Electricity in Quebec before Nationalization, 1919 to 1939.
The Canadian province of Quebec is one of the foremost producers of hydroelectricity in the world. Its vast network of high current rivers provides an accessible and enormous energy potential. In the early days of electricity, this potential was acknowledged and investments were made to harness it. The history of electricity in Quebec is tightly linked with the economic history of the province.
From the start of the industry in the late nineteenth century, the private electrical industry of Quebec was accused of being monopolistic (Bellavance 1998). The claim was supported by the contention that prices were higher than in neighboring Ontario. The most popular policy proposal was aimed at nationalizing the industry and reducing prices to encourage development. In the 1940s, a partial nationalization of the industry was enacted, and by the early 1960s the rest of the industry had been nationalized. The belief that Quebec's electrical industry was a monopoly during the pre-nationalization era is largely based on the contention that it was a natural monopoly (Dales 1957; Faucher 1992; Bellavance 1998; Bellavance et al. 1999).
In this paper, we argue that a mistakenly bleak portrait of the pre-nationalization era was painted. We do not attempt to assess whether nationalization was a good idea, nor do we attempt to assert that the private providers did not gouge consumers. Rather, the goal of this paper is to properly evaluate the outcomes of the private regime to establish the basis for future research on the process of nationalization and the outcomes of such a policy. We present this paper as a necessary first step for such future research.
The electrical industry in the United States was subject to similar criticism as the Quebec's industry from the late nineteenth century until the mid-twentieth centuiy, and numerous economists and historians have rejected or disputed this claim (Bradley 1996; Smith 1988; Rassenti et al. 2001; Emmons 1991, 1993). The downward trends in prices, the increasing supply of electrical services and the greater quality of the service noted by these authors suggest a more nuanced narrative where a sizeable share of technological improvements was passed on to consumers through lower prices. New evidence collected for this paper suggests that this was also the case in Quebec, where prices fell more rapidly than they did in the United States.
Furthermore, in the absence of legal barriers to entry, even a single firm can behave in a competitive manner since the market can be contested (Coursey et al. 1984), especially if there is a bidding process (Demsetz 1968). Given that, in Quebec, like in the United States, numerous contracts for street lighting and industrial clients were obtained through a bidding process, there is good reason to side with those who are skeptical of the applicability of the natural monopoly view of the electrical industry. Franchise granting, which means the presence of a single firm for a contractually fixed period of time, docs mean that a single firm operates in the market. This is what some denounced. However, Demsetz argued that if there is bidding for the franchise, then firms compete on rates and propose prices that are contractually fixed and similar to prices found in competition (i.e., the absence of monopoly profits). (1)
However, the granting of an exclusive franchise creates incentives to expend resources to obtain that franchise (Tullock 1967; Krueger 1974). Such expenditures could entail lobbying politicians to make the franchise-bidding process favorable to one particular firm. Conversely, the franchise opens the door to problems on the part of politicians. If a franchise has a duration inferior to the life of the capital assets associated with the project, it allows a politician to deny renewal. This denial then undervalues the firm's capital, which can then be acquired at bargain rates by an entering firm (in which case the profits are split with the politician) or by the government that decides to publicly operate the service. While there has been a substantive discussion of how localities managed utilities in the United States and how this affected rates (Emmons 1993; Kitchens and Jaworski 2017), (2) the literature on Quebec fails to identify whether or not there was a failure in franchising through corruption or holding out on the part of politicians (Troesken 2006).
We also find that proponents of nationalization provided misleading comparisons to make their cases (as did the opponents). They compared Quebec's private industry with Ontario's electricity--which is misleading since that province decided that the public sector should be a provider of electricity at cost. By doing so, Ontario subsidized lower prices and greater consumption, which were funded through a higher tax burden (Murray and Flood 1922; Fleming 1983, 1992). In addition, electricity at cost meant greater demand--which the domestic industry could not supply. The shortfall had to be imported from Quebec. This increased demand for electricity produced in Quebec, putting upward pressure on prices there. Thus, the complaints of high prices in Quebec were partly a spillover from Ontario's decision to nationalize. In addition, private firms in Quebec were among the most taxed within Canada.
Moreover, nascent electrical industries were locked in competition with the firewood, coal and gas industries (Stigler and Friedland 1962; Troesken 1996; Fouquet 2008). Quebec had by far the greatest level of per capita consumption of firewood in Canada from 1871 through 1951 (MacFayden 2016), (3) and larger proportions of total expenditures on light and fuel came from firewood (Dominion Bureau of Statistics 1941). (4) Moreover, the point that private and public ownership influences rate-setting across consumer classes quite differently has largely gone unconsidered in the case of Quebec (Peltzman 1971). In the interwar United States, private ownership tended to be cheaper for large energy consumers (Neufeld 1987; Kitchens and Jaworski 2017), which implies that industrial clients tended to gain. As large consumers, industries tend to have more pull on the electricity market--a point that was noted very early on in the case of Quebec by Dales, who pointed out that industrial electricity rates in Quebec were quite lower than those in Ontario for nonferrous industries and modestly inferior in terms of general manufacturing (Dales 1957). Moreover, many companies had the option of using thermoelectricity and self-generation to provide energy (Gelly 2010; Gelly 2003). As such, there were some constraints on monopoly power through these channels.
These facts present a different portrait of the pre-nationalization situation than the one often portrayed. Although it was imperfect in a context where rent-seeking was present, the private market for electricity in Quebec was able to deliver increasing quantities of electricity across a wide territory at lower prices.
