93	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1953	109.29	120.44	Note ...: not applicable	114.94	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1954	109.29	120.62	Note ...: not applicable	113.34	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1955	109.29	116.60	Note ...: not applicable	112.05	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1956	109.29	115.13	Note ...: not applicable	111.00	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1957	109.29	113.10	Note ...: not applicable	110.14	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1958	109.29	111.68	Note ...: not applicable	109.45	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1959	109.29	109.69	Note ...: not applicable	108.89	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1960	109.29	107.91	Note ...: not applicable	108.43	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1961	109.29	107.39	Note ...: not applicable	108.05	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1962	109.29	107.36	Note ...: not applicable	107.74	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1963	109.29	104.07	Note ...: not applicable	107.49	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1964	109.29	99.16	Note ...: not applicable	107.29	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1965	109.29	100.03	Note ...: not applicable	107.12	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1966	109.29	102.71	Note ...: not applicable	106.98	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1967	109.29	97.75	Note ...: not applicable	106.87	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1968	109.29	95.93	Note ...: not applicable	106.78	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1969	109.29	97.45	Note ...: not applicable	106.70	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1970	109.29	96.23	Note ...: not applicable	106.64	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1971	109.29	93.57	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1972	109.29	100.00	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1973	109.29	123.35	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
1974	109.29	144.53	Note ...: not applicable	Note ...: not applicable	144.53	Note ...: not applicable	Note ...: not applicable
1975	109.29	136.83	Note ...: not applicable	Note ...: not applicable	136.74	Note ...: not applicable	Note ...: not applicable
1976	109.29	132.17	Note ...: not applicable	Note ...: not applicable	130.67	Note ...: not applicable	Note ...: not applicable
1977	109.29	124.97	Note ...: not applicable	Note ...: not applicable	125.89	Note ...: not applicable	Note ...: not applicable
1978	109.29	126.89	Note ...: not applicable	Note ...: not applicable	122.11	Note ...: not applicable	Note ...: not applicable
1979	109.29	139.53	Note ...: not applicable	Note ...: not applicable	119.10	Note ...: not applicable	Note ...: not applicable
1980	109.29	151.85	Note ...: not applicable	Note ...: not applicable	116.69	151.85	Note ...: not applicable
1981	109.29	147.46	Note ...: not applicable	Note ...: not applicable	114.75	142.38	Note ...: not applicable
1982	109.29	129.84	Note ...: not applicable	Note ...: not applicable	113.19	135.07	Note ...: not applicable
1983	109.29	125.62	Note ...: not applicable	Note ...: not applicable	111.92	129.36	Note ...: not applicable
1984	109.29	118.42	Note ...: not applicable	Note ...: not applicable	110.90	124.86	Note ...: not applicable
1985	109.29	105.14	Note ...: not applicable	Note ...: not applicable	110.06	121.29	Note ...: not applicable
1986	109.29	88.63	Note ...: not applicable	Note ...: not applicable	109.38	118.45	Note ...: not applicable
1987	109.29	96.12	Note ...: not applicable	Note ...: not applicable	108.83	116.17	Note ...: not applicable
1988	109.29	101.26	Note ...: not applicable	Note ...: not applicable	108.38	114.33	Note ...: not applicable
1989	109.29	101.59	Note ...: not applicable	Note ...: not applicable	108.01	112.85	Note ...: not applicable
1990	109.29	97.94	Note ...: not applicable	Note ...: not applicable	107.71	111.64	Note ...: not applicable
1991	109.29	86.19	Note ...: not applicable	Note ...: not applicable	107.47	110.67	Note ...: not applicable
1992	109.29	84.57	Note ...: not applicable	Note ...: not applicable	107.27	109.88	Note ...: not applicable
1993	109.29	83.96	Note ...: not applicable	Note ...: not applicable	107.10	109.23	Note ...: not applicable
1994	109.29	84.10	Note ...: not applicable	Note ...: not applicable	106.97	108.71	Note ...: not applicable
1995	109.29	87.71	Note ...: not applicable	Note ...: not applicable	106.86	108.28	Note ...: not applicable
1996	109.29	91.48	Note ...: not applicable	Note ...: not applicable	106.77	107.93	Note ...: not applicable
1997	109.29	85.85	Note ...: not applicable	Note ...: not applicable	106.69	107.65	Note ...: not applicable
1998	109.29	72.17	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	107.41	Note ...: not applicable
1999	109.29	77.26	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	107.22	Note ...: not applicable
2000	109.29	95.22	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	107.07	Note ...: not applicable
2001	109.29	82.57	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	106.94	Note ...: not applicable
2002	109.29	77.10	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	106.83	Note ...: not applicable
2003	109.29	91.07	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	106.75	Note ...: not applicable
2004	109.29	107.41	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
2005	109.29	130.97	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
2006	109.29	134.24	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
2007	109.29	139.42	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable
2008	109.29	164.17	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	164.17
2009	109.29	110.48	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	151.78
2010	109.29	133.06	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	142.33
2011	109.29	153.02	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	135.03
2012	109.29	141.02	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	129.33
2013	109.29	138.79	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	124.84
2014	109.29	134.07	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	121.27
2015	109.29	85.43	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	Note ...: not applicable	118.43
... not applicable Source: Statistics Canada, author's calculations.

The deflated long-run BCPI shows periods of more rapid acceleration than the general price level during the periods when growth in the long-run BCPI accelerates. The periods of relative price increase are followed by (often longer) periods of decline towards the original relative price structure. The pattern of a rapid increase in relative commodity prices followed by a return to a more normal relative price structure resembles a time series in which periodic shocks perturb a system that then returns towards equilibrium. Chart 11 demonstrates this for the periods of relative commodity price decline following significant positive shocks.

A value for the “normal” relative price of commodities is estimated based on the long-run averages of all data points. An autoregressive, or AR(1), model, is estimated using the complete sample, and the projected path from the peak of each regime change back to the average is shown in Chart 11.^{Note 7} In the case of the 1970s transition, the path from the 1973 and 1979 peaks are shown. The estimated path to return to equilibrium fits well with the path of the relative price of commodities during its return towards equilibrium, except during the period covered by Bretton Woods after the Second World War. However, the simple AR(1) model is unable to match the overshooting that occurs in each period.

Nevertheless, the AR(1) model is compatible with a system where periodic shocks (in this case major changes to monetary policy regime) create a relative price change. That relative price change is then worked out of the system over a period of 9 to 11 years (roughly half a generation). This period is quite long, well in excess of more typical periods of adjustment associated with monetary shocks.