Quebec's Electrical System before Nationalization
Quebec is graced with numerous rivers eminently suitable for hydroelectric generation, especially since storage dams and reservoirs could be built at small expense in order to better control the flow from rivers that generated electricity (Dales 1957). This potential was understood and harnessed late in the nineteenth century, as some small generation plants were established close to large potential markets. Production moved quickly inland--notably along the Gatineau River and the rivers that vein through the Mauricie region (Fig. 1). Nonetheless, the "proximity of the power sites to the demand centers ... help[ed] keep transmission costs low" "(Dales 1957, p. 29)", which meant the mileage of transmission lines was quite small in comparison with other provinces, notably Ontario. (5) During the interwar period on which we focus, the industry also consolidated progressively so that only a few highly vertically integrated firms remained (Hogue et al. 1979; Dales 1957). More than 90% of the output of the hydroelectric industry in Quebec was generated by five firms, each with its own geographical market: Montreal Light Heat and Power (MLHP), Shawinigan Water and Power (SWP), Gatineau Power (GP), Saguenay Power (SP) and Southern Canada Power (SCP). (6) There were few municipally-owned generating stations (20 in 1921, 12 in 1928 and 16 in 1940), which accounted for a small fraction of total output (Dales 1957). There were also few public distributors that had to buy electricity from private firms. In 1922, 38 municipalities (representing 7.06% of the population) operated their own distribution network (Department of the Interior 1923). (7) With some localized exceptions, as in the case of Quebec City (where it was deemed ineffectual), there was no wide-ranging regulation of electricity rates until the province-wide regulations of 1935, which have been considered largely irrelevant in affecting prices (Dales 1957; Dupre et al. 1996). (8)
What attention has been paid to the state of this industry before nationalization has centered on the issue of unproductive "trusts" which gouged consumers and municipal governments. Dubbed the "electrical question" (Faucher 1992, p. 417), the general contention is that they charged prohibitive prices, especially for farmers and small citydwellers (Faucher 1992; Boutet 1999). To substantiate this claim, numerous historians have pointed out the fact that prices were lower in the neighboring province of Ontario where electricity distribution and generation had been nationalized (Dorion 2000; Boutet 1999; Dales 1957; Faucher 1992; Piche 1937). If Quebec could nationalize its industry, they claimed, it would follow Ontario's economic path.
In his efforts to accelerate the industrialization of the province, Premier Louis-Alexandre Taschereau supported the development of the electrical industry believing it would stimulate growth. Yet, even within his own Liberal party, he faced conflict over the issue of nationalizing electricity with members of his party favoring the creation of a publicly owned distribution system. Eventually, a dissenting group of Liberals split from the party. Dr. Philippe Hamel is probably the most interesting supporter of nationalization. A dentist by training, he spent hours reading bulletins from the Federal Trade Commission in the United States, believing that the industry in Quebec was a trust (Faucher 1992). To support his claim, he provided evidence that rates in Quebec were considerably higher than in Ontario. In 1944, the liberal government had nationalized MHLP. In 1962, the remaining firms would be nationalized. (9) The general opinion is that this nationalization process was fairly positive (Hogue et al. 1979; Fleury 2004) or that, at the very least, it was an improvement relative to the state of affairs when the industry was privately owned. This is the accepted narrative.
However, a first and very brief glance at the evidence indicates several problems with the account of the industry prior to nationalization. Quebec was one of the poorest provinces in Canada (Geloso 2017; Brown and Macdonald 2015). Reasonably, we can assert that a province whose citizens were poorer than the average citizen should also have had fewer customers for electricity relative to the national level. However, as Table 1 illustrates, this was not the case. The number of electricity consumers per inhabitant in Quebec as a percentage of the same in Canada was actually very high at 98.5%, A surprising figure given that its personal income per capita was roughly 20% below the Canadian average. Of course, this does not provide sufficient proof to invalidate the current dominant opinion, but it does create enough doubt to warrant further inquiry.
There are further causes for doubting the conventional narrative, all of which revolve around the constant contemporary comparison made with Ontario. This comparison, which has inspired numerous justifications of the benefit of nationalization, is flawed.
The first is that the nationalized system of Ontario was bound to show lower prices for consumers. A part of this is explained by the lower borrowing costs resulting from public ownership. Work by Emmons (1997) (10) on the United States during the interwar period showed that prices were 5.3% lower in large cities served by publicly-owned firms than their counterparts served by regulated private firms and more than 10% lower than unregulated private firms. Emmons found that half of the price differential between public and private was explained by this capital cost advantage.
Secondly, Dales (1957) suggests that the electrical industry in Quebec actually did offer lower prices with regard to large clients. As we will see below, Ontario's nationalized industry which had to provide electricity at cost did not match the low industrial prices of Quebec. This was also observed in the United States whereby residential clients with high levels of consumption enjoyed substantially lower prices (Kitchens and Jaworski 2017). Given the importance of energy-intensive industries in Quebec, like aluminum, paper factories and pulp factories, this is an important feature (Dales 1953). In addition, lower industrial rates may result from the potential for self-generation. Quebec's important industries in the 1920s and 1930s were in the metallurgy, non-ferrous metals and pulp and paper sectors. In these industries, there was a potential for relying on self-generation, which meant that their demand for electricity produced from central generators was more elastic (Neufeld 1987; Bradley 1996: 61). Allain Gelly (2003, 2010: 208-286) surveyed the practices of industries in Quebec and argued that they could reasonably consider thermoelectricity (coal and gas) as viable alternatives, which made their demand more elastic. (11) Moreover, the price discrimination might have been necessary for the expansion of the network--the volume of consumption in small communities is so small that alone they cannot support the network. Thus, high rates for clients with low volume and low rates for clients with high volumes must be secured simultaneously for the network to exist and expand.
More importantly, when one compares the two provinces, one compares a partially nationalized market that was mandated to provide electricity at a cost (Ontario) to an unregulated market (Quebec). In fact, Ontario was unique in Canada with its completely public system "(Dupre and Patry 1998, p. 120)". Set up in the early years of the twentieth century, the provincial electric utility, the Hydro Electric Power Commission (HEPC), was established to provide power at cost. At first, HEPC bought electricity from private providers and distributed electricity at cost on its transmission lines. It would then buy the city-owned distributors, with the cost difference made up through taxation (Bliss 1987). However, after the First World War, HEPC started to acquire private generators and construct its own generators. In 1921, it made a "clean-up deal" where it bought the main remaining private competitors (Dupre and Patry 1998: 126-129). By the end of the 1920s, most of the private providers in rural areas had closed down in the face of "the expanding Hydro monolith" "(Fleming 1992, p. 126)." In essence, HEPC had virtually been substituted for private industry. By 1930, the bulk of Ontario's energy was supplied by 22 hydroelectric plants owned and operated by HEPC (Dominion Bureau of Statistics 1930). (12) By charging electricity at a cost, the systematic comparison between Quebec and Ontario with regard to electricity markets is biased in favor of finding lower prices in Ontario.
In 1922, generating costs in Quebec were cheaper by 32% than in Ontario, even if domestic rates were lower in Ontario (Murray and Flood 1922). (13) Thus, even though the federal government exempted municipal services and Crown corporations from paying taxes, which put the Ontario electrical system at an advantage (Boutet 1999), it was relatively less efficient than Quebec's. (14) Moreover, taxpayers had to pay, since the financial liabilities incurred by the Hydro Electric Power Commission rose from less than $4.3 million in 1911 to just below $98.4 million by 1920 (Murray and Flood 1922). Ontarians paid for this by having higher taxes than Quebec (ibid). When comparing 35 Quebec cities with 66 Ontario cities, we see that the 1920 per capita tax burden was significantly higher in Ontario (ibid) as a result of HEPC's activities. Comparisons with Ontarian cities HEPC had not yet penetrated in 1920 also show that the tax burden was lower with lower per capita debt burden (ibid). We can use the personal income and personal disposable income data made available by Brown and Macdonald (2015) to measure the difference in the tax burden. Although not a perfect measure, the difference between the personal income measure and personal disposable income measure gives us the effective tax burden of the average person. Using that approach, we find that this effective tax burden stood at 2.59% in Quebec between 1926 and 1939 compared with 4.52% in Ontario.
More importantly, complaints about high prices are rarely considered possible outcomes of government policies rather than as straightforward market outcomes. On the one hand, in Quebec, the taxes that private electrical firms had to pay stood well above those in the rest of Canada. Table 2 illustrates the share that taxes represented within the total expenses of the industry. As one can see, the tax burden in Quebec was higher than in Ontario and the rest of Canada, and it roughly doubled between 1925 and 1939 to the point of representing close to one-third of total expenses, a factor that would have weighed in determining interprovincial price differences. (15)
On the other hand, the legislative process leading to the granting of franchises should be considered. Franchises were contracts between a utility and a municipality where rights and obligations were contractually specified. This acted as a mild form of regulatory control. Franchises could be considered as competition ex ante (or competition for the field), where competitive auctions for the exclusive right to operate would attract firms to the bidding process. As long as the process was fair and open, the utility that won the auction would offer prices equal to those in a competitive setting, even if it had a franchise (Demsetz 1968). (16) However, if the process is rigged in favor of a firm, then noncompetitive outcomes can emerge. Moreover, there is also the chance for political opportunism, as politicians might be tempted to create holdup problems when the time comes to renew the franchise (Troesken 2006). A franchise whose duration is shorter than the life of capital assets invested by a firm could be denied renewal, which would lead to the undervaluation of its capital. Then, the firm's capital could be acquired at bargain rates by the new franchisee (or by the government), if it decides to publicly produce the service. These problems arise because the monopoly franchise creates the incentive to rent-seek (Tullock 1967). If the resources expended in attempting to obtain the exclusive franchise equal the value of the economic rent (i.e., complete rent dissipation), then monopoly profits from the acquired franchise should be regarded as social losses (Hillman 2015).
The city of Quebec, which was also the main breeding ground for the nationalization movement and was reputed to have the highest prices in the province (Boutet 1999)--illustrates all of these possibilities. While most cities tended to confer franchises lasting 25 years, as in the case of Montreal, Quebec City offered a franchise of ten years (ibid. 1999), which was subjected to intense political bargaining. By 1912, there had been several consolidations that had led Quebec Railway, Light, Heat & Power (QRLHP), which was presided over by a sitting Conservative Member of Parliament, to become the last remaining firm. However, in 1912, the city's liberal mayor refused to renew the 1902 franchise and offered its first franchise of ten years to a new firm, Dorchester Electric, which had lobbied for the franchise.
It quickly proved unreliable and was bought back by SWP (and the franchise was extended to 1925). QRLHP was unable to compete with the franchised firm, which had the exclusive right to light streets and public places and had to declare bankruptcy, only to be bought at a bargain by the franchised firm (Hogue et al. 1979). (17) Then, the city of Quebec created a regulatory commission in charge of overseeing rates (the Commission des Services Publics de Quebec), which historians have described as being under the control of the franchised firm (Boutet 1999). Later, at mid-course of the franchise that lasted from 1925 to 1935, municipal politicians sought legal advice to see if they could repudiate the franchise. They also lobbied the provincial government to obtain an amendment to the municipal charter that would allow them to municipalize the service. All of this occurred while they were trying to negotiate lower rates. The case of Quebec City suggests that there is a problem caused by the lack of attention devoted to the issue of how franchises were conceded in Quebec as a whole. Heated political battles around the franchising process may have led to political holdup, corruption in the attribution of contract and rent dissipation in the rent-seeking process, all of which would have been vectors for discontent toward private firms.
This is where we should also question the sources of the arguments against private firms. As mentioned above, many of the criticisms came from self-interested politicians who played to their constituencies. One should also consider the importance of farmers in this debate. The number of farms reported in the 1921 census was equal to 31 % of all families in the province compared with 25% in the 1931 census. This was a substantial share of voters who desired cheap electricity. (18) Farms in Quebec had electrified more slowly than farms in Ontario, a point about which farmers frequently complained and proposed redress for through special financial measures (Dorion 2000). The share of electoral districts in rural areas being considerable, it incited politicians to engage the issue, especially during the Great Depression. In such an environment, facts might have been discarded in favor of political rhetoric. For farmers, mandating governments to provide cheaper electricity was a form of rent-seeking, and given their vested interest and their importance in the debate, we should be skeptical of the reliability of the narrative they advanced to justify their demands .A more cautious and nuanced look is required to assess the pre-nationalization market.