In summary, there are a number of periods when changes in monetary policy regime coincide with important changes in the relative price of commodities, and the behaviour of commodity prices relative to the U.S. FDE deflator could be interpreted to follow an autoregressive-model-type adjustment pattern. There is support for the hypothesis that factors beyond changes in the general price levels are important for commodity price growth and that changes in relative commodity prices may correspond with changes in monetary policy regime.

While there is support for the notion that major changes in monetary policy regime can influence the relative price of commodities, in the context of this paper, there are not sufficient degrees of freedom to disentangle the effect of monetary policy regime changes from the effect of historical events. Major changes affecting economic systems, such as the two world wars or the emergence of the Euro and BRIC nations, can cause significant change throughout economies. It may therefore be unsurprising that major relative price adjustments and changes in policy stance occur simultaneously. An inference problem thus arises, where it becomes difficult to identify whether the change in monetary policy regime was responsible for the change in relative commodity prices, or whether the changes in relative commodity prices and monetary policy regime changes are due to more general changes brought about by major historical events.

4.4.4 Super cycles

An alternative to the role of monetary policy is the presence of a cycle in the long-run real BCPI that has a multi-year duration. This type of “super cycle” has recently been described for particular prices (see, for example, Jacks 2013; Cuddington and Jerrett 2008; Jerrett and Cuddington 2008). Super cycles are said to occur as periods of increased demand (often argued to stem from mass industrialization or from urbanization) cause accelerated price growth that only returns to trend over decades because of constraints in the supply response (Jacks 2013; Erten and Ocampo 2013).

In the context of the long-run real BCPI, the two world wars coincided with periods of rapid, mass industrialization in North America, but also with a great destruction of production capacity in Europe and Asia. The role of mass industrialization in commodity demand in global markets is, therefore, unclear for the long-run real BCPI. The 1970s coincided with a broad-based commodity price increase, but, because of the role of the oil shocks, this is often characterized as a period of supply disruption and de-industrialization in developed economies.

During the 1870-to-2015 sample period, the two most significant periods of urbanization in the United States occurred early: in the late 1800s and early 1900s, and in the 1950s. However, these correspond to periods of decline or stagnation in the long-run real BCPI. The period from the mid-1990s to 2015 also represented a period of important urbanization, but one that stemmed from changes in the BRIC nations rather than in the United States.

Relative to historical events, the super-cycle hypothesis based on mass industrialization or urbanization is less consistent with known events. Nevertheless, super cycles are presented here for two reasons.

First, a number of periods of significant demand and supply changes coincide with periods of more rapid commodity price growth. War demand, supply disruption and the integration of the BRIC nations into the global economy in the 1990s and 2000s all represent periods when real economy influences, rather than monetary effects, may be responsible for the pattern of acceleration followed by calm found in the long-run real BCPI, even if their timing does not directly align with periods of urbanization or mass industrialization.

Second, the periodic acceleration found in the long-run real BCPI has a noteworthy degree of cyclicality. Beginning with the First World War, there is a period of accelerated commodity growth roughly every 30 to 35 years. So, while the sources may be different, there may also be a set of responses in economies that is stable across time and that affects economic performance for the following generation.

Super cycles are measured here using the asymmetric band-pass filter developed by Christiano and Fitzgerald (2003). This follows the practice of Jacks (2013) and Cuddington and Jerrett (2008). The band-pass filter is used to decompose the logarithm of a price series into a long-run trend, a super-cycle component spanning 20 to 70 years, and another component spanning less than 20 years, as follows:

$$\ln P\equiv \ln P^{Trend}+\ln P^{Supercycle}+\ln P^{Other}$$

The trend component is illustrated relative to the natural logarithm of the long-run real BCPI in Chart 12, while the deviation from the long-run trend and the super cycle are illustrated in Chart 13. The long-run trend is not required to be constant through time and is permitted to adjust as the low-frequency behaviour of the series evolves. In the case of the logarithm of the long-run real BCPI, the long-run trend first declines during the period of generally falling prices in the 1870s and 1880s, before turning positive until the late 1960s. It then declines modestly until the end of the sample period.

Data table for Chart 12

Data table for Chart 12
Table summary
This table displays the results of Data table for Chart 12. The information is grouped by Year (appearing as row headers), Real BCPI, total, all commodities and Long-run trend, calculated using BCPI/PFDE U.S. units of measure (appearing as column headers).

Year	Real BCPI, total, all commodities	Long-run trend
	BCPI/PFDE U.S.
1870	122.53	116.38
1871	108.07	116.02
1872	106.99	115.66
1873	115.51	115.30
1874	119.10	114.94
1875	117.98	114.58
1876	115.48	114.22
1877	105.28	113.86
1878	98.72	113.50
1879	103.86	113.15
1880	111.75	112.80
1881	133.75	112.45
1882	113.30	112.11
1883	111.71	111.77
1884	105.17	111.43
1885	108.32	111.10
1886	102.94	110.78
1887	109.22	110.47
1888	111.02	110.16
1889	90.50	109.86
1890	112.02	109.56
1891	102.92	109.28
1892	102.19	109.00
1893	102.64	108.74
1894	99.57	108.48
1895	92.81	108.23
1896	93.82	107.99
1897	106.76	107.76
1898	97.30	107.55
1899	99.32	107.34
1900	104.77	107.14
1901	111.02	106.95
1902	110.26	106.78
1903	108.32	106.61
1904	101.35	106.45
1905	103.10	106.31
1906	102.46	106.17
1907	110.39	106.05
1908	116.65	105.94
1909	121.81	105.84
1910	114.68	105.75
1911	119.91	105.67
1912	112.88	105.60
1913	120.28	105.54
1914	123.51	105.49
1915	116.96	105.45
1916	144.91	105.42
1917	164.65	105.40
1918	147.41	105.39
1919	141.20	105.39
1920	107.67	105.40
1921	87.02	105.42
1922	86.41	105.45
1923	83.64	105.49
1924	94.50	105.54
1925	108.21	105.60
1926	98.59	105.66
1927	97.31	105.74
1928	87.28	105.82
1929	95.16	105.91
1930	80.87	106.01
1931	66.90	106.12
1932	64.78	106.23
1933	81.30	106.36
1934	88.86	106.49
1935	96.18	106.62
1936	106.61	106.77
1937	101.12	106.92
1938	89.99	107.07
1939	94.74	107.24
1940	91.92	107.40
1941	101.77	107.58
1942	106.17	107.76
1943	116.23	107.94
1944	114.71	108.12
1945	115.66	108.31
1946	123.66	108.51
1947	137.49	108.70
1948	128.74	108.90
1949	121.39	109.10
1950	125.88	109.30
1951	130.04	109.50
1952	127.52	109.70
1953	120.44	109.90
1954	120.62	110.10
1955	116.60	110.30
1956	115.13	110.49
1957	113.10	110.68
1958	111.68	110.87
1959	109.69	111.05
1960	107.91	111.23
1961	107.39	111.40
1962	107.36	111.56
1963	104.07	111.72
1964	99.16	111.87
1965	100.03	112.01
1966	102.71	112.14
1967	97.75	112.25
1968	95.93	112.36
1969	97.45	112.45
1970	96.23	112.53
1971	93.57	112.60
1972	100.00	112.65
1973	123.35	112.69
1974	144.53	112.71
1975	136.83	112.72
1976	132.17	112.71
1977	124.97	112.68
1978	126.89	112.63
1979	139.53	112.57
1980	151.85	112.48
1981	147.46	112.38
1982	129.84	112.25
1983	125.62	112.11
1984	118.42	111.95
1985	105.14	111.76
1986	88.63	111.56
1987	96.12	111.33
1988	101.26	111.09
1989	101.59	110.82
1990	97.94	110.53
1991	86.19	110.23
1992	84.57	109.90
1993	83.96	109.55
1994	84.10	109.19
1995	87.71	108.80
1996	91.48	108.40
1997	85.85	107.98
1998	72.17	107.54
1999	77.26	107.09
2000	95.22	106.61
2001	82.57	106.13
2002	77.10	105.63
2003	91.07	105.11
2004	107.41	104.59
2005	130.97	104.05
2006	134.24	103.50
2007	139.42	102.95
2008	164.17	102.38
2009	110.48	101.81
2010	133.06	101.23
2011	153.02	100.64
2012	141.02	100.06
2013	138.79	99.47
2014	134.07	98.87
2015	85.43	98.28
Notes: BCPI stands for Bank of Canada commodity price index. PFDE U.S. stands for U.S. final domestic expenditure price index. Source: Statistics Canada, author's calculations based on CANSIM table 383-0027.