Actual Performance of the Industry
To adequately measure the performance, it is necessary to rely on numerous sources of official data. Much of this data is drawn from the Central Electrical Stations booklets produced by the Dominion Bureau of Statistics (DBS). We complement this source with evidence from the the Quebec Statistical Year Book and other archived publications of the DBS. Taken together, these sources give cause for a cautiously more optimistic assessment of the industry's performance. The most damning piece of evidence comes from prices. Until the Great Depression, prices were falling in both nominal and real terms at a very fast pace. During the Depression, prices went up slightly and then went down again. By 1939, real prices were not as low as they had been during the 1920s, but they were near historical lows and declining progressively. But this first piece of evidence has some limitations since the price quotations prior to 1930 are collected from the work of MHLP (1934) and Boutet (1999) while the DBS statistics are a composite of prices across the province (Fig. 2). Furthermore, the fall in prices in Quebec was actually more pronounced than that seen for the rest of Canada as a whole between 1913 and 1939. Between 1910 and 1932, prices in Quebec had fallen (in nominal terms) by 50.2% compared with 40% and 44.4% for the Canadian and Ontarian figures (Dominion Bureau of Statistics 1927; Dominion Bureau of Statistics 1940: 4). Moreover, the evolution of those numbers is similar to that of the United States and the United Kingdom (Fig. 3). In terms of comparison with the United States, Quebec as a whole fared very well. Prices were much lower than in most American cities (Fig. 4). While Montreal appears to be a low-price city when compared with American cities, Quebec City appears to be a high-price one. Within Canada's domestic electricity service, Montreal was doing very well (Fig. 4). Moreover, these figures underscore the more important fact that the industry in Quebec tended to target industrial clients (large power) (Dales 1957). The revenues per 1000 kWh distributed to large power customers can be computed from 1933 onward. As can be seen in Fig. 5, in spite of being heavily taxed, the Quebec electrical firms charged between $1.23 and $1.95 less per 1000 kWh than HEPC. (19)
Moreover, a significant level of attention must be given to the relative evolution of the supply of electricity. Throughout the period, Quebec's electrical capacity increased at a much faster pace than Ontario's. In fact, Quebec started off producing less electricity than Ontario, but by 1926, was producing more (Fig. 6). It should be pointed out that, contrary to the situation in Quebec, Ontario's electrical consumption was being subsidized and barely taxed at all. As was pointed out in Table 2, taxes represented a growing share of total expenditures for the Quebec electrical industry--from 16.7% in 1925 (the first year where taxes are not reported as miscellaneous) to 32.8% in 1939. The vast expansion of the supply of electricity in Quebec occurred under a much heavier tax burden for developers and was not being subsidized, as was the case in Ontario (Table 2).
The two pieces of evidence most often presented in favor of nationalization and against the private production of electricity have been the number of households connected to the network and the price difference between Quebec and Ontario. Dorion (2000) pointed out that Quebec lagged well behind Ontario in terms of the percentage of households connected, and grew more slowly. However, this claim is problematic. The largest problem is that Ontario, under the policy of power at cost, provided roughly $850,000 worth of power below cost (Dupre et al. 1996) directly at taxpayers' expense (Nelles 1974). In comparison, Quebec's electrical firms paid taxes and provided electricity at market prices. HEPC's policy of power at cost basically meant subsidizing consumption and a greater level of consumption in the first place. A significant part of that consumption had to be imported from Quebec since HEPC was unable to meet the demand (Evenden 2009). The strategy of HEPC in Ontario to sell energy at a cost did not force Quebec producers to sell also at cost. Thus, if HEPC faced shortages, it had to import the energy from Quebec. In 1926, HEPC established a contract with four private companies from Quebec to supply Ontario with more than 500,000 HP at the price of $15 per horsepower year (Biss 1936).
To convince Quebec producers to sell them the energy, they had to take market prices and offer to buy electricity at prices that were more attractive than what Quebec providers could have obtained otherwise and then resell this at a lower price in Ontario. As Piche (1937) pointed out, 17.9% of Quebec's production of energy was exported either to Ontario or the United States. If you adjust for the percentage of electricity lost during transmission, this number jumps to 19.2%.
This is important because it means that the greater demand in Ontario meant greater demand for Quebec-produced electricity. Consequently, this meant that Quebec consumers would pay higher prices than they otherwise would have had Ontario not opted to publicly produce electricity. It also means that the complaints made against private electric companies were in fact the result of HEPC's pricing policy that subsidized greater levels of energy consumption which then had to be met by importing from Quebec, thus increasing prices there.
Need for a New Narrative
The different portrait than generally accepted, which we have painted above, suggests the need for a new research agenda. It suggests three paths to follow for future research: a) the franchising process in order to explain price differences; b) inter-industry competition and technological change in the broadly defined energy sector and; c) the effects of satisfying low-volume customers versus high-volume customers.
Economic theory suggests that if a private company has a monopoly, productivity should stagnate, innovation should slow down, production should be diminished and prices should increase. In no respect can we observe such phenomena in the case of Quebec before nationalization. The industry could have, nonetheless, extracted monopoly profits in spite of these trends. Indeed, Emmons (1993) shows that while all cities in the United States enjoyed a price decline, the levels were different. Cities with public firms competing with private firms had the lowest prices while cities with private monopolies had the highest prices. However, the comparisons of prices in Quebec City and Montreal should give us an indication of what could have been happening. The residential rates for Montreal, in U.S. dollars, were actually lower than the lowest prices in the United States (those for publicly-owned firms in competition with other firms). The prices for Quebec City were higher than those Emmons found for American cities with private monopolies. Could the franchising process explain the difference? In Montreal, franchises were granted for twenty-five years while their length stood at ten years in Quebec City. The differences could (partly) be explained by incentives caused in the franchising process. Little attention has been attributed to the franchising process, which is unfortunate given the details of the case of Quebec City discussed above.
There is also the need to consider the changes in technological constraints that could have hindered the status of natural monopolies. To maintain a status of dominance in a market and generate vast profits, companies must constantly innovate, especially if the product is new, and thus lower prices (McKenzie and Lee 2008). Competition in this business probably does not come from the usual area of operation, but from new substitute products, or from the mere threat of entry of new competitors. It may be that a few firms dominated the electrical market in Quebec, but all indications are that they sought constantly to innovate, grow and reduce prices to face the stress of their resources and their environment (Liebowitz 1999).
In a certain way, the electrical industry had strong competitors in anthracite coal (Emery and Levitt 2002) and hardwood (Dominion Bureau of Statistics 1941). Most households satisfied their daily needs for fuel and light by burning either anthracite coal or hardwood. In fact, Quebec had the highest level of per capita consumption of firewood in Canada (MacFayden 2016). These were cheap and available sources of energy that electricity had to replace in household budgets. Gas also tended to be cheaper in Quebec, and its price tended to increase at a slower pace than in Ontario (an increase of 116.2% between 1913 and 1937 as opposed to an increase of 32.8%) (Dominion Bureau of Statistics 1938).