The deviation from the long-run trend and the super-cycle component follow a similar path, with the difference between the two representing other shorter-term deviations that reflect idiosyncratic shocks and business cycles. The super cycle in the long-run real BCPI has peaks in 1915, 1949, 1980 and 2009, with 34 years between the first two peaks, 31 years between the second two peaks and 29 years between the last two. The troughs occur in 1898, 1932, 1966 and 1995, with 34, 34 and 29 years, respectively, between them. On average, the cycles take an average of 15.5 years to move from a trough to a peak and 16.3 years to move from a peak to a trough.

Across the cycles, 1950s is the mildest, while the trough that coincides with the commodity price weakness in the 1990s is the deepest. The highest peak is the final peak (2009), which has the greatest amplitude of all cycles in the sample.

Data table for Chart 13

Data table for Chart 13
Table summary
This table displays the results of Data table for Chart 13. The information is grouped by Year (appearing as row headers), Detrended real BCPI and Super cycle, calculated using deviation units of measure (appearing as column headers).

Year	Detrended real BCPI	Super cycle
	deviation
1870	6.15	0.74
1871	-7.95	0.21
1872	-8.67	-0.32
1873	0.21	-0.83
1874	4.15	-1.30
1875	3.40	-1.71
1876	1.26	-2.04
1877	-8.58	-2.30
1878	-14.78	-2.46
1879	-9.29	-2.55
1880	-1.05	-2.56
1881	21.30	-2.51
1882	1.20	-2.42
1883	-0.06	-2.33
1884	-6.26	-2.25
1885	-2.78	-2.22
1886	-7.84	-2.27
1887	-1.25	-2.42
1888	0.86	-2.69
1889	-19.35	-3.08
1890	2.45	-3.59
1891	-6.36	-4.20
1892	-6.82	-4.90
1893	-6.10	-5.64
1894	-8.91	-6.37
1895	-15.42	-7.04
1896	-14.17	-7.59
1897	-1.00	-7.94
1898	-10.25	-8.03
1899	-8.02	-7.81
1900	-2.37	-7.23
1901	4.06	-6.25
1902	3.48	-4.87
1903	1.71	-3.08
1904	-5.10	-0.91
1905	-3.21	1.58
1906	-3.72	4.33
1907	4.34	7.24
1908	10.71	10.21
1909	15.97	13.12
1910	8.94	15.83
1911	14.24	18.22
1912	7.28	20.18
1913	14.74	21.58
1914	18.03	22.34
1915	11.51	22.39
1916	39.49	21.68
1917	59.25	20.22
1918	42.02	18.01
1919	35.81	15.13
1920	2.27	11.65
1921	-18.40	7.69
1922	-19.04	3.38
1923	-21.85	-1.13
1924	-11.04	-5.66
1925	2.61	-10.07
1926	-7.08	-14.18
1927	-8.43	-17.85
1928	-18.54	-20.94
1929	-10.76	-23.37
1930	-25.15	-25.04
1931	-39.22	-25.91
1932	-41.46	-25.96
1933	-25.05	-25.22
1934	-17.63	-23.73
1935	-10.45	-21.56
1936	-0.16	-18.80
1937	-5.80	-15.58
1938	-17.09	-12.01
1939	-12.50	-8.22
1940	-15.48	-4.34
1941	-5.81	-0.51
1942	-1.58	3.17
1943	8.29	6.58
1944	6.59	9.64
1945	7.34	12.27
1946	15.16	14.41
1947	28.79	16.03
1948	19.84	17.10
1949	12.29	17.60
1950	16.58	17.56
1951	20.54	16.97
1952	17.82	15.89
1953	10.53	14.34
1954	10.52	12.38
1955	6.30	10.07
1956	4.64	7.49
1957	2.42	4.71
1958	0.81	1.83
1959	-1.37	-1.06
1960	-3.32	-3.85
1961	-4.01	-6.44
1962	-4.20	-8.72
1963	-7.65	-10.60
1964	-12.70	-11.96
1965	-11.97	-12.74
1966	-9.42	-12.86
1967	-14.51	-12.29
1968	-16.43	-11.01
1969	-15.00	-9.05
1970	-16.30	-6.44
1971	-19.03	-3.30
1972	-12.65	0.28
1973	10.66	4.12
1974	31.82	8.05
1975	24.11	11.87
1976	19.46	15.38
1977	12.29	18.37
1978	14.26	20.63
1979	26.97	22.02
1980	39.37	22.38
1981	35.08	21.63
1982	17.59	19.74
1983	13.51	16.72
1984	6.48	12.66
1985	-6.62	7.71
1986	-22.93	2.04
1987	-15.21	-4.08
1988	-9.83	-10.39
1989	-9.23	-16.58
1990	-12.60	-22.36
1991	-24.04	-27.41
1992	-25.33	-31.49
1993	-25.59	-34.36
1994	-25.09	-35.86
1995	-21.09	-35.88
1996	-16.92	-34.39
1997	-22.13	-31.44
1998	-35.37	-27.15
1999	-29.83	-21.71
2000	-11.39	-15.37
2001	-23.56	-8.43
2002	-28.52	-1.20
2003	-14.05	5.98
2004	2.82	12.78
2005	26.92	18.90
2006	30.73	24.07
2007	36.47	28.09
2008	61.79	30.80
2009	8.67	32.14
2010	31.83	32.11
2011	52.38	30.77
2012	40.96	28.25
2013	39.32	24.75
2014	35.19	20.50
2015	-12.86	15.75
Note: BCPI stands for Bank of Canada commodity price index. Source: Statistics Canada, authors' calculations.