Annually, the average French family in Montreal spent $24.80 on coal and coke, $12.00 on wood and $19.90 on electricity, while their total expenditures for fuel and light stood at $77.00 (Dominion Bureau of Statistics 1941). These shares, especially for firewood, were quite high. A similar reality applied to the industrial sector, where Gelly (2003, 2010) notes that firms could use thermoelectricity and self-generation, and electricity providers deployed important efforts to sway them (by offering them substantially lower prices): in 1924, close to 30% of electricity used by industrial establishments came from self-generation as opposed to less than 10% in 1934 (Gelly 2010). Thus, while it is true that electric companies in urban markets generally had very large market shares for electricity, they did not command the majority of the energy market. In order to profit, these companies had to be able to outcompete alternative sources of energy.
As pointed out above, prices declined while output increased in the 1920s, a period of rapid economic growth. The innovation of providing a large supply of electricity profited the rise of industries and increased welfare on the whole. This suggests that a more plausible story is that the nascent electrical industry targeted industrial clients rather than residential clients. Indeed, data from the Central Electric Stations series produced by Dominion Bureau of Statisticsfor (1939) suggest that 86.6% of all electricity consumed in Quebec in 1939 (and total consumption was superior to total consumption in Ontario) was directed toward large power clients compared with 59.3% in Ontario. (20) Electric companies often wrote complex contracts with large power clients that offered important discounts on rates (Gelly 2010). Dales (1957) pointed out that revenues per kWh distributed to these clients were lower than those in Ontario, where public ownership prevailed (Fig. 5). This finding is widely consistent with the literature on how public and private ownership affect rates differently for different classes of customers (Peltzman 1971; Kitchens and Jaworski 2017). The effect of differential pricing has gone largely ignored in Quebec, with the exception of Gelly (2010), and a more proper consideration of the effects that this may have had on the province should be produced.
The most important conclusion we derive in this paper is that the claim of monopoly regarding Quebec's electrical industry during the interwar period is a weak one. One of the reasons for that is the justification for the claim is based on comparisons with Ontario where electricity had been nationalized. However, the pricing at cost of electricity in Ontario encouraged greater levels of consumption which HEPC could not satisfy alone. To satisfy this increased demand, electricity had to be imported from Quebec which put upward pressures on Quebec prices. In addition, the Ontario public firm was not operating the heavy tax burden that Quebec private firms were. This is a flawed justification.
While it is true that these firms had large market shares, this should not be confused with market power (Armentano 1996; McKenzie and Lee 2008). As pointed out above, prices were falling and output was increasing in Quebec. Price levels for residential services compared relatively well within North America and even within Canada, and rates for industrial clients were substantially lower than elsewhere.
This offers an important avenue for future research. The first would be to consider the importance of the fact that the electrical industry was locked in such a competition. If it was trying to replace coal, wood and gas for households and thermoelectricity and self-generation for industries, did this constitute a strong constraint on market power? Other factors might have conflated to create resentment toward the industry which are worthy of future research efforts. One would have been the aforementioned side effect of the Ontario public monopoly, another was the high burden of taxes, and yet another could be the effects of rent-seeking in the process of acquiring exclusive franchise for services.
Although this article in no way pretends to settle the debate, it did establish the foundations for a well-articulated research agenda that would provide adequate nuances regarding the pre-nationalization electricity market of Quebec. At the very least, it did provide motivation for such an agenda.
Acknowledgements Acknowledgements go to Alexandra Foucher, Mike Guetta, Robert Rogers and Art Carden, Jean-Thomas Bernard, Herbert Emery, Edward Stringham, Pierre Desrochers, Michael Giberson, Gerard Belanger and the participants of the research seminar at Texas Tech University.
Armentano, D. (1996). Antitrust and monopoly: Anatomy of a policy failure. Oakland: Independent Institute.
Bellavance, C. (1998). L'Etat, la 'houille blanche' et le grand capital: L'alienation des ressources hydrauliques du domaine public quebecois au debut du XXe siecle. Revue d'histoire de l'Amerique francaise, 51(4), 487-520.
Bellavance, C., Levasseur, R., & Rousseau, Y. (1999). De la lutte antimonopoliste a la promotion de la grande entreprise. L'essor de deux institutions: Hydro-Quebec et Desjardins, 1920-1965. Recherches sociographiques, 40(3), 551-578.
Biss, I. (1936). The contracts of the hydro-electric power commission of Ontario. Economic Journal, 40(183), 549-554.
Bliss, M. (1987). Northern Enterprise: Five centuries of Canadian business. Toronto: McClelland and Stewart.
Boutet, D. (1999). Le mouvement d'opposition au monopole de l'electricite a Quebec dans l'entre-deuxguerres. MA thesis: Universite du Quebec a Trois-Rivicres. Available online at http://depot-e.uqtr.ca/3358 /1 /000658919.pdf.
Bradley Jr., R. (1996). The origins of political electricity: Market failure or political opportunism? Energy Law Journal, 77(1), 59-102.
Brown. M., and Macdonald, R. (2015). Provincial convergence and divergence in Canada, 1926 to 2011. Ottawa: Statistics Canada. Available online at http://www.statcan.gc.ca/pub/11f0027m/11f0027 m2015096-eng.htm.
Bureau of Labor Statistics. (1939). Changes in retail prices of electricity, 1923-1938. United States Department of Labor. Available online at https://fraser.stlouisfed.org/title/4172.
Clark, G. (2017). What were the British earnings and prices then. Measuring Worth, Available online at http://www.measuringworth.com/ukeamcpi/.
Coursey, D., Isaac, M., & Smith, V. (1984). Natural monopoly and contested markets: some experimental results. Journal of Law & Economics, 27(1), 91-113.
Dales, J. H. (1953). Fuel, power and industrial development in central
Canada. American Economic Review, 43(2), 181-198.
Dales, J. H. (1957). Hydroelectricity and industrial development: Quebec, 1898-1940. Cambridge: Harvard University Press.
Demsetz, H. (1968). Why regulate utilities? Journal of Law & Economics, 11(1), 55-65.
Department of the Interior. (1923). Central electric stations in Canada, Part II: Directory, Nov. 1, 1922. Ottawa: Department of the Interior, Government of Canada.
Dominion Bureau of Statistics. (Editions from 1919 to 1940). Census of industry: Central electric stations in Canada. Ottawa: Department of Trade and Commerce.
Dominion Bureau of Statistics. (1927). Index numbers of rates for electricity for residence lighting and tables of monthly bills, 1913, 1924, 1925, 1926. Ottawa: Government of Canada.
Dominion Bureau of Statistics. (1938). Domestic rates for manufactured & natural fuel gas, 1913-1937. Ottawa: Government of Canada.
Dominion Bureau of Statistics. (1940). Index numbers of cost of electricity for domestic service, commercial light and small power. Ottawa: Government of Canada.
Dominion Bureau of Statistics. (1941). Family income and expenditure in Canada, 1937-1938: A study of urban wage-earner families, including data on physical attributes. Ottawa: Government of Canada. Available online at http://publications.gc.ca/collections/collection_2017/statcan/CS62-D-66-1941-eng.pdf.
Dorion, M. (2000). L'elcctrification du monde rural quebecois. Revue d'histoire de l'Amerique francaise 54(1), 3-37.
Dupre, R., & Patry, M. (1998). Hydroelectricity and the state in Quebec and Ontario: Two different historical paths. In G. Zaccour (Ed.), Deregulation of electric utilities (pp. 119-147). Boston: Kluwer.
Dupre, R., Patry, M., & Joly, P. (1996). The politics and regulation of hydroelectricity: The case of Quebec in the thirties. Montreal: CIRANO.
Emery, H., & Levitt, C. (2002). Cost of living, real wages and real incomes in thirteen Canadian cities, 19001950. Canadian Journal of Economics/Revue canadienne d'Economique, 35(1), 115-137.
Emmons, W. (1991). Private and public responses to market failure in the US electric power industry. 18821942. Journal of Economic History, 51(2), 452-454.
Emmons, W. (1993). Franklin D. Roosevelt, electric utilities, and the power of competition. Journal of Economic History, 53(4), 880-907.
Emmons, W. (1997). Implications of ownership, regulation, and market structure for performance: evidence from the US electric utility industry before and after the new deal. Review of Economics and Statistics, 79(2), 279-289.
Evenden, M. (2009). Mobilizing rivers: Hydro-electricity, the state, and world war II in Canada. Annals of the Association of American Geographers, 99(5), 845-855.
Faucher, A. (1992). La question de l'electricite au Quebec durant les annees trente. L'Actualite Economique, 68(3), 415-432.
Fleming, K. (1983). The uniform rate and rural electrification issues in Ontario politics, 1919-1923. Canadian Historical Review, 64(4), 494-518.
Fleming, K. (1992). Power at cost: Ontario hydro and rural electrification, 1911-1958. Montreal: McGillQuccn's University Press.
Fleuty, J.-L. (2004). Les porteurs de lumieres: l'histoire de la distribution de l'electricite au Quebec. Montreal: Editions Multimondcs.
Fouquet, R. (2008). Heat, power and light: Revolutions in energy services. London: Edward Elgar Publishing.
Gelly, A. (2003). A precipitous decline, steam as motive power in Montreal: a case study of the Lachine Canal industries. IA : The Journal of the Society for Industrial Archeology, 29(1), 65-85.
Gelly, A. (2010). Vapeur, thermoelectricite et hydroelectricite comme force mortice le long du corridor industriel du canal de Lachine, des annees 1850 a la Seconde Gueire Mondiale. PhD thesis, Department of History, Laval University. Available online at http://theses.ulaval.ca/archimede/fichiers/27029/27029.pdf.
Geloso, V. (2017). Rethinking Canadian economic growth and development since 1900. London: Palgrave MacMillan.
Hannah, L. (1979). Electricity before nationalisation: A study of the development of the electrical supply industry in Britain to 1948. London: McMillan.
Hillman, A. (2015). Rent-seeking as political economy. In A. Hillman & R. Congleton (Eds.), Companion to the political economy of rent-seeking (pp. 10-18). London: Edward Elgar Publishing.
Hogue C., Bolduc, A.. & Larouche, D. (1979). Quebec: un siecle d'electricite. Montreal: Libre Expression.
Kitchens, C., & Jaworski, T. (2017). Ownership and the price of residential electricity: evidence from the United States, 1935-1940. Explorations in Economic History, 64, 53-61.
Krucger, A. (1974). The political economy of the rent-seeking society. American Economic Review, 64(3), 291-303.
Liebowitz, S. (1999). Winners, losers and Microsoft. Oakland: Independent Institute.
MacFayden, J. (2016). Hewers of wood: A history of wood energy in Canada. In R. W. Sandwell (Ed.), Powering up Canada: The history of power, fuel, and energy from 1600 (pp. 129-161). Montreal and Kingston: McGill-Queen's University Press.
Marchand, S. (1988). L'impact des innovations technologiques sur la vie quotidienne des Quebecoises du debut du XXe siecle (1910-1940). Material Culture Review/Revue de la culture materielle, 28(Fall). 1-14.
McKenzie. R., & Lee, D. (2008). In defense of monopoly: How market power fosters creative production. Ann Harbor: University of Michigan Press.
Ministere de l'Industrie et du Commerce. (From 1918-1944). Quebec Statistical Year Book. Quebec: Ministere de l'Industrie et du Commerce.
Montreal Light, Heat and Power (MHLP). (1934). Diminution et revision des tarifs d'electricite: annonce de la quinzieme diminution. L'Ordre, September, 7, 3.
Murray, W. S., & Flood, H. (1922). Government owned and controlled compared with privately owiied and regulated electric utilities in Canada and the United States. New York: National Electric Light Association.
Nelles, H. V. (1974 ). The Politics of Development: Forests, Mines, and Hydro-Electric Power in Ontario, 1849-1941. Kingston and Montreal: McGill-Queen's University Press.
Neufeld, J. (1987). Price discrimination and the adoption of the electricity demand charge. Journal of Economic History, 470), 693-709.
Officer, L. (2017). Exchange rates between the United States dollar and forty-one currencies. Measuring Worth, Available online at http://www.measuringworth.com/exchangeglobal/.
Officer, L., and Williamson, S.H. (2017). The annual consumer price index for the United States, 1774-2015. Measuring Worth, Available online at http://www.mcasuringworth.com/uscpi/.