Data table for Chart 14

Data table for Chart 14
Table summary
This table displays the results of Data table for Chart 14. The information is grouped by Year (appearing as row headers), BCPI, total, all commodities, Energy, Metals and minerals, Agriculture and Forestry, calculated using deviation units of measure (appearing as column headers).

Year	BCPI, total, all commodities	Energy	Metals and minerals	Agriculture	Forestry
	deviation
1870	0.74	-0.59	Note ...: not applicable	0.22	2.26
1871	0.21	-1.86	Note ...: not applicable	-0.52	1.85
1872	-0.32	-2.78	Note ...: not applicable	-1.26	1.38
1873	-0.83	-3.32	Note ...: not applicable	-1.97	0.88
1874	-1.30	-3.45	Note ...: not applicable	-2.64	0.39
1875	-1.71	-3.20	Note ...: not applicable	-3.25	-0.05
1876	-2.04	-2.61	Note ...: not applicable	-3.79	-0.39
1877	-2.30	-1.78	Note ...: not applicable	-4.25	-0.63
1878	-2.46	-0.79	Note ...: not applicable	-4.62	-0.73
1879	-2.55	0.25	Note ...: not applicable	-4.92	-0.70
1880	-2.56	1.23	Note ...: not applicable	-5.15	-0.56
1881	-2.51	2.07	Note ...: not applicable	-5.33	-0.33
1882	-2.42	2.69	Note ...: not applicable	-5.46	-0.06
1883	-2.33	3.05	Note ...: not applicable	-5.59	0.20
1884	-2.25	3.14	Note ...: not applicable	-5.72	0.39
1885	-2.22	2.99	Note ...: not applicable	-5.89	0.47
1886	-2.27	2.63	8.39	-6.12	0.37
1887	-2.42	2.14	10.12	-6.42	0.07
1888	-2.69	1.59	11.28	-6.81	-0.46
1889	-3.08	1.08	11.79	-7.28	-1.24
1890	-3.59	0.68	11.60	-7.83	-2.23
1891	-4.20	0.46	10.70	-8.43	-3.42
1892	-4.90	0.45	9.17	-9.06	-4.73
1893	-5.64	0.67	7.11	-9.67	-6.11
1894	-6.37	1.11	4.67	-10.20	-7.46
1895	-7.04	1.70	2.06	-10.60	-8.70
1896	-7.59	2.38	-0.53	-10.78	-9.72
1897	-7.94	3.05	-2.86	-10.68	-10.46
1898	-8.03	3.60	-4.75	-10.24	-10.84
1899	-7.81	3.92	-6.02	-9.38	-10.81
1900	-7.23	3.92	-6.53	-8.07	-10.33
1901	-6.25	3.52	-6.22	-6.26	-9.40
1902	-4.87	2.69	-5.08	-3.95	-8.05
1903	-3.08	1.41	-3.16	-1.13	-6.31
1904	-0.91	-0.29	-0.59	2.14	-4.25
1905	1.58	-2.33	2.45	5.82	-1.95
1906	4.33	-4.60	5.73	9.81	0.48
1907	7.24	-6.98	9.00	14.00	2.97
1908	10.21	-9.31	12.00	18.26	5.39
1909	13.12	-11.43	14.45	22.42	7.68
1910	15.83	-13.19	16.12	26.32	9.74
1911	18.22	-14.46	16.84	29.80	11.53
1912	20.18	-15.13	16.47	32.68	12.99
1913	21.58	-15.13	14.97	34.80	14.11
1914	22.34	-14.44	12.35	36.01	14.87
1915	22.39	-13.08	8.72	36.20	15.30
1916	21.68	-11.11	4.25	35.27	15.41
1917	20.22	-8.65	-0.82	33.19	15.23
1918	18.01	-5.81	-6.20	29.94	14.78
1919	15.13	-2.75	-11.60	25.57	14.10
1920	11.65	0.37	-16.68	20.15	13.21
1921	7.69	3.39	-21.14	13.83	12.12
1922	3.38	6.17	-24.72	6.78	10.83
1923	-1.13	8.58	-27.20	-0.79	9.35
1924	-5.66	10.53	-28.42	-8.64	7.66
1925	-10.07	11.97	-28.31	-16.49	5.76
1926	-14.18	12.86	-26.88	-24.06	3.63
1927	-17.85	13.21	-24.20	-31.08	1.28
1928	-20.94	13.05	-20.44	-37.28	-1.28
1929	-23.37	12.44	-15.81	-42.41	-4.02
1930	-25.04	11.44	-10.58	-46.26	-6.88
1931	-25.91	10.14	-5.04	-48.66	-9.80
1932	-25.96	8.60	0.50	-49.49	-12.68
1933	-25.22	6.90	5.77	-48.69	-15.40
1934	-23.73	5.11	10.49	-46.29	-17.85
1935	-21.56	3.28	14.45	-42.34	-19.89
1936	-18.80	1.48	17.49	-36.99	-21.41
1937	-15.58	-0.25	19.50	-30.43	-22.28
1938	-12.01	-1.87	20.46	-22.90	-22.42
1939	-8.22	-3.32	20.40	-14.71	-21.78
1940	-4.34	-4.55	19.41	-6.16	-20.32
1941	-0.51	-5.50	17.62	2.40	-18.08
1942	3.17	-6.11	15.22	10.64	-15.12
1943	6.58	-6.30	12.40	18.24	-11.56
1944	9.64	-6.03	9.37	24.91	-7.53
1945	12.27	-5.23	6.31	30.39	-3.24
1946	14.41	-3.89	3.42	34.48	1.13
1947	16.03	-2.00	0.84	37.07	5.35
1948	17.10	0.39	-1.34	38.08	9.22
1949	17.60	3.19	-3.03	37.52	12.55
1950	17.56	6.26	-4.23	35.47	15.17
1951	16.97	9.40	-4.95	32.09	16.96
1952	15.89	12.38	-5.25	27.57	17.86
1953	14.34	14.93	-5.20	22.18	17.85
1954	12.38	16.76	-4.91	16.20	16.96
1955	10.07	17.59	-4.49	9.93	15.30
1956	7.49	17.17	-4.03	3.69	13.01
1957	4.71	15.29	-3.64	-2.22	10.25
1958	1.83	11.85	-3.37	-7.53	7.24
1959	-1.06	6.83	-3.28	-12.01	4.18
1960	-3.85	0.33	-3.37	-15.49	1.28
1961	-6.44	-7.41	-3.65	-17.83	-1.28
1962	-8.72	-16.03	-4.06	-19.00	-3.33
1963	-10.60	-25.06	-4.55	-18.99	-4.78
1964	-11.96	-33.92	-5.02	-17.88	-5.56
1965	-12.74	-41.96	-5.38	-15.79	-5.66
1966	-12.86	-48.54	-5.55	-12.89	-5.12
1967	-12.29	-53.01	-5.42	-9.39	-4.05
1968	-11.01	-54.80	-4.93	-5.51	-2.56
1969	-9.05	-53.46	-4.02	-1.49	-0.82
1970	-6.44	-48.72	-2.68	2.44	1.00
1971	-3.30	-40.49	-0.90	6.07	2.73
1972	0.28	-28.92	1.28	9.22	4.20
1973	4.12	-14.39	3.78	11.74	5.27
1974	8.05	2.48	6.50	13.54	5.86
1975	11.87	20.86	9.32	14.55	5.89
1976	15.38	39.77	12.08	14.77	5.37
1977	18.37	58.09	14.63	14.23	4.33
1978	20.63	74.69	16.81	13.02	2.84
1979	22.02	88.42	18.47	11.22	1.03
1980	22.38	98.26	19.46	8.98	-0.96
1981	21.63	103.37	19.68	6.42	-2.98
1982	19.74	103.14	19.04	3.70	-4.87
1983	16.72	97.26	17.50	0.95	-6.50
1984	12.66	85.74	15.05	-1.71	-7.73
1985	7.71	68.95	11.74	-4.16	-8.47
1986	2.04	47.58	7.67	-6.33	-8.68
1987	-4.08	22.65	2.96	-8.15	-8.35
1988	-10.39	-4.60	-2.22	-9.60	-7.51
1989	-16.58	-32.72	-7.66	-10.67	-6.23
1990	-22.36	-60.18	-13.13	-11.36	-4.61
1991	-27.41	-85.44	-18.39	-11.72	-2.79
1992	-31.49	-107.04	-23.21	-11.76	-0.89
1993	-34.36	-123.72	-27.34	-11.55	0.93
1994	-35.86	-134.49	-30.58	-11.13	2.56
1995	-35.88	-138.67	-32.76	-10.54	3.87
1996	-34.39	-135.96	-33.75	-9.83	4.79
1997	-31.44	-126.48	-33.45	-9.01	5.27
1998	-27.15	-110.71	-31.86	-8.12	5.30
1999	-21.71	-89.50	-29.02	-7.16	4.89
2000	-15.37	-64.03	-25.02	-6.15	4.11
2001	-8.43	-35.68	-20.02	-5.07	3.04
2002	-1.20	-6.00	-14.23	-3.94	1.77
2003	5.98	23.43	-7.90	-2.75	0.43
2004	12.78	51.04	-1.29	-1.51	-0.87
2005	18.90	75.44	5.29	-0.22	-2.03
2006	24.07	95.45	11.56	1.09	-2.96
2007	28.09	110.17	17.24	2.40	-3.60
2008	30.80	119.05	22.10	3.69	-3.90
2009	32.14	121.89	25.94	4.90	-3.87
2010	32.11	118.85	28.61	6.00	-3.51
2011	30.77	110.42	30.04	6.94	-2.88
2012	28.25	97.38	30.20	7.70	-2.04
2013	24.75	80.74	29.14	8.22	-1.07
2014	20.50	61.64	26.98	8.50	-0.06
2015	15.75	41.31	23.89	8.52	0.91
... not applicable Note: BCPI stands for Bank of Canada commodity price index. Source: Statistics Canada, authors' calculations.