Peltzman, S. (1971). Pricing in public and private enterprises: Electric utilities in the United States. Journal of Law & Economics, 14(1), 109-147.
Peltzman, S. (1976). Toward a more general theory of regulation. Journal of Law & Economics, 19(2), 211-240.
Piche, P.-E. (1937). Regimes et tarifs d'electricite. Actualite Economique, 13(2), 17-30.
Rassenti, S., Smith, V., & Wilson, V. (2001). Using experiments to inform the privatization/deregulation movement in electricity. Cato Journal, 21(3), 515-544.
Smith, V. (1988). Electric power deregulation: background and prospects. Contemporary Economic Policy, 6(3), 14-24.
Statistics Canada. (2015). Table 384-5000: Data on long-run provincial and territorial economic performance, annual. Ottawa: Statistics Canada. Available online at http://www5.statcan.gc.ca/cansim/a26 ?lang=eng&id=3845000.
Stigler, G., & Friedland, C. (1962). What can regulators regulate? The case of electricity. Journal of Law & Economics, 5, 1-16.
Troesken, W. (1996). Why regulate utilities? The new institutional economics and the Chicago gas industry, 1849-1924. Ann Harbor: University of Michigan Press.
Troesken. W. (2006). Regime change and corruption: A history of public utility regulation. In E. Glaeser & C. Goldin (Eds.), Corruption and reform: Lessons from America's economic history. Cambridge: National Bureau of Economic Research and University of Chicago Press.
Tullock, G. (1967). The welfare costs of tariffs, monopolies, and theft. Economic Inquiry, 5(3), 224-232.
Wright, G. (2006). Electrical energy--retail prices, residential use, and service coverage: 1902-2000. Table Db 234-241. In S. B. Carter, S. S. Gartner, M. R. Haines, A. L. Olmstead, R. Sutch, & G. Wright (Eds.), Historical Statistics of the United States, Earliest Times to the Present: Millennial Edition. New York: Cambridge University Press.
(1) Troesken (2006) explains this franchise in the case of electricity for the United States.
(2) Emmons (1993) indicated that private monopolies, private firms in competition, public monopolies and public firms in competition all had different price levels in spite of sharing a common downward trend.
(3) It is also worth pointing out that electrical appliances, which were complements to electricity services (Fouquet 2008), were expensive, and the relatively poorer Quebeckers tended to acquire the less expensive wood--and coal-buming home appliances (Marchand 1988), which created pressures on electricity companies to reducc priccs in order to increase adoption rates (Gelly 2010).
(4) Firewood represented 15.6% and 44.6% of fuel and light expenditures for French households in Montreal and Quebec City ("British" households in Montreal reported expenditures equal to 5.6% of all expenditures on fuel and light). Overall, "British" families in Canada expended 10.5% of their fuel and light budget on firewood as opposed to 23.9% for "French" families (Dominion Bureau of Statistics 1941).
(5) This was largely a result of the fact that Quebec had greater population density in cities. The largest two cities of the province of Quebec (Montreal and Quebec City) accounted for 31.0% and 33.0% of the province's population in 1921 and 1931 as opposed to 21.7% and 22.9% for Ontario's two largest cities, Toronto and Hamilton (Dominion Bureau of Statistics, 1932).
(6) MLHP represented a minor share of total electricity output (4%) in Quebec, and 30% of the electricity it distributed was bought from SWP. Its power sites were in the Upper Saint-Lawrence region. SWP was located in central Quebec (the Maurice region) and was the largest single producer with numerous distribution subsidiaries, which notably served Quebec City and Trois-Rivieres. GP was an imported producer, but it sold most of its output to local distributors, the pulp and paper industry that was prominent in the region and the neighboring province of Ontario. As for SP, originally known as the Duke-Price Power Co., its output was laigely oriented toward the aluminum and pulp and paper industries. However, it did consolidate with other regional companies, which allowed it to become the main distributor for many localities of the Saguenay and Lac Saint-Jean region. SCP was the smallest of them and operated largely in the eastern townships in southcentral Quebec (Dales 1957; Dupre et al. 1996; Hogue et al. 1979).
(7) The Directory reported the population of the different municipalities (with the exception of one, Shipton in the eastern townships, which was linked to the municipality of Danville for distribution), which allowed us to arrive at this figure. The Directory reports one cooperative, operating in Rimouski, which served 0.1% of Quebec's population.
(8) Dupre et al. (1996) have analyzed the regulations of the Quebec Electricity Commission (1935-1937) and the Provincial Electricity Board (after 1937) and found that, as Stigler and Friedland (1962) and Pcltzman (1976) did, the regulatory boards were laigely ineffective.
(9) The exception was self-generators for large industries.
(10) It is worth pointing out that in an earlier version of his work, Emmons (1991) found a much larger price differential (28%), half of which could be explained by capital cost advantage. The refinements in Emmons' work from his 1991 version to the 1997 version led to a reduction in the price advantage estimate to this much lower figure.
(11) While it is beyond the scope of this paper to determine this, there is the possibility that lower rates for larger power consumers increased their productivity, which allowed them to offer higher wages. Thus, workers would have benefitted though this channel instead of through lower prices.
(12) Historian Keith Fleming, who studied HEPC extensively, summarized the fixation of rates as follows:
Rales charged by the municipalities had to be sufficient to cover the cost per horsepower incurred by the HEPC (...). Also to be considered in the final price were the municipality's proportionate share of transmission costs, operation and maintenance costs, interest charges on any money borrowed for installation, depreciation reserves, a reserve fund for obsolescence and contingencies and sinking fund reserves for the amortization of local commission's debts. Ultimately, a higher power cost resulting from longer transmission distances, smaller population density and lighter power load had to be borne solely by the contracting municipality (Fleming 1983, p. 486).
(13) The Murray and Flood report came from an association of electricity manufacturers. However, the statistics from the Dominion Bureau of Statistics' Central Electric Stations series point in the same direction. In 1923, the generation of 1000 kWh cost $4.74 as opposed to $7.15 in Ontario. In fact, this is probably a conservative estimate. Until 1932, the Dominion Bureau of Statistics did not publish data regarding interprovincial trade of electricity, so we have to divide expenses over electricity generated rather than electricity generated and distributed. This is a problem because Ontario tended to buy substantial quantities of electricity from Quebec, which it then distributed at a cost (Quebec imported a few hundred thousand kWh from the United States). After 1933, this is not a problem, as we can divide over electricity distributed and generated (pages have been lost in the 1932 edition, meaning we have to jump to 1933). In 1933, the cost per 1000 kWh serviced in the province of Quebec stood at $1.43 as opposed to $6.13 in Ontario.
(14) Moreover, the quality of transmission inferior in Ontario than in Quebec, which probably worsened the shortfall. Indeed, Dominion Bureau of Statistics (1940) data suggest that losses of energy in Quebec in 1939 represented 6.95% of total consumption compared with 16.63% in Ontario.
(15) The rest of Canada is defined here as Canada minus Ontario and Quebec.
(16) Especially if the price schedule were fixed in the franchise contract.
(17) The franchise acquired by die new firm, the Public Service Corporation of Quebec, only extended to electricity. QRLHP had managed to retain only its franchise for gas and tramways.
(18) City dwellers were well-connected to electricity by 1931. In Montreal, it was estimated that between 89% and 99% of households were connected (Gelly 2010).
(19) Quebec had some of the lowest rates in Canada. For example, the Dominion Bureau of Statistics for 1939 places Quebec in second position (slightly behind Manitoba) for the lowest rates (as proxied by revenues per 1000 kWh) for large power customers. This result is more or less in line with the points made by Peltzman (1971) and Kitchens and Jaworski (2017), who argue that public and private ownership influence rate setting across different consumer classes since private firms tend to be better able to price discriminate.
(20) Large power only accounted for 63.8% of revenues in Quebec (Dominion Bureau of Statistics 1939), so that, at $30,177 of revenue per large power customer, Quebec was more three times above the Canadian average.
Published online: 21 February 2018
Vincent Geloso  * Germain Belzile 
[mail Vincent Geloso
 Texas Tech University, Lubbock, USA
 HEC Montreal, Montreal, Canada
Caption: Fig. 1 Generation and transmission of electricity in Quebec, 1922. Source: Department ofthe Interior (1923). Note: Readers interested in the evolution of this map over time should consult Dales (1957), who produced a similar map for 1940
Caption: Fig. 2 Residential electricity rates in Quebec, adjusted for inflation (1913 $), 1910-1939. Sources: Montreal prices: Montreal Light, Heat and Power (1934); Quebec City prices: Boutet (1999); provincial average: Dominion Bureau of Statistics, Central Electric Stations (1939); deflator: city-specific price indexes taken from Emery and Levitt (2002). The provincial average was deflated by the Montreal price index
Caption: Fig. 3 Residential electricity rates and revenues per customer kilowatt hour (kWh) in Britain, Quebec and the United States (1929 = 1). Sources: Montreal prices: Montreal Light Heat and Power (1934); Quebec average: Dominion Bureau of Statistics, Central Electric Stations (1939); American prices: Wright (2006); British revenues per customer-kWh: Hannah (1979); deflators: for Montreal and provincial prices, the Montreal deflator constructed by Emery and Levitt (2002) was used. For British and American prices, the deflators at http://www.measuringworth.com/uscpi/ and http://www.measuringworth.com/ukearncpi (Officer and Williamson 2017; Clark 2017) were used
Caption: Fig. 4 Residential electricity rates in Quebec, Canadian and American cities (1923-1938) with inter-provincial comparisons within Canada (Quebec areas highlighted). Sources: American prices: Bureau of Labor Statistics (1939); Canadian provincial averages: Dominion Bureau of Statistics, Central Electric Stations (1939). Note: For comparisons with American cities, prices are expressed in U.S. dollars. The exchange rates were taken from http://www.measuringworth.com/exchangeglobal (Officer 2017)
Caption: Fig. 5 Revenues per 1000 kWh from power provided to large power customers in Quebec and Ontario, adjusted for inflation, 1933-1939. Sources: For revenues per 1000 kWh: Dominion Bureau of Statistics, Central Electric Station (numerous editions); deflators: Emery and Levitt (2002)
Caption: Fig. 6 Capacity of power plants installed in Quebec and Ontario (1000s HP). Source: Ministere du Commerce et de l'Industrie (numerous editions)
Table 1 Per capita income and customers per inhabitant (selected provinces) as a share of Canada with shares of Canadian population and customers, 1926 Per capita Customers income per inhabitant Quebec as % 79.9% 98.50% Quebec of Canada PEI as % of Canada 47.1% 33.31% PEI Nova Scotia as % 60.9% 59.78% Nova Scotia of Canada New Brunswick 61.1% 57.78% New Brunswick as % of Canada Ontario as % 119.6% 125.61% Ontario of Canada Population Share of Canadian share of Canada customers Quebec as % 27.5% 27.1% of Canada PEI as % of Canada 0.9% 0.3% Nova Scotia as % 5.4% 3.3% of Canada New Brunswick 4.2% 2.4% as % of Canada Ontario as % 33.5% 42.1% of Canada Source: Brown and MacDonald (2015) and Dominion Bureau of Statistics, Central Electric Stations (1926). Note: The income scries from Brown and MacDonald is also available from the Canadian Socioeconomic Information Management System (CANSIM) table 384-5000 (Statistics Canada 2015) Table 2 Taxes as a share of total expenses, selected years, for Quebec, Ontario and the rest of Canada Quebec Ontario Rest of Canada 1925 16.5% 4.1% 4.7% 1929 16.5% 3.9% 7.3% 1931 20.0% 3.4% 8.3% 1936 24.9% 3.6% 12.5% 1939 32.8% 3.4% 15.2% Source: Dominion Bureau of Statistics, Central Electric Stations (numerous editions). Note: Rest of Canada is Canada without Quebec and Ontario. 1925 is the first year with data
|Printer friendly Cite/link Email Feedback|
|Author:||Geloso, Vincent; Belzile, Germain|
|Publication:||Atlantic Economic Journal|
|Date:||Mar 1, 2018|
|Previous Article:||Does Competition Lead Firms to Bribery? A Firm-Level Study in Southeast Asia.|
|Next Article:||Investing in Health: A Macroeconomic Exploration of Short-Run and Long-Run Trade-Offs.|