Across the sub-indexes of the long-run real BCPI, the super-cycle components display some synchronicity. The peaks and troughs of the super cycles for agriculture and forestry often occur around the same time, although their amplitudes can be quite different. The super cycles for metals and minerals occur around the same time as those for agriculture and forestry in the first and final cycles, but behave differently from the 1930s to the 1960s. The energy super cycle in the first half of the period is negatively correlated with those for the other sub-indexes, but all sub-indexes become positively correlated in the second half of the period. The change in behaviour coincides with an increase in the amplitude of the energy cycle.

4.4.5 Commodity price changes in Canada

The long-run BCPI measures price changes in U.S. dollars for commodities important for Canada. Therefore, it does not reflect price changes relevant to Canadian producers when the U.S.–Canada exchange rate changes. For most of the period in question, Canada has had an exchange rate with the United States that is at or near parity. However, during the last 40 years, in the post-Bretton Woods period, Canada had a floating exchange rate. This creates a divergence between the change in the price on world markets and the change in the price relevant to Canadian producers. To illustrate this point and the different behaviour of the post-Bretton Woods period, the long-run real BCPI is presented in terms of U.S. and Canadian dollars in Chart 15.

Data table for Chart 15

Data table for Chart 15
Table summary
This table displays the results of Data table for Chart 15. The information is grouped by Year (appearing as row headers), BCPI (CAN$) / PFDE Canada and BCPI (US$) / PFDE United States, calculated using index units of measure (appearing as column headers).

Year	BCPI (CAN$) / PFDE Canada	BCPI (US$) / PFDE United States
	index
1870	155.3	128.8
1871	134.8	113.6
1872	118.0	112.4
1873	126.3	121.4
1874	132.0	125.2
1875	126.3	124.0
1876	122.2	121.4
1877	122.1	110.6
1878	114.0	103.7
1879	117.4	109.1
1880	128.9	117.4
1881	147.0	140.6
1882	121.3	119.1
1883	116.5	117.4
1884	118.5	110.5
1885	117.7	113.8
1886	112.2	108.2
1887	112.5	114.8
1888	120.3	116.7
1889	95.1	95.1
1890	117.2	117.7
1891	107.2	108.2
1892	105.3	107.4
1893	108.7	107.9
1894	109.8	104.6
1895	102.5	97.5
1896	100.8	98.6
1897	111.5	112.2
1898	99.9	102.2
1899	105.8	104.4
1900	108.8	110.1
1901	114.7	116.7
1902	112.0	115.9
1903	110.5	113.8
1904	103.5	106.5
1905	103.9	108.3
1906	104.6	107.7
1907	109.6	116.0
1908	112.1	122.6
1909	115.8	128.0
1910	112.5	120.5
1911	114.4	126.0
1912	106.2	118.6
1913	111.2	126.4
1914	116.2	129.8
1915	109.5	122.9
1916	134.2	152.3
1917	153.0	173.0
1918	144.0	154.9
1919	147.0	148.4
1920	122.6	113.2
1921	95.6	91.4
1922	88.7	90.8
1923	87.7	87.9
1924	100.4	99.3
1925	116.0	113.7
1926	106.0	103.6
1927	104.4	102.3
1928	92.3	91.7
1929	100.0	100.0
1930	83.2	85.0
1931	68.0	70.3
1932	69.9	68.1
1933	84.0	85.4
1934	86.7	93.4
1935	97.2	101.1
1936	105.6	112.0
1937	101.8	106.3
1938	89.2	94.6
1939	96.1	99.6
1940	95.6	96.6
1941	104.0	107.0
1942	111.7	111.6
1943	123.6	122.1
1944	122.3	120.5
1945	123.7	121.5
1946	140.4	130.0
1947	151.7	144.5
1948	134.7	135.3
1949	121.9	127.6
1950	133.2	132.3
1951	124.8	136.7
1952	115.7	134.0
1953	110.6	126.6
1954	108.9	126.8
1955	110.5	122.5
1956	106.4	121.0
1957	101.8	118.9
1958	102.2	117.4
1959	99.0	115.3
1960	98.5	113.4
1961	102.7	112.9
1962	108.0	112.8
1963	104.6	109.4
1964	98.6	104.2
1965	97.8	105.1
1966	98.4	107.9
1967	92.3	102.7
1968	90.3	100.8
1969	91.6	102.4
1970	88.7	101.1
1971	83.2	98.3
1972	86.3	105.1
1973	104.9	129.6
1974	117.4	151.9
1975	113.2	143.8
1976	102.9	138.9
1977	103.7	131.3
1978	112.4	133.4
1979	126.7	146.6
1980	138.3	159.6
1981	134.5	155.0
1982	117.3	136.5
1983	111.1	132.0
1984	109.9	124.5
1985	102.1	110.5
1986	85.8	93.1
1987	87.8	101.0
1988	85.2	106.4
1989	81.9	106.8
1990	77.9	102.9
1991	66.9	90.6
1992	69.5	88.9
1993	74.1	88.2
1994	78.8	88.4
1995	83.1	92.2
1996	86.4	96.1
1997	82.2	90.2
1998	73.3	75.8
1999	79.1	81.2
2000	97.5	100.1
2001	88.3	86.8
2002	83.0	81.0
2003	87.8	95.7
2004	97.1	112.9
2005	111.8	137.6
2006	108.1	141.1
2007	106.8	146.5
2008	125.5	172.5
2009	89.2	116.1
2010	97.2	139.8
2011	107.6	160.8
2012	99.8	148.2
2013	101.8	145.9
2014	105.1	140.9
2015	78.7	89.8
Notes: BCPI stands for Bank of Canada commodity price index. PFDE stands for final domestic expenditure price index. Sources: Statistics Canada, author's calculations.

As Chart 15 illustrates, the change in relative commodity prices as measured by the long-run real BCPI is essentially the same in U.S. dollars and Canadian dollars between 1870 and 1970. This is a result of having an exchange rate regime that was, for the most part, fixed at parity with the U.S. dollar, or that remained within about 10 cents of parity during periods when it was floating. The post-Bretton Woods period, however, saw a depreciation of the exchange rate through the 1980s and 1990s to reach $0.636 in 2002, before climbing above parity in the late 2000s.

These exchange-rate movements had the effect of dampening differences in commodity price growth across the years, consistent with the expectation that a floating exchange rate acts as a shock absorber for foreign shocks. As a result, real commodity price changes in Canadian dollar terms are more muted in Canada than in the United States under the current exchange-rate regime. Consequently, the force of commodity price changes in global markets does not exert as much short-run relative price pressure in Canada as it does in the United States.

5 Conclusion

Historical estimates for the BCPI and its sub-indexes have been produced for the period from 1870 to 1971 to link with the current BCPI. The estimated values reflect important historical events, from major economic shocks to the emergence of oil production and the electrification of the production process. Therefore, they contain information relevant for understanding processes related to emerging commodity production and structural change.

They also contain significant information about the timing of pressure exerted by commodity prices in Canada. The long-run BCPI, which links historical estimates to the BCPI, correlates with export prices and wholesale prices, suggesting that the magnitude and timing of events captured by the long-run index are appropriate. However, because the BCPI is chain-weighted, it more accurately measures growth rates than historical fixed-weight sources.

The long-run BCPI and its sub-indexes have growth-rate distributions characterized by skewness and kurtosis relative to a normal distribution. They are, in other words, subject to large innovations, with positive innovations being relatively larger. As a result, measures of average growth, variance, correlation, covariance and regression estimators, such as ordinary least squares, are biased because of the presence of these influential observations. The extent to which this may create problems for analysis and modelling depends on the use to which the data are put.

The long-run BCPI, its sub-indexes and the prices used to construct them also show a process of short periods of pressure, followed by periods of calm. This super-cycle-type pattern is most pronounced in the real long-run BCPI after the end of the classical gold standard at the outbreak of the First World War and often coincides with major changes in monetary regime.

References

Amano, R., and S. van Norden. 1993. “A Forecasting Equation for the Canada-U.S. Dollar Exchange Rate.” In The Exchange Rate and the Economy. Proceedings of the 1992 Bank of Canada Conference, Ottawa, June 22–23, 1992. Ottawa: Bank of Canada. p. 226–271.

Bailliu, J., and M.R. King. 2005. “What Drives Movements in Exchange Rates?” Bank of Canada Review (Autumn): 27–39.

Baldwin, J., and R. Macdonald. 2012. Natural Resources, the Terms of Trade, and Real Income Growth in Canada: 1870 to 2010. Economic Analysis Research Paper Series, no. 76. Statistics Canada Catalogue no. 11F0027M. Ottawa: Statistics Canada.

Balke, N., and M.A. Wynne. 2003. The Relative Price Effects of Monetary Shocks. Research Department Working Paper no. 0306. Dallas, Texas: Federal Reserve Bank of Dallas.

Bertram, G., and M. Percy. 1979. “Real Wage Trends in Canada 1900-26: Some Provisional Estimates.” The Canadian Journal of Economics 12 (2): 299–312.

Bordo, M.D. 1993. The Gold Standard, Bretton Woods and other Monetary Policy Regimes: An historical appraisal. NBER Working Paper Series no. 4310. Cambridge, Massachusetts: National Bureau of Economic Research.

Bordo, M.D. 2008. An Historical Perspective on the Crises of 2007-2008. NBER Working Paper Series no. 14569. Cambridge, Massachusetts: National Bureau of Economic Research.

Brown, M., and R. Macdonald. 2015. Provincial Convergence and Divergence in Canada, 1926 to 2011. Economic Analysis Research Paper Series, no. 96. Statistics Canada Catalogue no. 11F0027M. Ottawa: Statistics Canada.

BLS (Bureau of Labor Statistics). n.d.a. PPI Commodity Data including “headline” FD-ID indexes (for Processed foods and feeds – Unprocessed finfish). Available at: https://download.bls.gov/pub/time.series/wp/wp.data.3.ProocessedFoods (accessed April 21, 2017).

BLS (Bureau of Labor Statistics). n.d.b. PPI Commodity Data including “headline” FD-ID indexes (for Pulp, paper and allied products – Newsprint). Available at: https://download.bls.gov/pub/time.series/wp/wp.data.10.Pulp (accessed April 21, 2017).

Cayen, J.-P., D. Coletti, R. Lalonde, and P. Maier. 2010. What Drives Exchange Rates? New Evidence from a Panel of U.S. Dollar Bilateral Exchange Rates. Bank of Canada Working Paper no. 2010-5. Ottawa: Bank of Canada.

Cheung, C. 2009. Are Commodity Prices Useful Leading Indicators of Inflation? Bank of Canada Discussion Paper no. 2009-5. Ottawa: Bank of Canada.

Christiano, L.J., and T.J. Fitzgerald. 2003. “The Band Pass Filter.” International Economic Review 44 (2): 435–465.

Clarke, T.E. 1999. “The Responses of Prices at Different Stages of Production to Monetary Policy Shocks.” The Review of Economics and Statistics 81 (3): 420–433.

Coletti, D. 1992. “The Long-run Behaviour of Key Canadian Non-energy Commodity Prices (1900 to 1991).” Bank of Canada Review (Winter 1992–1993): 47–56.

Cranstone, D. 2002. A History of Mining and Mineral Exploration in Canada and Outlook for the Future. Ottawa: Metal and Minerals Sector, Natural Resources Canada.

Cuddington, J.T., and D. Jerrett. 2008. Super Cycles in Real Metals Prices? International Monetary Fund Staff Papers, volume 55, no. 4. Washington, D.C.: International Monetary Fund.

Dailami, M., and P. Masson. 2009. The New Multi-polar International Monetary System. World Bank policy research working paper, no. 5147. The World Bank, Development Economics, Development Prospects Group. Available at:
http://documents.worldbank.org/curated/en/470961468041977091/pdf/WPS5147.pdf (accessed April 28, 2017).

Diewert, W.E. 1992. “Fisher Ideal Output, Input, and Productivity Indexes Revisited.” Journal of Productivity Analysis 3 (3): 211–248.

Duguay, P. 2006. Productivity, Terms of Trade, and Economic Adjustment. Remarks to the Canadian Association for Business Economics, Kingston, Ontario, August 28, 2006.

The Economist. 2005. “How China Runs the Global Economy.” July 28. Available at: http://www.economist.com/node/4223552 (accessed January 7, 2016).

EIA (U.S. Energy Information Administration). 2017. U.S. Crude Oil First Purchase Price, Dollars per Barrel (table and chart). Last release date: April 3, 2017. Available at: https://www.eia/gov/dnav/pet/hist/LeafHandler.ashx?n=PET&s=F000000_3&f=A (accessed April 10, 2017).

Erten, B., and J.A. Ocampo. 2013. “Super Cycles of Commodity Prices since the Mid-nineteenth Century.” World Development 44 (C): 14–30.

Fisher, I. 1922. The Making of Index Numbers: A Study of Their Varieties, Tests, and Reliability. Boston: Houghton Mifflin.

Francis, M. 2008. “Adjusting to the Commodity-price Boom: The Experiences of Four Industrialized Countries.” Bank of Canada Review (Autumn): 29–41.

Harris, R., I. Keay, and F. Lewis. 2015. “Protecting Infant Industries: Canadian Manufacturing and the National Policy, 1870 to 1913.” Explorations in Economic History 56 (C): 15–31.

Hirsch, T. 2003. “Updating the Bank of Canada Commodity Price Index.” Bank of Canada Review (Spring): 31–33.

Holmes, A.D. n.d. “Section K: Price Indexes,” Historical Statistics of Canada. Statistics Canada Catalogue no. 11-516-X. Last updated July 2, 2014. Available at: http://www.statcan.gc.ca/pub/11-516-x/sectionk/4057753-eng.htm

Issa, R., R. Lafrance, and J. Murray. 2008. “The Turning Black Tide: Energy Prices and the Canadian Dollar.” Canadian Journal of Economics 41 (3): 737–759.

Jacks, D.S. 2013. From Boom to Bust: A Typology of Real Commodity Prices in the Long Run. NBER Working Paper Series, no. 18874. Cambridge, Massachusetts: National Bureau of Economic Research.

Jerrett, D., and J.T. Cuddington. 2008. “Broadening the Statistical Search for Metal Price Super Cycles to Steel and Related Metals.” Resources Policy 33 (4).

Kelly, T.D., and G.R. Matos, compilers. 2014. Historical statistics for mineral and material commodities in the United States (2016 version). U.S. Geological Survey Data Series 140. Available at: https://minerals.usgs.gov/minerals/pubs/historical-statistics/ (accessed April 10, 2017).

Kohli, U. 2006. “Real GDP, Real GDI and Trading Gains: Canada, 1981-2005.” International Productivity Monitor 13 (Fall): 46–56.

Kolet, I., and R. Macdonald. 2010. The Fisher BCPI: The Bank of Canada’s New Commodity Price Index. Bank of Canada Discussion Paper no. 6-2010. Ottawa: Bank of Canada.

Macdonald, R. 2007. Real GDP and the Purchasing Power of Provincial Output. Economic Analysis Research Paper Series, no. 46. Statistics Canada Catalogue no. 11F0027M. Ottawa: Statistics Canada.

Macdonald, R. 2008. The Terms of Trade and Domestic Spending. Insights on the Canadian Economy, no 18. Statistics Canada Catalogue no. 11-624-M. Ottawa: Statistics Canada.

Macdonald, R. 2010. Real Gross Domestic Income, Relative Prices and Economic Performance Across the OECD. Economic Analysis Research Paper Series, no. 59. Statistics Canada Catalogue no. 11F0027M. Ottawa: Statistics Canada.

Maier, P., and B. DePratto. 2008. The Canadian Dollar and Commodity Prices: Has the Relationship Changed over Time. Bank of Canada Discussion Paper no. 2008-15. Ottawa: Bank of Canada.

Manthy, R.S. 1978. Natural Resource Commodities—A Century of Statistics: Prices, Output, Consumption, Foreign Trade, Employment in the United States 1870–1973. Baltimore: Johns Hopkins University Press.

McDonald, J., A.P. O’Brien, and C. Callahan. 1997. “Trade Wars: Canada’s Reaction to the Smoot-Hawley Tariff.” The Journal of Economic History 57 (4): 802–826.

Morse, N.H. n.d. “Section N: Fisheries,” Historical Statistics of Canada. Statistics Canada Catalogue no. 11-516-X. Last updated July 2, 2014.

NBER (National Bureau of Economic Research). n.d.a. "m04093a, U.S. Wholesale Price of Newsprint Paper 01/1890 –12/1915," NBER Macrohistory: IV. Prices.  Available at: http://www.nber.org/databases/macrohistory/contents/chapter04.html (accessed April 10, 2017).

NBER (National Bureau of Economic Research). n.d.b. "m04093b, U.S. Wholesale Price of Newsprint Paper Rolls, Contract, F.O.B. Mill 01/1914 –08/1919, 03/1920–12/1928," NBER Macrohistory: IV. Prices.  Available at:
http://www.nber.org/databases/macrohistory/contents/chapter04.html (accessed April 10, 2017).

NBER (National Bureau of Economic Research). n.d.c. "m04093c, U.S. Wholesale Price of Newsprint Paper 01/1926 –12/1951," NBER Macrohistory: IV. Prices.  Available at: http://www.nber.org/databases/macrohistory/contents/chapter04.html (accessed April 10, 2017).

NBER (National Bureau of Economic Research). n.d.d. "m04093d, U.S. Wholesale Price of Newsprint Paper 01/1947 –02/1958," NBER Macrohistory: IV. Prices.  Available at: http://www.nber.org/databases/macrohistory/contents/chapter04.html (accessed April 10, 2017).

O’rourke, K. and J.G. Williamson. 1994. “Late Nineteenth-Century Anglo American Factor Price Convergence: Were Hecksher and Ohlin Right?” The Journal of Economic History 54 (4): 892–916.

Osborn. B.S. n.d. “Section L: Lands and Forests,” Historical Statistics of Canada. Statistics Canada Catalogue no. 11-516-X. Last updated July 2, 2014. Available at:
http://www.statcan.gc.ca/pub/11-516-x/sectionl/4147441-eng.htm (accessed March 13, 2017).

Powell, J. 2005. A History of the Canadian Dollar. Ottawa: Bank of Canada.

Quirin, J.D. n.d.a. “Section P: Mining,” Historical Statistics of Canada. Statistics Canada Catalogue no. 11-516-X. Last updated July 2, 2014. Available at:
http://www.statcan.gc.ca/pub/11-516-x/sectionp/4147442-eng.htm (accessed March 13, 2017).

Quirin, J.D. n.d.b “Section Q: Energy and Electric Power,” Historical Statistics of Canada.Statistics Canada Catalogue no. 11-516-X. Last updated July 2, 2014. Available at: http://www.statcan.gc.ca/pub/11-516-x/sectionq/4057756-eng.htm (accessed March 13, 2017).

Statistics Canada. n.d.a. Historical Statistics of Canada. Statistics Canada Catalogue no. 11-516-X. Last updated July 29, 1999. Available at: http://www5.statcan.gc.ca/olc-cel/olc.action?ObjId=11-516-X&ObjType=2&lang=en&Limit=1 (accessed March 13, 2017).

Statistics Canada. n.d.b. Table 001-0010 Estimated areas, yield, production and average farm price of principal field crops, in metric units, annual (table). CANSIM (database). Last updated December 7, 2016. Available at:
http://www5.statcan.gc.ca/cansim/a26?lang=eng&id=0010010&p2=33 (accessed March 20, 2017).

Statistics Canada. n.d.c. Table 381-0009 Inputs and outputs, by industry and commodity, L-level aggregation and North American Industry Classification System (NAICS), annual (dollars x 1,000,000) (table) (archived). CANSIM (database). Last updated November 7, 2011. Available at: http://www5.statcan.gc.ca/cansim/a26?lang=eng&id=3810009&p2=33 (accessed April 10, 2017).

Statistics Canada, n.d.d. Table 152-0005 Principal statistics of mineral industries, by North American Industry Classification System (NAICS), annual (dollars unless otherwise noted) (table) (archived). CANSIM (database). Last updated July 30, 2010. Available at: http://www5.statcan.gc.ca/cansim/a26?lang=eng&id=1520005&p2=50 (accessed April 28, 2017).

Statistics Canada, n.d.e. Table 152-0002 Principal statistics of the mineral industries, annual (dollars unless otherwise noted) (table) (archived). CANSIM (database). Last updated January 23, 2001. Available at:
http://www5.statcan.gc.ca/cansim/a26?lang=eng&id=1520002&p2=33 (accessed May 4, 2017).

Statistics Canada. 2015. Exploring the First Century of Canada’s Consumer Price Index. Statistics Canada Catalogue no. 62-604-X. Ottawa: Statistics Canada.

Trant, G.I. n.d. “Section M: Agriculture,” Historical Statistics of Canada. Statistics Canada Catalogue no. 11-516-X. Last updated July 2, 2014. Available at:
http://www.statcan.gc.ca/pub/11-516-x/sectionm/4057754-eng.htm (accessed March 13, 2017).

Urquhart, M.C. 1993. Gross National Product, Canada: 1870-1926: The Derivation of the Estimates. Montréal and Kingston: McGill-Queen's University Press.